ALBERT R. MANN 
 LIBRARY' -' 
 
 New York State Colleges 
 
 OF 
 
 Agriculture and Home Economics 
 
 AT 
 
 Cornell University 
 
Cornell University Library 
 HV 6046.F4 
 
 A study of women delinquents in New Yorit 
 
 3 1924 013 786 961 
 
The original of tiiis book is in 
 tine Cornell University Library. 
 
 There are no known copyright restrictions in 
 the United States on the use of the text. 
 
 http://www.archive.org/details/cu31924013786961 
 
A STUDY OF 
 
 WOMEN DELINQUENTS 
 
 IN NEW YORK STATE 
 
IPubltcatlons of the Bureau of Social Ibggiene 
 
 A Study of 
 
 Women Delinquents 
 
 in New York State 
 
 by' 
 
 MABEL RUTH FERNALD 
 
 Assistant Professor of Psychology, University of Minnesotaj 
 Formerly Director, Laboratory of Social Hygiene 
 
 MARY HOLMES STEVENS HAYES 
 
 Member of The Scott Company^ Formerly Psychologist, 
 Laboratory of Social Hygiene 
 
 ALMENA DAWLEY 
 
 Supervisor, Department of Social Investigation, Pennsylvania School for 
 Social Service; Formerly Sociologist, Laboratory of Social Hygiene 
 
 With Statistical Chapter by 
 
 BEARDSLEY RUML 
 
 Assistant to the President, Carnegie Corporation; 
 Secretary, The Scott Company 
 
 Preface by 
 
 KATHARINE BEMENT DAVIS 
 
 General Secretary, Bureau of Social Hygiene 
 
 NEW YORK 
 
 THE CENTURY CO. 
 
 1920 
 
Copyright, 1920, by 
 Ths Century Co. 
 
AUTHORS' NOTE 
 
 To a peculiar degree the investigation of which this book is the 
 report was a composite piece of work, the difficulties and problems of 
 which were shared by all the individuals of the group concerned in 
 its accomplishment. It would be impossible to acknowledge with any 
 adequacy the special contribution which each person made. We may, 
 however, indicate the general lines along which the work was divided. 
 The following members of the staff were responsible for the psycho- 
 logical examinations : Mary H. S. Hayes, Mabel R. Fernald, Jessie J. 
 Taft, Buford J. Johnson, Margaret V. Cobb. The task of social in- 
 vestigation was carried through by the following: Almena Dawley 
 (in charge of investigations for Bedford and Auburn), Virginia P. 
 Robinson (in charge of investigations for the city institutions), Veda 
 Elvin, Marie Lawrence, Mabel C. Huschka, Grace Massonneau, Maude 
 W. Moore. The general plan of statistical treatment was mapped 
 out by Beardsley Ruml, who acted as consultant on statistical problems 
 arising throughout the study. The detailed work of statistical analysis 
 and computation was under the supervision of Mary A. Clark. The 
 following members of the staff assisted in this phase of the work: 
 Agnes Crowley, Christine Brigham, Helen Towey, Marjorie Taft, 
 Louise Russell. 
 
 For the statements as actually formulated in this book, the three 
 authors assume entire responsibility. While there has been necessarily 
 much overlapping of work in the compilation of the book. Miss Dawley 
 was mainly responsible for the sociological portions, Mrs. Hayes and 
 Miss Fernald for the psychological. 
 
 We have noted in Chapter II our indebtedness to the authorities 
 in the various institutions who made possible this investigation through 
 their interest and cooperation. 
 
PREFACE 
 
 So long as prisons were used merely for the purpose of punish- 
 ment or for holding in safety those who were dangerous to society, 
 the behavior of prisoners was a matter of relatively small importance. 
 They could be kept in subjection by force if necessary and tractability 
 on their part was desirable chiefly because it made life easier for 
 those in charge. 
 
 With the coming of the reformatory and the adoption of the prin- 
 ciple of indeterminate sentence and release on parole behavior within 
 the institution assumed supreme importance. This could not be other- 
 wise for there were no other criteria in the hands of prison authorities. 
 If a prisoner were obedient, observed all rules cheerfully, was respect- 
 ful to his superiors, in the institutional school showed himself willing 
 to learn, in the shop or at other tasks worked faithfully and more or 
 less well, if he expressed to his spiritual adviser contrition for his 
 wrong doing and his resolve to amend his ways, he was a "good 
 prisoner." Another prisoner "kicked against the pricks" from the 
 moment of entrance; he was moody or sullen, or quick tempered, or 
 stupid, or resentful, disrespectful to his keepers, quarrelsome among 
 his mates. He was unwilling to take advantage of any opportunities 
 for self -improvement, possibly he exhibited not only no inclination to 
 learn but showed absolute incapacity for so doing. He was a 
 malingerer or he did his work so poorly that no one wanted him 
 in the gang. He was a "bad prisoner." The former received privileges 
 within the institution and an early parole. The latter was held and 
 daily became a worse prisoner until the time came when under the 
 law he had to be released. Sometimes, not infrequently, the "good 
 prisoner" failed to make good outside. Sometimes to the surprise 
 of the authorities the "bad prisoner" made good. Such disappointments 
 and surprises could have but one effect. Slowly but surely there 
 dawned the -recognition of the need of an intensive study of the 
 individual, not only of his behavior in the institution but why his 
 behavior was what it was. What were the causes of which his behavior 
 was the symptom ? 
 
 When the first reformatories for adults opened their doors we 
 
vm 
 
 PREFACE 
 
 had no such thing as applied psychology. Psychology did not concern 
 itself with practical problems of behavior but restricted itself to 
 theoretic discussions on mental processes and the like. To be sure 
 we had alienists, but there was little recognition of those peculiar men- 
 tal conditions which do not amount to insanity, but which if not treated 
 make their victims unfit for society outside or inside the prison. There 
 was no such thing as a science of sociology, nor any hint of the need 
 of careful study of the pre-institutional life of the prisoner with a 
 view to understanding him. 
 
 In a number of states we had schools and asylums for the feeble- 
 minded but much more was expected from education of these per- 
 sons than has been shown by experience to be possible. That the 
 feeble-minded were not suitable inmates of a reformatory was recog- 
 nized as early as the First Annual Report of the Reformatory Prison 
 for Women at Sherborji, Mass.,^ when the managers recommended 
 "changes in existing laws" authorizing the transfer of certain unsuit- 
 able persons committed for minor offenses, who among other things 
 were "weak minded." 
 
 New York State in establishing its reformatory institutions for 
 women sought to profit by the experience of the two other institutions 
 for women by providing in the State Charities Law that Boards of 
 Managers might return to committing judges any person who "is insane 
 or mentally incapable of being materially benefited by the discipline 
 of the institution." The difficulties of securing her admission to 
 crowded custodial institutions and the undesirability of turning her 
 back into a community in which she had already proved herself a 
 social menace, together with differences of opinion as to whether 
 in a given case a woman was so far incapacitated as to be "mentally 
 incapable of being materially benefited" have worked together to 
 prevent in our own state the relief which was intended for the reforma- 
 tories. In other types of institutions and in many other states there 
 is not even this possibility of transfer. 
 
 The Laboratory of Social Hygiene maintained for six years by 
 the Bureau of Social Hygiene in cooperation with the State Reforma- 
 tory for Women at Bedford Hills, grew out of the recognition of a 
 
 'First Annual Report of the Commissioners and Advisory Board of the 
 Reformatory Prison for Women at Sherborn, Mass., Oct., 1878. The Woman's 
 Prison at Sherborn, Mass., was the second prison exclusively for the care of 
 women in the United States. The Indiana Reformatory for Women and Girls 
 opened in 1873. 
 
PREFACE ix 
 
 practical need in the actual handling of delinquent women if a max- 
 imum number were to be returned to society prepared to lead a self- 
 supporting, law-abiding life^As a result of experience it was believed 
 that before one could apply methods of treatment with any certainty 
 it was necessary to have an accurate diagnosis of the indiv idual cas e, 
 taking into consideration social, physical an d mental factors. Such 
 case studies of the social and mental aspects begun with the women 
 committed to the State Reformatory at Bedford Hills and extended 
 to other groups of delinquent women in New York State forms the 
 basis of the work presented in this volume. These ca se studie s have 
 a threefold value. First as pointed out above, their~imrnediate prac- 
 tical importance is in indicating the actual tre atmen t_gf the case within 
 the institution. If it were possible to put in use a careful clearing 
 house method for treating the wards of the state in accordance with 
 their own needs and the best interests of society such case histories 
 would be the proper basis of scientific classification. Pending this 
 their use is limited to the soecialinstitution in which the case is studied 
 and to the existing possioilities of transfer. The value of the pub- 
 lication of case histories is discussed in the first chapter of this book. 
 The use of the material determined upon for this study does not pre- 
 clude a later publication of these histories. 
 
 Second, the value of standards of comparison is clearly pointed 
 out in chapter one. Although without more exten^ed~'studies of the 
 general population we cannot make accurate comparisons with the 
 delinquent groups, it is of great importance to have a standard by 
 which we can measure the inmates within a given institution or 
 compare the inmate population of one institution with another. Even 
 though this study fails to show a distinct criminal type it seems to 
 clearly indicate that the average woman prisoner of New York State 
 falls somewhat below the average individual in society in mentality 
 and economic efficiency. 
 
 Unfortunately no comparable data of the physical states of the 
 groups studied were available, nor are there comparable studies of the 
 health and physical condition of the general population. A detailed 
 study of 200 non-selected individuals at the State Reformatory for 
 Women at Bedford Hills^ made prior to the studies presented in this 
 
 ^Physical States of Criminal Women. A study made at the Laboratory 
 of Social Hygiene, Bedford Hills, by Alberta S. Guibord, M.D., Journal of 
 Criminal Law and Criminology, May i, 1917- 
 
X PREFACE 
 
 volume showed a very high degree of physical deficiencies "primarily 
 to a large extent preventable in that they are the result of faulty 
 nutrition, bad hygiene, bacterial infection and other concomitants 
 of unintelligence and poverty." 
 
 Even if it were conceivably possible to select and study an absolutely 
 representative group of non-delinquent women and the results showed 
 the same, or practically the same, averages and the same dispersion 
 of the particular factors, the standard would still be necessary for pur- 
 poses of classification and discussiouc 
 
 A third use to which it was hoped our studies could be put was 
 the pointing out with greater definiteness the causes of delinquency. 
 If any deductions can be made, our studies would seem to indicate 
 the great complexity of causal factors in individual cases as well as 
 in the various groups. 
 
 It is not possible, as yet at least, to single out a few factors to 
 which we can point as predominantly causative, whether these lie in the 
 social and economic conditions which govern environment, public 
 health, education and recreation, or in the constitutional factors which 
 are the most obvious in individual cases; but in a study carefully 
 made we believe it is not without significance that no one or two out- 
 standing environmental or constitutional causes were discoverable. 
 
 Katharine Bement Davis, 
 General Secretary, Bureau of Social Hygiene. 
 
CONTENTS 
 CHAPTER I 
 
 PAGE 
 
 Introduction: General Orientation with Reference to 
 
 Problems and Methods of Treatment , 
 
 Necessity of More Adequate Definition of Conceptions Regard- 
 ing Women Delinquents 3 
 
 Inadequacy of Material from Other Sources and Limitations on 
 Completion of Desired Comparisons with Other Groups . . 5 
 
 a. Comparison with Non-delinquent Women 5 
 
 b. Comparison ot Delinquent Women with Delinquent Men 7 
 
 c . Intercomparisons Within our Group of Delinquent Women 9 
 Use of Mass Methods of Treatment Rather Than Case History 
 
 Methods in Present Study 9 
 
 CHAPTER II 
 
 General Account of Delinquent Groups Studied ... 13 
 
 General Make-up of Group 13 
 
 Descriptive Account of Groups 15 
 
 a. State Reformatory for Women, Bedford Hills, N. Y. . 15 
 
 h , St ate Prison for Women. Auburn. N. Y 17 
 
 a ThS^ew York Magdalen Home (Inwood House) . . 20 
 
 d. The New York County Penitentiary 20 
 
 e. The New York City Workhouse 21 
 
 f. Probation Cases from Women's Night Court of Manhat- 
 
 tan and the Bronx 25 
 
 Groups of Women of this State Who Are Not Included in 
 
 Present Study 25 
 
 Reason for Presenting Data Classified by Institutional Groups 29 
 
 CHAPTER III 
 
 Statistical Methods ... 31 
 
 The Description of the Qualities of a Group 32 
 
 Comparisons and Relationships 33 
 
 Statistical Reservations 34 
 
 Technical Notes 35 
 
 xi 
 
xii CONTENTS 
 
 CHAPTER IV 
 
 Methods of Psychological Examination 
 
 Method of Estimation Versus Method of Examination . . _ . 
 Factors Affecting Choice, of Tests for Use in Examination 
 
 Method 
 
 Establishment of Methods on Bedford Group 
 
 a. Conditions of Testing • 
 
 b. Standardization of Methods of Giving and of Scoring 
 
 Tests 
 
 c. Preliminary Intensive Test Series 
 
 Selection of Tests for Present Investigation 
 
 a. Yerkes-Bridges Point Scale and Stanford-Binet . . . 
 
 b. Tests of the WooUey Series 
 
 c. Additional Tests 
 
 Use of Pearson Scale for Judging Mental Ability as Criterion of 
 
 Reliability of Tests Used 
 
 Selection of Tests for Special Consideration in Present Study 
 
 CHAPTER V 
 
 Methods op Social Investigation 
 
 General Plan of Study and Comparison with Other Types of 
 
 Studies of Delinquents 
 
 The Initial Interview 
 
 Method of Investigation 
 
 a. General Procedure of Investigation 
 
 b. Form of Report of Investigation 
 
 1. Information Concerning the Delinquent .... 
 
 2. Family History . . 
 
 3. Correspondence 
 
 Summary of Data. Verified History Blank. Method of Con- 
 densing Material, Indicating Sources of Information and 
 Reducing Work to Code Form for Use on Statistical Cards . 
 
 CHAPTER VI 
 
 Nature and Extent of Delinquency 
 
 Form of Classification of Offenses 
 
 Nature of Present Offense . . . 
 
 a. New York City Police Department Classification . . 
 
 b. Felonies and Misdemeanors 
 
 c. Felonies and Misdemeanors in New York City Police 
 
 Department Classification 
 
 d. Comparison of Nature of Offense of Female Felons and 
 
 Male Felons in Sing Sing and the New York State 
 Reformatory at Elmira 
 
CONTENTS xiii 
 
 PAGE 
 
 Extent of Delinquency 102 
 
 a. Number of Previous Convictions 105 
 
 1. Relationship Between Age and Number of Previous 
 
 Convictions 105 
 
 2. Institutional Differences in Number of Previous Con- 
 
 victions 107 
 
 3. Number of Previous Convictions Among Felons and 
 
 Misdemeanants 116 
 
 4. Number of Previous Convictions Among Offenders 
 
 Against Chastity and Offenders Against Property . 119 
 
 b. Number of Previous Terms Served 121 
 
 c. Length of Time Served in Penal Institutions . . . . 122 
 
 d. Number of Times on Probation 124 
 
 e. Recidivists and First Offenders 126 
 
 1. Recidivists and First Offenders Among Institutional 
 
 Groups . 126 
 
 2. Recidivists and First Offenders Among Felons and 
 
 Misdemeanants 126 
 
 3. Recidivists and First Offenders Among Female Felons 
 
 and Male Felons 127 
 
 First Contacts with the Law 128 
 
 a. Age at First Conviction 128 
 
 1. Age at First Conviction Among Institutional Groups 128 
 
 2. Comparison with English Male Convicts . . . . 129 
 
 3. Age at First Conviction Among Felons and Misde- 
 
 meanants 131 
 
 4. Age at First Conviction Among Offenders Against 
 
 Chastity and Offenders Against Property Rights . 131 
 
 5. Relation Between Age at First Conviction and Nimiber 
 
 of Convictions 133 
 
 b. Nature of First Offense 135 
 
 c. First Sentence 139 
 
 d. Juvenile Delinquents 140 
 
 CHAPTER VII 
 
 Miscellaneous Considerations 147 
 
 Physical Data: 
 
 a. Age 147 
 
 b. Habits: Alcoholism, Drug Addiction and Use of To- 
 
 bacco 153 
 
 c. Record of Commitments to Hospitals for the Insane . 159 
 Certain General Social Factors 161 
 
 a. Civil Condition 161 
 
 b. Age at Marriage and Age of Husband at Marriage . . .162 
 
 c. Religion 163 
 
3dv CONTENTS 
 
 CHAPTER VIII 
 
 PAGE 
 
 Nativity and Color in Relation to Delinquency ... 167 
 
 Birthplace and Color of Delinquents ...... 168 
 
 a. Comparison of Nativity and Color Between Delinquent 
 Women and General Female Population in New York 
 
 City 171 
 
 Age at Present Conviction, Classified by Nativity and Color . 174 
 
 Nature of Present Offense . . . 177 
 
 a. Classified by New York City Police Department Classi- 
 fication 177 
 
 b Classified by Felonies and Misdemeanors 180 
 
 Extent of Delinquency 181 
 
 a. Number of Previous Convictions. Comparison of Dif- 
 
 ferences Among Nativity Groups 182 
 
 b. Recidivists and First Offenders by Nativity Groups . . 188 
 
 c. Relationship Between Length of Time in United States 
 
 and Number of Previous Convictions 191 
 
 1. Number of Previous Convictions and Years in United 
 
 States 191 
 
 2. Number of Previous Convictions and Age at Coming 
 
 to United States 191 
 
 First Contacts with the Law 196 
 
 a. Age at First Conviction 196 
 
 b. Nature of First Offense 198 
 
 c. First Sentence 201 
 
 CHAPTER IX 
 
 Significant Factors in Early Home Conditions and Family 
 
 Status 205 
 
 Method of Estimating Home Conditions During Childhood and 
 Adolescence and Distribution of Estimates by Institutional 
 Groups 205 
 
 a. Estimate of Economic Status in the Home .... 207 
 
 b. Estimate of Moral Standards in the Home . . . .210 
 
 c. Estimate of Parental Supervision in the Home . . . 212 
 
 d. Total Estimate of Home Conditions 214- 
 
 e. Comparison Between Verified and Unverified Data on 
 
 Home Conditions 216 
 
 f. Presentation of Cases Representing Various Classes of 
 
 Home Conditions 219 
 
 g. Relationship Between Estimate of Home Conditions and 
 
 Age at First Conviction 228! 
 
 h. Relationship Between Estimate of Home Conditions and ' 
 
 Number of Convictions 228 
 
CONTENTS XV 
 
 PAGE 
 
 Family Status 228 
 
 a. Nativity of Parents 228 
 
 b. Age of Parents at Time of Subject's Birth .... 23 1 
 
 c. Age of Subject at Time of Parents' Death .... 232 
 
 d. Number of Children in Fraternity 235 
 
 e. Order in Fraternity 237 
 
 f. Number of Full Term Children Bom to Subject . . .237 
 
 g. Number of Pregnancies of Subjects 238 
 
 Hereditary Factors. Defective Physical and Mental Strains, 
 
 Alcoholism and Criminality. Distribution by Institutional 
 
 Groups 240 
 
 CHAPTER X 
 
 Educational Background 245 
 
 Views Regarding Relation of Education to Crime .... 245 
 Lack of Data on Education for General Population .... 249 
 Distinction Between Educational Opportunities and Educa- 
 tional Attainments 250 
 
 Verification of School Record 252 
 
 Educational Opportunities 254 
 
 a. Kind of School Attended 254 
 
 b. Regularity of Attendance 255 
 
 c. Age at Leaving School 256 
 
 1. Data on Institutional Groups and on Total . . . 257 
 
 2. Comparative Data on Non-delinquent Groups . . . 259 
 
 3. Relationship Between Age at Leaving School and Age 
 
 at First Conviction 261 
 
 d. Number of Years in School 264 
 
 1. Data on Total Delinquent Group and on Institutional 
 
 Sub-groups 264 
 
 2. Data on Groups Classified by Nativity and Color . . . 265 
 
 3. ImpossibiUty of Comparison with General Population 268 
 
 4. Relationship Between Years in School and Number of 
 
 Convictions 270 
 
 Educational Attainments 270 
 
 a. Grade Finished 270 
 
 Data on Institutional Groups and on Total . . . 271 
 
 Data on Groups Classified by Nativity and Color . . 275 
 
 3. Relationship of Grade to Age Distinctions . . . . 275 
 
 4. Relationship of Grade to Age at Leaving School . .278 
 
 5. Relationship of Grade to Number of Years in School . 280 
 
 6. Relationship of Grade to Nature of Present Offense 284 
 Relationship of Grade to Number of Previous Convic- 
 tions 
 
 286 
 
xvi CONTENTS 
 
 PAGE 
 
 8. Relationship of Grade to Age at First Conviction . 288 
 
 b. Amount of Illiteracy 288 
 
 c. Results of Educational Tests 292 
 
 1. Reading — Trabue Scale 294 
 
 2. Spelling — Buckingham 294 
 
 3. Handwriting — ^Ayres Scale 294 
 
 4- Arithmetic — Coturtis 298 
 
 \ CHAPTER XI 
 
 Occupational History and Economic Efficiency .... 304 
 
 Age at Starting Work 306 
 
 Kind of Work Done 311 
 
 a. Classification of Kind of Work 311 
 
 b. Band of Work Done at Specified Times 313 
 
 Kind of Work in First Job 313 
 
 Kind of Work in Latest Job 315 
 
 3. Kind of Work at First Conviction 316 
 
 4. Kind of Work at Present Conviction 317 
 
 5. Prevailing Work 318 
 
 c. Comparison of Prevailing Work of Delinquents and Gen- 
 eral Female Population 320 
 
 d. Relationship Between Kind of Work and Grade Finished 326 
 Wage 328 
 
 a. Wage Earned at Specified Times 330 
 
 1. Wage in First Job 330 
 
 2. Wage in Latest Job 335 
 
 3. Wage at First Conviction 337 
 
 4. Wage at Latest Conviction 339 
 
 5- Average Wage for Prevailing Work 341 
 
 b. Relationship Between Average Wage and Social Factors 348 
 
 1. Relationship Between Average Wage and Grade Fin- 
 ished 348 
 
 2. Relationship Between Average Wage and Number of 
 Convictions 351 
 
 c. Brief Comparison with Wage of Certain Groups in Gen- 
 eral Population 3^8 
 
 Estimate of Regularity of Work. Distribution for Institutional 
 
 Groups and by Nativity and Color 361 
 
 Total Estimate of Work Record 364 
 
 a. Distribution for Institutional Groups 368 
 
 b. Distribution for Nativity Groups 369 
 
 c. Relationship to Factors within Work Record . . . .370 
 
 d. Relationship to Other Social Factors 372 
 
CONTENTS xvii 
 CHAPTER XII 
 
 PAGE 
 
 History of Sex Irregularities 380 
 
 First Sex Experiences 380 
 
 a. Age at First Sex Offense. Distribution by Institutional 
 
 and Nativity Groups 380 
 
 b. Rape or Consent . . 381 
 
 c. Sex Instruction 383 
 
 Extent of Sexual Irregularity 385 
 
 a. Percentages of Prostitutes, Those Sexually Irregular (Not 
 
 for Money), Occasional Sex Offenders and Non-sex 
 Offenders. Distribution by Institutional and Nativity 
 
 Groups .... 385 
 
 b. Length of Time in Prostitution 389 
 
 c. Length of Time Sexually Irregular 390 
 
 Special Study of Prostitutes 391 
 
 a. Age at Entering Prostitution by Institutional and Nativ- 
 
 ity Groups. Relationship to Number of Convictions, 
 
 Grade Finished and Home Conditions 391 
 
 b. Consideration of Other Social Factors in Lives of Prosti- 
 
 tute Group 400 
 
 Venereal Disease 408 
 
 CHAPTER XIII 
 
 Mental Capacity: Comparison with the General Popula- 
 tion 413 
 
 Comparison Based on Stanford-Binet 416 
 
 Comparison Based on Yerkes-Bridges Point Scale ..... 424 
 
 Comparison by Means of Woolley Series 432 
 
 CHAPTER XIV 
 
 Mental Capacity: Inter-Comparison of Delinquent Groups 435 
 
 Comparison of Groups Considered in Present Investigation . 435 
 
 a. Comparison of Total English-speaking Portions of Groups 43 7 
 
 b. Comparison of Institutional Groups: White Women 
 
 Only 443 
 
 Comparison of Our Delinquent Groups with Others . . . 447 
 
 a. Bedford and Albion Reformatories 447 
 
 b. Bedford Reformatory and Chicago House of Correction 449 
 
 c. Auburn Women and Auburn Men 450 
 
 d. New York Penitentiary Women and Ohio Penitentiary 
 
 Men 4SS 
 
 e. Workhouse Women (New York) and Workhouse Men 
 
 (Ohio) 4S6 
 
xviii CONTENTS 
 
 CHAPTER XV 
 
 PAGE 
 
 Mental Capacity: Related to Various Factors Which 
 
 Affect the Delinquent Group 460 
 
 Mental Capacity in Relation to Criminal Career .... 460 
 
 a. Mental Capacity and Nature of Offense 460 
 
 b. Mental Capacity and Extent of Criminal Record . . 464 
 
 c. Mental Capacity and Factors Connected with Beginnings 
 
 of Criminal Career 470 
 
 Mental Capacity in Relation to General Factors .... 480 
 
 a. Mental Capacity and Age 480 
 
 b. Mental Capacity and Classification by Color and Nativity 480 
 
 c. Mental Capacity and Use of Drugs and Alcohol . . . 485 
 Mental Capacity in Relation to Home Conditions .... 489 
 
 a. Mental Capacity and Estimate of Home Conditions . . 489 
 
 b. Mental Capacity and Hereditary Factors 491 
 
 Mental Capacity in Relation to Educational Background . . 493 
 
 Mental Capacity in Relation to Work Record 498 
 
 a. Mental Capacity and Age at Starting Work .... 498 
 
 b. Mental Capacity and Kind of Work 500 
 
 c. Mental Capacity and Wage 504 
 
 d. Mental Capacity and Estimates of Work Record . . 507 
 Mental Capacity in Relation to Sex History 510 
 
 a. Mental Capacity and Age at First Sex Offense . . . 510 
 
 b. Mental Capacity and Classification as Sex Offenders . 512 
 
 c. Mental Capacity and Venereal Disease 516 
 
 d. Mental Capacity and Age at Entering Prostitution . . 519 
 
 CHAPTER XVI 
 
 Summary and Conclusions 524 
 
 Bibliography ^oi 
 
 Index 537 
 
A STUDY OF 
 
 WOMEN DELINQUENTS 
 
 IN NEW YORK STATE 
 
STUDY OF WOMEN DELINQUENTS IN 
 NEW YORK STATE 
 
 CHAPTER I 
 
 INTRODUCTION: GENERAL ORIENTATION WITH REF- 
 ERENCE TO PROBLEMS AND METHODS OF 
 TREATMENT 
 
 QOME knowledge of the characteristics of those who make up the 
 S-' criminal and delinquent population of any country seems a slight 
 thing to demand as a prerequisite for the development of adequate 
 methods of handling the problem which this group represents. The 
 recognition of this demand as elementary may account for the fact 
 that ready generalizations regarding the criminal have been so abun- 
 dant. The work of Lombroso introduced certain tendencies which had, 
 at least, not been clearly defined, in connection with earlier expressions 
 of opinion. In the first place, these earlier statements made no pre- 
 tense of being scientific in the sense of being based on investigation, an 
 ipse dixit being considered quite sufficient to establish their claims. 
 Lombroso introduced at least the appearance of investigation, puerile 
 though it was for the most part. In the second place, these accounts 
 were offered, in the main, merely as descriptive literature of some in- 
 terest to the general population, but not as the basis on which the 
 treatment of the criminal should be determined. The insistence upon 
 the fact that th^ understanding of the criminal should play a larger 
 part in determining his treatment than should the facts concerning his 
 crime, constitutes Lombroso's great achievement; an achievement 
 which stands even after it has become clear thsft the majority of the 
 conclusions which he proffered are unsound. 
 
 NECESSITY OF MORE ADEQUATE DEFINITION OF CONCEPTIONS REGARDING 
 WOMEN DELINQUENTS 
 
 There is at present abundant literature regarding the criminal 
 group, much of which is not even worth citing because it rests, at its 
 
 3 
 
4 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 best, upon the most casual and superficial observation and at its worst 
 upon what the writer thinks he would find on observation. Possibly 
 no great harm was done by reliance upon such generalizations, when 
 little attention was-paidto the treatment oi eitl^er meii or-womeu.oi- 
 fenders^ the problem of their incarceration being considered as solveid 
 if two demands #ere met, ms.: the dematid of punishment for the 
 crime and the demand of protection of the state from further crimes. 
 With the development within recent years, first, of an insistence on 
 humanitarian treatment of the criminar merely on the ground that he 
 is a ."human_being"_^_and, more recently, of a tendency to insist that 
 ttie'J)enal institutions must be thought of as places for readjust- 
 ment of individuals to society, it becomes of primary importance to 
 know of what nature this human material j§. The system of penal 
 institutions of any state g.nd the plan of adrrjinistratjotj of each giyeji 
 institution are determined, ' in part at least, by certain conceptions re-; 
 garding the persons who are to be handled. These conceptions may 
 be vague and unforrnulated, or they may be definite and dogmatically 
 propounded. In either case they forni one important element in the 
 determination of procedure and so become of practical importance. 
 
 The acceptance of this point of view has become almost universal 
 among those who interest themselves in modern penology from either 
 its theoretical or its more practical aspects. As has been indicated 
 in the preface, the Laboratory of Social Hygiene was a direct out- 
 growth of Dr. Davis' conviction regarding the importance of a more 
 intelligent study of delinquent women,^ a conviction, it should be noted, 
 which represented a plea for more information from one who was 
 daily in personal contact with such women and constantly facing the 
 practical problems of their management. More recently we find Os- 
 borne, who likewise approached the problem from the practical side, 
 making the following statement : 
 
 "To lay the foundation of a new and genuinely scientific penology,- 
 we must make absolutely clear to ourselves the real nature and char- 
 acter of the men whp populate our prisons ; there must be no uncertainty, 
 in the minds of any of us as to what is meant when we refer to 'criminals,' 
 'convicts,' or 'prisoners.' "' ' ,' 
 
 'The work of Elizabeth Frye in England represented the most significant" 
 development of the humanitarian trend as applied to women delinquents 
 
 'See Davis, Katharine Bement. "The Rational Treatment of Women Con- 
 victed df- CTitrlein the Courts of New York City." ' - •- 
 
 T, ." *^^^Sr!?^' Thomas Mott. "Society and Prisons," p. IS, Yak Universihr- 
 Press, 1916. - ■ - , -' ■ 
 
PROBLEMS AND METHODS OF TREATMENT , ^ 
 
 Instances indicative of the same poiiit of view could be miiltiplied 
 ty reference to the discussions of any of the many conferences called 
 for the consideration of penal problems. It has shown itself in most 
 important fashion, also, by the actual introduction into prisons, re- 
 formatories, juvenile institutions, and courts of experts capable of 
 carrying out investigations designed to throw more light on the charr 
 acter of the individuals under observation. 
 
 The purpose of the present study may be stated as an attempt to 
 furnish a scientific basis for the conceptions regarding woman of? 
 fenders, through an investigation concerning the(distinguishing char* 
 "acteristics of women convicted of either serious crimes or minor 
 offenses in New York State^ It has not been our object either to 
 defend any specific thesis or to combat established ideas regarding 
 the characteristics of these woman. We have merely aimed to deter- 
 mine, so far as possible, the facts. 
 
 INADEQUACY OF MATERIAL FROM OTHER SOURCES AND LIMITATIONS ON 
 COMPLETION OF DESIRED COMPARISONS WITH OTHER GROUPS 
 
 Our . efforts were directed first toward making a survey of the 
 group of women delinquents under consideration, with a view to secur- 
 ing as much information as possible with regard tp their mental_cagaci- 
 ties and -the main facts of their pergonaLand environmental histories.: 
 With that material at hand we have attempted to make such compari- 
 sons as might be enlightening on the basis of our own data, and also 
 such other comparisons as were made possible by the data available; 
 as the result of other investigations. In this study, physical and 
 medical facts have been almost entirely disregarded since the condi- 
 tions of our inquiry have been such as to make impossible the securing, 
 of adequate medical data for most of the groups studied. Though we 
 recognize this omission as unfortunate, it has seemed to us preferable 
 to have our study incomplete in this direction rather than to use 
 information so inadequate that misleading conclusions might be drawn.; 
 
 The first comparison which seems called for is that between our 
 group of women ^delinquents and women in general, to determine 
 whether the group of delinquent women represents in any of. its 
 asped:s, a special selection out of the general female population, or 
 whether we may think of them as a mere random sample from the 
 total, differentiated at present by their criminal careers but not other- 
 wise. Specifically, we wish to know, for example, whether the delin- 
 quent group is markedly different in respect to mentality or whether 
 
6 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 it merely shows distribution of high and low mentality around a central 
 tendency with no significant difference from the distribution of the 
 general population. Similarly, we wish to know whether those who 
 break the law have differed markedly from the sum total of women in 
 their general environmental background as represented by their early 
 home conditions, their educational opportunities, the economic stress 
 to wtiMTthey have been subjected, and numerous other conditions 
 Which have acted upon them. Again, they should be compared with 
 the whole female population with reference to physical condition and 
 the presence orTibsence of disturbing hereditary factors. Such a com- 
 parison of the delinquent group with the whole population of women 
 is obviously of first importance for the. determination of the signifi- 
 cance to be attached to these various conditions as causative factors 
 in connection with criminality. In fact, one of the great sources of 
 fallacy in the literature concerning pathological conditions of all kinds; 
 whether medical or social, has been the tendency to study only the 
 deviating group and to assume that conditions present here in large 
 numbers have been factors occasioning the abnormality of the group, 
 without first determining whether. they may not be present in equal 
 measure in an unselected sample of the general population. 
 
 A mere recital of some of the factors which it is important to study 
 should be sufficient to suggest that, unfortunately, one side of the com- 
 parison, that represented by the non-delinquent population, will be 
 lacking for the great majority of points on which we wish enlighten-, 
 ment. Such information as is given in the United States Census re- 
 ports is the most comprehensive material available, but this is presented 
 under such general headings that its usefulness for our purpose is 
 limited. Other less extensive sources of information give a certain 
 amount of scattering data regarding special groups of women, but 
 none of these can be taken as adequately representative of women as a 
 whole. The fact is, therefore, that at present the comparison Sug- 
 gested as of prime importance, namely, that between the general 
 female population and our special group of offenders, cannot be carried 
 through in any satisfactory fashion. 
 
 As approximations to that result, two makeshifts are possible. The 
 first of these is the one just mentioned, namely, a comparison of this 
 group with such general data as are available on the whole female popu- 
 lation through the census reports. For all such information, however, 
 not only are we restricted to the most general data, but we must refer 
 back to the 13,th census which represents facts which were true of the 
 
PROBLEMS AND METHODS OF TREATMENT 7 
 
 population in 1910. We cannot be certain t^bat— F&k4icinshi£S-held the 
 same for the years covered by our study, 1915-7, especially in the light 
 of the war conditions of this period. The second possibility is a com- 
 parison with data on separate groups of women for such information 
 as may be available. Such comparisons may be suggestive, though not 
 final, and we shall not hesitate to make use of these sources whenever 
 possible. 
 
 Even in the face of these limitations on the satisfactory rounding 
 out of our comparisons, it has appeared desirable to present, in con- 
 siderable detail, the phase of the comparison which we are able to 
 furnish as the result of our investigations, in order that this may be 
 accessible at such time in the future as more complete data concerning 
 the non-delinquent female population may have been secured. 
 
 The second comparison which naturally suggests itself is almost 
 as much out of the question in any complete form at present as is the 
 preceding, namely, the comparison of women oifenders with similar 
 groups of men offenders. It is true that most of the work which 
 has been done on the subject of crime and criminals has been the 
 result of the study of men criminals. Goring's important work on 
 the English convict * is concerned entirely with men. Dr. Glueck's 
 report on a year's admissions to Sing Sing prison ^ gives comparative 
 data of special interest to us since it is concerned with men convicted in 
 this state. Dr. Heacox has put at our disposal certain unpublished 
 results of his study of men at Auburn prison which also furnish valu- 
 able material for comparison. These and other studies of men of- 
 fenders will be made use of for purposes of comparison, but it should 
 be noted that they are confined almost entirely to the study of men 
 convicted of relatively serious offenses and committed to state prisons. 
 We lack similar extensive studies of adult men convicted of minor 
 offenses and committed to work-houses, county penitentiaries and re- 
 formatories. It will be apparent from our account of the character 
 of the offenses which the women under consideration have committed 
 that these fall in the latter class of relatively minor offenses. The 
 group of women that is in any way comparable with the type of men 
 offenders who have been most extensively studied is so. small as seri- 
 ously to hamper the working out of any extensive comparison. We 
 note, for example, that while the total number of men committed to the 
 •Goring, Charles. "The English Convict. A Statistical Study." T. Fisher 
 
 "'Glueckr Bernard. ' "A Study of 608 Admissions to Sing Sing Prison." 
 Mental Hygiene, Vol. 2, No. 1, Jan., 1918. 
 
8 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 state prisons of New York during the fiscal year ending July 1, 1917, 
 is 1403^ the total number of women received in Auburn prison, tlie 
 only prison of the state which receives women, during the same period 
 of time is 27/ A fuller discussion of this phase of the comparison 
 will be offered in detail under our account of the nature of offenses 
 committed by women. For any close comparisons between our group 
 and the men convicted of crime, studied in the investigations mentioned 
 above, we feel it necessary to restrict ourselves to those women who 
 have been convicted of felonies, of whom we have 126 cases. These 
 are not entirely comparable with the men sentenced to state prisons,; 
 since they include, in addition to Auburn commitments, cases corns 
 mitted for felonies to the State Reformatory at Bedford Hills, th| 
 New York County Penitentiary, and the Magdalen Home. They do! 
 represent, however, cases all of whom might have been sentenced to 
 state prisons, so that comparison seems reasonably justifiable. The 
 number of commitments of women to the State Prison is so small 
 that it would be necessary either to spread the study over the whole, 
 country or to extend it over a long period of time. To have secured 
 500 cases of women given State Prison terms in New York State woull 
 have required an investigation covering the commitments during a 
 period of twelve years, whereas this number of cases can be secured in 
 less than six months from the men's prisons of the state. 
 
 With reference to this comparison, therefore, of women delinquents 
 as a whole with men delinquents as a whole, we are in practically the 
 same position as for the first comparison suggested ; that is, we must 
 offer data on one side of the comparison without being able to secure 
 the complementary information on the group to be compared. Since 
 we feel certain that studies of male misdemeanants and of male felons 
 not committed to State Prisons will be completed before many more 
 years, we have no hesitation in offering our data as useful for this 
 comparison at some future time. 
 
 ""Report of the Superintendent of State Prisons." New York 1917 p 20 
 table 5; p 116, statement 1 ; p. 232, table 8. The total number gi'ven above is 
 obtamed by combmmg the figures given in the tables specified as the number 
 of cases received m each prtson from the district assigned to that brison A 
 combination of the total number received during the year in all the prisons ot 
 the state would be misleading since a given individual might thus appear sev- 
 eral times,_ being recorded, for example, as received at Sing Sing from Sing 
 Sing district and received at Auburn later from Sing Sing prison, and pos- 
 sibly again received at Clinton or at Great Meadow prison from Aubirn prison 
 Ibid p. 198 table 1 (number received exclusive of those returned for vio- 
 lation oi parole). 
 
PROBLEMS AND METHODS OF TREATMENT 9 
 
 The third type of comparison which requires consideration is more 
 lirectly possible from our own data. We refer to the study of the 
 ^ter-relations observed within the group of delinquent women them- 
 selves. yThe total group may be divided into numerous subdivisions 
 selected now on one basis of classification, now on another, in accord- 
 ince with the special interest of the moment. For example, there is 
 first the division with which we began ; namely, of groups in terms of 
 the institution to which the women have been sentenced. It seems 
 desirable to determine what general principles of selection, if any, have 
 been operative to determine commitment to one institution rather than 
 to another. Cutting across this classification is that in terms of nature 
 af offense committed, since we are equally concerned to know whether 
 individuals committing one type of offense are distinguished in any 
 important ways from those committing other offenses. Similarly we 
 may reclassify our whole number for special consideration into groups 
 according to nationality and color, or according to criminal record, or 
 sex history, or record of juvenile offenses. These and numerous other 
 groupings which suggest themselves offer fields wherein inquiries re- 
 garding the characteristic marks of the special group as distinguished 
 from the whole group of delinquent women are pertinent. 
 
 It is important to bear in mind, with reference to all discussion of 
 such inter-relationships, that we offer our data as applying only to 
 relationships within the group of women dehnquents under considera- 
 tion. We make no claims to their extension to other groups, as for 
 example to men criminals who may differ in many respects from 
 women criminals, or to juvenile delinquents, who constitute another 
 quite distinct problem. 
 
 USE OF MASS METHODS OF TREATMENT RATHER THAN CASE HISTORY 
 METHODS IN PRESENT STUDY 
 
 1 The methods of approach to these problems naturally divide them- 
 selves into methods of investigation and methods of analysis of results. 
 These are described in detail in Chapters III, IV and V. A few points 
 of general significance may, however, be noted here. 
 
 First of these js the fact that the methods of investigation have con- 
 sisted of careful individual studies, while the methods of analysis have 
 treated the results as mass data and handled these statistically. 
 
 From the standpoint of the present , study the main object in the 
 detailed individual studies was the increased accuracy of the inform> 
 
10 STUDY OF WOME^J DELINQUENTS IN NEW YORK 
 
 tion secured.' The fact that our numbers are not large made us feel 
 the necessity of making our data as reliable as possiblfe- The more 
 thoroughly an investigator comes to know a given woman, the more 
 likely is she to acquire the main facts of importance concerning her 
 and to pass judgment satisfactorily on the trustworthiness of the facts. 
 Moreover, the mere knowledge on the part of the inmate that facts are 
 to be verified tends of itself to reduce the amount of false statement. 
 And, in addition to these aids to accuracy, there is always the pos- 
 sibility that a thorough investigation will unearth clues which will lead 
 to clearing up the case and giving an entirely different picture of 
 significant factors from that offered by the inmate herself. 
 
 In our account of the sociological methods (Chapter IV) the fact 
 is emphasized that there was much difference in the completeness of 
 this investigation in the various groups studied : that it was most satis- 
 factory for the Bedford group and least so for the Workhouse. We 
 feel, however, that even for the latter group our data are considerably 
 more reliable than any information based on an inmate's statement 
 alone. 
 
 Having accumulated the material by individual methods the ques- 
 tion may arise regarding the reasons for applying mass methods of 
 treatment to them. These methods were chosen consciously and in 
 spite of full recognition of the importance of careful study of the 
 individual for understanding of that individual. Wherever practical 
 direction of the treatment of the individual case has been an issue 
 we have acted on this principle. At the same time we are not in 
 sympathy with the attitude of distrust toward all methods approaching 
 the statistical, as we find this manifested by many modern leaders in 
 the field of criminology, especially those of the medical profession. 
 
 In the first place," we believe that some background of knowledge 
 regarding many individuals is essential even for the adequate under- 
 standing of a given individual. For example, if we leave the field of 
 criminology for the moment and turn to the province of medicine 
 which has been particularly wary of mass methods, it will probably be 
 admitted that a patient suffering with a given combination of symptoms 
 at a given time will feel more confidence in a physician who not only 
 studies the individual case before him "as an individual" with thor- 
 oughness, but who brings to such study a. background of knowledge 
 
 ' It should be remembered that there was another entirely different rea<!on 
 for prosecuting the individual studies with thoroughness. This was the fact 
 that, m the case of the Bedford women especially, this information was relied 
 upon m the mstitution as a basis for treatment of the case. 
 
PROBLEMS AND METHODS OF TREATMENT 11 
 
 of some cases other than the present one. Some such background may 
 be acquired through the physician's own experiences and the anecdotal 
 material which he may acquire from professional friends. Such in- 
 formation, not organized nor tabulated in any way, may contribute 
 materially to his skilful handling of the given patient at the given 
 time. However, there are few of us who would elect to trust our- 
 selves to the physician with only that background. Some contact with 
 the generalized experience of others is the least that we can accept 
 from our physician. 
 
 The method of case histories has been the favorite medical device 
 for securing this background, and it has unquestionably its uses. In 
 our opinion it has also its dangers, in its very lack of objectivity. It is 
 too susceptible to the whim or prejudice of the person who selects the 
 cases to be used and compiles their histories. It is not possible, 
 through case histories, to prove or disprove any facts, but it is possi- 
 ble to convey impressions and establish convictions. Whether or not 
 these are trustworthy depends entirely on the background, the insight 
 and the intellectual honesty of the person who presents them. It would 
 be quite possible so to select cases that they would convey, more or less 
 convincingly, the impression that women convicted of larceny are blue- 
 eyed. It would be only necessary that we cite only blue-eyed women, 
 and that we always mention this characteristic and that we happen to 
 mention other characteristics with less regularity. Lombroso's work 
 exemplifies this treatment, carried almost to the degree of absurdity 
 suggested above, in the attempt to establish the physical stigmata of 
 the criminal. 
 
 It has seemed to us that the dangers involved in mere individual 
 treatment, or mere series of case studies, were so great that the more 
 organized mass presentation must precede any special studies of indi- 
 viduals. Without the background of such a study there is the risk 
 of making much of facts which are in no sense unusual and at the same 
 time disregarding facts of importance which genuinely distinguish this 
 individual from his fellows. 
 
 We would not urge that it is possible to say which are the more im- 
 portant,— the common, characteristic facts or the unusual distinguish- 
 ing traits. The essential thing is that the correct poin of view be 
 associated with each. The traits which occur infrequently in a delin- 
 quent group are especially important to note as a basis for differentiat- 
 ing the treatment of the individual or the small group of individuals in 
 whom this trait occurs from the whole group. 
 
12 STUDY OF WOMEN DELINQUENTS IN NEW YORk| 
 
 ^ On the other hand it is important for the treatment of delinquent 
 groups as a whole both that common characteristics be noted and that 
 they be recognized as common. As we have already stated, methods 
 of treatment are inevitably affected by the preconceptions of the gov- 
 erning bodies concerning the characteristics of the group with whicb 
 they deal. Obviously the plan of organization of the penal system 
 will be influenced by the existence of any of the following assump- 
 tions : that all criminals are congenitally degenerate ; that they are prfr 
 dominantly feeble-minded; that they are abnormally clever; that they 
 are victims of early environmental conditions; that they are produced 
 by economic necessity; that they are predominantly uneducated, and 
 that this lack of education has been a factor in their criminal career; 
 that they have more than the average education for their social or racial 
 group and that this fact has affected them adversely. These, and 
 many other generalizations have figured in the minds of persons dealing 
 with delinquents and have contributed toward the penal systems which 
 have been inaugurated. Many of them are open to scientific study, 
 as Goring has so conclusively demonstrated with reference to Lom- 
 broso's theories of the physical degeneracy of criminals. Presumably 
 the problem of the delinquent will not prove so simple as to be sum- 
 marized in any of the above general formulae, but, if we can avoid 
 being misled into assuming that a condition which exists commonly 
 must be a cause, the discovery that a state of affairs is to be expected 
 with frequency in any delinquent group should have at least some bear- 
 ing on their treatment. 
 
 The great advantage of the case history method is the fact that it 
 presents a picture of the whole personality, which the present type of 
 study fails to show. In our opinion both types of study should be 
 used since each can throw light on the other. We have chosen the 
 mass method as particularly applicable to our data and as meeting, in 
 our opinion, a definite need in the study of criminology. 
 
CHAPTER II 
 
 GENERAL ACCOUNT OF DELINQUENT GROUPS STUDIED 
 
 ' I ^HE total group under consideration is made up of ^omen con- 
 ■*- victed of offenses against the law in the courts of iSTew York 
 State. V We shall refer to this constantly as an adult group, since all 
 cases have a legal age ^ of sixteen years or over. There is no upper age 
 limit for the individuals~included although there is a tendency toward 
 such an upper limit for certain of the groups. 
 
 Though any oflfense against the law is technically a crime, and any 
 person committing a crime may be called a criminal, we have preferred 
 to refer to these women as "delinquents," or, using a most general 
 term, as "offenders." The term "criminal" connotes to most persons 
 an individual guilty of one of the more serious offenses and is approxi- 
 mately synonymous with the term "felon." Since the major part of 
 the group of women whom we are studying is not of this type, we have 
 preferred to avoid the term "criminal." It seems a far cry to desig- 
 nate as "criminal" the girl of sixteen or seventeen who' may be con- 
 victed of "associating with vicious and dissolute persons and being in 
 danger of becoming morally depraved," or the woman who is con- 
 victed, even for the twentieth time, of intoxication, or, in fact, even the 
 common prostitute whose offense against the law is made possible only 
 by the participation of men, who are not even considered accomplices 
 in a criminal act in the eyes of the law. The very fact that, there is 
 such extreme lack of uniformity between this and other countries, 
 and even between different sections of this country, as to whether these 
 acts constitute even legal offenses, furnishes another reason for hesi- 
 tating to brand those committing them by the term "criminal." 
 
 GENERAL MAKE-UP OF GROUPS, WITH A VIEW TO SECURING REPRESENTA- 
 TIVES OF ALL TYPES 
 
 In planning the make-up of the total number to be studied our main 
 
 desire was to secure, as our composite group, one which should con- 
 
 *We distinguish between "legal" age and "'actual" age, the former being 
 the age which the court accepts at the time of conviction. There are a few 
 cases, in our total number, where the actual age was under sixteen. 
 
 13 
 
14 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 tairt(Tepresentatives of all types and degrees of delinquency ordinarily 
 found among women offenders. It seemed hopeless to plan to obtain 
 these in the actual proportions in which they occur, but we hoped, by 
 proper weighting of the groups, to be able to offer eventually a satis- 
 factory picture of the sum total of the women offenders of the state. 
 We realized that any particular institutional group could not be con- 
 sidered representative, since numerous factors, many of which would 
 be impossible of evaluation, enter into the selection of cases to be sent 
 to any given institution. This was, in fact, the main consideration 
 which led us to the extension of our study beyond the scope of the 
 Reformatory at Bedford Hills, where the work was initiated. 
 
 The first problem was to determine how we could best secure the 
 representative total desired. The two possibilities which suggested 
 themselves were the following : ( 1 ) to secure the women at the time of 
 conviction as they came through the various courts of the state; or 
 (2) to take them after commitment as they were to be found in the 
 various institutions. The former method, had it been feasible, would 
 have had many advantages over the latter, since it would have made 
 possible the elimination of the factors of selection represented by the 
 sentence to one institution or another. However, it appeared utterly 
 impracticable to approach our problem from the avenue of the court, 
 under present conditions. If a clearing house or a series of clearing 
 houses were in existence, this method of approach would be possible, 
 but in the absence of some such plan, whereby the women could be 
 held, subject to study, for a short time at least, conditions are too 
 variable and not adapted to the successful carrying out of any plan 
 of investigation. Women would be shifted from one place to another 
 while their investigation was in progress, which would mean endless 
 waste of time in the mere attempt to keep track of their location. 
 Moreover, the majority of city and county jails offer no places which 
 would be evep reasonably satisfactory for interviews and mental ex- 
 aminations. \^he approach by way of the institutions to which the 
 cases had been committed seemed therefore the only practicable 
 methods) This line of approach we have accordingly followed. 
 
 Within the several institutions we have made it our practice to take 
 consecutive commitments, in order that the individuals used might be 
 representative of the group, with no further selection than that which 
 was made by the court when the commitment was determined upon. 
 In our account of the various groups we shall indicate to what extent 
 we have been able to follow this plan absolutely, and where we have 
 
ACCOUNT OF DELINQUENT GROUPS STUDIED 15 
 
 had to make adjustments to meet special conditions. We feel justified 
 in saying that tendencies toward special selection have been reduced 
 to a minimum within our groups. In one case only, that of the New 
 York City Workhouse, have these tendencies proved very serious. In 
 that case, as we shall see, we have found it necessary to eliminate a 
 part of the very large group under consideration and to treat our cases 
 as typical of a specially defined portion of the Workhouse rather' than 
 of the whole. 
 
 Since it was not feasible to make a survey of all the institutions of 
 the state it became necessary to make a selection from among the avail- 
 able institutions and delinquent groups. In the effort to include all 
 types and degrees of offenders the following groups were chosen : ( 1 ) 
 The New York State Reformatory for Women at Bedford Hills ; (2) 
 The State Prison for Women at Auburn; (3) The New York Mag- 
 dalen Home (Inwood House) ; (4) The New York County Peni- 
 tentiary; (5) The New York City Workhouse; (6) A group of women 
 placed on probation through the Night Court for women for the bor- 
 oughs of Manhattan and the Bronx (Ninth District Court). A brief 
 description of the general characteristics of these groups may serve to 
 make clear the reasons for their selection. Detailed analysis of the 
 personnel of the various institutions will constitute one of the problems 
 to be discussed later in the text. 
 
 DESCRIPTIVE ACCOUNT OF GROUPS 
 
 (a) The New York State Reformatory for Women, Bedford Hills, 
 
 N. Y. 
 
 The State Reformatory for women at Bedford Hills receives 
 women from the eastern part of the state.^ The Western House of 
 Refuge at Albion is the reformatory for the rest of the state and re- 
 ceives women under exactly the same conditions as does the reforma- 
 tory at Bedford Hills. The institution at Bedford Hills was chosen 
 as representative of both reformatories of the state and no study was 
 made of the institution at Albion. 
 
 These institutions receive women convicted of a wide range of of- 
 fenses, including both felonies and misdemeanors. Of the felonies, 
 
 " Commitments to the New York State Reformatory for Women at Bedford 
 Hills are from the first, second, third and nmth judicial districts. This in- 
 cludes 19 counties : the five counties within Greater New York, the counties on 
 Long Island, and the counties bordering the Hudson River as far north as 
 Rensselaer County. 
 
16 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 murder in the first or the second degree is the only oiFense for which 
 one cannot be committed, except that a woman cannot be sentenced to 
 the reformatory for a felony after having had any previous conviction 
 for a felony. Of the misdemeanors and other minor offenses there 
 are very few for which one cannot be committed to a reformatory, ex- 
 cept in case of offenses for which no institutional commitment is pro- 
 vided. Accordingly the Bedford group has the possibility of includ- 
 ing almost the whole range of offenses for which women are impris- 
 oned. All cases committed to these reformatories are given an inde- 
 terminate sentence not to exceed three years, power being vested in 
 the Board of Managers to parole or discharge at any time before the 
 expiration of the three years.V No cases are received under sixteen 
 years of age, and it was formerly true that none were received over 
 thirty. There is at present legal provision for the commitment to the 
 reformatory of individuals over thirty, convicted of specified offenses 
 in certain parts of the district covered. There is, however, a tendency 
 to think of thirty as an upper age limit for the reformatory group^ 
 
 From many points of view our study of the Bedford women is more 
 satisfactory and complete than is our study of any of the other groups. 
 For the understanding of this group we had more background than 
 for any of the others included because of the fact that the work of 
 the Laboratory had been centered here from its beginning. In fact 
 this type of study had been inaugurated by Dr. Davis, in her "Study 
 of Prostitutes at Bedford" published in 1913.^ The investigation 
 made by Dr. Weidensall, reported in "The Mentality of the Criminal 
 Woman," * dealt also with this group. In addition to this general back- 
 ground of understanding there was the fact that all the conditions 
 made possible here a particularly thorough and reliable investigation. 
 The physical conditions surrounding the work, which are important 
 for their influence on the success of both interviewing and examining, 
 are much better than in any of the other institutions. Moreover, the 
 work was so well established that it was accepted as a regular part of 
 the routine of admission to the institution. In addition to these 
 factors, there is the fact that the organization of the Laboratory made 
 possible a greater thoroughness of investigation here than for any of 
 the other institutions. For all these reasons we shall tend to consider 
 
 'Kneeland, George J. "Commercialized Prostitution in New York City." 
 Chapter VIII by Katharine Bement Davis. "A Study of Prostitutes Committed 
 to the State Reformatory for Women at Bedford Hills." 
 
 * Weidensall, Jean. "The Mentality of the Criminal Woman," 1916. 
 
ACCOUNT OF DELINQUENT GROUPS STUDIED 17 
 
 the Bedford group as of rather central niterest, and shall use it fre- 
 quently as a point of reference for other groups. 
 
 The Bedford commitments under consideration for the main part 
 of the present inquiry consist of 102 consecutive cases committed be- 
 tween September 1, 1916, and August 9, 1917. In addition to these, 
 whom we have designated the "Contemporary Group," we have used 
 for much of the material on the psychological tests other cases ex- 
 amined prior to the study of this group. These have been used for 
 the greater certainty which they offered that the Bedford group is 
 , actually a fair sample of commitments to Bedford in general. The de- 
 tails as to numbers used will be given at a later point. We have not 
 used the earlier groups on the social data, since the methods used for 
 securing such data were not entirely the same during the early part 
 of the Laboratory work and the forms for recording data were very 
 different. This latter difference made the work of including the early 
 Bedford groups so great as to be prohibitive within the time limits of 
 the present investigation. 
 
 (b) The State Prison for Women, Auburn, N. Y. 
 
 The state prison for women at Auburn represents for women of- 
 fenders of this state the same- type of institution as is furnished in 
 Sing Sing, Auburn, Clinton, and Great Meadow prisons for men felons. 
 We have already commented on the great discrepancy between the 
 number of men and the number of women committed to institutions 
 of this type. Any woman convicted of a felony may be sentenced to 
 Auburn prison. No cases other than felonies are legally committable 
 there. It is to be expected that in the case of offenses for which there 
 is a range of choice regarding the institution to which they shall be 
 sent, commitment to state prison will be used where the offense is 
 especially grave or where the offender appears to be a particularly dan- 
 gerous member of the community. Sentence to state prison is com- 
 monly thought of as — by its very nature — the most severe sentence 
 which can be given, although it is possible for one to receive a shorter 
 term in state prison than that made possible by the indeterminate re- 
 formatory sentence with its three-year maximum. No person under 
 sixteen years of age can be committed to state prison, but there is no 
 upper age limit. 
 
 We have already commented on the slow rate of admission of 
 women to Auburn prison. For this reason it was necessary to begin 
 our study with cases whose admission antedated by a considerable 
 
18 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 period of time the actual beginning of our investigation. In doing this 
 we lost entirely from our series eight cases, five of whom had served 
 their time and left the institution before we began our work, one of 
 whom had died, and two of whom refused to talk about their cases or 
 take any examinations.^ The first case of our regular Auburn series 
 was committed February 4, 1915, and the last case, June 27, 1917, 
 making the range of time covered by the commitments of this group 
 more than two years. The total number of cases received during this 
 time was 88. These numbers were supplemented, for certain of the 
 tests where we were particularly desirous to increase our numbers, by 
 additional cases preceding and following this regular group. 
 
 There were many difficulties inherent in carrying on such an inves- 
 tigation with this group. In fact without the cordial cooperation which 
 we received from the authorities of the prison it would have been en- 
 tirely impracticable to attempt this study.° Those women especially 
 whose commitments dated back one or two years before the beginning 
 of our study, were naturally inclined to resent' going over the facts 
 of their cases again. Moreover, on general principles they were in- 
 clined to object to the inauguration of any such a careful study of them 
 as individuals. It required much tact and ingenuity to meet this atti- 
 tude of mind. We were fortunate in finding an opportunity to be of 
 some real value to certain of the women in the course of our investiga- 
 tions, by clearing up misunderstandings with members of their family 
 and establishing more friendly relations between them. This helped 
 toward the creation of a more friendly feeling toward our work. 
 
 (c) The New York Magdalen Home'' 
 
 The New York Magdalen Home is one of several institutions 
 under private management which receive women committed through 
 the courts. The main other institutions of the state coming under this 
 
 "The cornmitment data were available on these cases and they are accord- 
 ingly included in the tables which give information concerning the offenses for 
 which committed. 
 
 "We cannot express too warmly our appreciation of the cooperation which 
 we received. Dr. Frank L. Heacox, physician of the Prison, aided us at every 
 turn. Not onl> did he permit us the use of his offices in the women's prison 
 for our work, but his friendly attitude of cooperation helped much to over- 
 come the antagonism which the inmates felt toward any strange innovation 
 He also permitted us the use of his records. For all the medical data which 
 we offer on this group of women we are directly indebted to Dr. Heacox. To 
 Mrs. Margaret E. Daly, matron of the women's prison, Miss Curtis first as- 
 sistant, and Mrs. Stone, head of the school, we are also indebted 'for con- 
 stant cooperation and assistance. 
 
 'This institution is now known as Inwood House. 
 
ACCOUNT OF DELINQUENT GROUPS STUDIED 19 
 
 classification are the following: the House of the Good Shepherd 
 (New York City), the House of the Good Shepherd in the City of 
 Brooklyn, the House of Mercy (New York City), the Wayside Home 
 of the City of Brooklyn (now at Valley Stream, Long Island), the Asy- 
 lum of Our Lady of Refuge (Buffalo), the Mount Magdalen School 
 of Industry and Reformatory of the Good Shepherd (Troy), St. 
 Anne's School of Industry and Reformatory of the Good Shepherd 
 (Albany), the Shelter for Unprotected Girls (Syracuse). 
 
 Since it was obviously not feasible to obtain representatives from 
 all o,f the above institutions, we hoped to secure a fairly adequate 
 representation through one group of 100 cases, including 50 cases from 
 the New York Magdalen Home and 50 cases from the New York 
 House of the Good Shepherd. The latter institution, however, refused 
 our request for permission to undertake the study, stating that it 
 was contrary to their method of government "to subject our inmates 
 to scrutiny relative to their past." Accordingly it was decided to 
 accept the group from the Magdalen Home as typical of commit- 
 ments to the private institutions. Observation of the women as they 
 pass through the courts has convinced us that this is not entirely true, 
 and that the House of the Good Shepherd, at least, has probably 
 a somewhat distinctive group. Specifically it appears to receive more 
 cases of older women, and of women convicted of offenses of the in- 
 toxication type, than does the Magdalen Home. We regret that it 
 was impossible to include these in our study. The fact remains, how- 
 ever, that it would be difficult to designate with certainty any one or 
 two of these institutions as characteristic of the group, since, by their 
 very nature, there is much scope for individuality in their organiza- 
 tion. Accordingly, while we hv ve chosen the Magdalen Home as the 
 representative of this group oi institutions, we do not claim that its' 
 cases are necessarily characteristic of commitments to private institu- 
 tions as a whole. 
 
 The Magdalen Home receives women between 16 and 35 years of 
 age from New York City and from a few of the neighboring counties. 
 The range of misdemeanors for which one may be committed is very 
 similar to that for the State Reformatories. The law does not provide 
 for conviction to this institution for felonies, though occasional cases 
 so convicted have been sent here, when the individual seemed espe- 
 cially likely to profit by the lighter sentence.^In such cases the procedure 
 has amounted practically to suspension of sentence by the judge under 
 the condition that the woman voluntarily commit herself to the in- 
 
20 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 stitution. One important characteristic of the private institutions is 
 their privilege of returning to the court, for recommitment elsewhere, 
 cases who prove to be undesirable members of their community. This 
 introduces a tendency toward selection which, although exercised after 
 commitment in particular cases, may well react upon the judges, lead- 
 ing them to refrain from committing cases who are likely to be misfits 
 in these institutions. 
 
 Gases are committed for three different kinds of terms: (1) for 
 definite periods not exceeding one year; (2) for the period of their 
 minority; (3) for an indeterminate period not to exceed three years, 
 power of discharge at any time prior to this being vested in the Board 
 of Managers. The last-mentioned type of commitment is the most 
 frequent, but as a rule women are discharged or paroled after a shorter 
 period of time than is commonly required at the Reformatory at Bed- 
 ford Hills. This fact operates in the same direction as the principle 
 just mentioned : zAz., toward the selection of more promising and less 
 confirmed offenders, in so far as it affects the judges making the com- 
 mitments. 
 
 We include in our study 76 consecutive cases received at the Mag- 
 dalen Home between September 29, 1916, and June 15, 1917. It was 
 not necessary to omit completely any case from this series, though sev- 
 eral cases were omitted from a part of the study. These will be dis- 
 cussed later.* 
 
 (d) The New York County Penitentiary 
 
 The New York County Penitentiary is one of the five penitentiaries ' 
 of the state which receive women. Commitments are made for speci- 
 fied offenses through City Magistrates Court and for felonies and mis- 
 demeanors through courts of Special Sessions, General Sessions, Su- 
 preme Court or County Court. At present as a result of the parole 
 law, all commitments from New York County are given an indeter- 
 minate sentence not to exceed three years, although other neighboring 
 counties may commit women for definite terms. There is no age limit 
 other than the lower limit of sixteen years. 
 
 ' We are greatly indebted to the management of the Magdalen Home for the 
 cordiality with which they granted us permission to carry out this investiga- 
 tion. We are especially grateful to Mrs. Mary E. Paddon, Executive Secretary, 
 and to Miss Janet Macchonachie, Superintendent, for their assistance and co- 
 operation. 
 
 ° Since the completion of our study, the Westchester County Penitentiary 
 has been opened, but this receives only commitments of men at present. 
 
ACCOUNT OF DELINQUENT GROUPS STUDIED 21 
 
 This institution is characterized by an inability to offer anything in 
 the way of constructive training during the period of imprisonment. 
 Formerly there was no provision for employment other than the ordi- 
 nary work of the place. That need was met at the time of this study 
 by a sewing room which accommodates about thirty women at a time. 
 Beyond this there was no organized training or occupation. We might 
 expect therefore that there would be a tendency to commit to the Peni- 
 tentiary unpromising cases who could profit little by the opportunities 
 offered in an institution of the Reformatory .type, or older cases whose 
 habits and characters were so "set" that there was little prospect of 
 altering this. This tendency is even more to be expected here than in 
 the case of the Workhouse, since there are no short term sentences to 
 take care of the more trivial offenders. 
 
 Though our investigation was not started until August, 1916, it was 
 made retroactive beginning with the first case committed under the 
 Parole Law. No definite sentence cases were included. Our study 
 covered a total of 110 cases committed consecutively between January 
 12, 1916, and May 14, 1917. Four cases were lost from this total be- 
 cause they had already been paroled at the time that we began our 
 study. The commitment data concerning the present offenses of these 
 four cases, being available and trustworthy, were used, however. 
 
 (e) The New York City Workhouse 
 
 The New York City Workhouse receives only misdemeanants, but 
 can receive almost all varieties of these. The terms for which they 
 may be committed vary from definite sentences of from one day to six 
 months duration to the indeterminate sentence, for which the maximum 
 is two years. By observation we are convinced that the Workhouse 
 receives individuals of a wide variety of types. As is to be expected 
 from the fact that it is primarily simply a place for holding persons in 
 confinement to serve their time, and that opportunities for constructive 
 individual readjustment along educational, vocational and medical lines 
 are conspicuously lacking, it is preeminently the place for "old round- 
 ers" of all sorts, for cases who have been given opportunities and have 
 failed and whose prognosis is considered most unfavorable. Under 
 the Parole Law it is made imperative that persons having as many as 
 two previous convictions within two years or three at any time shall 
 be sentenced for the indeterminate term, if sentenced to the Workhouse 
 at all. This law of itself tends to divert the stream of old offenders 
 among the misdemeanants to the Workhouse. 
 
22 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 But there is an entirely different principle operative also in deter- 
 mining Workhouse commitments, since the Workhouse offers a short 
 sentence of a few days or weeks which is, from point of view of pun- 
 ishment, the most lenient penalty, except probation, suspended sentence 
 or fines, that can be offered to a trivial offender.^" Accordingly a 
 short Workhouse term is very frequently given a first offender with 
 the idea that her offense is not, serious enough to justify a longer sen- 
 tence. This leads to the inclusion within the total Workhouse commit- 
 ments of at least two quite disparate groups : vis. (1) the old offenders 
 who are given for the most part terms of more than ten days, and (2) 
 fairly new offenders who are frequently given the shortest sentences 
 possible for their offenses. 
 
 There is, however, still another element in the Workhouse to com- 
 plicate the situation, vis.: the group of old intoxication cases. The 
 majority of cases convicted of intoxication in the New York City 
 Courts are given one of the following three penalties: (1) they may 
 be fined not more than $10 with the alternative of serving time, a day 
 for a dollar, in the city prison; (2) they may be given suspended sen- 
 tence" especially used with first offenders, and (3) they may be sen- 
 tenced to the Workhouse for periods of one day to six months. Of the 
 above mentioned three possibilities the Workhouse term is given least 
 frequently to first offenders, being used for such cases as a rule only 
 when it is evident that they cannot take care of themselves, and that 
 they have no one else to take care of them if turned loose. 
 
 Of the first two groups — those fined and those given suspended 
 sentence^ — we have no direct knowledge. With the intoxication group 
 committed to the Workhouse we had some experience, enough to con- 
 vince us that they were in many respects different from the rest of 
 the Workhouse population. Most notably did they differ in degree 
 of recidivism. All the longest records of previous convictions were 
 found among this group and the mean number of previous convictions 
 was larger than for any other group. In addition to this it is an older 
 group than the Workhouse group as a whole, the range of nationalities 
 included is very different and the mentality as measured at the present 
 time at least, is lower than that of the rest of the group. 
 
 _"> By many of the women themselves a short workhouse sentence appears 
 definitely preferable to even a chance on probation. The custom of imposing 
 fines IS now discontniued for all offenses involving prostitution, though they 
 are^still imposed for intoxication and disorderly conduct. 
 
 "^ A small number are given probation, but this is 'infrequent. Suspended 
 sentence is more commonly used for intoxication cases 
 
ACCOUNT OF DELINQUENT GROUPS STUDIED 23 
 
 The Workhouse then may be thought of as made up of at least 
 three quite distinct portions: (1) General misdemeanants, exclusive of 
 cases of intoxication, who give the impression of .being hardened and 
 unpromising or who have failed on previous opportunities; (2) new 
 and relatively hopeful cases who are let off with a few days at the 
 Workhouse, and (3) the intoxication cases who are given Workhouse 
 terms. Of these three groups we confined our study almost entirely to 
 the first. 
 
 The exclusion of the other two groups was brought about by two 
 quite distinct conditions. The second group, made up of the new and 
 more hopeful offenders, had been given for the most part terms of ten 
 days or less. The present policy of the Department of Correction is 
 not to transfer such cases to the Workhouse on Blackwell's Island at 
 all but to allow them to serve their terms in the various city prisons. 
 The physical impracticability of their locations together with the tem- 
 poral difficulties introduced by the shortness of their terms and the con- 
 sequent difficulty in completing their study before they went out made 
 it out of the question to include them in our investigation. 
 
 The intoxication group introduced a different complication. 
 Though all cases committed for a term longer than ten days were actu- 
 ally in the Workhouse they made, of all our groups, the most difficult 
 cases to examine or interview. In the first place, when they had -been 
 coming to the "Island" for almost as long as they could remember, 
 they resented the innovation of such a study, just as one might resent a 
 change in his favorite hotel. In the second place, they frequently 
 appeared to be still suffering more or less from the effects of their 
 latest debauch, during the major part of their term. Where this 
 was not the case they often showed so much evidence of senility 
 and possible deterioration resultant upon their alcoholism that it was 
 impossible to estimate what their original mental ability might have 
 been. Their interviews showed either marked lack of veracity or 
 a slurring of the accuracy of their memory for facts which made 
 their accounts both incomplete and unreliable. The fact that they 
 probably represent the most drifting elements of the population adds 
 to the difficulty of securing even reasonably satisfactory information 
 concerning them. Since this group of alcoholic cases constituted at 
 this time over' fifty per cent of the cases committed to the Workhouse 
 for periods of over ten days it became apparent to us that our study 
 of the Workhouse, which as we shall see later was the least satisfactory 
 of all our groups even without this handicap, would be made almost 
 
24 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 entirely valueless by the erratic qualities of this group. The desirable 
 solution of the problem would have been to make separate studies 
 of these two Workhouse groups, including 100 in each. That was 
 not feasible and accordingly when nineteen intoxication cases had been 
 studied it was decided to restrict our studies for the remaining cases 
 to individuals whose present conviction did not involve intoxication in 
 any form.'^^ 
 
 Owing to the great rapidity with which cases are received at the 
 Workhouse it seemed impracticable to attempt to cover 100 con- 
 secutive cases, especially if both intoxication and other cases were 
 included. Our first plan, therefore, was to take them in four groups 
 of 25 each. We started our first group in accordance with this plan 
 in December, 1916, and covered 27 cases within seven days. By this 
 time we had come to appreciate the special difficulties incidental to 
 work with the group of intoxication cases and had decided to dis- 
 continue testing them. Eliminating these left thirteen cases received 
 consecutively at the Workhouse during this period. After excluding 
 the intoxication cases we found it possible tO' cover the remaining 
 number of cases as one group of consecutive admissions. During the 
 latter period 96 cases were received, making the total number received 
 at the Workhouse and available for study under the conditions as 
 defined above, 109 cases. Of these it was necessary to omit entirely, 
 except for commitment data, seven cases,^' making a total of 102 cases 
 studied more or less completely. 
 
 The special group of women committed for intoxication included 
 twenty-one cases, one of whom was committed to Matteawan and one 
 died before there was opportunity to give mental tests. These twenty- 
 one cases include ten committed for public intoxication, seven for dis- 
 orderly conduct, and four for vagrancy, the latter two charges both 
 signifying intoxication in these cases. This small group of cases 
 studied will be mentioned at various points in our study for the light 
 which they may throw on the characteristics of cases of this type. The 
 numbers are, however, too small to justify the same sort of statistical 
 
 " In addition to the cases specified as public intoxication we have eliminated 
 cases committed for disorderly conduct or vagrancy where the account of the 
 offense makes clear that it was primarily a case of intoxication. 
 
 "The reasons for omission of these seven cases were as follows: One was 
 transferred to Matteawan, two were transferred to other city jails as work- 
 ers, one was in the hospital during the whole of her term, two were taken out 
 on bail pending the results of an appeal and did not return during the period 
 of investigation, one though received at the Workhouse was a ten-day com- 
 mitment and was therefore not included. 
 
ACCOUNT OF DELINQUENT GROUPS STUDIED 25 
 
 treatment as the rest of the material. Whenever mentioned they will be 
 clearly referred to as "Workhouse Intoxication Group." In consulting 
 the tables it is important to remember that the so-called "Workhouse 
 Group" does not include any representatives of the large portion of 
 the Workhouse population who have been committed for intoxication. 
 
 (/) Probation cases from the Women's Night Court of Manhattan 
 
 and the Bronx 
 
 Our probation group is made up of women placed on probation 
 through the Women's Night Court of Manhattan and the Bronx. 
 The original plan was to secure a total of 100 probation cases made 
 up of women sentenced through the three types of courts in New 
 York City: viz., the City Magistrates' Courts, the Courts of Special 
 Sessions and the Courts of General Sessions. Since the probation 
 cases from the Night Court are the largest and most important 
 group placed on probation through the City Magistrates' Courts, it 
 was decided to consider these representative of the whole group 
 of women probationers from the Magistrates' Courts. 
 
 We chose this as the probation group with which to begin our 
 study, considering it entirely problematical whether such an investiga- 
 tion was practicable under the conditions of probation. After we 
 had established the fact that such a study could be made to advantage 
 from our point of view and yet without disturbing the friendly re- 
 lations existing between the probationer and her probation officer, we 
 asked permission to begin a study of a smaller group from the Court 
 of Special Sessions. This request was not granted. Accordingly it 
 was decided to restrict our study to the group of probationers from 
 the Night Court with whom its practicability had been demonstrated. 
 We regretted the necessity of omitting the other two probation groups, 
 since it prevented our total probation group from including all types 
 of offenders who are put on probation." 
 
 "We wish to express our gratitude to Mr. Edwin J. Cooley, Chief Pro- 
 bation Officer, City Magistrates' Courts, for his ready interest in the carrying 
 out of this study and his willingness that it be undertaken. To Miss Alice 
 Smith, Probation Officer of the Night Court, we are indebted for constant and 
 detailed cooperation which was mainly responsible for making the work prac- 
 ticable. She devised an arrangement which operated, we feel sure, to the 
 advantage of the probation officer and the woman under consideration as well 
 as to that of the Laboratory study. According to this plan women who seemed 
 likely to be selected for probation were assigned to members of the Laboratory 
 staff for investigation, just as, at other times, they are assigned to various 
 volunteer organizations who assist the probation officer. In this way we had 
 a perfectly natural avenue of approach to the women, friendly relations were 
 established, and it was easy to keep in touch with them and complete our in- 
 
26 STUDY OF WOMEN DELINQUENTS" IN NEW YORK 
 
 The women placed on probation through the Night Court may be 
 of any age over sixteen years, though the opportunity for probation is 
 naturally given more frequently to the younger women. By virtue of 
 the selection which the court itself exercises, they fall into the general 
 groups of offenders against chastity, intoxication cases, and various 
 forms of disorderly conduct and of incorrigibility. All are mis- 
 demeanants, but the group does not include all types of misdemeanants, 
 since there are certain of these, such as offenders against property, 
 drug users, etc., who cannot be convicted through the Magistrates' 
 Courts. The group tends to be made up of first offenders or of 
 cases who offer special promise of making good, either because' of 
 their own personal qualities or because of unusually satisfactory 
 conditions in which they can be placed. There is a general tendency, 
 quite definitely stated as .such, to give all first offenders an oppor- 
 tunity on probation, unless there are very clear reasons why they 
 need the supervision and protection of an institution. 
 
 Cases were taken in consecutive order as they were placed on 
 
 probation by the court, our actual procedure being to have at least 
 
 one member of the Laboratory staff in court each night, ready to 
 
 interview each woman referred by the judge to the probation officer 
 
 as a probable case for probation. There were 102 cases admitted 
 
 to probation between the beginning of the study in March and its 
 
 close in July, 1917. Of these, eleven were omitted almost completely 
 
 from our study, only the information which we received through 
 
 the probation officers' record, covering the woman's present offense, 
 
 her criminal record, and a few other scattering bits of information 
 
 being available. These omissions were due to a variety of reasons, 
 
 such that we feel satisfied that no single type of cases was being 
 
 selected out.^^ The only exception to this is the omission of the 
 
 vestigations after they had been placed on probation. Report of our prelimi- 
 nary investigations in each case was made to Miss Smith, making it unnecessary 
 that this work be duplicated. In many cases it was possible for us to make 
 a more complete investigation than would have been possible for the proba- 
 tion officer with the many demands upon her time for supervision of probation- 
 ers as well as for investigation. 
 
 " The reasons for these omissions were as follows : Four were sent out of the 
 city to relatives, either the same night that they were placed on probation or 
 so shortly after this that no investigation was practicable; four were older 
 women whom we omitted intentionally since we felt that we should have too 
 great difficulty in_ securing their cooperation ; one case was committed to Belle- 
 vue for observation the day after her admission to probation, and was later 
 committed to a hospital for the insane; two were omitted because it happened 
 that their initial interview was taken entirely by the probation officer and it 
 seemed unwise in their cases to transfer any of the work of investigation to an- 
 other person. 
 
ACCOUNT OF DELINQUENT GROUPS STUDIED 27 
 
 four older women. Because of this omission our figures fail to show 
 this small, but possibly fairly constant, proportion of more mature 
 women in this group. We should add that, in addition to these eleven 
 complete omissions, there are a considerable number of cases on 
 whom our study is incomplete in some direction. Several of the cases, 
 in order to be included in the series at all, had to have all the work 
 that was to be done, including the testing, completed within one 
 or two days, since they were being sent to parts of the country too 
 far away for visits to be practicable. 
 
 GROUPS OF WOMEN OF THIS STATE WHO ARE NOT INCLUDED IN PRESENT 
 
 STUDY 
 
 In the foregoing statement we have shown what groups of de- 
 linquent women were included in our study. It is doubtless important 
 to indicate also what groups were entirely omitted. We have al- 
 ready mentioned the groups, closely similar to those selected for 
 our study, which were omitted to prevent duplication. These omis- 
 sions were the following: Of the two reformatories, the Western 
 House of Refuge at Albion ; of the private institutions, eight or more 
 institutions of the state listed on page 19 ; of the County Penitentiaries, 
 those for Albany, Erie, Monroe and Onondaga Counties. There were 
 no omissions of Workhouses so-called, though we did omit two por- 
 tions of the Workhouse total, viz., the very short term commitments 
 and the intoxication cases. For the rest of the state, outside of New 
 York City, county jails doubtless come nearer than anything else to 
 serving the function of the Workhouse. These it was necessary to 
 omit entirely. They probably add no cases different in type from 
 those committed to the Workhouse, though they might materially 
 alter the proportions of different types. The other important omis- 
 sion from the institutional groups is that of the State Farm for 
 Women at Valatie.^® This institution receives women over 30 years 
 of age who have had at least five previous convictions within two 
 years. The commitments are largely for intoxication, so that this 
 group corresponds closely with that of the alcoholics omitted from 
 the Workhouse study. 
 
 In addition to these omissions from the institutional groups there 
 
 are certain obvious omissions from the total number of delinquents, 
 
 from among those who are not institutional cases. These are: (1) 
 
 The cases fined. The numbers of these are large and cover a great 
 
 "This institution has been closed since the above account was written. 
 
28 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 variety of offenses, though fines cannot be imposed for the most 
 grave offenses, and are not given at present in New York City for 
 offenses involving prostitution. Fines may be given to first offenders 
 or, for certain offenses, notably intoxication and disorderly conduct, 
 to very old offenders. (2) Cases given suspended sentence. This 
 form of treatment also is very common and is applied to a wide range 
 of offenses. It is given mainly in the case of either first or nearly 
 first offenders. In the various court reports these cases are not 
 always clearly distinguished from the group of those placed on pro- 
 bation. (3) Cases placed on probation through Courts of General 
 and Special Sessions, through the Supreme Court and County Courts, 
 and through local police courts. The one group of probation cases 
 which we secured from the Women's Night Court of New York City 
 cannot be considered as representative of all groups. 
 
 It should be noted that when we speak of these various omissions 
 we are speaking of present convictions only. If we consider the 
 total range of sentences which our women have had in the course of 
 their careers it is probable that all types mentioned above have some 
 representative in our total. 
 
 From a consideration of all our groups and from a certain 
 amount of observation of courts and of other groups we feel that 
 we have, among our selected total, representatives of all important 
 types of women delinquents of this state, with the single exception 
 of those cases committed for intoxication, of whom we have an 
 inadequate representation. Aside from this, we have women com- 
 mitted for all types of offenses, through all types of courts, given 
 sentences varying from probation to a term of twenty years or life. 
 There are first offenders and recidivists. There are women com- 
 mitted for prostitution in some form, women who are prostitutes 
 but who are committed for some other offense, and women who have, 
 so far as we can learn, never been sexually irregular. There are 
 women whose court careers began in childhood and women whose 
 first offense came late in life. There are women of widely different 
 ages and from many social classes. We feel convinced, therefore, 
 that all the usual types of women delinquents are included in our study, 
 although we cannot claim that the several types are present in correct 
 proportions. 
 
 We may note further, among the omissions from this study, the 
 women offenders, who are known to exist in appreciable numbers, 
 who escape conviction altogether. How large their numbers may be 
 
ACCOUNT OF DELINQUENT GROUPS STUDIED 29 
 
 and what their dominant characteristics we have no means of 
 determining. That they represent, on the whole, the more efficient 
 members of the delinquent group seems self-evident, since they manage 
 to pursue their chosen careers without interruption at the hands of 
 the law. This may be due to the fact that they evade detection more 
 cleverly, or that they secure protection more effectively, or that, 
 even if arrested, they marshal their forces more skilfully to escape 
 conviction. 
 
 REASONS FOR PRESENTING DATA CLASSIFIED BY INSTITUTIONAL GROUPS 
 
 The question may be raised, on considering the data offered in the 
 chapters which follow, as to why prominence is given to the classifica- 
 tion by institutional groups, when these may be largely arbitrarily 
 determined and are local in their interest. Two reasons may be offered, 
 of which the first is the more, important. 
 
 (1) The separate institutional groups are the actual concrete 
 units with which we started. Selection within any given group is 
 as nearly random as it could possibly be, whereas the make-up of 
 the total was necessarily determined by the relative numbers in 
 the various sub-groups, as well as by the choice of sub-groups. With 
 the material presented in this way it is possible for other investigators 
 to re-group the data in other ways in accordance with their special 
 interests. For example, probation cases might be disregarded and 
 only institutional commitments considered, institutions of the re- 
 formatory type might be thrown together, or still other special 
 combinations made. 
 
 (2) Some interest attaches, locally at least, to the attempt to 
 discover what principles of selection were effective in determining 
 commitments at the time when this investigation was made. The 
 answer to this question may well contribute to the solution of the 
 further question as to whether the present system of commitment 
 is adequate, or whether there is need for some additional device, 
 such as that offered by a clearing-house, for more careful study of 
 women offenders before commitment. 
 
 The interest, from this point of view, need not be exclusively 
 local, since other communities contain institutions corresponding in 
 type with those here considered, though any one community may 
 fail to contain all types. With this in mind we might well have 
 listed the groups by type names, discarding the more specific designa- 
 tions, using some such terms as these: (a) State Reformatory, (b) 
 
30 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 State Prison, (c) Private Reformatory Institution, (d) County Peni- 
 tentiary, (e) City Workhouse, (f) Probation Group (Sex Offenders). 
 Additional data from similar groups in other parts of the country 
 should make it possible eventually to determine whether any uniform 
 tendencies are operative affecting the selection of cases for particular- 
 types of institutions, or whether selection is entirely determined by 
 changing local conditions. If the latter should prove to be the case 
 no permanent significance can be attached to the comparison of the 
 local groups. 
 
CHAPTER III 
 STATISTICAL METHODS 
 
 THIS investigation attempts to discover the extent to which cer- 
 tain social, economic and psychological factors are associated in 
 a delinquent group. If relationships are to be found whose significance 
 is general, it is clear that a general group must be studied. Par- 
 ticular cases, dramatic though they may be, cannot give justification 
 in themselves for action affecting the group as a whole. General 
 relationships can be established only through the study of a sample 
 selected at random from the continual flow of women into the 
 prisons and reformatories. 
 
 Such a truly representative group of delinquent women can only 
 be approximated. We have selected our cases by studying consecutive 
 commitments to a number of typical institutions. Using consecutive 
 commitments seemed the surest way to obtain a sampling of the 
 general group of delinquent women that would be free from the 
 influence of arbitrary selection whether conscious or unconscious. 
 The presence of an individual in our group is due to no other fact 
 than that she entered an institution during the period of investigation. 
 
 The concepts underlying a study of the facts and relationships 
 characteristic of such a representative group are fairly simple. In 
 the first place, the various qualities of the group observed must be 
 described. We waijt to know something of the intelligence of the 
 group, the amount of recidivism, the economic condition prior to 
 conviction. In the second place, relationships must be pointed out 
 where they exist. We are , interested both in comparisons between 
 groups of individuals, as for example in the intelligence of delinquent 
 women as compared with the general population, and in relationships 
 between the factors we are studying, as between intelligence and 
 number of commitments or between age at first conviction and type 
 of offense. In the third place, there must be some statement of 
 what reservations must be made in extending generalizations valid for 
 the particular group under observation to the larger group of which 
 the cases studied are only a chance sample. 
 
 31 
 
32 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 These three concepts include all that is involved in the statistical 
 treatment of our material. The concepts themselves are not statistical. 
 They are too matter of fact for that. The statistics are merely 
 a method of realizing as far as possible the scientific ideal embodied 
 in the concepts. 
 
 THE DESCRIPTION OF THE QUALITIES OF A GROUP 
 
 In describing a quality or factor, intelligence for example, as 
 it occurs in the group of women which we have studied, we shall 
 ordinarily state two things about it. First, we shall giye a figure 
 which indicates a center about which the measurements tend to cluster, 
 an average of the group in the quality. In some cases we shall give 
 the arithmetic mean, in other cases the median, sometimes perhaps 
 the mode. Each of these statistical terms has a special meaning 
 of its own; but the function of all is to indicate a somewhat central 
 point of reference for the group. Second, we shall give a figure 
 which indicates the amount the individual cases are scattered about 
 the average or central tendency. The scatter or dispersion is a most 
 important consideration, since this shows the diversity or variability 
 of the group. Sometimes groups that are of the 'same average intelli- 
 gence differ greatly in variability, and, of course, when the variability 
 is high, extreme cases are much more likely to be found. Variability 
 or the dispersion of a group is indicated by statistical constants, some- 
 times by the average deviation, more frequently by the standard 
 deviation (o"). These two constants have each a special significance, 
 but for most purposes of interpretation, they may be thought of as 
 meaning the same thing. Roughly, about two-thirds of the cases 
 measured will be found within the distance of the standard deviation 
 from the central tendency. Thus, if the average intelligence of the 
 Bedford women is 80 points and the standard deviation is 6 points, 
 about two-thirds of them will normally fall between 74 and 86, 
 that is within 6 points on either side of 80. Cases as far removed from 
 the average as two times the standard deviation rarely occur, and 
 cases at a distance of three times this constant are most exceptional. 
 In the illustration just imagined, very few women indeed would 
 be found to score as low as 62 or as high as 98. 
 
 Thus the average or central tendency and the standard deviation 
 or index of variability are the statistical constants we shall most 
 need to describe a quality or trait of our groups. Both are important 
 in judging the nature of any quality. 
 
STATISTICAL METHODS 33 
 
 COMPARISONS AND RELATIONSHIPS 
 
 After a. quality of one group of individuals has been described, 
 it is often desirable to compare it with the same quality of another 
 group. In this way, we may discover that the average intelligence 
 of women sent to Bedford is lower than that of the general popula- 
 tion. When we compare standard deviations, we may find that the 
 variability of the Bedford group in intelligence is not so great as 
 that of the general population. 
 
 These comparisons are easily made, the one precaution necessary 
 being to make certain that the quality or trait is being observed 
 under like circumstances for the two groups. 
 
 Relationships are only more extended comparisons. We may 
 find by comparison that the Workhouse women are lower in in- 
 telligence than Bedford women; and again that Bedford women are 
 lower than Penitentiary; and Penitentiary lower than Magdalen. If 
 we put all three comparisons together, we have clearly a relationship 
 between type of institution and intelligence. 
 
 The interpretation of relationships is a practical matter. It is 
 through relationships that the many tangled threads of social causes 
 can be partially straightened out. (^'Relationships can be made the 
 basis of prediction and of action; and the closer the relationship, 
 the more definite the prediction and the surer the action. For this 
 reason, we require a method of exjpressing closeness of relation. 
 
 The common ways of stating degree of relationship are the 
 correlation coefficient, the correlation ratio and the contingency co- 
 efficient. All are misleading in that they give an unwarranted 
 impression of the possibility of accurate prediction. These coeffi- 
 cients vary from to 1, for a relation such as would occur in 
 the long run by chance, 1 for a perfect agreement. The coefficients 
 are usually expressed as decimals, as .18, .37, .53. The reader should 
 be warned that a coefficient of .50 by no means indicates that 50 
 per cent of the error of prediction has been removed. 
 
 The correlation coefficient and correlation ratio refer to different 
 types of mathematical relationships between the variables related. 
 If there is a large difference between these constants, it means 
 that the methods of prediction must be studied mathematically with 
 great care if the best is to be found. For interpretative purposes 
 in this investigation, it is usually not necessary to distinguish be- 
 tween the correlation ratio and correlation coefficient. Usually both 
 
34 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 are relatively low, the number of cases is not large, and under these 
 circumstances the differences between these two measures of relation- 
 ship are less significant. 
 
 The partial correlation coefficient is occasionally used in this in- 
 vestigation. This coefficient varies from to 1 as does the correlation 
 coefficient proper. It is used to express the relation between two 
 variables for constant values of a third (or more) which cannot be 
 experimentally controlled. Care should be taken to ascertain exactly 
 what this third variable is in every case, and to appreciate just what 
 it would mean to have all individuals equal in this variable. The 
 meaning of the partial relationship of the two variables in question 
 can be more easily understood in this way. 
 
 STATISTICAL RESERVATIONS 
 
 When we find the average intelligence of a group which we 
 measure to be 80 points, we want to be able to make some state- 
 ment about the intelligence of the larger group of which the in- 
 dividuals we actually did observe are only a small, chance sample. 
 How accurate a measure of the total group's intelligence is 80 ? What 
 reservations must be made? 
 
 The amount of reservation is stated as plus or minus a certain 
 amount. That is, the true value may be more or less than that 
 observed. Thus, the average intelligence of the total group might be 
 stated as 80 i 2. The 2 is really an index of reservation, — it is 
 a sort of standard deviation of the average. 
 
 The ± 2 is computed mathematically, — its purpose is to give 
 us a definite point to start from. The amount of reser-\'ation we 
 make depends on how cautious we must be. With intelligence stated 
 as 80 ± 2, the chances are somewhat better than 2 to 1 that the 
 average of the whole group, judging from the measurements on our 
 sampling of individuals, is not less than 78 nor more than 82. The 
 chances are about 20 to 1 that it lies between 76 and 84; they are 
 more than 300 to 1 that it lies between 74 and 86. Thus it is fairly un- 
 likely that the true constant lies outside a range of twice the ± 2, and it 
 is almost certain not to lie outside a range of three times this figure. 
 
 An observed difference between two averages also must be stated 
 with reservations. The difference between Bedford women and women 
 in general in intelligence might be 20 ± 4, in favor of the general group. 
 ± 4 is our starting point, and we can qualify our statements as much 
 as we like. We are reasonably sure in this example that women in 
 
STATISTICAL METHODS 35 
 
 general excel Bedford women by more than 12 points, and we are 
 practically certain that the difference is more than 8, these figures 12 
 and 8 being respectiv.ely 2 and 3 times the ± 4 from 20 points — the 
 difference observed. We are sure that the difference in intelligence is 
 not 0, and that the Bedford women do not excel. 
 
 Correlations are also stated with reservations, as .32 ± .05. Is 
 there any chance that if we had measured all possible cases instead of 
 only our sample, we would have found a zero relationship? Prac- 
 tically none, for this correlation coefficient .32 is more than 6 times 
 the index of reservation. 
 
 When we are studying the differences between averages or be- 
 tween standard deviations or between correlation coefficients, we 
 always find out how many times larger the observed difference is 
 than the index of reservation. We compute the ratio of the difference 
 to this index. If the. ratio is 2 we are fairly sure the difference is 
 really in the direction we observed it, if it is 2.5 we are very confident, 
 and if it is 3 we are practically certain. No arbitrary ratio can 
 be set as indicating absolute certainty, for after all certainty is at 
 best a matter of probabilities. 
 
 TECHNICAL NOTES 
 
 1. The (7 of statistical constants is used throughout instead of 
 the P. E. There appears no virtue in the further reduction to 
 P. E. since a tables are now available. 
 
 2. In the histograms, showing distribution of groups according 
 to various categories, the mean and a have been mdicated, respectively, 
 by a dot, whose location indicates the position of the mean, and arrows 
 leading from the dot, thus -< — • — ^~. The distance from the 
 dot to the tip of either arrow equals <r. The entire figure illustrated 
 above, therefore, represents 2cr and indicates graphically the range 
 of variation from the mean implied in the expression ±(r. 
 
 3. The probabilities favoring the sign of an observed difference 
 are twice as great as those favoring inclusion within ±0-, due to the 
 fact that only one-half the outlying probabilities are critical for the 
 argument. 
 
 4. Testing by Blakeman's criterion was done extensively, but little 
 light came from it. Where inexcusable non-linearity was found 
 by test, the regression was frequently excessively irregular, with 
 no indication of the form of the relation, while a few cases of ap- 
 preciable non-linear form tested linear. 
 
CHAPTER IV 
 METHODS OF PSYCHOLOGICAL EXAMINATION 
 
 METHOD OF ESTIMATION VERSUS METHOD OF EXAMINATION 
 
 IN studying the mental capacity of a given group there are two 
 methods possible: the method of estimation and the method of 
 examination. 
 
 The first method figures mainly where the groups studied are very 
 large or so conditioned that individual examinations are not prac- 
 ticable. This is the method used by Goring^ in his study of 3000 
 English convicts, by Lombroso and the older penologists, and in most 
 of the "surveys" of large districts. 
 
 Its necessity is evident under certain conditions where the groups 
 are too large or the time too short to make individual examining 
 feasible — although the system of group testing employed in the Army 
 has opened a way to the solution of many of these difficulties. There 
 are certain other conditions which render examining impracticable, as 
 for example, where one is dealing with a group over which one has 
 no control, as in the house-to-house canvass method of many of the 
 surveys, or the grading of large industrial groups where the novelty 
 and seeming tendency of the examination method toward paternalism 
 may, for a time yet, force us to have recourse to the older method of 
 mere estimation. But on the whole this method is fast being sup- 
 planted by the more exact one. 
 
 The method of the intelligence examination is comparatively recent 
 in development but not so recent but that one is surprised to find in 
 /^so important a work as Goring's no mention made of it. He, in fact, 
 specifically deplores the lack of instruments which would allow one 
 to give to mental measurements the same degree of exactness that the 
 tape and the callipers afford for physical, but does not even suggest 
 that anything is being done toward the realization of such an object. 
 More clearly, perhaps, than any one else, he recognizes the pitfalls of 
 
 ' Op. cit. 
 
 36 
 
METHODS OF PSYCHOLOGICAL EXAMINATION 37 
 
 the method which he is employing and makes every effort to safe- 
 guard himself. He says: 
 
 "Estimates of mental qualities can be made: and every day the world, 
 colloquially, does make them, with a more or less broad degree of ac- 
 curacy. Many judgments of the kind, truly, are quite valueless; for, 
 in so many cases, distinctions of mental and moral characters are 
 animated by personal feeling. Yet, because opinions, biassed by feelings 
 of generosity or malice, must be ignored, that is no reason for disregard- 
 ing the judgments of a just critic on the grounds that mental charac- 
 ters are beyond the range of legitimate observation. Personal estimations 
 of both mental and physical attributes, if carefully made and recorded 
 by an unbiassed and disinterested investigator, whose personal equation 
 can be estimated and allowed for, represent evidence of substantially the 
 same character, and of equal value scientifically, as that produced by 
 measurement. And it is precisely for this reason — because, starting as 
 they did from preconceived theories, we cannot be certain that crimino- 
 logical investigators in the past were not prejudiced, that their observa- 
 tions upon criminals were not warped by the desire to fit fact to theory — 
 it is precisely for this reason, and not because we discredit the human 
 power of estimating broad differences of degree or magnitude without 
 the aid of callipers and tape, that we asserted at the commencement of 
 the present work that criminal anthropology must be built upon measure- 
 ments, and upon measurements only. Measurements are relatively free 
 from the personal equation of the investigator; they record the actual 
 dimensions of an object, and not its apparent size, as viewed by a biassed 
 observer : and, consequently, the atmosphere of disquieting doubt is 
 absent from generalization or theory based upon measurements. But 
 coming now to mental characters, we do not say that, because non- 
 measurable, these therefore cannot be observed; we do, however, contend 
 that broad generalizations upon the mental characters of the criminal 
 must leave the critical mind cold and unconvinced until they have been 
 based upon facts, more unquestionably accurate, and more concise in state- 
 ment, than those general impressions of the truth which have been 
 presented in the past by the leading exponents of criminal anthropology."^ 
 
 Turning now to the other method, that of examination, it seems 
 inconsistent at the first glance to insist that the method of estimation 
 is less satisfactory, since we are compelled to admit that the cri- 
 terion of value for a mental test is in the last analysis its agreement 
 with the judgments of competent observers. It may, therefore, seem 
 illogical to argue for the supplanting of the method of estimation 
 by the establishment of a system of mental examinations, but its justi- 
 " Op. cit., page 237. 
 
38 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 fication rests on two facts. First, it enables one to obtain in a brief 
 period information which it would take many times as long to obtain 
 by the other method ; and, second, it possesses a quality of uniformity 
 very diificult to obtain by estimates. Even if one could assume the 
 judgment of the observer to be entirely intelligent and free from any 
 emotional bias it would still happen, in all probability, that each 
 observer would tend to make his judgment in terms of the persons 
 with whom he has been in contact. That is, the college professor might 
 judge as "stupid" such of his students as fell markedly below the 
 others in ability, while the manager of a group of unskilled laborers 
 might rank as "bright" individuals who excelled their fellows but who 
 actually fell below the mental level of even the most stupid of the 
 other group. Such a scale. as that of Pearson, to be described later, 
 aims to reduce such discrepancies by presenting an absolute scale on 
 which judgments should be made. Adequate evidence has not been 
 offered, however, to show that judgments based even on such a scale 
 are free from the influence of personal bias of one sort or another. 
 
 The method of examination, therefore, while it calls for the serv- 
 ices of a special examiner and the time necessary to make the exami- 
 nation, and requires as well a certain amount of cooperation from the 
 individuals to be tested, outranks the older method in that it enables 
 a judgment of mentality to be made within a few days which it would 
 take weeks or months to reach by the judgment method, and in that 
 it provides a uniform and impersonal grading extremely difficult and 
 in many instances impossible to obtain in any other way. It thus 
 makes possible comparison of groups of very different mental levels. 
 
 FACTORS AFFECTING CHOICE OF TESTS 
 
 The selection of a series of psychological tests which will satisfac- 
 torily diagnose the mentality of a particular, group of individuals is, 
 however, even at the present time, a condition which is still devoutly 
 to be sought. The work that has been done toward establishing norms 
 for one test or another has been carried on mostly on some specific 
 group and is applicable mainly to that group. Due to the fact that 
 this problem has been in the past mainly of interest to the schoolmen, 
 we find that the "subjects" for their studies have been largely recruited 
 from the ranks of college students, or grade school children. But with 
 the recognition of the importance of such examinations there is fast 
 piling up an accumulation of data in other fields, less scholastic, which 
 will furnish a more nearly unselected group for comparison. 
 
METHODS OF PSYCHOLOGICAL EXAMINATION 39 
 
 There are two important aspects of the field of mental testing: 
 first, the diagnosing of the mentality of a developing child and de- 
 termining his mental ability as proportionate to his physical age, and 
 second, the estimating of the mental capacity of an individual from a 
 particular group with reference to his rank in that group. This lat- 
 ter problem will, of course, remain unfinished until the time when 
 each specific group can itself be compared with figures for the general 
 population, since any individual may be thought of as belonging to an 
 indefinite number of groups of which the most comprehensive is the 
 whole population. As we have earlier stated, ours is a group which 
 may fairly be assumed to be mentally adult so that the first problem 
 is not properly applicable here. The "mental ages," used in express- 
 ing the results of the Stanford-Binet, we think of as merely a quan- 
 titative expression of the score. 
 
 Recognizing that ours is clearly an aberrant group, we have aimed, 
 in so far as possible, to employ only tests which have been previously 
 standardized with more nearly unselected subjects. It is because of 
 their wide application to a relatively heterogeneous population that 
 we have given so much weight to the tests of the Binet type and those 
 of the Woolley series. In view of the eccentricity of our group and 
 our lack of facilities for establishing norms we have ourselves devised 
 no new tests. 
 
 ESTABLISHMENT OF METHODS ON BEDFORD GROUP 
 
 The data on which our establishment of measures of the mental 
 status of these individuals was mainly based were obtained from a 
 more intensive study on a large group of women committed to the 
 Bedford Reformatory between March, 1915, and August, 1917. This 
 group includes the 102 cases which we have called the "Contemporary 
 Bedford Group" which were studied during the year 1916-1917 and 
 used in comparison with our other prison groups. The number is 
 raised to 343 by the addition of 241 cases which were committed to 
 Bedford between the dates of March, 1915, and September, 1916, 
 They represent a practically consecutive series of commitments be- 
 tween the dates mentioned. (The actual number of commitments dur- 
 ing this time was 352.) 
 
 Conditions of Testing 
 
 The conditions for testing this group were exceptionally good. The 
 women on their arrival at Bedford are taken directly to the Reception 
 
40 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 Cottage where they are held in strict quarantine for a minimum of two 
 weeks. During this time each woman is locked in a separate room, 
 and is allowed no communication with the other inmates. This re- 
 sulted very happily for our work, as it not only cut out the factor of 
 one subject coaching another on the problems she was to be given, 
 but also enabled us to deal with each individual alone, without the 
 background of her position among her fellows. One is tempted to 
 emphasize this point after working with a group where the subjects 
 were taken directly from the general prison population. We have one 
 case in mind which had to be dropped from our series because an initial 
 "grouchiness" was augmented to such a degree by her desire to "save 
 her face" before her mates that it developed into so determined an 
 obstinacy as to make the results of our testing worthless. 
 
 The first contact between the Bedford girl and the Laboratory was 
 made by the sociologist who established friendly relations by showing 
 a sympathetic interest in her case and a willingness to do whatever she 
 could to help her to adjust herself to the new conditions. After this, 
 one of the psychologists brought the girl to the Laboratory for a series 
 of testings. These series generally covered a period of five or ten 
 days, depending on whether the longer or shorter series was used. 
 
 The physical conditions of testing were excellent. The Laboratory, 
 being located in the country, is spared all the street noises ordinarily 
 unavoidable. We have allowed no visitors to be present during the 
 testing and have never attempted to use the examination as anything 
 like a clinic. 
 
 The examiners always aimed to maintain a cordial and friendly 
 attitude. Introducing the -"play element" so much recommended in 
 dealing with children is of course quite out of place here. We en- 
 deavored particularly to dispel the bugaboo of a "brain test," and of 
 ourselves as "the doctors that send you to the crazy house." 
 
 We tried in general to get our subjects into an attitude that was 
 friendly, responsive and eager, without being hampered by self-con- 
 sciousness or the nervousness which results from an over-anxiety 
 to do well. We benefited indirectly by the quarantine period in that 
 by the time a girl had been alone in her room for a few days she quite 
 enjoyed her visit to the Laboratory and welcomed the labor of work- 
 ing at tests for an hour in preference to the tedium of sitting idle in 
 her room. If the examination was interrupted or any serious error in 
 administering a test occurred the results were dropped from the series. 
 
 On the whole we feel satisfied that the showing which these women 
 
METHODS OF PSYCHOLOGICAL EXAMINATION 41 
 
 made on the tests given them represents pretty closely what they are 
 capable of doing under optimal conditions. Attempts to malinger or 
 try to appear stupid or erratic on the tests have not made any serious 
 trouble for us. In the possible two instances where it was tried, it 
 was very clearly an attempt to avoid taking the tests without openly 
 refusing, rather than an overt effort to deceive. There is of course 
 no real incentive to malingering in our problems. We encountered 
 one psychopathic girl who was a "practical joker," to whom we doffed 
 our hats. She impressed upon us the conviction that once a subject's 
 sense of humor is allowed to run rampant, the end of the science of 
 mental testing is in sight. 
 
 Standardisation of Methods of Giving and of Scoring Tests 
 
 In the giving of tests throughout our work we aimed at a close 
 uniformity of method. To this end the instructions for giving each 
 test were written out in detail and committed to memory by each 
 one of the examiners. In this way we endeavored to make certain 
 that the personal equation which enters into every mental examination 
 was reduced to a point where its effects were negligible. 
 
 The methods of scoring the results were likewise worked out in 
 most complete detail and reduced to writing. Each examiner marked 
 the results on this basis. During the examination the subject's re- ' 
 sponse was taken down verbatim on blanks specially prepared for 
 each test. This slowed down the progress of the examination some- 
 what, but we felt that the gain in accuracy obtained thereby justified it, 
 since this enabled a second examiner to check from the actual data 
 the marking of the first. We feel that this method gives results fully 
 as uniform as Terman's method of having all the evaluating done by 
 one person. 
 
 Preliminary Intensive Test Series 
 
 Recognizing that in selecting a series of tests for final use, it was 
 an easier task to prune down from an over-extensive series than to 
 build up from one too meager, we started our work with a very in- 
 tensive study of 120 successive commitments. This study covered a 
 period of at least ten days' individual testing— each day's period 
 being an hour in length — and approximately three days of group 
 testing. The limits of the present report force us merely to list 
 here the tests used and reserve any description of them or of the 
 results obtained thereon for later publication. 
 
42 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 Wherever it was practicable to do so we attempted to gage the 
 reliability, of a test by giving two series of the same test, e.g., two lists, 
 of opposites, two forms of canceling unfamiliar figures. 
 
 We made it a basic principle that in giving a test we should, as far 
 as possible, make sure that the subject understands what she is 
 expected to do. To this end we made extensive use of exa,mples 
 and sample sheets. It has been contended that this capacity to 
 understand quickly is in itself a part of what is being tested, but 
 such arguments seem to us to cloud the issue, because in that case 
 one would be consistently overweighting this factor by complicating 
 every test with its presence. It is quite true that in some tests, like 
 the Capacity to Follow Directions, a stupid subject often loses track 
 of what he is expected to do and finishes the test, so to speak, ad lib., 
 and that this is in itself an essential part of the test. But we feel 
 that this is an entirely different performance from that of a subject 
 who is entirely at sea from the start and who, in consequence, confuses 
 his record by demanding more instructions, or, becoming discouraged 
 at his inability to understand, gives up the test without a real effort. 
 
 To speak first of the tests of the Binet type — ^we had been quite 
 puzzled at the conflicting figures obtained therewith by various ex- 
 aminers and on closer examination were equally appalled at the 
 variations in the technique of giving and method of scoring them. 
 We were finally driven to the necessity of reducing to tabular form 
 the specific instructions for giving and evaluating these tests, as formu- 
 lated by the principal exponents of the Binet system in this country. 
 After we had finished this time-consuming task we became convinced 
 that a problem of considerable interest would be a comparison of the 
 same individuals when tested by the original Binet method and by the 
 two chief modifications of this, viz., the Yerkes-Bridges Point Scale 
 and the Stanford-Binet; We accordingly drew up a series of instruc- 
 tions that enabled us to score an individual by each of the three 
 methods without repeating separate tests where these were identical 
 or nearly so. The process will be described in detail in a later paper. 
 
 Dr. Weidensall,^ formerly psychologist at this Laboratory, has 
 made an extensive study of the results of testing 88 Bedford women 
 by the scale devised by Dr. Woolley* as compared with the results 
 obtained by Dr. Woolley in her study of the working children of 
 
 'Op. cit. 
 
 * Woolley, Helen Thompson, and Fischer, Charlotte R., "Mental and Physical 
 Measurements of Working Children." Psychol. Monog., vol. 18, no. 1, 1914. 
 
METHODS OF PSyCHOLOGICAL EXAMINATION 43 
 
 Cincinnati. We regard Dr. WooUey's norms as the best data avail- 
 able for comparison with our women, and we accordingly dupli- 
 cated Dr. Weidensall's work, in that we used practically all of the 
 tests reported by Dr. Woolley in her paper of 1914. It had been 
 our intention to make a comparison of our results with those of 
 Dr. Woolley by using the same method of treatment, i.e., the method 
 of percentile groupings, but since one such report has already come 
 out from this Laboratory it seemed wiser to postpone such a study. 
 We list briefly the tests which were given in our intensive series : 
 
 I. Test Series Used.' 
 
 1. Binet-Simon Scale 1911 (both Binet (Town translation) and God- 
 dard forms). 
 
 2. Yerkes-Bridges Point Scale. 
 
 3. Stanford-Binet Scale. 
 
 4. Woolley Series. 
 
 a. Sentence completion. 
 
 b. Association by opposites. 
 
 c. Cancelation of a's. 
 
 d. Memory for digits after the auditory-visual-articulatory 
 
 method of presentation. 
 
 e. Substitution or symbol digit test. 
 
 f. Strength of grip. 
 
 g. Steadiness of hand. 
 
 h. Rapidity of hand movement. 
 i. Card sorting. 
 j. Visual and auditory acuity. 
 Height and weight records were taken from the measurements made 
 by the Reformatory physician. Tests of vital capacity were not made. 
 
 II. Individual Tests Used. 
 
 The remaining tests have been drawn from many sources. Some are 
 standard tests of very wide application and others are quite recent and 
 have been in less general use. 
 
 "For account of the procedures used in these series see the following ref- 
 erences : 
 
 Binet, A. and Simon, Th., "A Method of Measuring the Development of 
 the Intelligence of Young Children." Translated by Town, C. H. 
 
 Goddard, H. H. "The Binet-Simon Measuring Scale for Intelligence.'' Re- 
 vised. The Training School, Vol. VIII, No. 4, June, 1911. 
 
 Goddard, H. H. "Standard Method for Giving the Binet Test." Training 
 School, Vol. X, No. 2, 1913. 
 
 Yerkes, R. M., Bridges, J. W., and Hardwick, R. "A Point Scale for 
 Measuring Mental Ability." 1915. 
 
 Terman, L. M. "The Measurement of Intelligence." 1916. 
 
 Woolley, H. T. and Fischer, C. R. Op. cit. 
 
44 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 1. Memory for digits after the visual method of presentation. 
 
 2. Memory for digits after the auditory method of presentation. 
 Same type of number series used for 1 and 2 as for auditory-visual-articu- 
 latory presentation (Woolley series). 
 
 3. Memory for ideas of a logical passage visually presented (Healy).« 
 
 4. Memory for ideas of a logical passage auditorily presented 
 (Healy).6 
 
 5. Memory for objects in a series (Ellis— unpublished data). 
 
 6. Knox cubes. Memory for movement in a serial order. Both 
 the original Knox series and the series devised by Pintner were used.^ 
 
 7. "Aussage" tests — suggestibility and fidelity of report. Two pic- 
 tures used— "The Disputed Case" and "The Doctor.''^ 
 
 8. Free association test. Kent-Rosanoff list.^ 
 
 9. Controlled association — action-agent. Two lists taken from Wood- 
 worth and Wells." 
 
 10. Controlled association — mixed relations. Two lists taken from 
 Woodworth and Wells.^" 
 
 11. Cancelation of single digits.^" 
 
 12. Cancelation of number groups containing a specified pair of 
 digits." 
 
 13. Cancelation of words containing a specified pair of letters.^^ 
 
 14. Cancelation of unfamiliar figures. Two sets designed by Franz.^^ 
 
 15. Trabue completion test. Both the preliminary set of 56 sentences 
 and the later Language Scale A were used.^^ 
 
 16. Pictorial completion (Healy).^* 
 
 17. Construction puzzles A and B (Healy and Fernald).^^ 
 
 18. Ellis Island constrtiction puzzles : Knox's "Moron" and "Diamond 
 Frame," and Kempf's "Diagonal." ^^ 
 
 ■Healy, W. and Fernald, G. M. "Tests for Practical Mental Classifica- 
 tion," Psychol. Monog., 1911, vol. 13 (no. 54), pp. 34-40. 
 
 'Knox, H. A. "A Scale Based on the Work at Ellis Island for Estimating 
 Mental Defect." J. of the Amer. Med. Assoc, vol. 62, 1914, p. 742. Pintner, R. 
 and Paterson, D. G. "A Scale of Performance Tests," 1917, pp. 67-69. 
 
 'Whipple, G. M. "Manual of Mental and Physical Tests," 1915, Part II, 
 pp. 17-42. 
 
 "Kent, G. H. and Rosanoff, A. J. "A Study of Association in Insanity." 
 Amer. J. of Insanity, 1910, vol. 47, pp. 37-96, 317-390. 
 
 "Woodworth, R. S. and Wells, F. L. "Association Tests." Psychol. 
 Monog., vol. 13, whole no. 57, 1911, pp. 24-29, 32-41, 62-67. 
 
 "Norsworthy, N. "The Psychology of the Mentally Deficient." 
 
 ■= Franz, S. I, "Mental Examination Methods," 1912, pp. 130-132. 
 
 "Trabue, _M. R. "Completion-Test Language Scales." Teachers College, 
 Columbia University, Contributions to Education, no. 77. 1916 
 
 "Healy, W. "A Pictorial Completion Test." Psychol. Rev., vol. 21, 1914, 
 pp. 189-203. 
 
 Pintner, R. and Anderson, M. M. "The Picture Completion Test." Educ. 
 Psychol. Monog. No. 20. 
 
 " Op. cit., pp. 14-17. 
 
 " Op. cit., pp. 743-4. 
 
METHODS OF PSYCHOLOGICAL EXAMINATION 45 
 
 19. Sequin form-board by the tactual method." 
 
 20. Mirror drawing test.*' 
 
 21. Hard and easy directions test. (Woodworth and Wells.)" 
 
 22. Woolley ingenuity test.^" 
 
 23. Test of mechanical construction. (Stenquist.)^"^ 
 
 24. Test of selective judgment. (Bonser.)^^ 
 
 25. Steadiness of movement.^* 
 
 26. Bogardus factory test.^* 
 
 III. Educational Tests. 
 
 A series of educational tests were given with a view to determining 
 the extent of formal education the subject had received and retained. 
 These were given by Mary A. Clark of the Laboratory staff. The 
 women came to the Laboratory after the period of quarantine was over 
 and were tested in groups of from three to eight, the aim being to 
 avoid having women of very varying ability in the same group. The 
 tests extended generally over a period of four days. The tests given 
 were as follows: 
 
 1. Arithmetic. 
 
 Courtis Test Series A, forms 1 and 3, and Series B, form 1." 
 
 2. Spelling. 
 
 A combination made by Thorndike of some of the words standard- 
 ized by Buckingham. 2* 
 
 " See Goddard, H. H. "The Form Board as a Measure of Intellectual De- 
 velopment in Children." Training School, 1912, vol. 9, pp. 49-52. 
 
 Sylvester, R. H. "The Form Board Test." Psychol. Monog., vol. IS, no. 
 
 65, 1913. 
 
 "Whipple, op. cit.. Part II, pp. 119-133. 
 
 ^' Op. cit., pp. 68-72. 
 
 ""Thompson, H. B. "Psychological Norms in Men and Women,' 1903, pp. 
 
 '^ St'enquist, J. L., Thorndike, E. L., and Trabue, M. R. "The Intellectual 
 Status of Children Who Are Public Charges." Arch, of Psychol, 1915, vol. 5, 
 
 "°^B'on^er F. G. "The Reasoning Ability of Children of the Fourth, Fifth 
 and Sixth School Grades." Teachers College, Columbia University, Contribu- 
 tions to Education, no. 37, 1910, pp. S-6. ■ , rr, , „ ,nlA T, ^ T 
 
 ^Whipple, G. M. "Manual of Mental and Physical Tests, 1914, Part 1, 
 
 ^^''" Bogardus E S. "The Relation of Fatigue to Industrial Accidents," Amer. 
 J. of SocioL, 1911-12, vol. 17, pp. 1-69. , ^. . . c • .t. ^ <■■ 
 
 " Courtis, S. A. "Manual of Instruction for Givmg and Scormg the Courtis 
 
 Standard tests." , ^ ,, ^ j t->- .. -u ,.• >> 
 
 " Buckingham, B. R. "Spelling Ability, Its Measurement and Distribution 
 Teachers College, Columbia University, Contributions to Education, Wo. 59. 
 1913. 
 
46 STUDY OF WOMEN DELINOUENTS IN NEW YORK 
 
 3. Handwriting. 
 
 A passage to be scored by both the Ayres and the Thorndike 
 
 Scales.^^ 
 
 A test of accurate copying. 
 
 An original letter. 
 
 4. Reading. 
 
 Thorndike's Scale A and early form of Scale Alpha.^* 
 
 5. History. 
 
 A list of questions compiled by Van Wagenen and not yet published. 
 
 6. Geography. 
 
 A selection from a series of questions being standardized by the 
 Boston Public Schools. 
 
 7. General information. 
 
 A brief set of questions on general information, compiled by us. 
 
 8. Practical knowledge. 
 
 Tests of ability to tell time, name money and make change, 
 use of tape measure, etc. 
 
 SELECTION OF TESTS FOR PRESENT INVESTIGATION 
 
 In August of 1916 it was decided to undertake the study of other 
 penal groups. For this, so extended a series of tests was impossible. 
 After careful study we accordingly reduced our series to those in- 
 cluded in the following sections. Our criteria for this reduction were 
 four: 
 
 1. Some tests we had found of more intrinsic value than others. 
 Our criteria of value will be discussed later. 
 
 2. Some tests were retained because they furnish a basis for com- 
 parison with the work of others. 
 
 3. Some tests were discarded because they were too cumbersome 
 to be easily applied outside of a laboratory. 
 
 4. Some tests were omitted to avoid duplication. 
 
 The conditions of testing in the other institutions were less ideal 
 than in our earlier work, but, for the most part, reasonably satisfactory. 
 The testing of the Probation Group was done in part at the Florence 
 
 =" Ayres, L. P. "A Scale for Measuring the Quality of Handwriting of 
 School Children." Russell Sage Foundation, 1912. 
 
 Thorndike, E. L. "Handwriting." Teachers College Record. 1910, vol. 11, 
 No. 2. 
 
 '"Thorndike, E. L. "Reading Scale and Visual Vocabulary Scale A," and 
 also the "Administration Scale A for Visual Vocabulary." Teachers College 
 Record, September, 1914. 
 
METHODS OF PSYCHOLOGICAL EXAMINATION 47 
 
 Crittenton Home where the women were sometimes held pending a 
 decision on their cases. In other instances they came to the city office 
 of the Laboratory. 
 
 The factor which caused the most trouble at the institutions other 
 than Bedford was the necessity of making the tests after the women 
 had already become an integral part of the prison population. The 
 business of testing was something then which was suddenly introduced 
 into their program, and was not an already established custom to which 
 they were expected to conform. Taking as we did the commitments 
 for the current year only, it left a goodly remainder of old inhabitants 
 who never had been called upon to be tested, and these often formed a 
 background of scoffers against which it was hard to contend. On the 
 whole, however, we fared very well in this matter of recalcitration. 
 One girl at the Magdalen Home refused to cooperate. Three of the 
 older and more intelligent women at Auburn argued cannily that the 
 less there was on record about them the better for them, and several 
 cases of old rounders at the Workhouse declined to put up with such 
 "new-fangled notions." 
 
 The matter of one inmate being coached by another was not so 
 serious a matter as we feared it might be. The really intelligent 
 women who might be the most effective teachers could generally be 
 made to see our point of view on the matter of communication, and 
 where the less intelligent attempted instruction it was often a case of 
 the blind leading the blind. There were instances where we felt that 
 the subjects had been tutored, but they were quite as likely to have mis- 
 understood their instructions as to have profited by them, so that the 
 total results were not materially affected. The testing covered gener- 
 ally a period of six days, of about an hour a day. 
 
 As in our earlier series, the instructions after being carefully 
 worked over were reduced to writing and committed to memory. The 
 methods of scoring were worked out in detail and each examiner was 
 provided with a copy and marked the tests in accordance therewith. 
 Where ambiguous answers occurred these were discussed in conference 
 and marked on the basis of a joint decision. In every case the tests 
 after being marked were checked by another examiner to catch up any 
 errors. 
 
 Yerkes-Bridges Point Scale and Stanford-Binet Scale 
 
 Both the Yerkes-Bridges Point Scale and the Stanford-Binet were 
 included, in spite of their similarity, because of the basis of comparison 
 
48 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 which these series afford with the work done by other investigators. 
 Our methods for giving the two were changed from the original meth- 
 ods of the authors in question at the following points, all of which 
 are minor ones, made to enable us to score the test for both series with- 
 out being forced to repeat ourselves. 
 
 In fourteen of the tests on the Point Scale there is duplication 
 which involves seventeen of the tests on the Stanford-Binet. In two 
 of these, viz., the defining of abstract words, and detecting absurdities, 
 the two methods overlap in part and it is only necessary to ask the 
 additional questions in order to score the test for both. In seven 
 tests the variation in the method of giving seemed to us so slight as 
 to be immaterial. These tests are comprehension of questions (easy 
 and difficult), the comparison of two objects from memory, aesthetic 
 comparison, comparison of two lines, sixty words, and the arranging 
 of weights. In the tests of copying a square, and copying a diamond, 
 the change in difficulty from drawing with a pencil (Yerkes-Bridges) 
 to drawing with a pen as required in the Stanford-Binet seemed so 
 great that, where either of these was included for the latter, we re- 
 peated the test, using pen and ink, and scored it on that showing. 
 
 In five tests, viz., the comparison of two weights, omissions in pic- 
 tures, counting backward, definition by use, and dissected sentences, 
 we used the Yerkes-Bridges form. In drawing the design from 
 memory we penalized the Yerkes-Bridges Scale somewhat in that we 
 followed the Stanford directions of giving only 10" exposure instead 
 of 15" as instructed by Yerkes. 
 
 Tests of the Woolley Series 
 
 We aimed to use as many as possible of the tests of the Woolley 
 series because of its extensive use on non-delinquent groups otherwise 
 corhparable with ours. The tests of steadiness and rapidity of move- 
 ment, involving as they do even a small amount of mechanical appa- 
 ratus, we discarded as impractical, because it was necessary under the 
 conditions of working away from the Laboratory to put all material 
 out of the way after each day's testing and the labor involved in con- 
 stantly re-setting was too time-consuming. 
 
 Throughout the giving of these tests we have used the material 
 specified by Dr. Woolley^' and followed carefully the instructions de- 
 vised by her. We feel, therefore, that our findings are comparable 
 with hers. The tests selected are as follows : 
 
 " Op. cit. 
 
METHODS OF PSYCHOLOGICAL EXAMINATION 49 
 
 1. Sentence completion test. (Series beginning "If some child".) 
 Here the sabject is required to complete the beginnings of thirteen 
 
 sentences. These are exposed one at a time to the subject who writes 
 an ending to each sentence. The test is scored on total time, individual 
 reaction time, and the number of ideas expressed. Our instructions 
 for giving are practically the same as those given by Dr. Woolley, and 
 our method of scoring follows the same lines as hers but has been 
 made a little more specific. 
 
 2. Association by opposites. (List 6, beginning "Inside".) 
 
 The responses are written by the subject and the test is scored 
 both for accuracy and for total time of the performance. 
 .3. Cancelation of a's. (Form beginning "hplg".) 
 
 Our directions for giving and evaluating are but little modified 
 from Dr. Woolley's. 
 
 4. Memory for digits, by the visual-auditoiy-articulatory method. 
 The numbers, printed in heavy type, are read aloud by the subject 
 
 and examiner together. Dr. Woolley used two series each of 7, 8, and 
 9 place numbers. We have added pairs of 10, 11, 12 and 6, 5, 4, and 3 
 place numbers where it was necessary to go in either direction in order 
 to ascertain the memory span. We have used her method in giving 
 and scoring this test. 
 
 5. Substitution or symbol digit test. 
 
 A series of nine geometrical figures are to be associated respectively 
 with the nine digits. After the association has been established by 
 means of three practice pages it is reproduced from memory on a 
 fourth page. Time and accuracy are both scored here. Our directions 
 for giving are practically identical with those of Dr. Woolley. 
 
 6. Card sorting test. 
 
 The Jastrow card sorting box with the colored cards described by 
 Dr. Woolley was used. Our instructions for giving were essentially 
 similar to hers but we varied her procedure in that we gave three trials. 
 We found the best time of the three to be the most satisfactory meas- 
 ure. 
 
 Additional Tests 
 
 1. Action-agent association. 
 
 The list we used is selected from the two lists used in our earlier 
 series. One of these was taken directly from the Woodworth and 
 Wells monograph and the other we compiled ourselves. We print 
 below the selected Hst: 
 
50 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 1. gallops 11- scratches 
 
 2. bites 12. flies 
 
 3. cuts 13. ticks 
 
 4. rings 14. creeps 
 
 5. explodes IS. blows 
 
 6. stings 16. evaporates 
 
 7. crows 17. rolls 
 
 8. aches 18. burns 
 
 9. growls 19. pricks 
 10. cries 20. roars 
 
 The responses were given orally and the reaction time of each taken 
 with a stop watch. The degree of accuracy was credited in whole and 
 half credits. 
 
 2. Mixed relations association. 
 
 Here again we made a selection of twenty words from two lists 
 used in our earlier work. Both of these are given by Woodworth and 
 Wells in the paper just cited. We print below the composite form 
 used: 
 
 1. Eye — see 
 
 Ear- 
 
 2. Monday — Tuesday 
 
 April— 
 
 3. Do— did 
 
 See- 
 
 4. Bird — sings 
 
 Dog- 
 
 S. Straw— hat 
 
 Leather — 
 
 6. Hammer — tool 
 
 Dictionary- 
 
 7. Dog— puppy 
 
 Cat— 
 
 8. House — room 
 
 Book- 
 
 9. Sky— blue 
 
 Grass — 
 
 10. Swim — water 
 
 Fly- 
 
 11. Once — one 
 
 Twice — 
 
 12. Cat— fur 
 
 Bird- 
 
 13. Pan- tin 
 
 Table— 
 
 14. Good— bad 
 
 Long— 
 
 15. Fruit — orange 
 
 Vegetable- 
 
 16. Sit — chair 
 
 Sleep — 
 
 17. Pen — write 
 
 Knife- 
 
 18. Nose — face 
 
 Toe— 
 
 19. Hungry — food 
 
 Thirsty — 
 
 20. Man — woman 
 
 Boy— 
 
 Our instructions for giving were but slightly modified from the direc- 
 tions of these authors. The accuracy here also was credited in whole 
 and half credits. 
 
 3. Cancelation of unfamiliar figures. 
 
 One of the sets of unfamiliar figures described by Franz was used. 
 This consists of eight uncommon but generally similar figures, each 
 repeated twenty-five times in haphazard arrangement. We are in- 
 debted to Dr. Franz for the use of his plates. Time and accuracy 
 were both recorded. 
 
 4. Directions test. 
 
 We used the form of hard directions devised by Woodworth and 
 
METHODS OF PSYCHOLOGICAL EXAMINATION 51 
 
 Wells and their form of easy directions which begins : "Cross out the 
 smallest dot." Time was recorded and the accuracy marked on a score 
 of twenty credits. 
 
 5. Knox's cubes. 
 
 Our material is comparable with Pintner's rather than with Knox's, 
 in that the blocks were uniform in color instead of being varied, and the 
 series devised by Pintner and Anderson was used. The measure 
 used in this study is the number of lines reproduced without error. 
 
 6. Rossolimo dissected pictures.'" 
 
 This test is taken directly from Rossolimo's profile method and 
 represents one of the tests of what he. designates "capacity to com- 
 bine." It consists of ten pictures, five in colors of familiar objects, 
 and five of conventional designs in black on a white ground. They 
 increase progressively in difficulty, depending, Rossolimo says, "on the 
 increasing complexity of the pictures and the increasing complexity 
 of the dissection." The pictures in order of increasing difficulty are 
 presented one at a time with the direction : "See if you can put these 
 pieces together to make a picture." The time for a successful com- 
 pletion is recorded. The time limit for each is three minutes. 
 
 7. Pictorial completion. (Healy.) 
 
 The blocks were always presented in a haphazard but definitely 
 specified order. Our directions are varied from those given by Healy 
 and by Pintner and Anderson chiefly by the addition of this definite 
 arrangement of blocks. We have used the system of evaluation de- 
 vised by Pintner and Anderson. 
 
 8. Trabue completion test. Scale A. 
 
 This test was given and scored in accordance with Trabue's direc- 
 tions. 
 
 9. Woody arithmetic test. 
 
 Form B of the Woody series'^ and all of the four fundamental 
 operations were given. They were scored according to Woody's di- 
 rections. 
 
 10. Courtis arithmetic test. 
 
 Form B of the Courtis Series was used. It was given and scored 
 in exact accordance with Courtis' instructions. Since it had appeared 
 
 " Rossolimo, G. "Die Psychologische Profile zur Methodik der quantitativen 
 Untersuchung der psychische Vorgange in normale und pathologische Fallen." 
 Klinik d. psych, u. nerv. Krankh., 1911, , vol. 6. We are indebted to Dr. 
 Whipple for the loan of his materials, from' which we had copies made. 
 
 " Woody, C. "Measurement of Some Achievements in Arithmetic." Teach- 
 ers College, Columbia University, Contributions to Education, No. 80, 1916. 
 
52 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 from our earlier use of the scale that the addition and the division tests 
 were a fair index of one's capacity in all four processes, only these 
 two were used in the present investigation. 
 
 11. Reading. 
 
 We used Thorndike's two reading scales, A and Alpha, em- 
 ploying his directions for giving and scoring. 
 
 12. Spelling. 
 
 For this test a list of sentences suggested by Dr. Thorndike was 
 used. They contain twenty-seven of the words standardized by Buck- 
 ingham for children ranging from third to super-eighth grade. Each 
 word misspelled, omitted or illegible was counted as an error. . 
 
 Tests 8, 9, 10, 11 and 12 were given as group tests when practic- 
 able, the number in the group varying from two to ten. 
 
 13. Handwriting. 
 
 This was meant to test both the speed and the quality of writing. 
 It was possible to grade the copy by both the Thorndike and the Ayres 
 scales. The speed was measured in terms of the number of letters 
 per minute. The subject was instructed to write repeatedly the words 
 one, two, three, up to ten over a period of one minute about as fast as 
 he ordinarily wrote in writing a letter. He was first allowed to prac- 
 tise on a sample sheet to make sure he understood what was expected 
 of him. 
 
 14. Practical information. 
 
 A series of tests of practical information was devised which con- 
 sists of the following simple operations : naming money ; making 
 change, telling time; counting the value of stamps; repeating days of 
 the week and months of the year; and measuring feet and inches with 
 a tape measure. 
 
 USE OF PEARSON SCALE FOR JUDGING MENTAL ABILITY AS CRITERION OF 
 RELIABILITY OF TESTS USED 
 
 As a criterion for measuring the reliability of a test we have made 
 use of the judgment scale devised by Pearson.^^ His method of obtain- 
 ing this scale may be briefly described as follows: He assumed the 
 hypothesis that grades of intelligence for the general population dis- 
 tribute their frequencies in the form of the normal probability curve. 
 He constructed a descriptive scale of intellectual levels, on the basis of 
 which he secured judgments from the instructors of approximately 
 
 ''Pearson, Karl. "On the Relationship of Intelligence to Size and Shape of 
 Head, and to Other Physical and Mental Characters." Biometrika, vol. S, p. 105. 
 
METHODS OF PSYCHOLOGICAL EXAMINATION 53 
 
 1000 Cambridge students and 5000 unselected English school children. 
 By this means he obtained the frequencies for each class and so deter- 
 mined the numerical equivalents for each interval on the scale. He 
 was thus able to assign quantitative values to different intellectual 
 classes obtained by judgmental methods, which made possible quanti- . 
 tative comparisons with other measurable factors. The form of the 
 scale as we used it is as follows : 
 
 Class M. Especially able. A mind especially bright and quick, both 
 in perception and in reasoning, not only about customary but about 
 novel facts; able and accustomed to reason rightly about things on 
 purely self-initiative. 
 
 Class N. Capable. A mind less likely than M to originate inquiry, 
 but quick in perception and reasoning about the perceived. 
 
 Qass O. Fair intelligence. A mind ready to grasp and capable of 
 perceiving facts in most fields; capable of good reasoning with moderate 
 effort. This group comprises, say, one-third of the population at large. 
 
 Class P. Slow intelligence. A mind slow generally, although possibly 
 more rapid in certain fields, but quite sure of knowledge once acquired. 
 
 Class Q. Slow. A mind advancing in general but very slowly; with 
 time and considerable effort not incapable of progress. Very slow in 
 thought generally, but with time understanding is reached. 
 
 Class R. Slow, dull. A mind capable of perceiving" relationships 
 between facts in some few fields with long and continuous effort but not 
 generally or without external assistance. 
 
 Class S. Very dull. A mind capable of holding only the simplest 
 facts and incapable of reasoning or grasping the relationship between 
 facts. 
 
 Class X. Judge a girl to be in Class X if you are not familiar 
 enough with her ability to justify an estimation of her intelligence. 
 
 Class Y. Judge a girl to be in Class Y if you feel that you are 
 prejudiced in favor of or against the girl, and are for that reason incapable 
 of judging her mental ability fairly. 
 
 We ourselves added Classes X and Y to the scale, not because we 
 intend to make statistical use of them but because we felt they would 
 relieve the judges from being forced into making a judgment where 
 they felt incapable of doing so. We have also adopted Ruml's^' 
 device of allowing the judges to add a plus or a minus sign to their de- 
 cision, indicating that they consider a subject to belong in a given 
 class but inclining toward a higher or lower interval. 
 
 °'Ruml, Beardsley. "The Reliability of Mental Tests in the Division of an 
 Academic Group." Psychol. Monog. vol. 24, no. 4, 1917. 
 
M + 
 
 471 
 
 mentaces 
 
 
 
 322 mentaces 
 
 9n 
 
 M 
 
 4S1 
 
 ft 
 
 O — 
 
 302 
 
 H 
 
 R + 
 
 M — 
 
 416 
 
 It 
 
 P + 
 
 282 
 
 " 
 
 R 
 
 N + 
 
 391 
 
 " 
 
 P 
 
 262 
 
 tt 
 
 R — 
 
 N 
 
 371 
 
 u 
 
 P — 
 
 242 
 
 tt 
 
 S + 
 
 N — 
 
 353 
 
 it 
 
 Q + 
 
 220 
 
 it 
 
 s 
 
 + 
 
 337 
 
 it 
 
 Q 
 
 192 
 
 it 
 
 
 54 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 The numerical equivalents, which Pearson calls "mentaces," for 
 each class are given below : 
 
 177 mentaces 
 
 157 
 
 130 
 
 116 
 
 96 
 
 62 
 
 We wish now to say something of thp qualifications of the persons 
 who made these judgments for us and to express our grateful appre- 
 ciation of their service. Our method has been to use only those cases 
 where we have been able to secure judgments from three observers. 
 We have then taken the numerical equivalents of these and used their 
 average as a measure to correlate with each test or series of tests. 
 
 At Bedford, which has served as the basis for most of this part 
 of our work, the judgments were made by persons who had unusually 
 good opportunity to size up the individuals under consideration. Miss 
 Ida Murphy was superintendent of the school and had direction also 
 of the assigning of the women to the various forms of employment 
 in the institution. In this way she came to study very carefully the 
 ability of each. Mrs. Louise Engle, the matron of the Reception Cot- 
 tage where the women remained for a period of at least two months 
 after their arrival, was in direct contact with them during this period 
 and thus had ample opportunity to form an adequate opinion. Miss 
 Dawley, the sociologist of the Laboratory, was the third of the main ob- 
 servers and she from her intensive study of each woman had occasion 
 constantly to make such judgments. In the great majority of cases 
 it is the average of these three judgments that we have used. To 
 supplement these for certain groups we secured judgments from Miss 
 Eleanor Miller, who had direction of all the agricultural work and 
 supervision of the "Farm Group," from Miss Winspere and Miss 
 Decker, two of the teachers in the institution, and from Mrs. Hoffman, 
 who was in charge of the work in the laundry. 
 
 Our confidence in the reliability of this group of judges is such 
 that we do not feel that the value of their estimates is seriously affected 
 by the fact that they had some knowledge of the test results. That they 
 should have this knowledge was unavoidable since each commitment 
 was discussed at a staff meeting within a month or so of her arrival. 
 It was impossible to secure judgments upon her prior to this discussion 
 since she was not yet known by most of the persons mentioned above. 
 
METHODS OF PSYCHOLOGICAL EXAMINATION 55 
 
 The influence of this previous knowledge was lessened by the fact that 
 several months had elapsed between the staff meeting and the time 
 that judgment was made, and that results of the tests were not actually 
 before the judges as they made their estimates. 
 
 At Auburn the judgments were made by Dr. Frank Heacox, 
 the prison physician ; by Mrs. Margaret E. Daly, the superintendent of 
 the women's prison ; by her assistant, Miss Curtis ; and by Mrs. Helen 
 P. Stone, the school-teacher of the women. Except in occasional in- 
 stances they had no knowledge of the outcome of the tests. We regret 
 that the smallness of the group prevented our making any extended 
 use of these figures. 
 
 At the Magdalen Home we were so fortunate as to obtain judg- 
 ments from Miss Janet Macconachie, the superintendent; from Mrs. 
 Mary E. Paddon, the executive secretary ; from Mrs. Nellie R. Crane, 
 the school-teacher ; and from the field worker. Miss Esther Cook. Like 
 the Auburn judgments, these were made without knowledge of the test 
 results, but were less serviceable than the Bedford data because of the 
 smallness of the group on which we were actually able to secure three 
 judgments. 
 
 On the other groups we have been unable to obtain enough reliable 
 judgments to make a satisfactory average. In the Workhouse, many 
 cases are not even known by sight except by the matron having direct 
 charge of them. Combining of results from the three institutions to 
 obtain larger numbers proved not to be feasible since there was an evi- 
 dent tendency to use different standards of reference in the different 
 institutions. In other words, the factor discussed on page 38 as 
 limiting the application of the method of estimates from one group 
 to another appeared here as a limitation on our comparisons. 
 
 SELECTION OF TESTS FOR SPECIAL CONSIDERATION IN PRESENT STUDY 
 
 With this scale as a criterion, we have selected from our data for 
 use in the present study a group of tests which, taken singly, give a 
 reasonably high correlation, as determined for the Bedford group, and 
 which, thrown into combination, give a still higher correlation. This 
 scheme is admittedly less satisfactory than the method of eliminating 
 tests by multiple correlation, but in view of the small number of cases 
 on which we were able to secure three reliable judgments we have felt 
 that we have refined our method as far as our data warranted. We 
 cannot say with certainty that, of the tests which we have given, we 
 have selected that combination which would give the highest possible 
 
56 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 correlation with our criterion, but we can say that by this method we 
 have been able to select a group of tests which give a correlation with 
 the judgments of competent observers sufficiently high to make it cer- 
 tain that they furnish a reliable measure of intellectual capacity. 
 
 It is not possible within the limits of this book to describe iii detail 
 the results of the correlation of all of our individual tests with the 
 judgmental scale. We hope at an early date to publish these data 
 in one of the psychological periodicals, but for the present can only 
 state the principle on which the tests were chosen and from there on 
 restrict our discussion to these tests. We should state, however, that 
 we have no intention of implying that the tests not included were found 
 unsatisfactory. Their range of efficiency as measured by the judg- 
 ment scale runs from zero to sixty odd. Some of them would thus 
 seem to be quite useless as measures of intelligence and others might 
 prove to be as useful as those we have chosen or even more so. 
 
 Considering first the correlations between the judgment scale and 
 the two measures of the Binet type — the Yerkes-Bridges Point Scale 
 and the Stanford Revision — we find the coefficients to be, in the case 
 of the Bedford group, .78±.042 and .81 ±.037 respectively — figures 
 sufficiently large to make our data another proof that these series fur- 
 nish reliable measures of intellectual capacity. 
 
 From our data we next assembled the four individual tests from 
 the psychological series whose correlation with the judgment scale was 
 highest and combined these into a group to serve as another measure of 
 intellectual capacity. These are: (1) accuracy bf response in the 
 opposites test, (2) accuracy of response in the mixed relations test, (3) 
 the vocabulary score (Stanford-Binet), and (4) the score on Trabue's 
 completion test. The coefficients of correlation of these with the judg- 
 ment scale, for the Bedford group, are respectively : 
 
 0pp. Ace. Mixed Rel. Ace. Vocab. Score Trabue Com. Score 
 Judgments .65 .69 .71 .69 
 
 The scores on these tests were combined by the method, suggested 
 by Woodworth, of expressing each individual score as a deviation from 
 the mean of the group as a whole and in terms of the standard devia- 
 tion as a unit, (-j In this way one is able to combine measures 
 
 regardless of whether they are expressed in terms of time, errors or 
 any other form of measurement which may have been employed. We 
 
 added these individual standings, f -Y unweighted, and thus obtained a 
 
METHODS OF PSYCHOLOGICAL EXAMINATION 57 
 
 standing on the group of tests. The series of these combined standings 
 gives a correlation of .81 with the judgment scale. We have accordingly 
 regarded this group of tests also as a reliable measure of intellectual 
 capacity. 
 
 In this combination we have consciously selected a group of tests 
 where the language factor is especially starred, designating the series 
 as the language group. Our justification for such a selection lies in 
 their high correlation with the judgment scale. If, as seems not im- 
 probable, an individual's capacity to handle language is one of the im- 
 portant determinants of the judgment which is made on his intellectual 
 capacity, then there would seem to be no impropriety in using such 
 tests as a measure. 
 
 In an effort to counteract this emphasis on the language factor, 
 we endeavored to evolve a series of performance tests, recognizing 
 the importance of such tests where one is dealing with those whose 
 knowledge of English is limited or whose emotional characteristics 
 are such as to hamper them in an effort to express themselves verbally 
 under conditions of examination. With a view to compiling such a 
 series we picked from our data four tests where the factor of language 
 is emphasized as little as possible and whose correlations with judg- 
 ments are among the highest of such non-language tests. These are: 
 (1) card sorting (Woolley), (2) memory for digits (Woolley, A. V. A. 
 method), (3) Knox cubes, (4) cancellation of unfamiliar figures 
 (Franz). Specifically, we used as a measure for the card sorting the 
 time of the best of three trials; for the memory for digits the sum of 
 the averages of the pairs of seven, eight and nine place numbers ; for 
 the Knox cubes the number of lines of the Pintner series correctly 
 repeated ; and for the cancellation of unfamiliar figures we used one- 
 half of the sum of the standings of time and errors; that is to say, 
 we obtained standings for time and for errors, added them, and, to 
 avoid weighting this test too heavily in relation to the others, divided the 
 sum by two. 
 
 We thus have four tests represented unweighted, although one of 
 these is made up of two measures. We combined these four tests in 
 the same way in which we combined the language group— by a simple 
 sum of their standings. 
 
 The coefficients of correlation of each individual test with the judg- 
 ment scale are respectively : 
 
 Card sorting AVA digits Knox cubes Cancel, time Cancel, score 
 Judgments .67 .61 .54 .40 ' .32 
 
58 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 By combining them into a group the correlation was found to be .17 ±. 
 .053, a coefficient sufficiently large to indicate that we have here 
 also a useful measure of intellectual capacity. 
 
 We have, now, four series which would seem to be reliable meas- 
 ures of intellectual ability in that they show high correlation with intel- 
 ligence as estimated in terms of the Pearson scale by the judgments of 
 three competent observers — ^the coefficients ranging from .77 to .81, 
 with standard deviations in no case large enough to cast doubt on their 
 significance. We figure below the table of inter-correlations of these 
 four with each other and with the judgments. 
 
 INTER-CORRELATION OF TESTS 
 
 
 Yerkes Stanford- 
 
 Language 
 
 Performance J 
 
 Fudgmi 
 
 
 Points Binet 
 
 Tests 
 
 Tests 
 
 
 Yerkes Points .... 
 
 — .87 
 
 .88 
 
 .63 
 
 .78 
 
 Stanford-Binet ... 
 
 — 
 
 .89 
 
 .66 
 
 .81 
 
 Language Tests . . 
 
 
 — 
 
 .68 
 
 .81 
 
 Performance Tests 
 
 
 
 — 
 
 :3i 
 
 Judgments 
 
 
 
 
 — 
 
 As measures of mentality available for purposes of comparison 
 with the work done by others, we have used chiefly the figures on the 
 results obtained by the Yerkes and the Stanford scales. For purposes 
 of inter-comparison on our own data, we combined the results on 
 the several series just described into a single measure which we 
 called "Test Aggregate." We obtained this by taking the average 
 of all four standings, where all of the series were given, and of 
 three where only three were available. For this we have used no 
 records where the subject had not taken at least three of the test 
 groups. This measure was correlated with the judgment scale and 
 found to give a coefficient of .86. The inter-correlations of this 
 measure with the others which are integral parts of the combination 
 are as follows: 
 
 
 Yerkes 
 
 Stanford- 
 
 Language 
 
 Performance 
 
 
 Point 
 
 Binet 
 
 Group 
 
 Group 
 
 Test Aggregate .. 
 
 .93 
 
 .94 
 
 .95 
 
 .82 
 
 This, then, is the measure which we shall figure for the most part as 
 our means of comparison of one group of our subjects with another, 
 whether the grouping be by institutions, by color, by crime committed 
 or any other mode of separation. This series which we have called Test 
 Aggregate, can, of course, only be used fairly with subjects who can 
 understand spoken English. In order to call a case "English-speab' 
 ing" in this work, we have arbitrarily ruled to limit the term to those 
 
METHODS OF PSYCHOLOGICAL EXAMINATION 59 
 
 born in this or any other English-speaking country or who have had at 
 least two years' schooling in this country. This ruling we felt, from 
 our personal knowledge of the subjects, served effectively to eliminate 
 all those with serious language difficulty.^* English-speaking illiter- 
 ates were not eliminated. Only one of the four groups of test is appre- 
 ciably affected by inability to read and write. An illiterate could there- 
 fore make his Test Aggregate score on the other three scales. 
 
 For those comparisons where we have sought to include the non- 
 English-speaking women as well, we have used the records of our 
 group of performance tests as a measure of mental capacity. The 
 great majority of our comparisons have, however, been made with 
 Test Aggregate, which means that we have been forced to omit from 
 such comparisons the non-English-speaking women. We recognized 
 that this limits the completeness of our comparisons, but purposely 
 chose this horn of the dilemma since Test Aggregate appears to give 
 a more satisfactory measure of intelligence than does our Performance 
 Group. We should be penalizing the whole in our efforts to include 
 a few, if we had carried through our comparisons on the basis of the 
 Performance Group. 
 
 *" In order to determine whether we had lowered the standing of our group 
 by including, even where language tests were used, some individuals for whom 
 English was an acquired tongue, we compared the total which we have called 
 "English-speaking" with a group selected on a much more conservative basis, 
 namely, individuals who were born in English-speaking countries themselves, 
 as were both of their parents. There was no appreciable difference between 
 the two groups as measured by Test Aggregate, but strangely enough, what 
 shade of advantage there was was in favor of the former. 
 
CHAPTER V 
 METHODS OF SOCIAL INVESTIGATION 
 
 IN discussing the method of social investigation of the six groups 
 of delinquent women with which we are dealing, it is important at 
 the outset to bear in mind that these women have committed offenses 
 against the law and that with the exception of one group they were 
 confined in institutions for varying lengths of time. Three hundred 
 and twenty-eight women or 54.6 per cent of the 587 cases studied have 
 had at least one previous conviction. Many of the women, especially 
 the sex offenders, have been violators of the law many times when 
 they were not arrested. This means that the majority of them have 
 important episodes in their lives which it is to their advantage not to 
 talk about to an investigator. The effort to keep back information 
 concerning a previous delinquency often tends to make a suspicious 
 woman untruthful about other events in her life which, however, may 
 have no bearing on the thing she is trying to conceal. 
 
 Our case studies, therefore, differ to quite an extent in the method 
 of approach and investigation from the case histories which most 
 social organizations are able to make. Miss Mary Richmond's book 
 on "Social Diagnosis," which is most comprehensive in outlining a 
 plan for case workers, presupposes that "case work . . . has for its 
 immediate aim the betterment of individuals or families, one by one, 
 as distinguished from their betterment in the mass." ^ The initiative 
 is taken, in many of these cases, by the individual or family who needs 
 help. The women in this study, on the contrary, were questioned en- 
 tirely without request on their part, and with the exception of the 
 probation group, they could hope to gain very little by talking freely 
 and frankly. To establish a friendly relationship and make it agreeable 
 for these women to give the information necessary to conduct an in- 
 vestigation was one of our most difficult and time-consuming problems. 
 
 'Richmond, Mary E. "Social Diagnosis." Russell Sage Foundation, New 
 York, 1917, page 25. 
 
 6o 
 
METHODS OF SOCIAL INVESTIGATION 61 
 
 GENERAL PLAN OF STUDY 
 
 In general, the form of social investigation used was intensive, with 
 an attempt to understand the individual thoroughly, not only through 
 obtaining her own reactions, but by complete investigation of all 
 outside sources which could help in making such an understanding 
 possible. A study of our results and comparison of the woman's story 
 alone with the full investigation would indicate that the facts are 
 obscured by using either plan exclusively, and that the intensive method 
 is essential in gaining a true insight into the individual and the factors 
 related to her delinquency. This work differs, therefore, from most 
 extensive studies of delinquency in that the subject's statement has 
 been corrected and added to by as thorough an investigation as was 
 possible. 
 
 The main trend of our sociological study was to obtain from each 
 woman delinquent studied an oral statement giving the details of her 
 personal history and her background, such as heredity and early en- 
 vironmental influences. From her initial statement a first-hand study 
 of these factors was made, by visits to and correspondence with the 
 various individuals or organizations which had come into contact with 
 her at any time. In all of the groups information given by the subject 
 was recorded separately from the verified data about her, in order that 
 the two kinds of material might be used for comparison and to deter- 
 mine whether there was any consistent tendency on the part of the 
 women studied to misrepresent or fail to give information concerning 
 any important factors in their lives. 
 
 To accomplish this work there were six trained investigators, with 
 occasional help from one other person. The general plan was followed 
 of having the same worker take the bulk of the cases in one group. 
 Such a plan is time saving and also is necessary in making the worker 
 known among the inmates of an institution, and in having the inves- 
 tigation appear to be more nearly a routine of the institution. In each 
 group, however, a number of the cases were worked on by individuals 
 other than the one who was doing the main part of the work in that 
 place. This was necessary at times when the work with one group 
 would become particularly heavy. Such an arrangement was also felt 
 to be of benefit in determining, by a liberal sampling of cases inves- 
 tigated by other workers, whether there was any personal bias enter- 
 ing into the work of the one who was mainly interested in any one 
 institution. 
 
62 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 In order to eliminate the personal equation as much as possible 
 and have the work on all of the groups comparable, a uniform set of 
 record blanks and the same general scheme of recording material was 
 used by each worker. A particular effort was made to keep distinct 
 and separate information obtained concerning the subject, and the 
 worker's impression of the informant. Both of these are necessary 
 information, but if carefully recorded as separate items, may be used 
 in a much more uniform scheme by one or two persons, who may 
 weigh the reliability of the information for all groups alike. 
 
 THE INITIAL INTERVIEW 
 
 Probably the most important part of any such investigation as this 
 is the initial interview with the subject. There are two main reasons 
 for the need of obtaining accurate information at this interview. 
 The first and most important perhaps, because it applies to a larger 
 number of cases is, that by obtaining reliable information to start 
 with, the work of the investigator will be considerably lessened, and 
 the final investigation will be more satisfactory. If a wrong address 
 is given, or if a woman says that her parents are living in a city miles 
 away, when they are, as a matter of fact, nearby, much of the work- 
 er's time is consumed in trying to trace them, and there is always a 
 chance that they cannot be found. This involves either questioning 
 the subject again, or losing much valuable information from outside 
 sources. A second reason of importance for getting a true initial 
 story is that this information may be all that can be obtained. There 
 are cases where no investigation is possible, and where the subject's 
 statement must be taken as the best information available. This is 
 true of many foreign-born women whose families are not in this coun- 
 try, or, of native-born who have no close friends. There is also much 
 information which can be obtained in no other way than through the 
 subject. This includes such s.ubjeclive^material as the attitude of the 
 subject toward persons and envifonmentalinfluences, her reasons for 
 doing certain things, and the resulting reactions. Usually, no amount 
 of investigation can discover these intimate personal facts which are 
 often important elements in the delinquency, and so it is very necessary 
 that these be obtained from the interviews with the women involved. 
 
 The possibility of obtaining reliable information from a woman 
 when she is in prison and the investigator is a stranger to her is de- 
 pendent largely upon the worker's ability to establish a personal rela- 
 tionship, and to win the subject's confidence. In some cases it was 
 
METHODS OF SOCIAL INVESTIGATION 63 
 
 necessary to promise not to investigate certain things which the woman 
 was willing to talk about because she trusted the worker's word. In 
 such cases there was undoubtedly more value in getting the story di- 
 rectly from the woman, but with no investigation, than in failing to 
 win her confidence sufficiently to get her story, and running the chance 
 of getting part of the information from another source. 
 
 Needless to say, it is difficult to create a personal relationship in a 
 brief interview under unfavorable surrounding^. Wherever it was 
 possible, therefore, particularly among the women in the Magdalen 
 Home, the New York County Penitentiary, Auburn State Prison and 
 the Reformatory at Bedford, the women were seen several times. If 
 she was in an antagonistic mood the first day, usually no effort was 
 made to continue the questioning at that time, but on another day she 
 was approached again. With the more intelligent and older women, 
 it was often necessary to explain the purpose of the questioning, and 
 to say that an investigation was being made for research purposes, but 
 that in each case the welfare and interest of the individual concerned 
 were considered first. 
 
 Though it is preferable to have a complete statement from the 
 woman before starting the investigation, this was not always possible, 
 especially in the Probation and Workhouse groups. In the -Probation 
 group, where often the girl was seen for only a few minutes in the 
 court room on the night of her conviction, and the investigation had 
 to be made before the next night when she was to be sentenced, it was 
 difficult to get from her more than a few important addresses and ref- 
 erences. In such cases the fuller "Statement of Girl" had to come after 
 the first investigation. The Workhouse group of women presented dif- 
 ficulties of a different nature. They are such a large, heterogeneous 
 group, composed of women sentenced from one day to two years, that 
 it is impossible for one or two workers to make any impression on 
 them in a short time. The women \who were there on a short term 
 or who were about to be freed from the institution were particularly 
 difficult to obtain satisfactory information from, because there was no 
 incentive for them to tell about their past histories. If they expected 
 to go back to their old life upon release, they did not want any worker 
 to be able to trace them, and if they expected to go to work they did 
 not care to have the employer know about their record, as he might if 
 there was a chance of the worker's coming to visit. There are also a 
 goodly number of older women among them, many of whom could 
 not remember details of their early life. Typical among them is the 
 
64 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 old lady, who, when questioned about some detail of her childhood, 
 replied— "God knows, dearie, I don't." Because of the inherent diffi- 
 culties involved in this group, no effort was made, in most cases, to 
 see a woman more than once, and at that time the best information 
 possible was obtained. This information, we reaHze, is much less re- 
 liable than the statements obtained from any other groups. 
 
 In each case, after the first interview, the information obtained was 
 recorded on the "Statement of Girl" folder. A copy of the folder is 
 given here with the statement of a girl committed to the Reformatory 
 at Bedford Hills. (See pages 65-8.) 
 
 Uniform methods of technique in questioning were used in all of 
 the cases. With age, for instance, the woman was asked how old she 
 was and was further asked if she was twenty-two or twenty-three, as 
 the case might be, on her last birthday, or whether she was to be that 
 age on her next birthday. Such further questioning we found was 
 essential in getting the correct age, since many persons, particularly the 
 foreign women, counted the age they were "going on" as their present 
 age. Many of them were unable to figure out the year of birth, so we 
 did not rely on their statement for that, but used only the statement 
 of number of years old, and day and month of birth. 
 
 The section on residences and addresses was used primarily as a 
 check on dove-tailing together the facts in the woman's history, and 
 as an aid to the investigator in visiting old home addresses. Such a 
 summary has been particularly helpful in working over the data be- 
 cause it gives in a condensed form the frame on which the other facts 
 of the history may be based. Under this section each year of the 
 woman's life was accounted for, and the information was grouped as 
 follows : 
 
 Assuming that the subject was born in 1890, 
 
 1890-1902, lived with parents at — E. 10th St., New York City. 
 1902-1906, lived with parents at — E. 176th St., New York City. 
 
 1906-1910, lived with employer, Mrs. , at . 
 
 1910-1915, lived with husband at . 
 
 April- July, 191 S, lived in furnished room at . 
 
 July, 1915 — present conviction, lived with consort at . 
 
 All of the section on family history was obtained with great care 
 from each woman. By so doing and by getting accurate addresses, 
 much time of the worker was saved later. This section, perhaps more 
 than any other, required tact and persistence in getting reliable infor- 
 mation. If the woman, did not want certain members of her family 
 
METHODS OF SOCIAL INVESTIGATION 
 
 65 
 
 Date of commitment 2/8/17 
 Date of interview 2/14/17 
 
 N0....1 
 
 Field worlcer A. B. . . 
 
 STATEMENT OF GIRL 
 
 NAME Am) LAST ADDRESS Jane Dow, furnislied rooms at 35 Pine St. and 64 Spruce St., 
 Alias Josette Raemaelcer 
 
 , N. Y. 
 
 AGE AND DATE OF BIRTH 19 years, 8 mos. Born 6/23/97 
 PLACE OF BIRTH R— , N. Y. 
 HOW LONG IN U. S. Always 
 RELIGION Methodist 
 
 MARITAL CONDITION Single 
 
 Under what name. 
 
 If married, when. 
 
 RACE American 
 COLOR White 
 
 Where 
 
 To whom 
 
 RESIDENCES AND ADDRESSES (In chronological order) 
 
 1897-1913, lived in R — — , with her parents. Worked in several places but always came home at 
 night, 
 
 1913- until 3 wks. before arrest, lived in N , N. Y., with her parents except for short period 
 
 of time in O— — and C— — where she was working. 
 3 weeks before arrest, lived in furnished rooms in B , N. Y. 
 
 FAMILY HISTORY 
 
 Name and address 
 Father Wm. Dow, 3 Center St., 
 
 N , N.Y. 
 
 Age Birthplace Time in U.S. Occupation Education 
 
 40 N.Y.C. Always Fireman in Reads and 
 
 boiler writes 
 factory 
 
 Naturalized 
 Mother Carrie Black, — Mrs. Dow, 
 
 3 Center St., N , N. Y. 
 
 Sibs 1. Girl, died at 2 days 
 
 2. Subject 
 
 3. Charles, N , N. Y. 
 
 4. John, died in 1907 at age of 
 
 5. Marietta, N , N. Y. 
 
 38 
 
 13 
 
 2 
 4 
 
 -, N.Y. 
 
 Left in 8th 
 grade 
 
 In 6th gr. 
 
 Husband None 
 
 Children None 
 
 Other relatives 
 
 Paternal uncle, R. M. Dow, F , N. Y. 
 
 Maternal aunts: Mrs. J. Street, M . N. Y. 
 
 Mrs. P. Gray, M , N. Y. 
 
 Institutional record 
 Relative 
 None 
 
 Institution 
 
 Dates 
 
 Complaint 
 
66 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 HOME CONDITIONS 
 
 During Childhood and Adolescence Lived in R , N. Y.,with her people until 1913. Always 
 
 bad a good home there. At first her people lived near the creek, then on M. Ave. and later on 
 S. St. In all of these places they had a separale house with quite a little ground so that they 
 could have a garden. Each of these houses nad at least 7 or 8 rooms. 
 
 Economic conditions Father used to earn $25 to $30 a week in a rubber factory in R . With 
 
 this income her mother was able to make a vnry comfortable home. 
 
 Moral standards Father has very strict standards. "Would never allow cider in the house even 
 to make mince-meat because he was afraid that the boys would learn to drink." Both her 
 mother and father are good church people. 
 
 Attitude toward family Seems to feel kindly toward her people but thinks that they should have 
 
 come to B when she was arrested. Says that her father is so severe that he will never fee] 
 
 the same toward her again because she has disgraced his name. 
 
 During Recent Life 
 Total income of family (Specifyitems) Has worked a little during the last two years but has been 
 at home most of the time. Her father is now earning $16 a week. They have vegetables which 
 they raised on their farm last summer and they have cows and chickens so that there is httle 
 necessity for buying much food. 
 
 Expenses of family 
 
 Rent Lease house and 45 acres of ground for $15 month. 
 
 Insurance All of family are insured. Jane is insured in Prudential and Colonial for $500 
 
 Benefit societies, etc. Father belongs to the Order of Foresters. 
 
 Miscellaneous "Mother has to squeeze every penny in order to get along. She will not 
 run a grocery bill and insists upon paying for the lease as soon as the money is due. I did 
 not think I could stay with them any longer unless I was working." 
 
 Character of locality (Factory, business, suburban, residence, etc.) Lives out in the coimtry 
 some distance from a raihoad station. Says that they have very good neighbors and that every- 
 body around them is respectable. 
 
 Type of dwelling (Tenement, separate house, etc.) 8 room, wooden house, in good condition. 
 Number of rooms 8 
 
 Number of persons in household (Specify lodgers) 3 adults and 2 children. 
 Sleepmg accommodations 4 bedrooms. Jane had a room to herself. 
 
 Moral standards See above. 
 
 Attitude toward family Feels very sorry for her mother who is pregnant again. Thinks that this 
 tSoughTe" "onfinemenr '" '""' "''"''' " *'""* ''"'" '" '^' *'^* ^^' "°*^^ '^'^^ °°* ''™ 
 
 EDUCATION 
 EngUsh-speaking Yes Read and write English 
 
 School Address (Last school) Public school, R , N. Y. 
 
 Age at starting school 5 yrs. Attendance Regular 
 
 Age at leaving school (If under i6, had she obtained an employment certificate) 12 yrs. 
 
 Grade at leaving school Lacked only 3 weeks of finishing grammar school. 
 
 Later schooling None 
 
 Language spoken at home English 
 
METHODS OF SOCIAL INVESTIGATION 
 
 67 
 
 WORK HISTORY (Places of employment in chronological order and in each:) 
 
 Kind of work Employer and Address Dates of WeeUy Reason for Disposition 
 
 employment Wage Leaving of Money 
 
 1. Operator L Embroidery fac- 6 wka. 1909 *5 wk. Work made Mother 
 
 tory, R , N. Y. 
 
 2. Gen'l housework Mrs. M , R- 
 
 N.Y. 
 
 4. " 
 
 5. Waitress 
 6. 
 
 her nervous 
 
 2H mos. 1913 S3 wk. Employer 
 "queer" 
 
 Mrs. M .Main St., 4 mos. 1913 $2K wk. Moved with 
 
 Mrs. M 
 N. Y. 
 
 -, N. Y. 
 
 H Sanitarium, 
 
 O , N. Y. 
 
 J Hospital, C- 
 
 N. Y. 
 
 parents 
 2 wks. 1914 J6 wk. Too nervous 
 
 2 wks. 1915 $18 mo. Emjjloyer too 
 domineering 
 
 2 wks. 1916 J20 mo. Too many 
 supervisors 
 
 SeU 
 
 INSTITUTIONAL HISTORY 
 Previous Court Record of Delinquency (Chronological order) 
 Date Place Complaint 
 
 None 
 
 Disposition 
 
 Commitment to Non-Penal Institutions (Chronological order) 
 
 Institution (As Orphan Asylum, etc.) Dates 
 
 None 
 
 Record of Conduct 
 
 Hospitals 
 
 None 
 
 Dates 
 
 Diagnosis 
 
 Attitude toward institutional experience Insists that she is innocent and that it is wrong for a girl 
 of her refinement and sensibilities to be forced to associate with the "street-girls" in a reformatory. 
 
68 STUDY OF WOMEN DELINQUENTS IN NEW YUtiif. 
 
 SEX HISTORY 
 First sex instructions 
 
 When 11 years old 
 
 From whom Mother. (Menstruation was established when 11 
 yrs. old) 
 
 Age at first sexual ofifense 17 yrs. 
 
 With whom Boy she knew in N- 
 Rape or by consent Consent 
 
 Approximate age at entering prostitution 
 ' Only means of support 
 Practised continually since 
 
 Never. Has had sexual intercourse twice, the first time 
 when she was 17 years, by consent, and the second time 
 when she was in O — ■ — . She had a "lonesome road" to 
 walk after she left the street car and one night a man who 
 had noticed her on the car, followed her and forced her to 
 have relations with him. 
 
 Practised where (Hotels, furnished roomSi etc.) 
 
 Average weekly earnings from prostitution 
 
 HEALTH (Diseases, accidents, operations, etc.) Has always been very nervous and has had the tic 
 of twitching her mouth since she was a little girl. As a child she had chicken-pox, measles, mumps 
 and rheumatism. She also had an abscess on her neck which her mother "drew up" with a poultice, 
 
 until it broke. Four years ago she had tubercular adenoids removed by Dr. B. in N . She 
 
 seems to be extremely nervous at present and is conscious all the time of the twitching of her fare 
 Is troubled with neuralgia a great deal. For several years has had spells when "everything goes 
 black for 10 or 15 minutes." 
 
 Habits: 
 
 Alcohol No 
 
 Drags No 
 
 Tobacco No 
 
 RECREATION (Opportunities, favorite amusements, friends, etc.) Says she has never had an oppor- 
 tunity for much recreation. When she was a child in R she always went directly home from 
 
 school and never knew other children very well. In N she became acquainted with some 
 
 young people who were very kind to her. She went out to church with them every week and 
 
 occasionally had supper with them at a small inn in N . Says she does not care much for 
 
 dancing and has never been to a theatre. 
 
 SUPPLEMENTARY FACTS AND EXPLANATION (Details of offense for which committed, etc.) 
 Jane says that she left the Hospital where she was working three weeks before her arrest because 
 
 there were too many supervisors. She went to B , New York, where she rented a furnished 
 
 room, and was hvmg there on the money she had saved when she was arrested. She says that on 
 the night ot her arrest she had been to a dance, the third one she had ever attended. She went 
 directly trom the dance to a hotel, engaged a room, and had just entered it when a detective 
 came in and said she was under arrest on a charge of vagrancy. She says she was going to work 
 the next day, and had enough money to pay for her room that night. 
 
METHODS OF SOCIAL INVESTIGATION 69 
 
 interviewed, her tendency at first was to be suspicious and refuse to 
 tell anything. Later when she felt she could rely upon the worker's 
 word, she would give a correct statement concerning her family with 
 the understanding that certain members were not to be seen. 
 
 The material on home conditions was intended to give a picture of 
 each woman's reactioq/toward her childhood and later home^ This 
 was not usually obtainecTHby direct questioning, but more often came 
 out in her story as she told it, uninterrupted. The other details such 
 as rent, income, kind of home, etc., were in response to direct ques- 
 tions. A particular effort was made to get the woman's attitude to- 
 ward her family, her treatment by them, and to find out whether the 
 conditions in her early home seemed in any way responsible for her 
 delinquency. 
 
 On the whole, very little antagonism was met in asking for infor- 
 mation concerning the school history. In a few cases the last school 
 address was refused, but usually it was possible to obtain this, if the 
 subject understood that in writing for the school record, no statement 
 would be made that she had been arrested. /The reliability of the in- 
 formation given by the women of the various groups concerning their 
 school attainments seems to be, as would be expected, in direct pro- 
 portion to the degree of personal relationship that could be established 
 between worker and subject^/ Here, again, the Workhouse group has 
 suffered most. Many of the first statements among them were so 
 unreliable that repeated efforts to locate school records were unsuccess- 
 ful. Many of the women professed to have forgotten the address of 
 the last school, and knowing that the record could not be found without 
 it, exaggerated the statement of their attainments. Among the younger 
 women, and in the groups where several interviews were possible, there 
 was rarely a case where information enabling us to verify the record 
 Was refused. 
 
 The work record was the most difficult information to obtain from 
 all of the women. An attempt was made to find out about each job she 
 had held and to get, especially, about each job, the kind of work done, 
 the length of time employed, wage received, reason for leaving and 
 the address of the employer so that the record could be verified. Ab- 
 solute accuracy in obtaining such a record is probably impossible, 
 especially with the older women who have worked during a number 
 of years, and have had many jobs. For the most part, however, the 
 material is quite satisfactory, and checked up by the verification not 
 only of the work record itself, but by visits to members of the family 
 
70 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 and friends who have known intimately of the subject's life. It seems 
 probable that records of most of the important jobs have been obtained. 
 The very irregular work records have been the ones to suffer most be- 
 cause of the woman's inability to remember about each of fifteen or 
 twenty jobs. But by persistent questioning as to the kind of work 
 she was doing at given times, it has been possible in many of even 
 these most irregular cases, to obtain from the women and to verify the 
 bulk of the jobs during her work period. With a few exceptions, such 
 questioning met little antagonism. It seemed a natural outgrowth of 
 friendly questions about the kind of work she was doing in the institu- 
 tion, and the work she would like to know how to do, tp her past work 
 experience. 
 
 The record of previous convictions and arrests was not difficult to 
 obtain from most of the women. The general feeling among them 
 seemed to be that a record of their convictions was on file at the court, 
 and that it was useless to contradict a finger-print record. In this 
 connection again, the older women in the Workhouse were least able 
 to give a satisfactory statement. Those who had had ten or twenty 
 previous terms would say : "I've been coming here off and on for the 
 last ten years — don't ask me how many times, because I can't remem- 
 ber." 
 
 Occasionally some woman refused to give the facts of her sex his- 
 tory, but as a rule no embarrassment seemed to be felt in talking of 
 these details. The method of approach varied in this with the different 
 groups. With the women at Bedford where the initial interview came 
 the day after the first physical examination, the most satisfactory way 
 of approaching the subject was by questioning first about previous 
 health, the age menstruation was established, whether at that time sex 
 instructions were given and from that to questions about the nature 
 of the first sex offense and later sexual irregularities. The greatest 
 difficulties in obtaining these facts came from the more intelligent 
 women in Auburn Prison and the New York County Penitentiary- 
 women who, possibly, had been sent to these institutions on larceny 
 charges and who had not been sex offenders more than a fevv times. 
 To many such a woman any implication that she was a prostitute was 
 most distressing, and had to be handled carefully. 
 
 The section on supplementary facts was made to include a brief 
 statement dove-tailing together the main facts otherwise indicated on 
 the folder, and such additional factors as conduct, abnormal traits, 
 attitude toward her delinquency, etc. 
 
METHODS OF SOCIAL INVESTIGATION 71 
 
 METHOD OF INVESTIGATION 
 
 With these data from the initial interview the field worker started 
 the investigation. In all of the groups except Bedford, where it was 
 necessary for one who had most intimate contact with the woman to 
 be near the institution all of the time, the worker investigated the cases 
 she had interviewed. This method of procedure is preferable where 
 the investigation per se is the object of interest, not only because it is 
 time-saving but also because the worker has a better knowledge of the 
 case if she has had the first interview. 
 
 (a) General Procedure of Investigation 
 
 The general plan of investigation was aimed to cover by visits and 
 letters verification of all points which are included on the "Statement 
 of Girl" folder, with such additional data as could be given by friends 
 or relatives on conduct, traits, etc. 
 
 A general order of investigation was followed in all cases where 
 it was feasible. Many times it was impracticable to follow any given 
 arrangement particularly when it happened that the worker had several 
 calls to be made on different cases in the same locality. 
 
 Every case that had at any time lived in New York or Brooklyn, 
 was first registered with the Social Service Exchange. In this way, 
 information giving the social agencies interested in the family was ob- 
 tained in many cases. 
 
 Wherever possible the court record, including the probation offi- 
 cer's report, was obtained early in the investigation. The probation 
 officer was also interviewed, so that the worker might know what 
 work had been done on the case before starting on her own investiga- 
 tion. If the finger-print record giving the number of previous con- 
 victions was. not included in the probation officer's report, this record 
 was obtained directly from the Finger-Print Bureau. With all of the 
 Penitentiary women and with the women sentenced to the Workhouse 
 on an indeterminate sentence, it was not necessary to get the court 
 and probation officer's records since these were included in the report 
 of the Parole Commission, through whose hands these women had 
 passed. . Our workers were given the benefit of all of the investiga- 
 tion done by the parole officers, and in this way were able to save much 
 duplication of work. 
 
 With women who were convicted in districts outside of New York 
 City, it was usually possible to interview the Chief of Police and the 
 
72 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 arresting officer who, in the smaller localities at least, had a fair knowl- 
 edge of the extent of the woman's delinquencies, and of her family 
 history. On the other hand, there is no probation officer in most of 
 these smaller places, and no record of finger-prints to determine num- 
 ber of past convictions. This necessitates going through many volumes 
 of inaccurately kept records, year by year, to get a woman's past rec- 
 ord, and if she has used several aliases it is very difficult to obtain her 
 complete record. 
 
 The difficulty in obtaining finger-print records for women committed 
 to Auburn State Prison was particularly marked. The women con- 
 victed in New York City are not usually finger-printed after convic- 
 tion until they are received in Auburn Prison. The Police Depart- 
 ment of New York finger-prints only a few felony cases who permit 
 this to be done before conviction. After conviction, the prisoner is 
 out of the jurisdiction of the Police Department and need not go to 
 the Police Headquarters. The finger-prints and Bertillon measure- 
 ments are taken, however, in Auburn State Prison, and two copies are 
 made. One is sent to the Bureau of Identification of the State Prison 
 Department in Albany, and one is filed at the prison. The subject's 
 statement of previous record is taken at the time of the finger print, 
 and is transmitted with the copies of the finger-prints to the Bureau 
 in Albany. This means that the Bureau of Identification has the actual 
 finger-print, any previous record in a state prison, and the subject's 
 statement except in occasional cases where the finger-print expert in 
 the prison has written to other cities to find out whether the woman has 
 a longer record than she is willing to admit. The Bureau of Identifica- 
 tion in Albany does not as a matter of routine have a record of mis- 
 demeanors and other less serious offenses which could be obtained by 
 sending a finger print to the Police Department in New York, so that 
 previous convictions there could be traced. 
 
 Therefore, in order to get a complete verified criminal record of the 
 women in Auburn Prison, it was necessary in each case to obtain the 
 statement filed with the Bureau of Identification at Albany and to 
 send for whatever record could 'be obtained from the Police Depart- 
 ment in New York. If the woman had used many aliases unknown 
 to our worker such a record could seldom be located in the index of 
 names in New York. If the woman had had a previous conviction 
 before being received at Auburn, however, and the exact date and 
 name used were known, her record in New York previous to that con- 
 
METHODS OF SOCIAL INVESTIGATION 73 
 
 viction could be obtained through her finger-print taken at that time. 
 In addition to applying to the two above-named bureaus, we wrote, in 
 every case where a woman had a record in another city, to the chief of 
 police in that city, to verify the woman's statement and to find out 
 whether there was any further record. 
 
 After obtaining the data from the courts and those connected with 
 the courts, the worker in most cases visited the social organizations 
 which had come into contact with the subject or her family. From 
 such organizations a full record was obtained. These data furnished 
 a background for understanding the.subject, and made the subsequent 
 visits to the immediate family more valuable. 
 
 There seem to be many advantages in thus obtaining records of the 
 court and social agencies before any members of the immediate family 
 are seen. Such information often enables the worker to question the 
 family about facts which she would know nothing of without first hav- 
 ing seen the records. Again, many families are loath to tell that the 
 woman under investigation has done certain reprehensible things and 
 had many convictions if they feel that the worker knows nothing of 
 these delinquencies, but if they understand that the worker has a 
 knowledge of what has occurred, they often tell more than could possi- 
 bly be obtained from any record. 
 
 After having seen members of the immediate family, the general 
 trend of the investigation was to interview other relatives who might 
 be able to throw light on either the subject or her family, and follow- 
 ing that to see the subject's husband or consort if there were one, 
 friends, landladies, and other persons who might be able to give any- 
 thing concerning important events which stood out in the subject's life. 
 On the whole, these latter informants are conside,red far less reliable 
 than any of those first mentioned. A landlady, for instance, often 
 gives inaccurate information concerning the number of men a girl 
 takes to her room because she realizes that she herself could be prose- 
 cuted for maintaining a disorderly house, if she admitted that she al- 
 lowed such things. In many cases, if the landlady felt the girl, had 
 given evidence which would incriminate her, she exaggerated the girl's 
 untruthfulness and dishonesty. 
 
 (b) Form of Report of Investigation 
 
 The information incorporated in the field worker's report was di- 
 vided throughout into three main parts : — 
 
74 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 I. Information concerning the subject. 
 II. Information concerning the subject's family. 
 III. Correspondence relating to the subject and her family. 
 
 This material was written up in detail, in the order in which the visits 
 were made, with all of .the information received from each informant 
 recorded separately. The first part of the report includes all of the 
 information obtained by the field worker, relating to the history of the 
 subject. Where an individual, not a member of the family, was inter- 
 viewed, a brief description of the informant was given, with the field 
 worker's impression of his reliability and accuracy. Aside from this, 
 no other extraneous data were included in Section I. A summary of 
 the order used in this section is as follows : — 
 
 I. Information Concerning the Subject. 
 
 A. Information obtained from court records, probation officer, finger- 
 prints. 
 
 B. Information obtained from social organizations which have 
 been interested in the subject. 
 
 C. Information of some member of immediate family, — mother, 
 father or older sister or brother as case may be. This informa- 
 tion, wherever possible, was made to cover the following groups 
 of facts. The grouping was very flexible, however, and in 
 many cases certain sections were omitted where there was 
 nothing of importance. 
 
 1. Infancy. Includes date and place of birth, whether or not 
 a full term child, health during infancy, age at beginning 
 to walk, talk, etc. 
 
 2. Childhood. Includes health, any abnormal traits, conduct, 
 development. 
 
 3. School Period. Age at starting and leaving school, the 
 grade reached, conduct in school, health. 
 
 4. Adolescent Period. Age menstruation was established, and 
 general health and conduct at that time. 
 
 5. Work Period. All jobs in chronological order, with kind 
 of work done, length of employment, wage, ability, and 
 disposition of wages for each job. 
 
 6. Adult Period. This section applies particularly to the older 
 women, and includes any important data on marital life, 
 delinquency, institutional commitments, etc. 
 
 7. Health. This section was used where it was desirable to 
 have a summary of health, such as cases of certain epileptics 
 in whom it was of interest to observe the development of 
 
METHODS OF SOCIAL INVESTIGATION 75 
 
 the disease. In many instances, where health was considered 
 under the various periods, it was not repeated here. 
 8. Conduct, Traits, Etc. This section was used particularly 
 where it seemed advisable to show development of traits and 
 behavior, as in many psychopathic cases, or where there was 
 a history of continued maladaptation. 
 
 D. Information of all other relatives. Facts from several inform- 
 ants were recorded separately in the order in which they were 
 obtained. 
 
 E. Information of consort or pimp. 
 
 F. Information of landlady and friends. All information under 
 this section and the following one was preceded by a state- 
 ment as to general impression and probable reliability of the 
 informant. 
 
 G. Information of certain employers. Includes information rela- 
 tive to kind of work done, ability, wage, conduct, etc. 
 
 II. Family History. All of the information under this section was 
 grouped by separate interviews, so that the statement of any unre- 
 liable or prejudiced informant could be selected from the rest. 
 
 A. Home conditions. 
 
 1. Early home during childhood and adolescence, including 
 description of economic and moral standards, parental 
 supervision, income, spirit in the home, etc. 
 
 2. Later home life, with information similar to that given 
 above, apd in addition more detailed facts regarding rent, 
 insurance, exact income, size of home, etc. 
 
 B. Information concerning the husband or consort. 
 
 1. Field worker's impression after interview. 
 
 2. Information of husband or consort, including statement 
 as to age, birthplace, education, occupation, habits, health, 
 wage, and personal history in brief. 
 
 3. Information of other informants, including court, institu- 
 tional and employment records, statements of friends, rela- 
 tives, etc. An effort was made to obtain information con- 
 cerning his characteristics, traits, and treatment of the 
 subject. 
 
 C. Information concerning the subject's children. The informa- 
 tion in this and in the other three sections follows the order 
 and general form used in Section B on Subject's Husband. 
 
 D. Information concerning the subject's father. 
 
 E. Information concerning the subject's mother. 
 
 F. Information concerning the subject's fraternity. 
 
 Wherever it was possible to do any more inten-sive work on 
 
76 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 family history, the order used by Dr. Charles B. Davenport^ 
 at Cold Spring Harbor, was followed: i.e. 
 Father's fraiernity. 
 
 Father's parents and their fraternities. 
 Mother's fraternity. 
 Mother's parents and their fraternities. 
 III. Correspondence. In general, letters were written to the following 
 sources of information, asking for the main facts m which we 
 were particularly interested. Many of these letters were written 
 not only to obtain information about the subject, but also, to ob- 
 tain institutional records, school records, etc., concerning other 
 members of the family. 
 
 A. Schools. Request was made for record card, or information 
 giving years in school, grade reached, age and date at leaving, 
 conduct. 
 
 B. Institutions, both Orphanages or Homes and Penal Institutions. 
 Request was made for cause of commitment, dates of entrance 
 and discharge, conduct. 
 
 C. Hospitals. A form letter was sent to hospitals asking for 
 length of stay, diagnosis, and any personal or family history 
 obtained. 
 
 D. Employers. Nearly all of the work records were verified by 
 correspondence, though occasionally the worker visited where 
 the subject had been employed for a long time in one place, 
 or where it was felt that the employer could give valuable 
 information bearing directly on the girl's delinquency. 
 
 E. Bureau of Records. An effort was made to verify each marriage, 
 and in many cases where there was confusion as to age, 
 to verify the date of birth. 
 
 SUMMARY OF STUDY 
 
 When the preceding information had been obtained, and the replies 
 to the letters had come in, all of the data were summarized, and re- 
 corded in condensed form on the Verified History Blank which is given 
 here. The case presented shows how the summarized verified data 
 varies from the girl's own statement which has previously been given. 
 
 This blank, it will be noticed, is comparable in form to the "State- 
 ment of Girl" blank presented in preceding pages, except that it lacks 
 certain sections on "Attitude" which were included to portray the sub- 
 ject's own feeling. This summary of the investigation was intended to 
 
 ' Davenport, Charles B. "The Family History Book.'' Bulletin No. 7, Eu- 
 genics Record Office. Cold Spring Harbor, N. Y. September, 1912. Appen- 
 dix 1, p. 95. 
 
METHODS OF SOCIAL INVESTIGATION 
 
 77 
 
 Date of commitment 2/ 8/17 
 Date of interview 2/14/17 
 
 NAME Jane Dow 
 
 VERIFIED HISTORY 
 
 N0...1 
 
 Field worker. .A. B. 
 
 OFFENSE Vagrant, haying no visible means of support, living without employment and having con- 
 tracted an inactions disease in practise of debauchery. 
 
 SENTENCE Indeterminate 3 years. 
 
 JUDGE Brown 
 
 COURT Recorder's Court, B , New York. X County 
 
 AGE AND DATE OF BIRTH 16 years, 7 months. Born June 23, 1900 ' 
 
 PLACE OF BIRTH R , New York ' RACE American 
 
 HOW LONG IN U. S. Always ' COLOR White 
 
 RELIGION Methodist i 
 
 MARITAL CONDITION Married « If married, when July 30, 1916 
 Under what name Jane Dow 
 
 Where C ,• New York 
 
 To whom John Gray 
 
 RESIDENCES AND ADDRESSES (In chronological order) 
 
 1900-1913, lived in R , New York, with parents. 
 
 1913- three weeks before arrest, lived in N , New York, with parents, except for short periods of 
 
 time in O , C— , S , and M , New York.' 
 
 Three weeks before arrest lived in furnished rooms in B , New York.* 
 
 FAMILY HISTORY 
 
 Name and address Age 
 
 Father William Dow, 3 Center St., 48 
 
 N , New York. Works steadily 
 
 and is respectable.^"'* 
 Naturalized 
 
 Mother Carrie Black, Mrs. Dow, 3 37 
 Center St., N^ — , New York. 
 Seems fairly intelligent. Is quite 
 nervous. Said to be unreliable.* 
 
 Bibs 1 
 
 1. Girl, died in infancy. 
 
 2. Subject. 
 
 3. Miscarriage at two months. 
 
 4. Charles, N— -, New York. Was 12 
 
 threatened with tuberculosis a 
 few years ago. Is said to be 
 troublesome and does petty 
 thieving. 
 
 5. John, died in 1907 at the age of 2 
 
 Had pneumonia and tubercular 
 abscess in liis neck. 
 
 6. Marietta, N , New York. 4 
 
 7. Child, born April 1917. 
 
 Birthplace Time in U. S. Occupation Education 
 New York Always In dye Reads and i" 
 
 State works writes 
 
 Left in 8th 
 grade 
 
 In 5th 
 grade 
 
 Husband John Gray, address unknown. 26 Utah 
 
 Brakeman 
 
 ?• 
 
 Children None ' 
 
 Other relatives Paternal uncle, R. M. Dow, F- 
 
 , New York.' 
 
 Institutional Record 
 
 Relative 
 None' 
 
 Institution 
 
 Dates 
 
 Complaint 
 
78 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 HOME CONDITIONS , -, imc a 
 
 During Childhood and Adolescence Haa lived with her parents most of the time until 191o- At 
 one time she went to her grandmother in Massachusetts where she remained for six months. It 
 seems probable that she has always had a fairly good home and that the mother has tried to give 
 Jane some training and supervision. The home spirit seems fairly good though the mother is 
 sometimes sharp in speaking to the children. Father has always worked steadily and has earned 
 
 a regidar wage. Two years ago the family moved from R into the country where they had a 
 
 house of eight rooms and paid *10 rent. They had a garden but did not raise much more than 
 enough for their own use. Jane always had her own room when she was a child.* A school 
 teacher who used to know the family saya that the mother was quite alack and did not keep the 
 home clean. 4 
 
 Economic conditions Probably fair because father worked steadily. Usually earned from $25 
 to $30 a week. 1 
 
 Moral standards Probably good. Parents have fairly good reputation and the father especially 
 seems to have been very strict in his moral standards. * 
 
 During Marital or Recent Life Has worked a little during the last two years but has been at home 
 the most of the time.^ When away from home last summer was married and shortly after that 
 was committed to an institution.^ Has been at home off and on since her release from the in- 
 stitution.' 
 
 Total income of family (Specify items) Father earns $3 a day. Jane earned a small and irregular 
 wage but spent it all for herself.'" 
 
 Expenses of family 
 
 Rent Parents pay $8 a month rent now.' 
 
 Insurance All of the family are insured, Jane in the Prudential & Colonial for $500.^ 
 
 Benefit societies, etc. Father belongs to Order of Foresters.^ 
 
 Miscellaneous 
 
 Character of locality (Factory, business, suburban, residence, etc.) Present home in unpreten- 
 tious but respectable neighborhood, about twenty minutes* walk from the railroad station, in a 
 country district. 2 
 
 Type of dwelling (Tenement, separate house, etc.) Two story, frame house, rather attractive 
 and in good repair.^ 
 Number of rooms 7 ' 
 
 Number of persons in household (Specify lodgers) 5, including patient.' 
 Sleeping accommodations Jane has always had a room to herself. ' 
 
 Moral standards See above. While away from home probably got in with a fast crowd of 
 people so that her moral standards became lowered.* 
 
 EDUCATION 
 
 English-speaking Yes Read and write English 
 
 School Address (Last school) Public school, R , New York.' 
 
 Age at starting school 5 years = Attendance Very poor ' 
 
 Age at leaving school (If under x6, had she obtained an employment certificate) 13 years. No 
 employment certificate. ^ 
 
 Grade at leaving school 5B gr^de.' 
 
 Reason for leaving school Moved. Was in a very nervous condition and the doctor advised that 
 she be taken out of school. ' 
 
 Later schooling None > 
 
 Language spoken at home English ^ 
 
METHODS OF SOCIAL INVESTIGATION 
 
 79 
 
 WORK HISTORY (Places of employment in chronological order and in each:) 
 
 Weekly Reason for Ability 
 Wage Leaving 
 
 Kind of work Employer and Address Dates of 
 
 employment 
 
 1. Operator Green's Embroidery Several times. $5 wk. 
 
 Factory, R , 6 or 7 weeks, 1913. 
 
 New York. 
 
 111. 
 
 2. Gen'l housework Mrs. M , R^— , 2M mos., 1913 
 
 New York. 
 
 $3 wk. 
 
 Employer 
 "queer." 
 
 ? A,i 
 
 Mrs. M , Main 4 mos., 1913. 
 
 St., R , New 
 
 York. 
 
 $2.50 wk. Moved from 7 *■ ' 
 R . 
 
 4. Housework 
 
 Mrs. M— , N^— , 5 days, Nov., 1916. $6 wk. 
 New York. 
 
 Discharged. Very poor 
 worker.^ 
 
 5. Waitress 
 
 Sanitarium, 1 week, 1016. 
 — , New York. 
 
 $18 mo. Discharged. No train- 
 
 ing. Not 
 clean. 
 Immoral. 
 
 6. Waitress 
 
 J Hospital, ( 
 
 New York. 
 
 2 weeks, Dec. 5- $20 mo. 
 23, 1916. 
 
 Disposition of Money Spent money on self.' 
 
 Discharged. Unsatis- 
 factory, 
 rude^ im- 
 pertinent 
 and indis- 
 creet with 
 men.2 
 
 niSTITDTIONAL HISTORY 
 Previous^Court Record of Delinquency (Chronological order) 
 
 Date Place Complaint 
 
 August IS, 1916 R . New York. Soliciting 
 
 Disposition 
 
 St. Ann's School.'Albany .' 
 Committed for 6 months 
 but served only 1 month. 
 
 Commitment to Non-Penal Institutions (Chronological order) 
 
 Institution (As Orphan Asylum, etc.) Dates 
 
 None 1 
 
 Record of Conduct 
 
 Hospitals 
 
 None 1 
 
 Dates 
 
 Diagnosis 
 
80 
 
 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 ^Irsf'sS ta^Lctions . When 12 years. (Menstruation was established at 12 years.)' 
 
 From whom Mother. * 
 
 Age at first sexual offense 14 years.* ,r i a 
 
 With whom Boy she knew in N , New York. 
 
 Rape or by consent Consent.*^ 
 
 Appro^o^te age at entering Prosti.ution^^Nev«^^HpJad^se.u^^^^ 
 
 ^Lrdo-^tit^K^a'nd 'o-^^^sr^g wrtdisi. "r„og 
 
 Practised contmually since .... men into empty freight cars. Chief of Pohce, . 
 
 Practised where (Hotels, furnished rooms, etc.) .... 
 Average weekly earnings from prostitution .... 
 
 HEALTH (Diseases, accidents, operations, etc.) (5' 7" tall and weighs 141H lbs. Has quite a high 
 palate. Has two carious teeth. Heart shows sUght systolic murmur. Vaginal examination 
 showed hypertrophy of external genitals and considerable yeUowish discharge. Wassermann and 
 complement fixation test for gonorrhea, negative.) = Had pamter;s colic when a baby. Cried 
 steadily for three days after birth. Has been troubled with constipation since she was a baby 
 Never had any diseases except measles and whooping cough. Began to be very nervous when 
 or 10 years old. Her face t-Adtched and she gave a little kick with one toot when she walked. 
 When 9 years old had her adenoids removed and examination of them showed presence of tu- 
 berculosis. 1 
 
 Hahits: Alcohol No *> i Drugs No *- 1 Tobacco No *. » 
 
 RECREATION (Opportunities, favorite amusements, friends, etc.) See Statement of Girl. 
 
 StJPPLEMENTARY FACTS AND EXPLANATION (Details of offense for which committed, etc.) 
 When Jane was a very young child she was inclined to be disobedient. Frequently she did not 
 come home from school promptly and would be whipped. Jane's aunt feels that because she was so 
 nervous she should not have been whipped so much and that her treatment only exaggerated her ner- 
 vous condition. She has always been quick-tempered and irritable.^'' Since leaving home in 
 1916, Jane has undoubtedly been promiscuous sexually. One employer writes of her that "she is 
 not personally physically clean and I am afraid tliat her morals are even worse." Another em- 
 ployer writes that she was very indiscreet in her attitude toward male employees. ^ She was mar- 
 ried to a man she had known only a short time and he left the night after they were married to go 
 to Texas with his regiment.^*'^ 
 
 Jane's heredity is fairly good. Her father is quite intelligent and steady. The mother 
 seems to be normal mentally and physically, though she may not have a very strict standud of 
 truthfulness. The mother has had seven children, two of whom died during infancy and one was 
 a miscarriage at two months. One of the children died of a tubercular abscess in his neck. 
 
 The details of Jane's life preceding her arrest are as follows: She did not tell her mother 
 
 that she was discharged from the J Hospital in C , New York, and after she had been home 
 
 for a short time she told her mother that she was going back to the hospital. Instead, she went 
 
 to B and had been there about a week before her arrest. She met some woman in the station 
 
 and asked her about getting a room. The woman referred her to the Y.W.C.A., which gave her 
 a list of boarding places. With this list, Jane went to one of the places and told the woman who 
 was running the house that she was doing investigations for the National Board of the Y.W.C.A. 
 She stayed with this woman for two nights. On the third night she did not come in. On the following 
 night she went to one of the other places about twelve at night and told the same story and was let 
 in. The night after that, she did not get in until three o'clock and the woman thought there 
 was something queer about it and reported it to the Y.W.C.A., who put the matter into the hands 
 
 of the police. She was found going into the D House, a poor hotel, about two o'clock at night 
 
 and although they had no very definite charge against her the police arrested her. At that time, 
 Jane told the Recorder that she had a venereal disease and so he comiriitted her here on her own 
 statement. 4 
 
METHODS OF SOCIAL INVESTIGATION 81 
 
 present a digest of the various opinions on the case. It was planned 
 primarily in an effort to get away from the usual form of having the 
 worker make out an arbitrary statement of what she considered were 
 probably the correct facts and interpretation of certain events in the 
 subject's life. It seemed necessary, in order to minimize the per- 
 sonal equation as much as possible, to have all of these data in the 
 form of a summary of the material obtained, rather than as a clear- 
 cut, one-sided picture of the subject's life summarized by the worker. 
 With the plan used, all of the cases were treated in the same way by 
 a few persons who finally worked over the data, choosing in each in- 
 stance where two contradictory statements were given, the one which 
 seemed most consistent with the general plan used for all of the cases. 
 On the "Verified History" folders, each important fact recorded was 
 followed by a small index number in red ink, which indicated by what 
 source that information had been given. This was particularly useful 
 in the data on sex history where there were often varying indefinite 
 opinions. The following illustration will show how many opinions 
 there may be on the same case : 
 
 Approximate Age at Entering Prostitution 
 
 Never (A). Eighteen years — one year ago when she ran away 
 from home (^). Nineteen years — four weeks before arrest began to be 
 wild, and stayed out late at night (*). Fifteen years — has been known 
 on the street for four years, as a prostitute (^). 
 
 Here, the subject (A) denies prostitution, while her mother (^) 
 admits that since the girl ran away a year before, she had prostituted. 
 The mother seemed frank and truthful in giving information and told 
 of places where the girl had worked steadily until a year before. These 
 we were able to verify, and found that up to that time she had been 
 doing good work, and there was no cause to suspect immoral conduct. 
 The landlady (*) with whom the girl had lived during the preceding 
 year, seemed evasive and suspicious. Not wishing to incriminate her- 
 self, she was unwilling to say that the girl had prostituted in her home, 
 but had to admit that recently she had had reason to suspect she was 
 not doing right. The chief of police ('), on the other hand, probably 
 exaggerated greatly the information he gave. For a year he had heard 
 of this girl, and perhaps before that he had noticed her, especially if 
 she had made undesirable acquaintances. In giving his information, he 
 unconsciously made her out to be much worse than she really was. 
 This we were able to prove by verification of her work and school rec- 
 
82 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 ords, which gave her a good reputation during the period the chief of 
 police in the small town had said she was prostituting. The best judg- 
 ment on the case seems then to be that she began to prostitute at eigh- 
 teen, after leaving home. By having all of these data recorded on the 
 folder, similar information on a dozen cases Would be decided alike 
 by the one working over the data, whereas a dozen persons working 
 indep.endently might have decided differently on each case. 
 
 The following index numbers were used on the verified history to 
 indicate sources of information : 
 
 A = Statement of girl. 
 
 1 = Information of mother. 
 
 (la) ^ Information of father. 
 
 (lb) = Information of sibs. If necessary to differentiate be- 
 tween conflicting information of different sibs, the follow- 
 ing distinction was used, 
 (lb 1) coldest sib. 
 (lb 2) ^next oldest sib, etc. 
 
 (Ic) =: Information of other relatives. If necessary, we indicated 
 in note which relative was informant: grandparent, aunt, 
 etc. 
 
 (Id) := Information of husband. 
 
 (le) = Information of children. 
 
 2 = Information received by directly writing to or visiting source 
 
 indicated, as: "Reached 5B grade^", indicating that information 
 
 is from the school record. 
 
 "Worked Paper Box Factory, 6 mos. $10 a week^", indicating that 
 
 we have employer's record. 
 
 "Arrest 6/5/12. T. H. L. 5 days 2", indicating that we have 
 
 finger-print record. 
 
 "Magdalen Home, 6/5/13— 6/S/14. .Good worker and behaved 
 
 welP", indicating that we have record from the Magdalen Home. 
 
 3 = Information of probation officer, as "Worked Paper Box Factory 
 
 6 mos. $10 a week^", showing probation officer has verified 
 that fact, and we are taking her verification. 
 
 (3a) Parole officer (Penitentiary records). 
 
 (3b) Detective. 
 
 (3c) Social agency. 
 
 4 = Information of landlady or persons who have lived in same 
 
 house or near the girl. 
 
 After the completion of the investigations, and the making up of 
 the "Verified History" folders, much of the material was reduced to 
 
METHODS OF SOCIAL INVESTIGATION 83 
 
 code form for use on statistical cards. This was done by a few per- 
 sons working together, so that the work was uniform. The general 
 considerations as to the classifications used for certain groups of facts 
 will be discussed later under the chapters dealing with such facts. 
 
 The sociological investigation, in brief, aimed to make an intensive 
 study of the women delinquents considered, not only through know- 
 ing the woman herself, but by becoming informed about the en- 
 vironment in which she grew up and her manner of living outside 
 the prison. A particular effort has been made to weigh carefully 
 the information obtained concerning her, and to use similar meth- 
 ods of recording and summarizing data on all cases. This has been of 
 great value in making the work in the various groups comparable so far 
 as methods of obtaining and evaluating the material are concerned. 
 
CHAPTER VI 
 NATURE AND EXTENT OF DELINQUENCY 
 
 THE study of the association of various environmental factors in 
 the life of a criminal with the nature and extent of crime must 
 necessarily be preceded by a clear understanding of the classification of 
 crime, and the measure of degree of criminality to be used. Many 
 criminal sociologists within the last thirty years have made studies of 
 the environmental influences in the lives of criminals, but have treated 
 their material in a descriptive manner, and have assumed that the fre- 
 quency of an abnormal environmental condition among a group of 
 criminals was proof that the condition was a cause of the delinquency. 
 Often no attempt has been made to show how the relationship varied 
 with any measure of the extent of delinquency, or with the frequency 
 of this condition in the general population. One of the first difficulties, 
 it must be granted, in treating in mass form the material in such a 
 study as this, is the inability to obtain comparative figures for the gen- 
 eral population. Until such figures can be obtained, however, it seems 
 futile to assume that the repeated occurrence of a condition within a 
 criminal group is a cause of criminality, when it may be a commonly 
 occurring condition among the general population as well. Not having 
 the figures for the larger group makes it all the more necessary that 
 we show the relationships of observed factors to the degree of delin- 
 quency and treat such relationships as concomitant with, not neces- 
 sarily causative of, delinquency. 
 
 In this study, a delinquent has been regarded as one who has been 
 convicted of an offense against the law. The measurement of the ex- 
 tent of delinquency might be based on one or more of four factors, 
 (1) the age at first conviction, (2) number of convictions, (3) number 
 of terrns served in penal institutions, or (4) length of time served in 
 penal institutions. This chapter aims to give a picture of the criminal 
 record of the women in the various groups studied, and will discuss 
 the nature of the offense committed by the women at the time of the 
 present study, the advantages or disadvantages in using one or more 
 of the above-named measures of delinquency, and the factors connected 
 
 84 
 
NATURE AND EXTENT OF DELINQUENCY 85 
 
 with the first contact with the law. We shall attempt to show the 
 relationships beween several of the elements entering into the criminal 
 record, and shall use the material in this chapter as a basis for later 
 showing the relationship between extent of delinquency and various 
 environmental conditions. 
 
 CLASSIFICATION OF OFFENSES 
 
 The problem of classifying the offenses is no small one. As Miss 
 Mary Conyngton^ writes, — "the classification of offenses is a complex 
 affair. Going from place to place, one finds identical terms used for 
 different offenses and identical offenses described by different terms. 
 Thus, in one locality if a woman is convicted of keeping a disorderly 
 house it may mean only that a considerable amount of loud talking, 
 singing, and quarreling goes on there which makes it a nuisance to the 
 neighbors. In another it invariably means that she maintains a house 
 of ill fame. 'Violation of city ordinance' is a comprehensive term,' cov- 
 ering offenses which may range from a serious infraction of public 
 • order to shaking a rug from a front window. Obviously any attempt 
 to classify according to the charges under which the prisoners are com- 
 mitted would lead to grouping together widely varying offenses." 
 Though our study did not cover convictions outside of New York 
 State, it was found that within one state alone the classification of of- 
 fenses varied considerably in different localities, and somietimes within 
 the same city. In particular, the use of "Disorderly Conduct" is strik- 
 ing, as meaning, even in New York City, drunkenness, soliciting, at- 
 tracting a crowd on the street, soliciting alms, etc. In all cases where 
 "Disorderly Conduct" unqualified was given as the offense, therefore, 
 it was necessary to find out from the arresting officer what specific 
 violation of the law had taken place. There are many other instances 
 of vague and indefinite returns of offense committed, which necessi- 
 tated careful investigation of the details of the affair, so that the classi- 
 fication might be as definite as possible. 
 
 The New York City Police Department classification of crime ^ 
 was chosen as being the best adapted to show the general relationships 
 in our groups of delinquent women. This system of classification was 
 made up on the basis of crimes which had passed through the New 
 York City Police Department, and, therefore, took account of the local 
 
 ^"Report on Condition of Woman and Child Wage-earners in the United 
 States." Vol. XV. "Relation between Occupation and Criminality of Women." 
 By Miss Mary Conyngton. Washington Gov't. Printing Office, 1911, p. 16. 
 
 ' Police Department, City of New York, Annual Report, 1916, pp. 30-64. 
 
86 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 idiosyncrasies in many laws, such as the specific sections of a general 
 law which might be lost sight of in a more universal classification. An 
 instance of this occurs in what is commonly known as the "Vagrancy 
 Law." Under a more general application, undoubtedly many of the 
 persons convicted under the general heading entitled "Vagrants" would 
 be so classified under a somewhat miscellaneous grouping. To one 
 who knows this particular law in New York State, however, it is evi- 
 dent that very specific charges can be made in each of the ten sections 
 of the law. The New York City Police Department classification has 
 taken cognizance of the fact that section 1 defines as vagrants those 
 who "living without employment, have not visible means to maintain 
 themselves ;" that section 2 defines habitual drunkards ; that section 3 
 refers to those who have contracted an infectious disease in the prac- 
 tice of debauchery, etc. ; and that section 4 defines those who are guilty 
 of prostitution and soliciting.' Recognizing the sections of the law 
 which are given in, the record of conviction, the Vagrancy heading has 
 been regrouped into "Vagrancy intoxication," "Vagrancy prostitu- 
 tion," etc., so that each falls into the large group where it would prop- . 
 erly belong. 
 
 TABLE 1 
 
 COUNTIES IN WHICH DELINQUENT WOMEN WERE 
 CONVICTED 
 
 
 Institutional Groups 
 
 Total 
 
 County 
 
 Bed- 
 ford 
 
 Auburn 
 
 Mag- 
 dalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Proba- 
 tion 
 
 Num- 
 ber 
 
 Per 
 
 cent 
 
 Counties included in 
 Greater New York — 
 
 Counties outside of 
 Greater New York 
 
 67 
 
 35 
 
 57 
 31 
 
 74 
 2 
 
 110 
 
 107 
 2 
 
 102 
 
 517 
 70 
 
 88.1 
 11.9 
 
 Total 
 
 102 
 
 88 
 
 76 
 
 110 
 
 109 
 
 102 
 
 587 
 
 100.0 
 
 
 
 Of the group of delinquent women under consideration, Table 1 
 shows that 517 or 88.1 per cent were convicted in the five counties in- 
 cluded in Greater New York. The Police Department has prisoners 
 from each of these counties, thereby making it possible to fit the ma- 
 
 ""Code of Criminal Procedure, State of New York." Title VI, § 887, sec- 
 tions 1, 2, 3, 4. 
 
NATURE AND EXTENT OF DELINQUENCY 
 
 87 
 
 jority of our cases into the details of a scheme which was worked out 
 from returns of convictions in the counties which represented our 
 group most largely. 
 
 TABLE 2 
 
 KIND OF COURT 
 
 Nxunber of Delinquent Women in Institutional Groups and Number 
 
 and Per Cent Distribution for Total, Classified by Kind of 
 
 Court in Which Convicted 
 
 
 Institutional Groups 
 
 Total 
 
 Kind of Court 
 
 Bed- 
 ford 
 
 Auburn 
 
 Mag- 
 dalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Proba- 
 tion 
 
 Num- 
 ber 
 
 Per 
 cent 
 
 1. Children's Court... 
 
 2. City Court 
 
 3. City Magistrates' 
 
 Court 
 
 3 
 
 47 
 7 
 
 5 
 
 7 
 9 
 9 
 
 13 
 
 2 
 
 33 
 31 
 
 23 
 
 1 
 1 
 
 66 
 2 
 
 6 
 
 2 
 12 
 
 32 
 
 63 
 1 
 
 97 
 12 
 
 102 
 
 1 
 4 
 
 314 
 52 
 
 70 
 
 7 
 9 
 9 
 
 94 
 26 
 
 .2 
 
 .7 
 
 53.6 
 
 4. Coimty Court 
 
 5. General Sessions 
 
 Court 
 
 8.9 
 11.9 
 
 6 . Justice of the Peace 
 Court 
 
 1.2 
 
 7. Police Court 
 
 8. Recorder's Court . . . 
 
 9. Special Sessions 
 
 Court 
 
 1.5 
 1.5 
 
 16.0 
 
 10. Supreme Court 
 
 4.4 
 
 Total 
 
 102 
 
 87 
 
 76 
 
 110 
 
 109 
 
 102 
 
 586 
 
 100.0 
 
 
 
 
 
 Table 2 shows the distribution of the women in our group by the 
 courts in which they were convicted. The City Magistrates' Courts, it 
 will be seen, have the largest number of cases, with Special Sessions, 
 General Sessions and County Court following. The one case con- 
 victed in Children's Court was very nearly sixteen years old, and 
 needed continued institutional supervision, so the judge committed her 
 to the Magdalen. 
 
88 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 NATURE OF PRESENT OFFENSE 
 
 (a) Classified by New York City Police Department Classification 
 Table 3 shows for each of the institutional groups studied, the 
 nature of the present offense, classified by the New York City Police 
 Department classification. The classification is flexible enough so that 
 various offenses may be included under either of two or more general 
 headings as necessity arises for doing this. Assault, for instance, is 
 ordinarily classified as an offense against the person. When the as- 
 sault, however, is with intent to commit rape, it becomes an offense 
 against chastity. In classifying our offenses, not only the offense of 
 which the delinquent was convicted was taken into consideration, but 
 also the actual nature of the crime, about which we were able to learn 
 by investigation. There is, therefore, no need of falling into the diffi- 
 culty cited by Dr. Glueck* when he says, — "there seems to be no valid 
 reason for following the legal classification. This classification is fre- 
 quently not in accord with the actual nature of the crime, as is the 
 case, for example, when a man who has committed rape is permitted 
 to plead to assault. Furthermore, the legal classification, already ex- 
 tensive, is growing constantly with the addition of new legislation, so 
 that data obtained to-day may lose in value by to-morrow. It has 
 been found, on the other hand, that as a result of the examination car- 
 ried out in these cases, a classification on the basis of motive was pos- 
 sible in practically all instances, and the inception of the motive could 
 be traced to one of the several fundamental human, instinctive atti- 
 tudes." 
 
 With the group of delinquent women studied, it was found to be im- 
 possible in many cases to attempt a classification of offenses on the 
 basis of motive. Whether or not the girl who solicits a man and takes 
 him to a "creep house" where he is robbed, is guilty of a crime having 
 its impulse in the instinct of acquisitiveness or of sex is very often 
 hard to determine. Another type of case difficult to decide upon is 
 found in the history of a Polish peasant woman sent, to Auburn for 
 twenty years for murder in the second degree. Her husband was a 
 drunkard and would not support her. Neither would he allow her to 
 work and send money to her four children in Russian Poland. After 
 he had threatened to kill her one night, and had scoffed at her for 
 worrying over her children, whom she had heard were starving, she 
 
 * Mental Hygiene, January, 1918, vol. II, no. 1. "A Study of 608 Admis- 
 sions to Sing Sing Prison," by Bernard Glueck, p. 92. 
 
NATURE AND EXTENT OF DELINQUENCY 89 
 
 killed him in a most brutal manner. The next day, before the mur- 
 der had. been discovered, she tried to find work so she could get money 
 for her children. The instinct that prompted the killing was difficult 
 to determine, since, viewed from various viewpoints, it may have been 
 self-protection, protection of children, or possibly a "crime which had 
 its impulse in the instinct of pugnacity." There are other numerous 
 examples among the offenses committed by the Auburn women, par- 
 ticularly, which were not easy to attribute to any one motive, so that a 
 plan such as that used by Dr. Glueck did not seem feasible. 
 
 In order to show the relative seriousness of the offenses committed, 
 a further classification into felonies and misdemeanors was used. Many 
 of the petty offenses, such as violations of city ordinances, which are 
 not, in a strictly legal sense, misdemeanors, were, however, classed as 
 such, since we wanted to show by felonies the type of crime whicJi 
 may be punishable by a state prison term or death, and by misde- 
 meanors the less serious offenses which call for sentences in pri- 
 vate institutions, reformatories, probation, etc., but which can not be 
 sentenced to a state prison. Often the distinction is only a question of 
 degree, as, for instance, if a man steals goods valued at $25 from a 
 store during the day time he is guilty of petit larceny, a misdemeanor 
 for which he can not be sent to a state prison. If, however, he takes 
 one cent more, making the amount over $25, from the home of the 
 owner of the store, at night time, he is guilty of grand larceny, first 
 degree, and may be punished by a term not exceeding ten years in state 
 prison. Obviously, the distinction between the two is absurd, so far as 
 motive, intent and real seriousness of offense are concerned. With the 
 cases which fall near the dividing line of felony and misdemeanor, it is 
 true that any attempt at classification into these two large groups is 
 worth little, except in a legal sense. It seemed advisable, however, to 
 use such a classification because the cases near the border-line of such 
 a distinction are comparatively few, and because by distinguishing the 
 felony cases, we have material comparable to the studies of men in state 
 prisons, as Sing Sing and Auburn, which include only felons. 
 
 Both the New York City Police Department classification of of- 
 fenses and the grouping into felonies and misdemeanors have been 
 used in considering the nature of the present offense and the first 
 offense. Table 3 indicates in detail the nature of the present offense 
 for the 587 women in our study, and shows what types of crimes are 
 included under the general headings. 
 
 Under "Offenses against the Person," Auburn Prison has delin- 
 
90 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 3 
 
 NATURE OF PRESENT OFFENSE 
 
 Number of Delinquent Women in Institutional Groups, Classified by 
 New York City Police Department Classification 
 
 Nature of Offense 
 
 Institutional Gbotjpb 
 
 Bed- 
 ford 
 
 Aiiburn 
 
 Mag- 
 dalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Proba- 
 tion 
 
 Total 
 
 Offenses against the Person . . . 
 
 Abortion 
 
 Assault 
 
 Manslaughter 
 
 Murder 
 
 Suicide, attempted 
 
 Offenses against Chastity 
 
 Abduction 
 
 Adultery 
 
 Associating with vicious and 
 dissolute persons 
 
 Committing lewd and in- 
 decent acts 
 
 Compulsory prostitution. . . . 
 
 Disorderly house, keeping. . . 
 
 Exposure of person 
 
 Incest 
 
 Loitering 
 
 Prostitution, general 
 
 Prostitution in tenement 
 house 
 
 Rape, 2nd degree 
 
 Sodomy 
 
 Soliciting 
 
 Vagrancy: contracting infec- 
 tious disease in practice of 
 debauchery 
 
 Offenses against Family and 
 Children 
 
 Abandonment of child under 
 fourteen 
 
 Bigamy 
 
 Endangering health and mor- 
 als of children 
 
 64 
 
 1 
 1 
 
 ii 
 1 
 
 29 
 
 8 
 
 22 
 
 59 
 
 18 
 1 
 
 "i 
 
 17 
 22 
 
 91 
 
 63 
 26 
 
 89 
 
 81 
 
 31 
 
 4 
 12 
 9 
 5 
 1 
 
 316 
 
 2 
 1 
 
 31 
 
 1 
 1 
 7 
 1 
 1 
 
 13 
 1 
 
 1 
 
 1 
 
 159 
 
 12 
 
NATURE AND EXTENT OF DELINQUENCY 
 
 91 
 
 TABLE 3— Continued 
 
 
 
 
 Nature of Offense 
 
 Bed- 
 ford 
 
 Auburn 
 
 Mag- 
 dalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Proba- 
 tion 
 
 Total 
 
 Offenses against Regulations for 
 Public Health and Safety. . 
 
 6 
 
 1 
 
 9 
 
 30 
 
 3 
 
 7 
 
 56 
 
 Habitual drunkard 
 
 "i 
 
 "i 
 
 1 
 
 2 
 
 "i 
 1 
 
 1 
 4 
 4 
 
 '27 
 
 3 
 
 "3 
 
 "7 
 
 ... 
 
 1 
 
 Intoxication 
 
 11 
 
 Possessing druss 
 
 38 
 
 Possessing and selling drugs. . 
 
 Violating liquor tax aw 
 
 Yelling and disturbing public 
 
 4 
 
 1 
 
 1 
 
 Offenses against Administra- 
 tion of Government 
 
 3 
 
 
 "2 
 
 17 
 
 1 
 54 
 
 4 
 
 62 
 
 5 
 
 ... 
 
 1 
 
 Perjury 
 
 2 
 
 Offenses against Property 
 
 Rights 
 
 142 
 
 
 
 Arson. 
 
 "■3 
 14 
 
 6 
 
 2 
 
 "i 
 
 1 
 
 40 
 1 
 3 
 6 
 
 "i 
 1 
 2 
 
 3 
 
 2 
 1 
 
 1 
 
 2 
 
 'is 
 
 41 
 1 
 
 "i 
 2 
 
 "5 
 10 
 
 6 
 
 4 
 
 Breaking window while in- 
 toxicated 
 
 1 
 
 Bringing stolen property into 
 
 1 
 
 Burelarv 
 
 2 
 
 Extortion 
 
 1 
 2 
 
 
 67 
 
 
 63 
 
 Receiving stolen property — 
 
 4 
 6 
 
 Throwing stones at street car 
 OpTiprfll C3riminalitv 
 
 1 
 27 
 
 
 
 Disorderly conduct, unquali- 
 fied 
 
 ■3 
 3 
 
 
 "3 
 
 "2 
 
 5 
 2 
 
 ■3 
 
 "6 
 
 5 
 
 Disorderly conduct, soliciting 
 alms 
 
 2 
 
 Posspssinff Distol 
 
 2 
 
 
 9 
 
 Vagrancy, unqualified 
 
 9 
 
 Total 
 
 102 
 
 88 
 
 76 
 
 110 
 
 109 
 
 102 
 
 587 
 
 
 
92 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 quents under each one of the sub-headings listed, and contains the 
 more serious offenses such as murder and manslaughter, which do not 
 appear in other groups.^ 
 
 Under "Oifenses against Chastity," the distribution through all the 
 groups is more general. But for a few exceptions, the Penitentiary 
 and Auburn Prison have the more serious oifenses of abduction, adul- 
 tery, compulsory prostitution, incest, sodomy and rape. The offenses 
 in the other four institutions fall largely into the prostitution offenses, 
 including soliciting, loitering, and violating the Tenement House Law. 
 
 Among "Offenses against the Family and Children," three types 
 of offenses occurred among our cases. Under the "Abandonment of a 
 child under fourteen" there were only two cases, one in Bedford and 
 one in Auburn. The Bedford girl was twenty years old, Polish, and 
 spoke very little English. She was of imbecile grade mentally, and 
 showed her lack of intelligence in the act which led to her arrest. Her 
 consort, the child's father, refused to work, and she thought that she 
 had to support him. She knew she would be able to do more work if 
 she did not have the care o£ the child, and so she took it to a vacant lot 
 near her home, when it was four weeks old, and buried it under a pile 
 of ashes. When the child was found several hours later, the mother 
 showed no feeling and was willing to tell just what she had done. The 
 Auburn case was that of a woman twenty-eight years old, who with her 
 husband abandoned their four children because they thought they 
 were not able to support them. Two of the babies were left in a hall- 
 way and one of them died because it was exposed to the rain and cold. 
 The other two were abandoned in a store and taken charge of by the 
 Children's Society. The Polish girl might have been sent to Auburn, 
 but probably because of her youth, her foreignness and stupidity, which 
 to a casual observer made her appear innocent and frightened, the 
 judge sentenced her to Bedford. Ruling out the intelligence factor, 
 her motive, however, was as reprehensible, since she intended to kill 
 her child, while the Auburn woman wanted to shift the responsibility 
 for the care of her child to some other person. 
 
 The bigamy case in Bedford differed in no radical way from the 
 three who were sentenced to Auburn, except that she was pretty, at- 
 tractive, and was clever enough to make her plea of her own good 
 
 "The one manslaughter case which appears in Bedford was first sentenced 
 to Auburn Prison. The two women against whom she had turned state's 
 evidence threatened to kill her while she was in the prison, and so she was 
 brought back to court for resentence and committed to Bedford. 
 
NATURE AND EXTENT OF DELINQUENCY 93 
 
 motives, but her great gullibility, carry weight with the probation officer 
 and judge. 
 
 "Endangering the Health and Morals of Children" included in the 
 Bedford group two women who had been prostitutes and heavy drink- 
 ers for a long time, but who could not be caught by the police until 
 one day when they were found drinking and entertaining several 
 women of bad character, in the presence of three small children. One 
 other Bedford case was an Austrian woman who persisted in locking 
 her two small children in her apartment for the day while she went 
 out to work, and neglecting them in other ways. The fourth Bedford 
 case was that of a girl sixteen years old who had sexual relations with 
 a strange man in the presence of her fourteen year old sister. The 
 Magdalen and Penitentiary cases are similar to the two Bedford cases 
 first noted. 
 
 "Offenses against Regulations for Public Health and Safety" are 
 supposed to include mainly the intoxication and drug cases. Since the 
 using of alcohol or drugs is never a felony, we should not expect to 
 find any of these cases in Auburn Prison. Selling drugs, however, is 
 a felony and we find one such case in Auburn. There are, it will be 
 noticed, no convictions for drug using in the Probation group. This 
 was determined by the court in which the Probation cases were studied, 
 since the Women's Night Court does not have jurisdiction over drug 
 charges.* 
 
 "Offenses against the Administration of Government" are repre- 
 sented by only two sub-heads in our group. The two perjury cases in 
 Bedford were both young girls who had perjured themselves in court 
 but who had been given a chance on probation. The perjury charges 
 were brought against them in the first place, largely because there was 
 no other charge on which to hold them, and it was felt that they 
 needed supervision. When they violated their probation rules they 
 were brought back to court and were resentenced to Bedford. The per- 
 jury involved was no more serious than is committed many times a 
 day by cases against whom no perjury charge is lodged. Accordingly 
 these offenses should not be thought of as being so unusual and im- 
 portant as they might seem from their title and the smallness of the 
 
 ' It is important to remember that the great bulk of intoxication cases who 
 would come under the head of "^Offenses against Regulations for Public Health 
 and Safety" were omitted from our group. Had these been included we 
 should have a second large sub-division in our Workhouse group. In Chart I 
 the bar representing these offenses would at least equal in length that indicat- 
 ing offenses against chastity, these two types of offenses making up the great 
 body of the Workhouse population. 
 
94 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 numbers. The bribery case in Auburn was that of an old woman who 
 gave an officer forty dollars with intent to influence him to release her 
 from custody. Her case was probably no more serious than either of 
 the two Bedford perjury cases, but the age factor made it obligatory on 
 the judge to send her either to the Penitentiary or to Auburn. 
 
 The group of "Offenses against Property Rights," we may note at 
 the outset, is not represented in the Probation group since the Women's 
 Night Court has no jurisdiction over such cases. The Workhouse has 
 only five cases and those all of petit larceny. In the Bedford group, 
 both grand and petit larceny cases are found. One of these grand 
 larceny cases was first sentenced to Auburn but the sentence was 
 changed later to a Reformatory term. The Auburn group of cases 
 includes various types of offense, such as arson, extortion, forgery, re- 
 ceiving stolen property, and robbery, in addition to grand and petit 
 larceny.'' The range of offenses against property rights in the Mag- 
 dalen is very small, covering only the three offenses of forgery, grand 
 and petit larceny. In the Penitentiary, the range covers air of the 
 offenses included in any other group, except extortion, forgery and 
 robbery. 
 
 The last group, called "General Criminality" in the New York 
 City Police Department classification, includes offenses difficult to 
 place in any scheme of classification. Such general terms as dis- 
 orderly conduct and vagrancy, with no qualifying statements, in 
 cases on which we were unable to get further information, could 
 not be put in any logical grouping. "Ungovernable child" should 
 rightly fall in more closely with a group 'of juvenile offenses, but 
 since the women under consideration are adults we have put the few 
 semi-juvenile offenses into the miscellaneous group. Possessing a 
 pistol and disorderly conduct, each represented by only two cases, also 
 seem to fit into none of the other divisions, and following the New 
 York City Police Department classification are treated as "General 
 Criminality." 
 
 Table 3 shows, then, that the specific offenses under the seven 
 main divisions are divided unevenly among the six institutional groups. 
 We should expect this, since the institutional groups we have selected 
 differ in the kind of commitments permitted them by law. That is, 
 Auburn may receive only felonies, the Workhouse and Probation only 
 
 'The one petit larceny case in Auburn was an illegal commitment, since a 
 person convicted of a misdemeanor may not be sent to state prison. This 
 woman was indicted on a charge of grand larceny, first degree, but pled guilty 
 to and was convicted of petit larceny. 
 
NATURE AND EXTENT OF DELINQUENCY 
 
 95 
 
 misdemeanors, and the other three institutions both felonies and mis- 
 demeanors, though the Penitentiary has a much larger percentage of 
 its cases felonies than have Bedford or the Magdalen. 
 
 In order to give in large the main divisions of the New York 
 City Police Department classification, Table 4 has been made on a 
 percentage basis giving only the totals of each division, in per cent, 
 for each institutional group and for the total group. 
 
 TABLE 4 
 
 NATURE OF PRESENT OFFENSE 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups, 
 
 Classified by Main Divisions of New York City Police 
 
 Department Classification 
 
 Nature of Offense 
 
 Institutional Groups 
 
 Bed- 
 ford 
 
 Auburn 
 
 Mag- 
 dalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Proba- 
 tion 
 
 Total 
 
 Offenses against the Person . . . . 
 Offenses against Chastity 
 
 Offenses against Family and 
 Children 
 
 Offenses against Regulations 
 for Public Health, Safety 
 and Policy 
 
 Offenses against Administration 
 of Government 
 
 Offenses against Property 
 Rights 
 
 General Criminality. 
 
 1.0 
 62.7 
 
 5.9 
 
 5.9 
 
 2.0 
 
 16.7 
 5.9 
 
 25.0 
 6.8 
 
 4.6 
 
 1.1 
 
 1.1 
 
 61.4 
 
 77.6 
 1.3 
 
 11.8 
 
 5.3 
 3.9 
 
 7.3 
 6.4 
 
 .9 
 27.3 
 
 56.4 
 
 1.8 
 
 83.5 
 
 2.7 
 
 4.6 
 9.2 
 
 87.2 
 
 6.9 
 
 5.9 
 
 5.3 
 53.8 
 
 2.0 
 
 9.5 
 
 .5 
 
 24.2 
 4.6 
 
 Total . 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Chart 1 shows in graphic form the percentage figures which are 
 given in Table 4. The most striking thing in the chart is, first of 
 all, the great irregularity within each group, and between any two of 
 the institutional groups which may be used for comparison. As has 
 been stated, this is due largely to the unevenness of distribution of 
 felonies and misdemeanors among the main divisions of the New 
 York City Police Department classification (See Chart 3), and the 
 
Ll 
 
 8 1 '■■■■■ 
 
 8 ■' ■ " 
 
 I .> 
 
 8 S. 
 
 1 I 
 
 
 
 
 
 
 
 
 
 ■ 1 . 1 1 
 
 d s 
 
 , ■■ 
 
 
 
 
 
 1 
 
 
 .. 
 
 
 o 
 
 
 
 
 
 
 
 
 
 
 
 .1 .. 
 
 iht in 
 
 °iUi ill 
 
 96 
 
NATURE AND EXTENT OF DELINQUENCY 97 
 
 distribution of felonies and misdemeanors among our institutional 
 groups. It will be seen at a glance that the offenses against chastity 
 constitute the largest percentage of offenses of any one group, except 
 in Auburn and the Penitentiary which have the highest percentage of 
 felony cases. In these two groups, again, the offenses against property 
 rights constitute the largest percentage. The second largest per- 
 centage of offenses in Auburn is offenses against the person, made 
 up of 93.5 per cent felonies. In the Penitentiary, offenses against 
 regulations for public health, etc., have second place, and they are made 
 up of only 7.1 per cent felonies. This high percentage in the Peni- 
 tentiary may be partly explained by the fact that though possessing 
 drugs, which is the largest factor in the offenses against regulations 
 for public health, is legally a misdemeanor, it is considered one 
 of the more serious misdemeanors and is therefore punished by one 
 of the more serious and longer sentences, i.e., the indeterminate 
 sentence at the Penitentiary. The Bedford group of cases which is 
 third highest in percentage of felons, has, as might be expected, a 
 representation in each division of offenses of the New York City 
 Police Department classification. The three groups of delinquents 
 having the highest percentage of misdemeanors, the Magdalen Home, 
 the Workhouse and the Probation group, have, naturally, few or 
 no cases in the divisions of the New York City Police Department 
 classification largely made up of felons. That is, in none of these 
 three groups are there any cases among the offenses against the 
 person or against the administration of government, and only in 
 the Magdalen group are there a small percentage of offenders against 
 property rights. 
 
 (b) Classified by Felonies and Misdemeanors 
 
 Table 5 in connection with Chart 2 will indicate how the in- 
 stitutional groups are divided into felonies and misdemeanors. 
 
 As stated in a footnote on page 94, the only misdemeanor case 
 
 in Auburn was not a legal commitment. This one petit larceny case, 
 
 therefore, gives a slightly erroneous impression in the graph, since 
 
 one would naturally expect to see the entire Auburn bar marked as 
 
 felony. Bedford includes only seven felony cases, or 6.9 per cent 
 
 of her total number.® The Magdalen with two women, or 2.6 pef 
 
 ' The decrease in percentage of felons in the Reformatory at Bedford 
 during the last few years is very marked. There have been as high as 32.2 per 
 cent of the Bedford women, felons, but since 1907 there has been a steady 
 and consistent decrease. 
 
i 
 i 
 
 
 
 ca 
 
 
 
 •a 
 
 
 
 o 
 
 
 
 
 
 
 « 
 
 
 
 In 
 
 
 
 >> 
 
 
 
 ^ 
 
 
 
 •^ 
 
 
 
 4) 
 
 
 
 59 
 
 
 
 
 
 P4 
 
 jfl 
 
 
 rrt 
 
 *S 
 
 
 S 
 
 u 
 
 
 
 •« 
 
 
 fo 
 
 Ct< 
 
 
 P^ 
 
 
 
 O 
 
 g 
 
 
 H 
 
 c^ 
 
 lO 
 
 fe 
 
 "ci 
 
 M 
 
 Cfl ' 
 
 Q 
 
 § 
 
 S 
 
 1 
 
 ei 
 
 1^ 
 
 
 
 (X4 
 
 rt 
 
 
 O 
 
 3 
 
 .a 
 
 
 ■H 
 
 Q 
 
 
 g 
 
 ^ 
 
 4) 
 
 0) 
 
 
 
 S 
 
 >o 
 
 UJ 
 
 o. 
 
 i 
 
 
 
 1 
 
 i-( 
 
 00 
 
 d 
 
 
 
 
 <N 
 
 l> 
 
 o 
 
 
 
 <o 
 
 
 
 i-H 
 
 c 
 
 
 eu 
 
 
 
 
 fc 
 
 
 
 
 e- 
 
 
 1 
 
 ;z; 
 
 to 
 
 »H 
 
 I> 
 
 
 
 N 
 
 to 
 
 00 
 
 
 
 1—1 
 
 ■* 
 
 lO 
 
 
 
 
 
 
 
 
 ■g 
 
 
 
 
 
 
 
 
 o 
 
 o 
 
 
 g 
 
 « 
 
 
 d 
 
 d 
 
 
 
 o 
 
 
 o 
 
 o 
 
 
 1 
 
 Ph 
 
 
 rH 
 
 i-H 
 
 (h 
 
 
 
 
 
 £ 
 
 1 
 
 
 
 2 
 
 
 ■la 
 
 
 
 
 
 
 a 
 
 
 o 
 
 o 
 
 
 £ 
 
 S 
 
 
 
 
 
 DQ 
 
 g 
 
 
 d 
 
 o 
 
 
 S 
 
 
 o 
 
 
 
 1 
 
 P^ 
 
 
 1— ( 
 
 1-t 
 
 »i< 
 
 
 
 
 
 o 
 
 1 
 
 
 m 
 
 OS 
 
 
 1 
 
 
 o 
 
 o 
 
 
 a 
 
 
 rH 
 
 1-H 
 
 
 
 z 
 
 
 
 
 
 ■a 
 
 
 
 
 
 !>»■ 
 
 
 CO 
 
 t^ 
 
 o 
 
 
 ^ 
 
 o 
 
 1> 
 
 IN 
 
 o 
 
 m 
 
 
 c 
 
 (M 
 
 t~ 
 
 o 
 
 ft 
 
 "S 
 
 Ph 
 
 
 
 T-l 
 
 p 
 
 « 
 
 
 
 
 
 S 
 
 
 1 
 1 
 
 O 
 
 O 
 
 o 
 
 
 CO 
 
 00 
 
 1-1 
 
 5 
 
 
 
 
 
 
 
 s 
 
 s 
 
 (N 
 
 
 o 
 d 
 
 
 13 
 
 s 
 
 
 OS 
 
 o 
 
 1 
 
 Ph 
 
 
 
 rH 
 
 .a 
 
 (N 
 
 ■* 
 
 CD 
 
 
 
 1> 
 
 1> 
 
 
 
 a 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 ti 
 S 
 
 05 
 
 r-t 
 
 o 
 
 
 g 
 
 00 
 
 iH 
 
 d 
 
 
 V 
 
 OS 
 
 
 o 
 
 
 p 
 
 Pm 
 
 
 
 i-( 
 
 
 1 
 
 1 
 1 
 
 S5 
 
 1-t 
 
 00 
 00 
 
 
 
 
 
 
 
 
 i 
 
 o> 
 
 iH 
 
 o 
 
 
 "O 
 
 u 
 
 S) 
 
 CO 
 
 d 
 
 
 g 
 
 s 
 
 
 05 
 
 o 
 
 
 <M 
 
 PlH 
 
 
 
 rH 
 
 1 
 
 t- 
 
 OS 
 
 § 
 
 
 
 B 
 
 
 
 T-H 
 
 
 
 iz; 
 
 
 
 
 
 
 
 
 
 
 
 
 a 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 o 
 
 
 
 
 
 
 
 •t-l 
 
 
 
 
 
 
 
 o 
 
 
 
 
 
 
 
 1 
 
 
 
 g 
 
 
 
 
 
 
 c 
 
 1 
 
 
 !? 
 
 
 
 ! 
 
 ■g 
 
 
 
 
 J 
 
 H 
 
 
 
 
 
 ■^ 
 
 ! 
 
 
 
 
 
 .? 
 
 
 
 
 
 1 
 
 : ^ 
 
 
 
NATURE AND EXTENT OF DELINQUENCY 
 
 99 
 
 cent of her total cases, felons, indicates that there is much less of a 
 tendency to send the more serious offenders to an institution which 
 is designed primarily for the younger and more hopeful girls. The 
 Penitentiary with thirty felons, or 27.3 per cent of its total, ranks next 
 
 BEDFORD 
 
 AUBURN 
 
 MAGDALEN 
 
 PENITENTIARY 
 
 WORKHOUSE 
 
 PROBATION 
 
 PER CENT 
 <2 60 80 ICO 
 
 T 
 
 MISDBMEANORS 
 
 P 
 
 I FELONIES L ' 
 
 Chart II 
 
 Nature of Present Offense 
 
 Per cent distribution of felonies and misdemeanors by institutional groups. 
 
 to Auburn in having the largest number of serious offenders. The 
 Workhouse and Probation groups are entirely made up of mis- 
 demeanants. 
 
 (c) Felonies and Misdemeanors in New York City Police Department 
 
 Classification 
 
 Table 6, presented graphically in Chart 3, indicates what per- 
 centage of each of the divisions of the New York City Police De- 
 partment classification are felons and misdemeanants. 
 
 As discussed on page 97, we may note that the offenses against 
 the person, a total of only 31 cases, have only two misdemeanants 
 or 6.5 per cent of its total number. The offenses against chastity, 
 however, with a liberal sample of 316 cases, contain 310 misdemeanants 
 or 98.1 per cent of its total. The group of offenders against the 
 family and children, as well as the offenders against the administration 
 of government, have very few cases on which to base any percentages, 
 but as the figures stand, the percentage of misdemeanants is SO per 
 cent in the one, and 33.3 per cent in the other. Out of 56 cases, 
 
100 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 6 
 
 NATURE OF PRESENT OFFENSE 
 
 Number and Per Cent of Felons and Misdemeanants Classified by New 
 York City Police Department Classification 
 
 
 Felons 
 
 Misdemeanants 
 
 Total 
 
 Nature of Offense 
 
 Number 
 
 Per cent 
 
 Number 
 
 Per cent 
 
 Number 
 
 Per cent 
 
 
 29 
 
 93.5 
 
 2 
 
 6.6 
 
 31 
 
 100.0 
 
 
 6 
 
 1.9 
 
 310 
 
 98.1 
 
 316 
 
 100.0 
 
 Offenses against Family and Children. 
 
 6 
 
 50.0 
 
 6 
 
 50.0 
 
 12 
 
 100.0 
 
 Offenses against Regulations for Pub- 
 lic Health, Safety and Policy 
 
 4 
 
 7.1 
 
 62 
 
 92.9 
 
 66 
 
 100.0 
 
 Offenses against Administration of 
 Government 
 
 2 
 
 66.7 
 
 1 
 
 33.3 
 
 3 
 
 100.0 
 
 Offenses against Property Rights 
 
 78 
 
 54.9 
 
 64 
 
 45.1 
 
 142 
 
 100.0 
 
 General Criminality 
 
 1 
 
 3.7 
 
 26 
 
 96.3 
 
 27 
 
 100.0 
 
 
 
 Total 
 
 126 
 
 21.5 
 
 461 
 
 78.5 
 
 687 
 
 100.0 
 
 52 or 92.9 per cent of the offenses against regulations for public 
 health, safety and policy are misdemeanants. Next to the offenders 
 against chastity, in size of group, are the offenders against property 
 rights, a total of 142 cases, 45.1 per cent of which are misdemeanants. 
 The last group of cases, the general criminality group, has out of 27 
 cases all but one, that is 96.3 per cent of the total, misdemeanants. Of 
 our total group of cases, 461 or 78.5 per cent are misdemeanants. 
 
 (d) Comparison of Nature of Offense of Female Felons, and Male 
 
 Felons in Sing Sing and the New York State 
 
 Reformatory at Elmira 
 
 Before discussing the extent of delinquency found in the various 
 institutional groups, it may be of interest to .note the comparison 
 nf nature of offense of the 126 felons in our group with the felons 
 as found in two of the institutions for male criminals in New York 
 State. For this comparison we have chosen Sing Sing Prison, which 
 may receive male felons of any adult age for any offense, and the 
 New York State Reformatory at Elmira, which receives only felons 
 
NATURE AND EXTENT OF DELINQUENCY 
 
 101 
 
 sixteen to thirty years of age who have never before been convicted 
 of a felony. The figures chosen for this comparison are the latest 
 figures obtainable in the annual reports of these two institutions and 
 include in each case, the consecutive commitments during the nine 
 months ending June 30, 1916." In the first comparison (Chart 4), 
 the most marked observation is that there is no striking difference in 
 the distribution of offenses in the two groups selected. It is particu- 
 larly interesting, however, to note that the female felons have a smaller 
 
 OFFENSES 
 AGAINST THE PERSON 
 
 CHASTITY 
 
 THE FAMILY 
 
 REG. FOR PUBLIC HEALTH 
 
 ADMINISTRATION OF GOV'T 
 
 PROPERTY RIGHTS 
 
 GENERAL CRIMINALITY 
 
 PER CENT 
 60 
 
 Chart III 
 
 Felonies and Misdemeanors 
 
 Per cent distribution among offenses classified by New York City Police 
 Department classification. 
 
 percentage, 4.8 per cent, of offenses against chastity than the male 
 felons with 8.0 per cent of their offenses in this group. In view 
 of the fact that of the total group of 587 women, 53.8 per cent are 
 offenders against chastity, it is important tO' observe the difference 
 between the female felons and either the total or the misdemeanant 
 group of females. The female felons have a slightly higher percentage 
 than the male felons in the offenses against the person, the family, 
 regulations for public health, and administration of government, but 
 
 ° "Twenty-second Annual Report of the State Commission of Prisons." 
 1916, pp. 413-415 and 446-447. The offenses as given in the report are listed 
 in alphabetical order with no attempt at classification. We have used the 
 N. Y. C. Classification in classifying these offenses, as we did for our group 
 of delinquent women. 
 
102 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 fall slightly below in the oifenses against property rights and general 
 criminality. 
 
 In comparing the female felons with the male felons in the Re- 
 formatory at Elmira (Chart 5), the same trend is shown as in Chart 4, 
 except that there are slightly more marked differences, especially in 
 offenses against the person, where there is a difference of 10.4 per 
 cent between the two groups, and offenses against property rights 
 where the male felons exceed the female by 16.7 per cent. The 
 
 OFFENSES 
 AGAINST THE PERSON 
 
 CHASTITY 
 THE FAMILY 
 
 PER CENT 
 
 40 60 
 
 f 
 
 kEG. FOR PUBLIC HEALTH 
 
 ADMINISTRATION OF GOV'T 
 
 PROPERTY RIGHTS 
 
 GENERAL CRIMINALITY 
 
 b 
 
 FEMALE FELONS 
 
 MALE FELONS 
 
 Chart IV 
 
 s 
 
 Percentage Comparison Showing: 1. Nature of Offense of 126 Female Fel- 
 ons in this Study; 2. Nature of Offense of 839 Male Felons Committed to 
 Sing Sing during the Nine Months ending June 30, 1916. 
 
 difference between the two groups in offenses against chastity is very 
 slight, only one-tenth of a per cent, which again is of interest since the 
 statement is often made that the percentage of offenders against 
 chastity is much higher among women in state prisons than among 
 men. 
 
 EXTENT OF DELINQUENCY 
 
 In considering the extent of the delinquency of the women in this 
 study, the measurement used has been for the most part the age 
 at first conviction and the number of previous convictions, keeping in 
 mind, wherever possible, the factor of present age, which determines 
 to some degree, the number of convictions. That is, a girl of six- 
 
NATURE AND EXTENT OF DELINQUENCY 103 
 
 teen has had less opportunity to be convicted many times than has a 
 woman of sixty. The number of months and the number of terms 
 served have also been used, but are felt to be of far less value than 
 the number of convictions, since the sentences in similar cases may 
 vary radically, depending upon the impression the woman makes on 
 the judge, while the conviction is usually a matter of law, likely to be 
 settled for all cases alike. The colored women, especially, would 
 have an unfair showing if any measure of term or length of time 
 
 OFFENSES 
 AGAINST THE PERSON 
 
 CHASTITY 
 
 THE FAMILY 
 
 REG. FOR PUBLIC HEALTH 
 ADMINISTRATION OF GOV'T 
 PROPERTY RIGHTS 
 
 GENERAL CRIMINALITY 
 
 
 
 PER CENT 
 
 
 20 40 60 80 100 
 
 5 
 
 1 
 
 - 
 
 
 
 ■ 
 
 
 
 
 
 r 
 
 
 
 
 
 I 
 
 ^ 
 
 
 ■ 
 
 
 i 
 
 
 
 1 
 
 
 I FEMALE FELONS 
 
 MALE FELONS 
 
 Chart V 
 
 Percentage Comparison Showing : 1. Nature of Offense of 126 Female Felons 
 in this Study; 2. Nature of Offense of 760 Male Felons Committed to the 
 New York State Reformatory at Elmira during the Nine Months ending Tune 
 30, 1916. 
 
 served were used, since the facilities for probation of colored women 
 in New York City are so poor that in most cases they are sentenced 
 to an institution for the same offense that would mean probation 
 to a white first offender. 
 
 In this study, a conviction has been interpreted as including those 
 who, in court, are convicted of the violation of any law. If, after 
 conviction, a woman was put on probation but violated her probation 
 within a week, was brought back to court on a bench warrant, and 
 resentenced to an institution, only one conviction would be counted, 
 since the probation was theoretically only a suspension of sentence 
 for a definite period of time. If, however, she had been put on 
 
104 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 probation for soliciting and while on probation stole, was brought 
 back to court and convicted again on the petit larceny charge, this 
 would be counted as a second conviction. Also, when returned to 
 court from an institution to which previously sentenced, and resentenced 
 on the same charge, only one conviction would be counted, though 
 in number of months the first institutional experience would count. 
 If, however, she was returned to court from an institution for being 
 incorrigible, etc., and sentenced to another institution on a new charge^ 
 this would be counted as a second conviction, and the first institutional 
 experience would count as a term served, and number of months 
 served. Time spent in a jail, hospital or Home preceding any con- 
 viction, no matter how long the time, was never counted as a term 
 served. Juvenile convictions have been counted as convictions, but 
 juvenile commitments to penal institutions as dependent children have 
 not been so counted. To those who may object to including juvenile 
 with adult convictions, it will be of interest to note the discussion at 
 the end of this chapter, showing that only 6.8 per cent of our cases 
 were convicted before the age of sixteen. 
 
 In this connection we might note the method used by Goring in his 
 study of the English convict,^" in "gauging a man's degree of criminal 
 tendency." He used the three standards of: (1) age at first con- 
 viction, (2) frequency of conviction per year of freedom, and (3) the 
 fraction of each year of life hitherto passed in confinement. The 
 second and third standards were obtained by noting the first and last 
 conviction of each convict, and taking the period of time between 
 the two as the length of his criminal career. "Next, the frequency 
 of his convictions to, and the length of time spent in, prison, during 
 this period between his first and last conviction, is noted^the fraction 
 of this period passed in freedom being obtained by subtracting from 
 it the length of time in prison. Finally, the number of convictions, : 
 including the first but excluding the present, divided by the years of 
 freedom, gives the frequency of convictions per year of freedom; 
 and the years in prison, divided by the total years of his criminal 
 career, give the fraction of each year spent in prison." 
 
 This method excludes all convicts who are at the time of the 
 study first offenders. Since 259 or 45.4 per cent of the women in 
 this study were first offenders, it seemed inadvisable to use a measure 
 of extent of delinquency which would at the outset eliminate nearly 
 
NATURE AND EXTENT OF DELINQUENCY 105 
 
 half of the group. There are also certain difficulties in applying 
 such a scheme to a group which contains as many kinds of sentences 
 as does the group under consideration. Goring did not have to 
 meet the problem of an offender who, though he may have had 
 three previous convictions, had been placed on probation each time 
 and had, therefore, never served any time in prison. Again, his 
 scheme makes no allowance for cases of offenders who may be con- 
 victed of larceny, sentenced to an institution and while there become 
 so troublesome that they are returned to court, reconvicted as in- 
 corrigible and sentenced to another institution. Here there are no 
 years in freedom, though there have been two previous convictions. 
 The great differences in sentences for relatively petty offenses may 
 not have had great weight in a study of English men prisoners twelve 
 years ago. In this study, however, there are numerous cases of women 
 who have been convicted of soliciting at some time and have received 
 either probation, five days in the Workhouse, an indefinite reformatory 
 sentence of three years, or in some cases, a sentence in a Home for the 
 period of minority which may be as much as five years. Such varying 
 sentences for the same offense obviously distort the value of a scheme 
 which is based on number of convictions per year of freedom, or frac- 
 tion of year passed in prison, since the individual who shows up worst 
 by the use of such a measure may be one whom the judge felt could 
 be benefited by a reformatory sentence. The more hopeless case, 
 on the other hand, who was given a short Workhouse term, would 
 figure as less criminal by this scheme, because of the short term. 
 
 (a) Number of Previous Convictions 
 
 (1) Relationship Between Number of Previous Convictions and 
 Age. — It is true that the criminal records of these women are in differ- 
 ent stages of completion because of their differences in age, and that in 
 order to make valid comparisons, some account must be taken of 
 the age factor. The correlation between number of previous con- 
 victions and present age is therefore of importance, both as an 
 indication of the extent of the direct relationship between these 
 two factors, and as furnishing a measure to be used in making al- 
 lowance for age in connection with other comparisons. Table 7 
 presents the data regarding this relationship. The coefficient of 
 correlation is found to be .23 with a standard deviation of .040. 
 This is sufficient to indicate the existence of a genuine, though small, 
 relationship. The correlation ratios are not appreciably larger than 
 
1 
 
 a 
 
 ■3 
 I 
 
 O 
 
 O 
 xo 
 
 § 
 
 4) 
 
 
 d 
 I 
 
 4> 
 
 a 
 
 o 
 
 S 
 
 o 
 
 Means 
 (Number of 
 
 Previous 
 Convictions) 
 
 t>- i> eo CO OS O CO 
 cii oi cd' C4 r-i ^ o 
 
 
 
 ^ 
 
 
 s 
 
 
 - -""■^"'ssssssssssess 
 
 1 
 m 
 
 § 
 
 1 
 
 M 
 
 i-t 
 
 -^ 
 
 CO 
 
 ■i 
 
 g 
 
 - 
 
 -< 
 
 2 
 
 
 
 00 
 
 T-l 1-1 
 
 IN 
 
 t* 
 
 
 
 
 
 
 lO 
 
 
 
 -* 
 
 D) 
 
 CSI 
 
 s 
 
 -* 
 
 1-1 
 
 T-l 
 
 ^ 
 
 (N 
 
 fH 
 
 ^ 
 
 N 
 
 o 
 
 Nrt 
 
 CC 
 
 
 OS 
 
 r-l rH i-lr^ 
 
 ■<1< 
 
 00 
 
 rt N iH 
 
 ■* 
 
 t> 
 
 N rHrt<Nr-(i-l 
 
 CO 
 
 CO 
 
 NrttN i-lrt^ 
 
 °°1 
 
 Ui 
 
 rt ^ i-i-iCON -*CC ^ 
 
 
 -* 
 
 1-1 W 1-1 i-t CO N M <N Tt< Cq N iH 
 
 S!]«- 
 
 CO 
 
 rHCOrHiOCO M lO l> M •<*( 
 
 S 
 
 J" 
 
 <N 
 
 iH .-(<N «INt-i<NWts.O3C0(NCO 
 
 CC 
 
 CO 
 
 1-1 
 
 ifH «(NCqU5U5l>.i-(MiN0iOC0cD 
 
 ? 
 
 
 o 
 
 1-1 N ■*■*'<*( CON «OO»«Mi0Tt<rHC000 
 
 s 
 
 OS 
 IN 
 
 
 
 1— " = 
 
 1 
 
 
 1 
 i 
 
 Bay 
 
 
 II II 
 
 
 OOCO 
 00 rH 
 
 MO 
 II 
 
 II 
 
 a a 
 
 ffi 4) 
 
 ca OS 
 
 oo 
 
 +I-H 
 
 o ; 
 
 II D 
 d o 
 
 §1 
 
 O (0 
 
 £ -oil 
 
 " i-Hi'S 
 |:SS6 
 
 ^ M II O P 
 SB 
 
 aiS 
 .2 3 8, 
 
 e.S 
 
 si 
 
 — oo 
 
 1 06 
 
NATURE AND EXTENT OF DELINQUENCY 107 
 
 che above coefficient, being .23 ± .040 for age on number of pre- 
 vious convictions and .25 ± .040 for number of previous con- 
 victions on age. Reference to the mean number of convictions for 
 the various age groups shows that there is a clear tendency to in- 
 crease in number of convictions with age up to about thirty-three 
 years of age, and that after this there seems no consistent tendency 
 toward increase, though results are somewhat erratic because of the 
 small numbers in the higher age groups. In any case, the relationship 
 between age and number of convictions is low enough to show that 
 age is not the sole nor even the main determinant of the number of 
 convictions. 
 
 It is of interest to note that Goring found a somewhat lower 
 correlation between age and number of convictions for his 2,225 
 men convicts: r = .1240.^^ This correlation coefficient is, however, 
 not entirely comparable with ours, since he has not included first 
 offenders in his table. While we admit that first offenders introduce 
 a considerable source of difficulty, since these individuals represent 
 all degrees of criminal tendency, from those who will never offend 
 again to those who are to be the most confirmed criminals, we have 
 not felt that our problem would be clarified by rejecting from con- 
 sideration nearly half of our group, especially since it appears to us 
 that the above-mentioned difficulty is found in only slightly lesser 
 measure for second and third offenders, and so on down the line. 
 Just as the first offender may either remain all his life a first offender, 
 or may go on to accumulate a large number of convictions, so the 
 second offender may remain only a second offender or may be 
 merely at the beginning of a long career of crime. 
 
 (2) Institutional Differences in Number of Previous Convictions. 
 — The distribution of the number of previous convictions, showing 
 the irregularity among the various institutional groups, is indicated 
 in actual percentages in Table 8 and graphically in Chart 6. 
 
 Those who have never had any previous convictions vary con- 
 siderably between the groups. The Probation group has the largest 
 number of first offenders, as would be expected, while in the order 
 of next highest percentage follow Auburn, the Magdalen, Bedford, 
 the Penitentiary, and last, the Workhouse with only 23.5 per cent 
 first offenders. The range of number of convictions is also of 
 interest among the various groups. Probation has the shortest range, 
 and is followed by Bedford, the Magdalen, Auburn, the Penitentiary 
 " Op. cit., p. 425. 
 
108 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 8 
 
 NUMBER (5F previous -CONVICTIONS 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups, 
 ■with Constants for Each Group 
 
 
 Institutional Groups 
 
 
 Previous 
 Convictions 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 Peniten- 
 tiary 
 
 Work- 
 house 
 
 Probation 
 
 Total 
 
 
 
 38.6 
 
 31.7 
 
 10.0 
 
 11.9 
 
 3.0 
 
 3.0 
 
 1.0 
 
 1.0 
 
 51.3 
 16.3 
 8.8 
 6.3 
 3.8 
 2.5 
 1.3 
 2.5 
 2.5 
 1.3 
 2.5 
 
 44.7 
 
 32.7 
 
 22.7 
 
 15.6 
 
 9.2 
 
 3.6 
 
 3.6 
 
 2.7 
 
 1.8 
 
 1.8 
 
 .9 
 
 .9 
 .9 
 
 \.% 
 .9 
 
 23.5 
 
 19.7 
 
 17.7 
 
 9.8 
 
 10.8 
 
 7.8 
 
 2.9 
 
 2.9 
 
 i.o' 
 i!6' 
 
 i;6' 
 
 83.3 
 
 13.7 
 
 2.0 
 
 i.o' 
 
 45.4 i 
 
 1 
 
 35 
 11 
 
 2 
 2 
 
 i 
 
 1 
 
 .5 
 .8 
 .6 
 .6 
 
 .'3' 
 .3 
 
 22.9 
 
 2 
 
 11.0 
 
 3 
 
 68 
 
 4 
 
 4.2 
 
 5 
 
 3.0 
 
 6 
 
 1.4 
 
 7 
 
 1.4 
 
 8 
 
 7 
 
 9 
 
 
 
 
 .7 
 
 10 
 
 5 
 
 11 
 
 4 
 
 12 
 
 
 
 .4 
 
 13 
 
 .2 
 
 14 
 
 4 
 
 18 
 
 .4 
 
 20 
 
 2 
 
 31 
 
 2 
 
 
 
 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cases 
 
 101 
 
 80 
 
 76 
 
 110 
 
 102 
 
 102 
 
 571 
 
 Moan 
 
 1.27 
 
 ±.149 
 
 1.76 
 + .312 
 
 1.03 
 ±.193 
 
 2.49 
 ±.359 
 
 2.82 
 ±.387 
 
 .216 
 ± .0565 
 
 1 63 
 
 
 ±.120 
 
 
 a 
 
 1.60 
 
 2.79 
 ±.221 
 
 1.68 
 
 3.77 
 
 ±.254 
 
 3.90 
 ±.273 
 
 .571 
 ±.0400 
 
 2.864 
 ±.0848 
 
 (To- 
 
 
 
 
 
 
 
 
 and Workhouse. The fact that Auburn changes place with Bed- 
 ford in the latter consideration would indicate that while there were 
 more first offenders among the Auburn women, the recidivists showed 
 a greater variability in the number of convictions. The Workhouse 
 in both considerations comes out at the worst end.^^ It is interest- 
 ing to observe that each of the eleven women in the total group who 
 
 "The W9rkhouse group, as used in these comparisons, does not include 
 the intoxication cases, which will be considered in the next chapter. If these 
 were added in the present comparison, the preponderance of recidivists with 
 many convictions in the Workhouse would be greatly increased. 
 
NATURE AND EXTENT OF DELINQUENCY 109 
 
 have had more than ten convictions have their cases complicated by 
 the factor of excessive alcoholism and previous convictions for in- 
 toxication, though the present offense does not involve intoxication. 
 The woman in the Workhouse, for example, who has had 31 previous 
 convictions had eighteen convictions for intoxication, seven convictions 
 for disorderly conduct, and six convictions for vagrancy. 
 
 Table 8 gives also, for each institutional group, the mean number of 
 convictions and the standard deviation which are represented graphi- 
 cally in Chart 6. This affords a basis for comparison of the institu- 
 
 TOTAL 
 
 H 
 
 CONVICTIONa 
 
 Chart VI 
 
 Number of Previous Convictions 
 
 Per cent distribution by institutional groups. 
 
 tional groups in terms of both their central tendencies and the amount 
 of dispersion within the groups. These figures are, however, insuffi- 
 cient, since we are not able to say, by mere inspection of the table, 
 whether a difference, which may be observed between two groups, is 
 large enough to be considered probably valid or not. A certain amount 
 of difference in constants is to be expected even if two samples are 
 taken entirely at random from the same total. Our first problem is, 
 therefore, to determine whether the differences observed are large 
 enough to justify us in consideritig the groups distinguished from one 
 another in respect to this factor, 'vis., number of previous convictions. 
 The statistical basis for this comparison is explained in Chapter 
 III. In general the method consists in finding the actual difference 
 between the constants of any two groups under consideration and 
 determining the ratio of this difference to the standard deviation of 
 
rL 
 
 BEDFORD 
 
 ■^ 
 
 I • t > 
 
 Ti l l 
 
 ^ r 
 
 AUBURN 
 
 PL 
 
 MAGDALEN 
 
 ^"^^^^ 
 
 CONVICTIONS 
 
 Chart VI— (Continued) 
 
 Number of Previous Convictions 
 
 Per cent distribution by institutional groups. 
 
 no 
 
PENITENTIARY 
 
 CONVtCTIONS 
 
 WORKHOUSE 
 
 C0NV1CTI0KS 
 
 PROBATION 
 
 .L 
 
 1- 
 
 ^ I I I I I I I I I I f 
 
 « ( T a 
 
 CONVICTIONS 
 
 ■r— I 
 
 Chart VI— (Continued) 
 
 Number of Previous Convictions 
 
 Per cent distribution by institutional groups. 
 
 Ill 
 
112 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 the differences^ Although no arbitrary dividing line can be laid 
 down, on one side of which there is certainty that a difference exists 
 and on the other side of which there is uncertainty, we shall 
 accept a ratio of 2 as critical for practical purposes. When 
 the ratio is 2 or more we shall feel a fair degree of confidence in the 
 existence of a difference between the two groups. A ratio of 3 we 
 shall accept as justifying practical certainty of the difference. When 
 the ratio is less than 2 the inference is, not that there is no difference 
 between the groups, but that our data fail to present convincing 
 evidence of their difference. The more closely the ratio appi-oaches 
 2 the greater the probability that the observed difference repre- 
 sents a genuine difference between the groups, even though the 
 ratio fails to measure up to the standard of probability which we 
 have accepted as critical. 
 
 In Table 9 we present the necessary data to supplement Table 8 and 
 make possible comparison of groups with reference to their average 
 number of convictions. This table gives, for all the different pairs 
 of institutions, the ratio of the difference, between any two given 
 means, to the standard deviation of the difference. 
 
 From these tables it appears that the Probation group, with a 
 mean number of convictions of .216, holds the lowest place. Com- 
 parison with each other group in turn, moreover, makes it clear 
 that the difference in favor of the Probation group is to be con- 
 sidered valid, since JL js decidedly over 3, which is sufficient for 
 confident prediction. The mean of the Magdalen is next in absolute 
 size, and is found to differ from each of the other groups except ' 
 Bedford by an amount large enough to justify confidence in the 
 genuineness of the difference. The difference between the Magdalen 
 and Bedford groups, however, being only .991 times the standard 
 deviation of the difference, does not justify more than the assertion 
 that there is a possibility that there may be a real difference between 
 the two groups. Bedford comes next with a mean number of con- 
 victions of 1.27. The differences between this value and the means for 
 the Penitentiary and the Workhouse are such as to justify assertion 
 of a difference. The difference between Bedford and Auburn, how- 
 ever, can not be stressed, since ~ is only 1.44. Though Auburn 
 and the Workhouse differ from one another, . by what appears to 
 
 " This ratio appears in the tables as ^, where d = the observed difference 
 between the measures and era = the standard deviation of the difference. 
 
NATURE AND EXTENT OF DELINQUENCY 
 
 113 
 
 TABLE 9 
 
 NUMBER OF PREVIOUS CONVICTIONS 
 
 Inter-comparison of Means of Institutional Groups in Terms of the 
 Ratio of the Difference Between the Means to the Stand- 
 ard Deviation of the Difference 
 
 
 Auburn 
 
 Magdalen 
 
 Penitentiary 
 
 Workhouse 
 
 Probation 
 
 Bedford 
 
 -1.4 
 
 .99 
 
 2.01 
 
 -3.15 
 -1.53 
 -3.60 
 
 -3.76 
 -2.13 
 -4.16 
 - .63 
 
 6.60 
 
 Auburn.. . . 
 
 4.88 
 
 Magdalen 
 
 4.04 
 
 Penitentiary 
 
 6.26 
 
 Workhouse 
 
 6.67 
 
 
 
 Explanation of table: The number in each space gives the value for — - for 
 
 the two institutional groups designated at the head of the column and to the left of 
 the row, respectively. A minus sign indicates that the mean of the institutional 
 group at the left is smaller than the mean of the group heading the column. 
 
 be a valid degree, in the direction of a smaller average number of 
 convictions for Auburn, neither of these groups can be distinguished 
 with a high degree of certainty from the Penitentiary, whose mean 
 falls between that for Auburn and that for the Workhouse. 
 
 Summarizing, the facts which seem at least reasonably well 
 established with regard to the difit'erences between the separate groups 
 in the matter of the criminal record as measured by average number 
 of convictions are as follows : 
 
 ( 1 ) The Probation group may be considered clearly differentiated 
 from all the other groups, by its small mean number of convictions, 
 a conclusion which was to be expected, since one of the prune grounds 
 for selection of a case for probation is the lack of a criminal record. 
 
 (2) The Magdalen group, while having a significantly larger 
 mean number of convictions than the Probation group, has a definitely 
 smaller mean number than Auburn, the Penitentiary or the Work- 
 house, but fails to be clearly distinguished from Bedford. 
 
 (3) The Bedford total is clearly distinguished, on the one hand, 
 from the Probation group, and, on the other hand, from the Peni- 
 tentiary and the Workhouse, but is not clearly distinguished from 
 either the Magdalen or Auburn. 
 
 (4) Auburn shows a clear difference, on the one hand, from 
 the Probation and Magdalen groups, and on the other hand, from 
 
114 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 the Workhouse, but is not certainly distinguished from either Bedford 
 or the Penitentiary. 
 
 (5) The Penitentiary is clearly distinguished, by its larger mean, 
 flumber of convictions, from all the groups except Bedford and 
 Auburn, in which cases the difference can not be considered estab- 
 lished as valid. 
 
 (6) The Workhouse is clearly differentiated, by its larger mean 
 number of convictions, from all groups except the Penitentiary. Be- 
 
 • tween the Workhouse and the Penitentiary the existence of a real 
 difference can not be proved. In this connection it is important 
 to remember again that our Workhouse group does not include 
 the cases committed for intoxication, the average number of convic^ 
 tions for which is very large. 
 
 The comparison of the mean number of convictions in the various 
 institutional groups may be summarized in another vvay by comparing 
 the mean of each group with the itiean for the total of all groups.^* 
 The results of this comparison are given in Table 10 supplementing 
 Table 8. From this it appears that Bedford, the Magdalen and 
 Probation are clearly distinguished from the total by a smaller mean 
 number of convictions, whereas the Penitentiary and the Work- 
 house show a definitely larger mean number of convictions. The 
 mean for Auburn is so close to the mean for the whole group 
 that the difference might easily be accounted for by chance. These 
 results are entirely in accord with those obtained by the more de- 
 tailed comparison, and have the advantage of presenting the situa- 
 tion more concisely. 
 
 We may compare, also, the amount of dispersion in the various 
 institutional groups, to discover, in the same way as in the case 
 of the comparison of the means, whether the apparent difference 
 constitutes a probably real difference. ^° Table 11 gives for the 
 various institutions, by pairs, the ratio of the difference between 
 any two given standard deviations to the standard deviation of the 
 differtoce. The results of this comparison may be summarized 
 
 " The formula for the standard deviation of the difference between the 
 mean of a sub-group and the mean of the total of which it is. a p art is given 
 
 by Pearson ("Biometrika," vol. V, p. 182) as i/ -+-(. -^"Y- p(M— m)' 
 
 r N^nV^ N/ N(N-n) 
 where S is the standard deviation of the total and N the number of cases in 
 the total, 0- the standard deviation of the sub-group and n the number of cases in 
 the sub-group, M the mean of the total and m the mean of the sub-groUp. 
 "The formula for the standard deviation of the difference between the 
 
 standard deviations is as follows : g- =4/ (T* + cr^ 
 
NATURE AND EXTENT OF DELINQUENCY 115 
 
 TABLE 10 
 
 NUMBER OF PREVIOUS CONVICTIONS 
 
 Comparison of Means of Institutional Groups with Mean of Total, 
 
 in Terms of the Ratios of the Difference Between the 
 
 Means to the Standard Deviation of the Difference 
 
 Institutional Group 
 
 Difference between 
 mean of sub-group 
 and m6an of total 
 
 d 
 
 Chances that real 
 difference doea not 
 exist are 1 in: 
 
 Bedford 
 
 Auburn .... 
 Magdalen. . . 
 Penitentiary- 
 Workhouse . 
 Probation. . . 
 
 - .39 
 .11 
 
 - .63 
 .83 
 
 1.17 
 -1.44 
 
 2.29 
 .37 
 3.09 
 2.74 
 3.53 
 11.38 
 
 91 
 
 3 
 
 1000 
 
 323 
 
 500 
 
 briefly as showing that the differences in the amount of dispersion 
 in the various groups may be considered valid, except in the case 
 of the comparison of Magdalen with Bedford, and that of the 
 Penitentiary with the Workhouse. In all the other cases the ratio 
 of the difference to the standard deviation of the difference is well 
 over 2. Probation shows the least variabi'ity within the group, and 
 the Workhouse the greatest, the order of variability being Probation, 
 
 TABLE 11 
 
 NUMBER OF PREVIOUS CONVICTIONS 
 
 Inter-comparison of Standard Deviations of Institutional Groups in 
 Terms of the Ratio of the Difference Between the Stand- 
 ard Deviations to the Standard Deviation of 
 the Difference 
 
 
 Au'oum 
 
 Magdalen 
 
 Penitentiary 
 
 Workhouse 
 
 Probation 
 
 Bedford 
 
 -5.30 
 
 -1.06 
 4.29 
 
 -8.27 
 -2.91 
 -7.25 
 
 -8.22 
 -3.17 
 -7.29 
 - .38 
 
 8.21 
 
 Auburn 
 
 9.90 
 
 Magdalen 
 
 7.80 
 
 Penitentiary 
 
 12.43 
 
 Workhouse 
 
 12.06 
 
 
 
 Explanation of table: 
 
 The number in each space gives the value for — for 
 
 the two institutional groups designated at the head of the column and to the left of 
 the row, respectively. A minus sign indicates that the mean of the institutional 
 group at the left is smaller than the mean of the group heading the column. 
 
116 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 Bedford, Magdalen, Auburn, Penitentiary, and Workhouse. It ap- 
 pears, thus, that the groups having the smallest mean number of con- 
 victions have also the lowest variability, a fact which is easily under- 
 stood, since the mean values of these groups approach the lower 
 limiting value of no previous convictions. 
 
 The above comparison of the groups, with reference to their mean 
 number of convictions, is likely to be somewhat misleading, since no 
 account is taken of the wide differences in actual ages of the groups, 
 with the consequent differences in opportunity for accumulation of 
 numerous convictions. From the point of view of a simple de- 
 scription of the groups as they stand, such crude comparison is not 
 only justifiable but useful, since we have a certain interest in knowing 
 how the groups compare with respect to the extent of criminal record 
 of their women, whatever the reasons for this difference may be. 
 For example, it is important to realize the marked tendency of the 
 Workhouse and the Penitentiary to include considerable numbers 
 of old offenders, because of the bearing which this fact would have 
 on the sending of young and impressionable first offenders into these 
 groups. On the other hand, the fact that Magdalen and Bedford 
 tend to receive the less extensive recidivists should be a reason 
 for considering well before sending into this group individuals with 
 frequent experience of conflict with the law. Comparison of norms 
 and of distributions of number of convictions is, therefore, of direct 
 importance for descriptive purposes. 
 
 When, however, we endeavor to make use of the number of con- 
 victions as at least an approximate measure of criminal tendency, 
 it becomes of great . importance that we make allowance for the 
 factor of age. In connection with the present inter-comparison of 
 institutional groups the fact that the three institutions having the 
 smallest mean numbers of convictions have also the lowest mean 
 ages'-^ may well raise a doubt as to whether there is any real difference 
 between the groups with respect to criminal tendency, or whether there 
 may not have been a selection primarily in terms of age which has 
 naturally resulted in a difference in the extensiveness of the criminal 
 records. 
 
 (3) Number of Previous Convictions Among Felons and Misde- 
 meanants. — It will be of interest here to make the comparison between 
 number of previous convictions among those convicted of felonies and 
 18 For the necessary data regarding ages of groups, see Chapter VII, p. 148. 
 
NATURE AND EXTENT OF DELINQUENCY 117 
 
 TABLE 12 
 
 Number and Per Cent of Previous Convictions Among |DeUnquent 
 Women Classified as Felons and Misdemeanants 
 
 
 Natobe of Offense 
 
 Total 
 
 Number of Convictions 
 
 Feloniea 
 
 Misdemeanors 
 
 
 
 Number 
 
 Per Cent 
 
 Number 
 
 Per Cent 
 
 Number 
 
 Per Cent 
 
 
 
 1 
 
 2 
 
 3 
 
 61 
 23 
 11 
 6 
 3 
 2 
 2 
 3 
 2 
 1 
 2 
 
 ■•■• 
 .... 
 
 51.7 
 19.4 
 9.2 
 5.1 
 2.5 
 1.7 
 1.7 
 2.5 
 1.7 
 .8 
 1.7 
 
 "!8 
 '".8 
 
 198 
 
 108 
 
 52 
 
 33 
 
 21 
 
 15 
 
 6 
 
 5 
 
 2 
 
 3 
 
 1 
 
 2 
 
 1 
 
 1 
 
 2 
 
 2 
 
 .... 
 
 43.7 
 
 23.8 
 
 11.5 
 
 7.3 
 
 4.6 
 
 3.3 
 
 1.3 
 
 1.1 
 
 .4 
 
 .7 
 
 .2 
 
 .4 
 
 .2 
 
 .2 
 
 .4 
 
 .4 
 
 '".2 
 
 259 
 
 131 
 
 63 
 
 39 
 
 24 
 
 17 
 
 8 
 
 8 
 
 4 
 
 4 
 
 3 
 
 2 
 
 2 
 
 1 
 
 2 
 
 2 
 
 1 
 
 1 
 
 45.4 
 
 22.9 
 
 11.0 
 
 6 8 
 
 4 
 
 5 
 
 6 
 
 7 
 
 8 
 
 9 
 
 10 
 
 11 
 
 12 
 
 13 
 
 14 
 
 18 
 
 20 
 
 4.2 
 
 3.0 
 
 1.4 
 
 1.4 
 
 .7 
 
 .7 
 
 .5 
 
 .4 
 
 .4 
 
 .2 
 
 .4 
 
 .4 
 
 .2 
 
 31 
 
 .2 
 
 Total 
 
 118 
 
 100.00 
 
 453 
 
 100.00 
 
 571 
 
 100.00 
 
 of misdemeanors, as well as to observe the difference among insti- 
 tutional groups. 
 
 Table 12 shows that a slightly larger percentage of the felons than 
 of the misdemeanants are first offenders and that the range is slightly 
 shorter in the felony group. The histogram which follows shows 
 graphically the number of previous convictions among the two groups, 
 and would seem to indicate that the seriousness of offense committed 
 does not have an important relationship in our total group to the 
 number of previous convictions. 
 
 Table 13, supplementing Table 12, makes possible a comparison 
 between felons and misdemeanants in terms of the mean number of 
 convictions and the amount of variations about these means in the two 
 groups. While a comparison of the means alone would suggest that 
 the felons as a group tend to have been convicted somewhat more often 
 than the misdemeanants, since theit central tendency is slightly larger. 
 
118 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 this point can not be safely maintained until we discover whether this 
 difference can be considered valid or whether it might have occurred 
 by chance. Since the ratio of the actual difference between the means 
 
 • nLOHt 
 
 I tOnnMBANAKTl 
 
 CONVICTIONt 
 
 Chart VII 
 
 Number of Previous Convictions 
 
 Percentage comparison between felons and misdemeanants 
 
 to the standard deviation of the differences is only .09, the chances are 
 even that the observed difference might have occurred by chance. Ac- 
 cordingly, we are justified in saying that there is no demonstrable dif- 
 ference between the felons and the misdemeanants with reference to 
 
 TABLE 13 
 
 FELONS AND MISDEMEANANTS 
 
 Comparison of the Means and Standard Deviations of Number of 
 
 Previous Convictions Among Felons and Misdemeanants 
 
 of Total Group 
 
 
 Felons 
 
 Misde- 
 meanants 
 
 Difference 
 
 d 
 
 Chances that 
 real differ- 
 ence does not 
 exist are 1 
 in: 
 
 Mean 
 
 1.65 
 
 + .277 
 
 1.63 
 + .133 
 
 .03 
 
 .09 
 
 2 
 
 (Tm 
 
 
 
 
 
 3.01 
 + .196 
 
 2.825 
 + .0939 
 
 .19 
 
 .85 
 
 5 
 
 (T^ 
 
 
 
 Cases 
 
 118 
 
 453 
 
 
 
 
 
 
NATURE AND EXTENT OF DELINQUENCY 119 
 
 their mean number of convictions." A comparison of the standard 
 deviations, or the measures of dispersion, of the two groups, indicates 
 a slightly wider scattering of cases in the felon group than in the mis- 
 demeanor. There is only a possible difference in the dispersion of 
 these two groups.^* 
 
 (4) Number of Previous Convictions Among Offenders Against 
 Chastity and Offenders Against Property. — As previously noted, the 
 two subdivisions of the New York City Police Department classifica- 
 tion which are represented by the greatest number of cases from our 
 group are the offenses against chastity and offenses against property. 
 
 TABLE 14 
 
 Number and Per Cent of Previous Convictions Among Delinquent 
 
 Women Convicted of Offenses Against Chastity and 
 
 Offenses Against Property Rights 
 
 
 Offenses Against: 
 
 
 
 
 
 
 Total 
 
 Number of Convictions 
 
 Chastity 
 
 Property Rights 
 
 
 
 
 Number 
 
 Per Cent 
 
 Number 
 
 Per Cent 
 
 Number 
 
 Per Cent 
 
 
 
 155 
 
 49.7 
 
 49 
 
 36.0 
 
 204 
 
 45.5 
 
 1 
 
 71 
 
 22.8 
 
 37 
 
 27.2 
 
 108 
 
 24.1 
 
 2 
 
 31 
 
 9.9 
 
 17 
 
 12.5 
 
 48 
 
 10.7 
 
 3 
 
 16 
 
 5.1 
 
 14 
 
 10.3 
 
 30 
 
 6.7 
 
 4 
 
 17 
 
 5.5 
 
 4 
 
 2.9 
 
 21 
 
 4.7 
 
 5 
 
 10 
 
 3.2 
 
 2 
 
 1.5 
 
 12 
 
 2.7 
 
 6 
 
 3 
 
 1.0 
 
 2 
 
 1.5 
 
 5 
 
 1.1 
 
 7 
 
 3 
 
 1.0 
 
 2 
 
 1.5 
 
 5 
 
 1.1 
 
 8 
 
 
 
 2 
 
 1.5 
 
 2 
 
 .5 
 
 9 
 
 2 
 
 .6 
 
 
 .7 
 
 3 
 
 .7 
 
 10 
 
 
 
 
 1.5 
 
 2 
 
 .5 
 
 11 
 
 i 
 
 .3 
 
 
 .7 
 
 2 
 
 .5 
 
 12 
 
 1 
 
 .3 
 
 
 
 1 
 
 .2 
 
 13 
 
 
 
 
 .7 
 
 1 
 
 .2 
 
 14 
 
 2 
 
 .6 
 
 
 
 2 
 
 .5 
 
 18 
 
 
 
 
 .7 
 
 1 
 
 .2 
 
 20 
 
 
 
 
 .7 
 
 1 
 
 .2 
 
 31 
 
 
 
 
 
 
 
 Total 
 
 312 
 
 100.00 
 
 136 
 
 100.00 
 
 448 
 
 100.0 
 
 "It is important to remember again, in connection with this comparison of 
 felons and misdemeanants with regard to their mean number of convictions, 
 that the intoxication commitments are omitted from the misdemeanant group. 
 We are convinced that if these had been included, the misdemeanants would 
 have shown a conclusively larger mean number of convictions. (See Chap- 
 ter VII, p. 154.) 
 
 " See Chapter III. 
 
120 STUDY OF WOMEN DELINQUKWlb liN i\iivv xKjr.a. 
 
 These two divisions which comprise 78 per cent of our total group 
 seem of sufficient importance to use for purposes of comparison. 
 Accordingly, we shall proceed to Table 14, which shows the 
 percentage of offenders against chastity and offenders against property 
 rights who have had the number of previous convictions specified in 
 the table. It would appear that among the first offenders there is con- 
 siderable difference between the percentage of chastity and property 
 offenders, 49.7 per cent as against 36.0 per cent. The range of number 
 of convictions for the offenders against property rights is also wider. 
 Turning from the frequency table above to Table 15, we may 
 consider whether or not the difference seemingly indicated in Table 
 14 is of any real significance. We may note, first of all, by comparing 
 the means, that there is a tendency for the offenders against property 
 rights to have a larger mean number of convictions than the offenders 
 against chastity. There is almost certainly a real difference between 
 the mean number of convictions of the two groups of offenders, 
 with a tendency, as stated above, for the offenders against property 
 rights to have more convictions. The difference between the standard 
 deviations of the two groups is even more marked, showing that 
 there is more variability in the number of convictions among the 
 offenders against property rights than among the offenders against 
 chastity. 
 
 TABLE 16 
 
 OFFENDERS AGAINST PROPERTY RIGHTS AND OFFENDERS 
 AGAINST CHASTITY 
 
 Comparison of the Means and Standard Deviations of Number of 
 
 Previous Convictions Among Offenders Against Property 
 
 Rights and Offenders Against Chastity of Total Group 
 
 
 Offenders 
 against 
 
 Property 
 Rights 
 
 Offenders 
 against 
 Chastity 
 
 Difference 
 
 d 
 0-d 
 
 Chances that 
 real differ- 
 ence does not 
 exist are 1 
 in: 
 
 Mean 
 
 2.00 
 
 ±.275 
 
 1.30 
 ±.120 
 
 .70 
 
 2.32 
 
 98 
 
 
 
 
 
 (7. . . . 
 
 3.20 
 ±.194 
 
 2.121 
 ±.0849 
 
 1.08 
 
 5.10 
 
 00 
 
 (Xa 
 
 
 
 
 Cases 
 
 136 
 
 312 
 
 
 
 
 
 
NATURE AND EXTENT OF DELINQUENCY 121 
 
 (b) Number of Previous Terms Served 
 In addition to the consideration of the number of previous con- 
 victions as an indication of the extent of dehnquency, we may note 
 briefly the distribution among the various groups of the number of 
 previous terms served. Table 16 will show this distribution. A term, 
 as we have used it, may be defined as a sentence of any length of time 
 to any Home or institution, as a delinquent. In comparison with Table 
 8, showing the number of previous convictions, it becomes evident, as 
 would be expected, that the number who have never served any pre- 
 vious term is always slightly larger than those who have never had any 
 previous convictions, since in each group there are a certain number 
 who have been given a chance on probation or suspended sentence. 
 The difference is fairly large in the Bedford group, i.e., 12.9 per cent. 
 
 TABLE 16 
 
 Per Cent Distribution of Number of Previous Terms Served in Penal 
 
 Institutions by Delinquent Women in Institutional Groups 
 
 
 Institutional Groups 
 
 
 Number of Terms 
 
 Bedford 
 
 Auburn 
 
 Magdaleu 
 
 Peniten- 
 tiary 
 
 Work- 
 house 
 
 Probation 
 
 Total 
 
 
 
 61. S 
 
 31.7 
 
 8.9 
 
 4.0 
 
 2.0 
 
 '26 
 
 67.6 
 12.5 
 
 7.6 
 
 6.3 
 
 2.6 , 
 
 5.0 
 
 2.5 
 
 1.3 
 
 2.6 
 
 '2.5 
 
 63 5! 
 
 38.2 
 
 21.8 
 
 14.5 
 
 6.4 
 
 4.6 
 
 2.7 
 
 3.6 
 
 .9 
 
 .9 
 
 .9 
 
 '".9 
 .9 
 .9 
 
 '".^ 
 .9 
 .9 
 
 24.8 
 
 19.8 
 
 16.8 
 
 12.9 
 
 7.9 
 
 6.9 
 
 3.0 
 
 3.0 
 
 i.'o 
 iio 
 
 '2;6 
 
 i^o 
 
 92.1 
 5.9 
 1.0 
 
 i.'o 
 
 63.8 
 
 19.5 
 
 9.3 
 
 6.6 
 
 3.2 
 
 2.5 
 
 1.9 
 
 .9 
 
 .7 
 
 .4 
 
 .5 
 
 4 
 
 1 
 
 25 
 5 
 3 
 
 .0 
 3 
 9 
 
 3 
 3 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 8 
 
 9 
 
 10 
 
 11 
 
 12 
 
 
 
 2 
 
 13 
 
 •J 
 
 14 
 
 4 
 
 16 
 
 
 16 
 
 
 17 
 
 2 
 
 18 
 
 2 
 
 20 
 
 31 
 
 .2 
 2 
 
 
 
 
 
 Total . . . 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of oases 
 
 101 
 
 80 
 
 76 
 
 110 
 
 101 
 
 101 
 
 669 
 
122 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 In the Auburn group we find a difference of 6.2 per cent, in the Mag- 
 dalen 18.5 per cent, in the Penitentiary 5.5 per cent, in the Workhouse 
 1.3 per cent, and in the Probation group 8.8 per cent. The Workhouse 
 has the smallest percentage, therefore, who have never served a pre- 
 vious term, and shows a smaller difference than any other group 
 between the number of previous convictions and the number of 
 previous terms served, indicating, as will later be shown in a table 
 giving the number of times on probation, that the Workhouse women 
 have been given sentences to institutions more than have the women 
 in any other group. The range of number of previous terms served 
 is very much like the range of number of previous convictions. 
 
 (c) Length of Time Served in Penal Institutions 
 
 If we turn to the length of time served in penal institutions,'' 
 the most striking thing in the distribution, as in Table 16, is the 
 small percentage in the Workhouse who have never served time previ- 
 ous to the present conviction, and the comparatively high percentage of 
 those who have served time amounting to less than a year altogether. 
 This is accounted for by the fact that at their first sentence, 91.2 per 
 cent of the women in the Workhouse were sentenced to the Work- 
 house, Penitentiary, or an institution like County Jails, in each of 
 which, in the past, a short term was possible. Since the Workhouse 
 has many commitments of five, ten, fifteen or twenty days, etc., it has 
 been possible for many of the women in our group to serve a large 
 number of terms, and still serve altogether a comparatively short time, 
 less than one year. It will be noted from Table 17 that while only 
 17.8 per cent of the women in the Workhouse had served more than 
 one year, though the mean number of convictions was 2.82, just the 
 same percentage, 17.8, of the Bedford women had served more than 
 one year, and the mean number of ^convictions was only 1.27. The 
 Penitentiary women, also, with a mean number of convictions of 
 2.49, only slightly less than the mean for the Workhouse group, 
 have 29.0 per cent of their total number who have served more than 
 one year in penal institutions. The Probation group has only two 
 cases, or 2.0 per cent of its total, who have served more than one year. 
 
 "We should expect the same percentage for those who had never served a 
 previous term, and those who had never served any time, since, of course, one 
 who has served even a few days has served a term, and conversely. The 
 difference in percentages for those never serving time (Table 17) or a previ- 
 ous term (Table 16) comes from the difference in the total number of cases 
 on which we were able to obtain data for both items. 
 
NATURE AND EXTENT OF DELINQUENCY 123 
 
 TABLK 17 
 
 LENGTH OF TIME SERVED IN PENAL INSTITUTIONS 
 PREVIOUS TO PRESENT CONVICTION 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups 
 
 
 
 
 Length of Time Served 
 
 Bed- 
 ford 
 
 Auburn 
 
 Mag- 
 dalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Proba- 
 tion 
 
 Total 
 
 None 
 
 51.5 
 30.7 
 7.9 
 5.9 
 3.0 
 1.0 
 
 59.0 
 20.5 
 7.7 
 9.0 
 1.3 
 1.3 
 1.3 
 
 63.2 
 
 25.0 
 
 7.9 
 
 2.6 
 
 1.3 
 
 39.1 
 31.8 
 
 *12.7 
 9.1 
 4.5 
 
 ■;9 
 
 1.8 
 
 24.5 
 
 57.8 
 
 12.8 
 
 2.0 
 
 2.0 
 
 i'o 
 
 93 n 
 
 54 1 
 
 Less than 1 yr 
 
 5 
 
 i 
 i 
 
 
 
 6 
 6 
 
 29 1 
 
 1 yr. to 2 jTS 
 
 8 3 
 
 2 yrs. " 3 " 
 
 4.9 
 
 3 " " 4 " 
 
 2 1 
 
 4 " " 5 " 
 
 .4 
 
 5 " " 6 " 
 
 .2 
 
 6 ■' " 7 " 
 
 .2 
 
 7 " " 8 " 
 
 .5 
 
 8 " " 9 " , 
 
 
 9 " " 10 " 
 
 .2 
 
 
 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 
 
 Number of cases 
 
 101 
 
 78 
 
 76 
 
 110 
 
 102 
 
 100 
 
 667 
 
 
 
 
 
 The comparisons between the three sets of tables of number of 
 previous convictions, number of terms served, and length of time 
 served in penal institutions, demonstrate that there is a very consid- 
 erable difference in the results, according to whether one or another 
 scheme of measurement of delinquency was used. In using number 
 of terms the factor of length of term must be considered. A woman 
 may have served twenty terms but each may have been for ten 
 days, while another woman with one previous term may have served 
 ten years. Again, the number of terms alone does not take cognizance 
 of the factor of probation and suspended sentence, which are as 
 important from the standpoint of a study in delinquency as is any 
 other sentence. The length of tifne in penal institutions, in the same 
 way, does not give a clear idea of one's record of delinquency, since 
 the committing judge and the kind of institution to which he is able 
 to send her are so variable in different communities. If a girl ap- 
 pears hardened and brazen, the judge may commit her to the Work- 
 house for ten days for soliciting, but if she seems young and re- 
 pentant he may commit her to Bedford for an indefinite term of a 
 possible three years. Obviously, the measure of extent of delinquency 
 
124 STUDY OF WOMEN DELINQUENTS IN NEW YUKK 
 
 should not operate to make the one appear more and the other 
 less criminal, as it would if the measure of length of time served 
 were used. 
 
 ((f) Number of Times on Probation 
 To show how much and how variable a factor probation or sus- 
 pended sentence has been in the several institutional groups, Table 18 
 is given. By the nature of the group, all the Probation cases have been 
 given at least one chance on probation if the present sentence is count- 
 ed as one. Among the other groups, since the present sentence is to 
 an institution, only past sentences will be considered in finding the 
 number of times placed on probation. It is evident that there is great 
 variability in this, and that the percentage who have never been given 
 a chance on probation or suspended sentence ranges from 63.2 per cent 
 in the Magdalen to 95.1 per cent in the Workhouse. In no institution, 
 except Bedford, has any woman been given more than two chances 
 in having her sentence suspended. This table becomes more significant 
 if taken in connection with Table 8, which shows the percentage of 
 each institutional group which has never had a previous conviction. 
 The Workhouse has the smallest percentage, 23.5 per cent, of first 
 offenders; it has the highest mean number of convictions, and the 
 smallest percentage who have been given a suspended sentence of 
 any kind. This would indicate that there is probably a process 
 
 TABLE 18 
 
 NUMBER OF TIMES PLACED ON PROBATION OR SUSPENDED 
 
 SENTENCE 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups 
 
 
 Institdtional Groups 
 
 
 N limber of Times 
 on Probation 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 Peniten- 
 ■ tiary 
 
 Work- 
 house 
 
 Probation* 
 
 Total 
 
 
 
 69.3 
 
 20.8 
 
 5.9 
 
 3.0 
 
 1,0 
 
 89.9 
 6.3 
 
 3.8 
 
 63.2 
 
 32.9 
 
 3.9 
 
 84.5 
 
 13.6 
 
 1.8 
 
 95.1 
 3.9 
 1.0 
 
 96! i 
 
 9.9 
 
 66 6 
 
 1 
 
 28 3 
 
 2 
 
 4 4 
 
 3 
 
 .5 
 
 4 
 
 2 
 
 
 
 Total . . . 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cases 
 
 101 
 
 79 
 
 76 
 
 110 
 
 102 
 
 101 
 
 569 
 
 '''Present probation is counted as one time on probation. 
 

 
 5< 
 
 
 
 ;) 
 
 
 m 
 
 
 
 
 (S 
 
 
 ft 
 
 ■s 
 
 
 s 
 
 a 
 
 
 
 1 
 
 
 o 
 
 •^ 
 
 
 H 
 
 & 
 
 
 Q 
 
 1 
 
 H 
 
 ^ 
 
 1 
 
 
 m 
 
 o 
 
 
 S 
 
 n« 
 
 
 
 
 S 
 
 M !> O 
 
 
 
 8 
 
 >0 ■>:1H O 
 
 
 
 Ih 
 
 U3 Ttl O 
 
 
 
 E 
 
 
 
 »-< 
 
 -S 
 o 
 
 Ph 
 
 
 
 1 
 
 1^ 
 
 
 
 
 H 
 
 J 
 
 00 05 1> 
 
 
 
 
 O ■* lO 
 
 
 
 1 
 
 C<3 N >0 
 
 
 
 Is 
 
 
 
 1 
 
 
 
 
 
 
 1 
 
 
 
 
 M 1> o 
 
 
 g 
 
 t4 
 
 CD M d 
 
 
 
 P 
 
 tH 00 O 
 
 
 1 
 
 P^ 
 
 
 
 1 -H 
 
 ft 
 
 
 
 
 
 g 
 
 .2 
 
 lo ^ 
 
 - (N 
 
 
 & 
 
 
 ■-I ^ 
 
 - O 
 
 
 
 P 
 
 
 
 
 
 
 z 
 
 
 
 
 
 
 
 
 
 
 
 g 
 
 o c 
 
 > o 
 
 
 1 
 
 
 1> Cf 
 
 3 O 
 
 
 
 ^ 
 
 t^ o 
 
 ) O 
 
 
 ■g 
 
 pL, 
 
 
 
 1— 1 
 
 
 ■g 
 
 h 
 
 
 
 
 
 o 
 
 ■9 
 
 f: ? 
 
 i S 
 
 
 
 Z ■ 
 
 
 
 
 
 -«J 
 
 
 
 
 
 & 
 
 8 
 
 CO Tl 
 
 ^ 9 
 
 
 <S 
 
 ^1 
 
 t^ ^ 
 
 1 o 
 
 [Q 
 
 
 B 
 
 CO Cf 
 
 
 P 
 
 fl 
 
 Ph 
 
 
 
 1—1 
 
 0) 
 
 1 
 
 
 
 
 g 
 
 '3 
 
 CO If 
 
 3 X 
 
 o 
 
 -< 
 
 1 
 p 
 
 l> CI- 
 
 i-H 
 
 
 
 
 
 
 - O 
 
 
 
 CO l> 
 
 . o 
 
 
 d 
 
 CD 
 
 its T 
 
 i 8 
 
 S 
 
 1 
 
 Oh 
 
 
 
 T— ( 
 
 K 
 
 1 
 
 tj T) 
 
 CD 
 
 
 s 
 
 ^ V 
 
 J 1> 
 
 
 
 3 
 
 
 
 
 
 
 !s 
 
 
 
 
 
 +3 
 
 
 
 
 
 
 d 
 
 00 c 
 
 O 
 
 
 s 
 
 « 
 (h 
 
 00 1- 
 
 d 
 
 
 CD 
 
 ■^ ir 
 
 o 
 
 
 g 
 
 Ah 
 
 
 
 T— ( 
 
 
 -s 
 
 Fh 
 
 
 
 
 
 ^ 
 
 1 
 
 OS 1- 
 
 o 
 
 
 CO Tt 
 
 00 
 
 
 
 
 
 
 
 
 
 
 Z 
 
 
 
 
 
 
 
 
 
 
 
 s 
 
 ■^ 
 
 CC 
 
 o 
 
 
 TJ 
 
 h 
 
 f_ 
 
 a 
 
 d 
 
 
 K 
 
 S 
 
 CD c<: 
 
 o 
 
 
 1 
 
 Ph 
 
 
 
 T— I 
 
 »H 
 
 
 
 
 
 
 1 
 
 § s 
 
 o 
 1-t 
 
 
 
 1 
 
 
 4 
 
 
 
 
 
 
 Ef 
 
 
 
 
 
 
 
 fix 
 
 
 
 
 
 
 
 a 
 
 
 
 
 
 
 
 TZ 
 
 "3 
 
 
 
 
 _g 
 
 P 
 
 ■2 
 
 
 
 
 
 c 
 
 ^ 
 
 
 
 
 !| 
 
 
 
 
 
 
 
 
 
 
 
 
 Q 
 
 t 
 
 
 
 
 
 p: 
 
 iz< 
 
 
 1 
 
 125 
 
126 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 of selection in the courts which sends these women to institutions 
 rather than giving them another form of sentence. One of the 
 most striking things in the table is the fact that so few of the women in 
 each group have, as the women express it, "been given a chance." This 
 is probably due to the fact that the adult probation system is a com- 
 paratively new thing, and that it has not been extended to any degree 
 to the higher courts which deal with the more serious offenders. 
 
 {e) Recidivists and First Offenders 
 
 (1) Recidivists and First Offenders Among Institutional Groups. 
 —To show briefly the trend of extent of delinquency among the insti- 
 tutional groups, it may be well to divide the offenders into recidivists 
 and first offenders. We have used the term "recidivist" to mean 
 an individual who has been convicted of an offense against the law 
 more than once. The kind of sentence after conviction has not 
 been taken into account, since with the development of probation 
 systems, it seems unfair to call a woman a recidivist if she was 
 sentenced to prison by one judge, while another woman with the 
 same charge might be given probation by a more lenient judge.'" 
 Table 19 shows that the percentage of recidivists is highest among 
 the Workhouse women, and in decreasing order follow the Peni- 
 tentiary, Bedford, Magdalen, Auburn, and the Probation group. 
 
 (2) Recidivists and First Offenders Among Felons and Misde- 
 meanants. — If we turn to the number of recidivists among felons and 
 misdemeanants (Table 20), we find that the percentage of recidivists 
 among the felons is considerably smaller than among the misdemean- 
 ants. It would seem from this that there is a tendency for a larger 
 number of those who commit the more serious offenses not to commit 
 further offenses. If we turn back to Table 13, however, we find that 
 there is no very significant difference, apparently, between the mean 
 number of convictions among felons and misdemeanants. From this, 
 it would seem that the recidivists among the felons tend to have a 
 larger percentage of convictions than the recidivists among the mis- 
 demeanants. 
 
 " It is obvious, of course, that there are marked limitations to the use of 
 ■'recidivist" as one who has been previously convicted by law. We might note 
 here the case of one Bedford woman 32 years old who had been a prostitute 
 in Chinatown for years and had smoked opium for some time. She had vio- 
 lated the law, undoubtedly, nearly every day during this time. She was not 
 arrested, however, because of her own cleverness and the protection of her 
 pimp, until she was sentenced to Bedford. Legally, she is a first offender, but 
 in every other sense of the word- she has violated the law for years. 
 
NATURE AND EXTENT OF DELINQUENCY 127 
 
 TABLE 20 
 
 RECIDIVISTS AND FIRST OFFENDERS 
 
 Number and Per Cent Among Felons and Misdemeanants 
 
 
 Felons 
 
 Misdemeanants 
 
 Total 
 
 
 Number 
 
 Per Cent 
 
 Number 
 
 Per Cent 
 
 Number 
 
 Per Cent 
 
 Recidivists 
 
 First Offenders 
 
 57 
 60 
 
 48.7 
 51.3 
 
 251 
 189 
 
 57.0 
 43.0 
 
 308 
 249 
 
 55.3 
 44.7 
 
 
 
 Total 
 
 117 
 
 100.0 
 
 440 
 
 100.0 
 
 557 
 
 100.0 
 
 (3) Recidivists and First Offenders Among Female Felons and 
 Male Felons. — It may be of interest to compare with the felons in our 
 group the percentage of recidivists which Dr. Glueck found in a study 
 of 608 male felons in Sing Sing Prison.^^ Among his total group, he 
 finds 66.8 per cent recidivists in comparison with the 48.7 per cent of 
 the female felons of our study. Dr. Glueck has defined a recidivist as 
 "an individual who in addition to his present term of imprisonment, has 
 served one or more previous sentences in penal or reformatory in- 
 stitutions." Since he has not based recidivism on number of previous 
 convictions, as we have, it is probably true that he has not counted 
 as recidivists many who would be so counted in our classification 
 and, therefore, that the difference between the percentage of recidivists 
 in the two groups is larger than would appear. Even as the two 
 stand, the difference is rather striking, that there should be 18.1 
 per cent more recidivists among the male than among the female 
 felon group. 
 
 A further comparison of the recidivists and first offenders in 
 our felony group with 200 consecutive admissions of male felons to 
 Auburn Prison" shows that the recidivists among these male felons 
 are 67.5 per cent of the total group, and the first offenders 32.5 
 per cent. These percentages approximate those among the male 
 felons in Sing Sing Prison and show, again, a much larger per- 
 centage of recidivists than we find among the female felons in our 
 group. 
 
 " These unpublished figures were obtained through the courtesy of Dr. 
 Heacox, physician in Auburn Prison. 
 
128 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 FIRST CONTACTS WITH THE LAW 
 
 Equally as important as a study of the extent of delinquency 
 is a knowledge of the circumstances which attended the first con- 
 tacts with the law. We have three groups of facts which serve in 
 a general way to supply this information, — namely, the age at first 
 conviction, the nature of the offense at the first conviction, and the 
 first sentence. 
 
 (o) Age at First Conviction 
 
 (1) Age at First Conviction Among Institutional Groups. — Table 
 21 shows the per cent distribution of delinquent women in the vari- 
 ous groups by the age at first conviction. Where there has been 
 no previous conviction, present age is used. Each age, given in 
 round numbers in years, includes the time between 6 months pre- 
 vious to and 6 months following that birthday. That is, 21 years 
 as given in this study, includes offenders who at the time of the 
 present conviction were of any age between 20 years, 6 months, and 
 21 years, 6 months. The range of years in the total group runs 
 from 8 to 72. There was only one woman, in the Probation group, 
 who was convicted at 8 years, and she as an incorrigible child. 
 The only case at 72 years was a woman in the Workhouse. The 
 range varies in the institutional groups, having a wide range in 
 the Workhouse, Auburn, and the Penitentiary, and a progressively 
 narrower range in the Probation, Magdalen, and Bedford groups. The 
 latter two, by the laws regulating the ages of the inmates, could not 
 take the very old cases which we find in the Workhouse, for instance. 
 We may note that the lowest mean and standard deviation, for the 
 age at first conviction, occurs in Bedford, and becomes increasingly 
 larger in the Magdalen, Probation and the Penitentiary. Auburn 
 follows next in the mean age and the Workhouse has the highest 
 mean age. The standard deviations of these two groups is reversed, 
 however, the Workhouse showing a less wide dispersion of ages than 
 which will be used for several correlations, we find the mean age at 
 wide range of ages, such as the Workhouse, Penitentiary, and Auburti, 
 we should expect a large standard deviation. In the total group, 
 which will be used for several correlations, we find the mean age at 
 first conviction 27.51 ± .453, and the standard deviation 10.68 
 ± .321. 
 
NATURE AND EXTENT OF DELINQUENCY 129 
 
 TABLE 21 
 
 AGE AT FIRST CONVICTION 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups, 
 with Constants for Each Group 
 
 
 Institutional Groups 
 
 
 Age at Firi 
 
 c 
 
 
 
 
 
 
 Total 
 
 Conviction 
 
 
 
 
 Peniten- 
 
 Work- 
 
 
 
 Bedford 
 
 Auburn 
 
 Mage 
 
 alen tiary 
 
 house 
 
 Probation 
 
 
 8 to 12 ye 
 
 irs 1.0 
 
 1.3 
 
 1. 
 
 5 
 
 
 1.1 
 
 0.7 
 
 12 " 16 
 
 8. 
 
 a 
 
 1.3 
 
 4. 
 
 3 4.6 
 
 2.0 
 
 2.2 
 
 4.0 
 
 16 " 20 
 
 47. 
 
 5 
 
 14.1 
 
 39. 
 
 5 3.7 
 
 8.2 
 
 20.7 
 
 21.7 
 
 20 " 24 ' 
 
 23. 
 
 B 
 
 12.8 
 
 27. { 
 
 3 19.3 
 
 13.3 
 
 28.3 
 
 20.8 
 
 24 " 28 ' 
 
 ' 11. 
 
 9 
 
 15.4 
 
 15. i 
 
 ? 18.4 
 
 16.3 
 
 25.0 
 
 17.2 
 
 28 " 32 ' 
 
 4. 
 
 ■) 
 
 9.0 
 
 6.( 
 
 J 18.4 
 
 9.2 
 
 15.2 
 
 10.7 
 
 32 " 36 ' 
 
 3. 
 
 3 
 
 12.8 
 
 4.( 
 
 ) 8.3 
 
 19.4 
 
 2.2 
 
 8.3 
 
 36 " 40 ' 
 
 
 
 9.0 
 
 1.; 
 
 1 11.0 
 
 8.2 
 
 1.1 
 
 5.2 
 
 40 " 44 ' 
 
 
 
 12.8 
 
 
 6.4 
 
 9.2 
 
 2.2 
 
 5.1 
 
 44 " 48 ' 
 
 
 
 2.6 
 
 
 
 1.8 
 
 5.1 
 
 2.2 
 
 2.0 
 
 48 " 52 ' 
 
 
 
 1.3 
 
 
 
 3.7 
 
 4.1 
 
 
 1.6 
 
 52 " 56 ' 
 
 
 
 5.1 
 
 
 
 ,. 
 
 2.0 
 
 
 1.1 
 
 56 " 60 ' 
 
 
 
 
 
 
 3.7 
 
 1.0 
 
 
 .9 
 
 60 " 64 ' 
 
 
 
 1.3 
 
 
 
 .9 
 
 1.0 
 
 
 .5 
 
 64 " 68 ' 
 
 
 
 1.3 
 
 
 
 
 
 
 .2 
 
 68 " 72 ' 
 
 
 
 
 
 
 
 1.0 
 
 
 .2 
 
 Total . 
 
 ... 100.0 
 
 100.0 
 
 100. C 
 
 ) 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cs 
 
 ises 101 
 
 78 
 
 76 
 
 109 
 
 98 
 
 92 
 
 554 
 
 Mean 
 
 ... 19.. 
 
 58 
 t50 
 
 31.1 
 + 1.29 
 
 21. i 
 
 !5 30.41 
 )69 +.971 
 
 32.1 
 ±1.10 
 
 23.96 
 ±.657 
 
 27.51 
 
 
 . . . +.' 
 
 + .453 
 
 
 
 
 (T 
 
 4., 
 
 52 
 518 
 
 11.41 
 ±.914 
 
 4.C 
 
 + A 
 
 »6 10.13 
 102 ±.686 
 
 10.84 
 
 + .775 
 
 6.30 
 
 ±.464 
 
 10.68 
 
 <Ja 
 
 + .; 
 
 + .321 
 
 
 
 
 
 
 (2) Comparison with English Male Convicts. — It is of some inter- 
 est to compare the figures for our group of delinquent women with 
 Goring's data for EngHsh men convicts, although his carry more 
 weight because of the greater numbers. He gives ^^ the means and 
 standard deviations of the age at first conviction, for 2,225 men, first 
 offenders being omitted from the numbers. From these data we com- 
 puted the standard deviation of the mean. His constants are as 
 follows : 
 
 Mean age at first conviction = 22.39 ± .194. 
 Standard • deviation of age at first conviction = 
 
 = 9.172. 
 
 ' Oi>. cit. Table 285, o. 424. 
 
130 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 Our data, as given in Table 21, are not comparable with Goring's, 
 since we have included first offenders. Dropping these cases we 
 find the following constants for 304 cases, all of whom have been 
 convicted at least once before the present conviction. 
 
 Mean age at first conviction = 25.41 ± .540. 
 
 Standard deviation of age at first conviction = 9.420. 
 
 Between the mean age at first conviction of Goring's group of 
 men and that of our group of women there is a difference of 3.02 
 years — more than five times the standard deviation of the difference— 
 which is sufficient to remove any uncertainty as to the genuineness 
 of the difference. It is evident, therefore, that the men studied by 
 Goring tended to have been convicted for the first time earlier 
 than were our women. Whether this indicates a tendency common 
 to men criminals as contrasted with women; whether it is due to 
 differences in conditions between England and this country; or 
 whether it is due to some peculiarity in the selection of the groups, 
 we can not say. It is evidently not due simply to the fact that 
 his group consists entirely of felons, while ours includes both felons 
 
 TABLE 22 
 
 AGE AT FIRST CONVICTION 
 
 Per Cent Distribution of Felons and Misdemeanants 
 
 Age at First Conviction 
 
 Felons 
 
 Misdemeanants 
 
 Total 
 
 8 to 12 years 
 
 .9 
 
 2.6 
 
 12.2 
 
 16.5 
 
 17.4 
 
 11.3 
 
 10.4 
 
 7.8 
 
 9.6 
 
 1.7 
 
 1.7 
 
 3.5 
 
 1.7 
 
 1.7 
 
 .9 
 
 .7 
 
 4.3 
 
 24,1 
 
 21.8 
 
 17.3 
 
 10.5 
 
 7.7 
 
 4.6 
 
 3.9 
 
 2.1 
 
 1.6 
 
 .5 
 
 .7 
 
 .2 
 
 "."2' 
 
 7 
 
 12 " 16 " 
 
 4 
 
 16 " 20 " 
 
 21 6 
 
 20 " 24 " 
 
 20 7 
 
 24 " 28 " 
 
 17.3 
 
 28 " 32 " 
 
 10 6 
 
 32 " 36 " 
 
 8 3 
 
 36 " 40 " 
 
 5 2 
 
 40 " 44 " 
 
 5 
 
 44 " 48 " 
 
 2 
 
 48 " 52 " 
 
 1 6 
 
 52 " 56 " 
 
 1 1 
 
 56 " 60 •' 
 
 9 
 
 60 " 64 " 
 
 5 
 
 64 " 68 " 
 
 2 
 
 68 " 72 " 
 
 2 
 
 
 
 Total 
 
 100.0 
 
 100.0 
 
 100 
 
 
 
 Number of cases 
 
 115 
 
 440 
 
 555 
 
 
NATURE AND EXTENT OF DELINQUENCY 131 
 
 and misdemeanants, since the mean age of first conviction of the 
 women felons (see Table 23) is higher than that of the women 
 misdemeanants. The discrepancy between our group and Goring's 
 would thus be increased rather than diminished if comparison were 
 made with the women felons alone. 
 
 (3) Age at First Conviction Among Felons and Misdemeanants. 
 — If we proceed to the consideration of age at first conviction as 
 found among the felon and misdemeanant groups, we note that the 
 range of ages is about the same in each group. 
 
 Table 23, however, which compares the means and standard 
 deviations of the age at first conviction in the two groups, shows that 
 there is a real difference between the central tendency of the two 
 groups in favor of the felons being first convicted at an older age. 
 There is also in all probability a valid difference between the dis- 
 persion in the two groups, with a tendency toward a wider scattering 
 among the felons in the age at first conviction. 
 
 TABLE 23 
 
 FELONS AND MISDEMEANANTS 
 
 Comparison of the Means and Standard Deviations of Age at First Con- 
 viction of Felons and Misdemeanants of Total Group 
 
 
 Felons 
 
 Misde- 
 meanants 
 
 Difference 
 
 d 
 
 Chances that 
 real differ- 
 ence does not 
 exist are 1 
 in: 
 
 Mean 
 
 30.3 
 ±1.07 
 
 25.51 
 + .438 
 
 4.84 
 
 4.19 
 
 71,943 
 
 Cm 
 
 
 
 (T 
 
 11.45 
 ±.755 
 
 9.20 
 ±.310 
 
 2.25 
 
 2.76 
 
 345 
 
 (Tn 
 
 
 
 
 Cases 
 
 115 
 
 440 
 
 
 
 
 (4) Age at First Conviction Among Offenders Against Chastity 
 and Offenders Against Property Rights. — The two largest groups of 
 delinquents in the New York City Police Department classification of 
 offenders, — the oifenders against chastity and the offenders against 
 property rights, — are of interest also in a consideration of the age at 
 first conviction. The frequency distribution in Table 24 indicates that 
 the range is somewhat shorter for both groups, than in the total group. 
 
132 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 since the two last age groups are lost, and only three cases fall in the 
 first age group. 
 
 Table 25 gives the comparison of the means and standard devia- 
 tions of the above table, and shows that there is a valid difference 
 between the means,— that is, that there is a tendency for the women 
 convicted of offenses against chastity to be first convicted younger 
 than the women convicted of offenses against property rights. There 
 is much less of a difference in the dispersion of the cases among these 
 two groups of offenders, but the ratio of 1.96 would indicate that 
 there might be a difference. It is of interest here to turn back to Table 
 15, and note the tendency for women convicted of offenses against 
 property rights to have a greater number of convictions than women 
 convicted of offenses against chastity. Table 14 shows that there is 
 a larger percentage of recidivists also among those who are convicted 
 of offenses against property rights. All of these facts together would 
 make it seem probable that the offenders against property rights, 
 though first convicted at a later age, are convicted more often after 
 that time than are the offenders against chastity. 
 
 TABLE 24 
 
 AGE AT FIRST CONVICTION 
 
 Per Cent Distribution of Offenders Against Chastity and Offenders 
 Against Property Rights 
 
 Age at First Conviction 
 
 Offenders against 
 Chastity 
 
 Offenders against 
 Property 
 
 Total 
 
 8 to 12 years 
 
 1.0 
 
 3.0 
 
 28.6 
 
 21.6 
 
 IS. 9' 
 
 8.6 
 
 7.3 
 
 3.3 
 
 3.3 
 
 1.3 
 
 1.3 
 
 .7 
 
 .7 
 
 .3 
 
 '6'7' 
 
 11.2 
 
 20.9 
 
 15.7 
 
 14.9 
 
 8.2 
 
 7.5 
 
 7.5 
 
 1.5 
 
 2.2 
 
 1.5 
 
 1.5 
 
 .8 
 
 .7 
 
 12 " 16 " 
 
 4 1 
 
 16 " 20 " 
 
 23.2 
 
 20 " 24 " 
 
 21 4 
 
 24 " 28 " 
 
 17.9 
 
 28 " 32 " 
 
 10 6 
 
 32 " 36 " 
 
 7 6 
 
 36 " 40 " 
 
 4 6 
 
 40 " 44 " 
 
 4 6 
 
 44 " 48 " 
 
 1 4 
 
 48 " 52 " 
 
 1 6 
 
 52 " 56 " 
 
 9 
 
 56 " 60 " 
 
 9 
 
 60 " 64 " 
 
 5 
 
 
 
 Total 
 
 100.0 
 
 100.0 
 
 100 
 
 
 
 Number of cases 
 
 301 
 
 134 
 
 435 
 
 
NATURE AND EXTENT OF DELINQUENCY 133 
 
 TABLE 25 
 
 OFFENDERS AGAINST PROPERTY RIGHTS AND OFFENDERS 
 AGAINST CHASTITY 
 
 Comparison of the Means and Standard Deviations of Age at First 
 
 Conviction Among Offenders Against Property Rights and 
 
 Offenders Against Chastity of Total Group 
 
 
 Offenders 
 against 
 
 Property 
 Eights 
 
 Offenders 
 against 
 Chastity 
 
 Difference 
 
 d 
 Cd 
 
 Chances that 
 
 real differ- 
 ence does not 
 exist are 1 
 in: 
 
 Mean 
 
 28.31 
 
 ±.871 
 
 24.70 
 ±.601 
 
 3.62 
 
 3.60 
 
 5,000 
 
 ITja 
 
 
 
 0" 
 
 10.09 
 ±.616 
 
 8.70 
 ±.354 
 
 1.39 
 
 1.96 
 
 40 
 
 (To- 
 
 
 
 
 Cases 
 
 134 
 
 301 
 
 
 
 
 
 
 (5) Relation Between Age at First Conviction and Number of 
 Convictions. — In order to determine whether among our total group of 
 delinquent women there is any relationship between age at first con- 
 viction and the total number of convictions,^* the following correlation 
 table (Table 26) is presented. We find that the coefficient of correla- 
 tion, +.02 gives no evidence of any significant relationship. Cor- 
 relation ratios were also determined which indicated the presence of 
 a slight degree of relationship, but which are rendered of dubious sig- 
 nificance by the irregularity in the lines of means. (See table, 
 noting the means given in the right-hand column and the bottom row of 
 the table.) This can not be considered an adequate indication of the 
 amount of correlation without a correction for the influence of present 
 age. While we should expect a tendency for individuals who were first 
 convicted when very young to acquire a larger number of convictions 
 than those first convicted late in life,=^ it is evident that, if any large 
 number of those who had their first convictions at an early age are still 
 relatively young, this tendency will not have an opportunity to show 
 itself to its full degree. 
 
 °*Our heading "Number of previous convictions" tends to be misleading in 
 this connection, the term "previous" signifying "previous to the present," obvi- 
 ously not "previous to the first." 
 
 "^Our expectation is based obviously on the assumption that the treatment 
 given at the time of the first conviction is not likely to cut short the career 
 of delinquency. 
 
•■e 
 1 
 
 O 
 
 a 
 
 .a 
 
 CO 
 
 § 
 
 •■C3 
 
 
 I 
 
 .a 
 § 
 
 s 
 
 o 
 u 
 
 Means 
 (Number of 
 
 Previous 
 Convictions) 
 
 lO ® •^. c> ■* -^ 00 
 
 O ci IN ci rt* iH « 
 
 
 
 H 
 
 
 3 
 
 
 '^-' "''^^"""SSISSSgSKSSSS"'^ 
 
 6 
 
 i 
 1 
 
 CO 
 
 '^ 
 
 iH 
 
 CO 
 
 
 
 
 s 
 
 1H 
 
 1-1 
 
 03 
 
 
 
 « 
 
 N 
 
 N 
 
 i-H 
 
 
 
 CO 
 
 
 
 U3 
 
 
 
 ■* 
 
 »H iH 
 
 CM 
 
 CO 
 
 ^ 
 
 i-i 
 
 N 
 
 tH ^ 
 
 Ol 
 
 
 « ^ 
 
 « 
 
 o 
 
 i-<rH rH 
 
 eo 
 
 O) 
 
 »-l iH rHiH 
 
 ■* 
 
 00 
 
 i-t NW 
 
 Tft' 
 
 
 I> 
 
 iH iH T-t i-Hr-l N 
 
 t- 
 
 CD 
 
 rHi-ICQ iHi-H N 
 
 O) 
 
 lO 
 
 .H i-t CO »-l CO »H i-l Tjl .H 
 
 CD 
 
 1-1 
 
 
 -* 
 
 N iHCO CON »ONrH«i-i 
 
 C4 
 
 « 
 
 rHCOCOCOrH'-iP'^WTjfCDfH 
 
 00 
 CO 
 
 M 
 
 s 
 
 IN 
 
 rt rti-<^^C0(N<NU3'<J(t^OO'-i(N 
 
 CO 
 
 00 
 
 iH 
 
 rt .HCOWi*lcOcO00OSI>-*NCiCOrt 
 i-ti-i(NiM 
 
 s 
 
 (N 
 
 o 
 
 I-t ^ ■* ■* CO CO -* •* rt 00 CO OS lO rH CD O) T-f 
 
 N 
 
 
 
 
 OQ 
 
 it 
 
 1^, 
 
 UOipiAUOQ (SJIJ )B 98v 
 
 t-CO ' 
 
 n'co 8 
 
 ""i 
 
 O^O" 
 
 .".»g 
 a a£ 
 SgS. 
 
 III 
 wfflS 
 
 "I 
 niig 
 
 O 
 .. •• n 
 
 d oV3 
 
 ■4a O d 
 
 SOS 
 
 S2'g 
 
 I". 
 Sis a 
 
 h «'^ 
 
 •~ a-B 
 °b5 
 S°° 
 
 rt m ** 
 
 a.2 3 
 b« " g > a 
 
 •a * ."to 
 I Mil SO^ 
 
 ..■s.S.a 
 
 ill 
 
 «, g 5 
 o « 
 
 S.2 8 
 
 eg I 
 
 si 1 
 
 C40 
 
 n II 
 
 b b 
 
 (OlO 
 II II 
 
 a a 
 a OS 
 
 I.. 
 
 ^1. 
 
 134 
 
NATURE AND EXTENT OF DELINQUENCY 135 
 
 To show the relationship between age at first conviction and total 
 number of convictions cleared of the influence of age, we have de- 
 termined the partial correlation coefficient of age at first conviction and 
 number of "previous" convictions for constant age^^ which is as fol- 
 lows : riK.A = — .619 ± .072. Although the size of the partial 
 coefficient is affected by the smallness of r, which can not be consid- 
 ered a wholly reliable value in view of the non-linearity of the rela- 
 tionship, it remains true that, even were r as large as either of the 
 ■q's, the partial would still be large enough to be significant and would 
 be negative in sign. In other words, data on the group under consider- 
 ation indicate clearly that the earlier the first conviction the greater 
 the likelihood of many convictions. 
 
 It is easier, however, to determine the fact of the relationship than 
 to account for it. There are several fairly obvious explanations. The 
 first is the simple fact that one convicted early has a longer period of 
 time ahead of him in which to accumulate a criminal record. Less 
 superficial is the assumption that an early conviction is symptomatic 
 of a marked criminal tendency, or, if one objects to the implications 
 of this term, at least of a combination of characteristics which makes 
 the individual particularly susceptible to pernicious influences. Still 
 another explanation might be offered on the assumption that the in- 
 fluences of the courts and the penal institutions may be so unfortunate 
 in their effect upon the young delinquent that they exaggerate, rather 
 than diminish, his criminal susceptibility. We do not pretend to de- 
 cide as to which of the above explanations is the more probable. 
 Very possibly each factor suggested is operative to some degree. 
 
 (b) Nature of First Offense 
 
 The nature of the first offense committed has been classified, as was 
 the present offense, by the New York City Police Department classi- 
 fication, and by felonies and misdemeanors. The first classification 
 as given in Tables 27 may be best interpreted in connection with Table 
 4, giving the nature of the present offense. It will be noted that the 
 first offense still has its highest percentage in the group of offenders 
 
 ""The data required for the above partial correlation are as follows: 
 Correlation coefficient of age at first conviction and number of "previous" 
 convictions : rm = .02. 
 
 Correlation coefficient of age at first conviction and present age : Ha = .942. 
 Correlation coefficient of number of "previous" convictions and present age: 
 tna = .23. 
 
 Tin — rNAt 
 
 The formula for the partial is rm.A = , 5 — / 5 — 
 
 yi— r^NAVi^r'^iA 
 
136 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 27 
 
 NATURE OF FIRST OFFENSE 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups, 
 
 Classified by Main Divisions of New York City Police 
 
 Department Classification 
 
 
 Institutional Ghoups 
 
 
 Nature of Offense 
 
 Bed- 
 ford 
 
 Auburn 
 
 Mag- 
 dalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Proba- 
 tion 
 
 Total 
 
 Offenses against the Person 
 
 Offenses against Chastity 
 
 Offenses against Family and 
 Ohildren 
 
 1.0 
 
 49.5 
 5.0 
 
 6.9 
 
 1.0 
 
 11.0 
 25.7 
 
 17.9 
 20.5 
 
 5.1 
 
 5.1 
 1.3 
 
 42.3 
 
 7.7 
 
 63.2 
 
 9.2 
 
 7.9 
 19.7 
 
 5.6 
 13.1 
 
 .9 
 12.1 
 
 47.7 
 20.6 
 
 42.4 
 
 9.1 
 
 2.0 
 46.5 
 
 81.5 
 
 6.5 
 
 1.1 
 10.9 
 
 3.8 
 44.3 
 
 1 8 
 
 Offenses against Regulations for 
 Public Health, Safety and 
 
 8.3 
 
 Offenses against Administration 
 of Government 
 
 .4 
 
 Offenses against Property 
 Rights 
 
 18.8 
 
 
 22.6 
 
 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 
 
 
 101 
 
 78 
 
 76 
 
 107 
 
 99 
 
 92 
 
 553 
 
 
 
 against chastity, but that the general criminality group takes second 
 place. This is undoubtedly due to the fact that the charges of in- 
 corrigible and ungovernable child, and the semi-juvenile offenses come 
 in this general group. Offenses against property rights have third 
 highest place instead of second place as among the present offenses. 
 Among the other four divisions of offenses there are slight but not 
 noticeably large differences from the nature of the present offense. 
 
 The nature of the first offense as classified by felonies and mis- 
 demeanors is also only very slightly different from the percentage of 
 felonies and misdemeanors at the time of the present offense. The 
 only group which changes more than one per cent is Auburn which has 
 29 women or 37.2 per cent who were misdemeanants at the time of the 
 

 
 a 
 
 
 o 
 
 
 a 
 
 
 m 
 
 
 <u 
 
 
 a 
 
 
 v 
 
 
 •rt 
 
 
 CO 
 
 
 •»^ 
 
 
 s 
 
 
 Id 
 
 
 d 
 
 
 OS 
 
 
 to 
 
 
 V 
 
 
 •a 
 
 
 o 
 
 
 
 
 « 
 
 
 Pk 
 
 
 >. 
 
 
 A 
 
 
 •d 
 
 
 0) 
 
 
 
 
 03 
 
 
 
 H 
 
 fS 
 
 t/3 
 
 O 
 
 S 
 
 
 M 
 
 o. 
 
 (>< 
 
 3 
 
 (z< 
 
 2 
 
 O 
 
 O 
 
 H 
 
 n 
 
 XfX 
 
 n 
 
 •^ 
 
 fe 
 
 S 
 
 fa -js 
 
 
 
 
 4a 
 
 m 
 
 t~ 
 
 o 
 
 1 
 
 
 
 % 
 
 lO 
 
 ■* 
 
 Q 
 
 
 
 M 
 
 tH 
 
 00 
 
 o 
 
 •-j 
 
 V 
 
 
 
 rH 
 
 1 
 
 fM 
 
 
 
 
 M 
 
 
 
 
 El 
 
 1 
 
 t> 
 
 o 
 
 t>. 
 
 
 
 00 
 
 OC 
 
 0= 
 
 
 
 a 
 
 
 ■* 
 
 U5 
 
 
 
 IS 
 
 
 
 
 
 
 43 
 
 
 
 
 
 
 
 
 o 
 
 O 
 
 
 
 
 
 
 O 
 
 
 
 s* 
 
 
 
 O 
 
 
 d 
 ^ 
 
 Ol 
 
 
 1-t 
 
 tH 
 
 
 
 
 
 
 
 2 - 
 
 <i> 
 
 
 IN 
 
 IN 
 
 
 Ph 
 
 
 O 
 
 O 
 
 
 
 § 
 
 
 T-H 
 
 i—t 
 
 
 
 ^; 
 
 
 
 
 
 
 
 
 
 
 
 s 
 
 
 o 
 
 O 
 
 
 
 
 
 
 
 
 1 
 
 
 
 o 
 
 o 
 
 
 s 
 
 
 o 
 
 o 
 
 
 Oh 
 
 
 T-H 
 
 1-H 
 
 {H 
 
 
 
 
 
 1 
 
 1 
 1 
 
 
 c 
 
 T-H 
 
 o 
 
 T-H 
 
 
 
 
 
 
 
 
 d 
 
 »i: 
 
 U3 
 
 o 
 
 
 £■ 
 
 s 
 
 
 
 
 
 
 t^ 
 
 IM 
 
 o 
 
 IS 
 
 •J 
 
 «* 
 
 <N 
 
 t~ 
 
 o 
 
 P 
 o 
 K 
 
 o 
 
 ■a 
 
 CL| 
 
 
 
 T-H 
 
 lU 
 CD 
 
 
 
 
 hJ 
 
 ^ 
 
 
 2; 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CC 
 
 ■* 
 
 o 
 
 g 
 
 g 
 
 o 
 
 (M 
 
 t^ 
 
 d 
 
 
 ■3 
 
 ■a 
 
 u 
 
 
 o- 
 
 o 
 
 s 
 
 Ah 
 
 
 
 T-H 
 
 »H 
 
 
 
 
 
 
 12; 
 
 (M 
 
 s 
 
 CD 
 
 
 
 
 
 
 
 
 3^ 
 
 OC 
 
 IN 
 
 O 
 
 
 e 
 
 u 
 
 <M 
 
 l> 
 
 d 
 
 
 % 
 
 CC 
 
 ec 
 
 o 
 
 
 3 
 
 Pk 
 
 
 
 T-H 
 
 
 .X3 
 
 b 
 
 
 
 
 
 ^ 
 
 09 
 
 § 
 
 § 
 
 00 
 
 
 
 
 
 
 
 
 1 
 
 o- 
 
 tH 
 
 o 
 
 
 •s 
 
 u 
 
 IC 
 
 -* 
 
 d 
 
 
 <s 
 
 
 o- 
 
 o 
 
 
 £ 
 
 CL| 
 
 
 
 tH 
 
 
 t3 
 
 tH 
 
 
 
 
 
 
 a 
 
 cc 
 
 § 
 
 T-H 
 
 O 
 tH 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ia 
 
 
 
 
 
 
 
 O 
 
 
 
 
 
 
 
 H-l 
 
 
 
 
 
 
 
 O 
 
 
 
 
 
 
 
 
 
 
 KT 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 1 
 
 
 
 
 J 
 
 H 
 
 
 
 
 c 
 
 •^ 
 
 
 
 
 
 c 
 
 .a 
 
 
 
 
 
 ^ 
 
 S 
 
 
 1 
 
 137 
 

 
 
 tS-5- 
 
 
 
 
 
 
 
 
 
 SiS-J 
 
 lOlO 00 ««■*(» 
 
 :S 
 
 
 
 
 S^l 
 
 S^iSSS^SSS 
 
 (O 
 
 
 
 
 
 
 
 
 2ag 
 
 
 
 
 
 
 
 •<o 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 •3 
 
 «COOIO««CO 
 
 »H 
 
 
 
 
 
 
 
 o 
 
 
 
 s 
 
 
 Zi-69 
 
 1-t 
 
 -- 
 
 
 
 69-99 
 
 ^ 
 
 iH 
 
 
 
 
 
 
 
 
 
 
 99-69 
 
 
 
 
 
 
 
 
 
 
 
 o 
 
 -•J 
 
 *^4 
 
 
 89-09 
 
 
 CO 
 
 
 
 09-1.9 
 
 N « 
 
 ■* 
 
 
 
 L9^K 
 
 N^ ^ 
 
 ■<* 
 
 
 
 
 
 
 
 
 o 
 
 £ 
 
 re-I9 
 
 « ^^ 
 
 ^ 
 
 
 
 IS-8^ 
 
 iHeo <H CO 
 
 » 
 
 
 
 
 
 
 
 
 1 
 
 g 
 
 8*-S* 
 
 WIN M ^^ 
 
 00 
 
 
 
 
 
 
 
 
 g 
 
 B* Zf 
 
 i-H (NO rHCO 
 
 CO 
 
 
 
 o 
 
 Z*-6E 
 
 i-CCD ■* COtO 
 
 
 11 
 
 
 68-98 
 
 (M<Di-lCC t*-«*l 
 
 S2 
 
 b 
 
 
 1 
 
 
 
 
 
 
 s 
 
 98-€8 
 
 CQION-* 0O(N 
 
 s 
 
 s 
 
 1 
 
 9 
 
 88-08 
 
 iHioNrjf osr* 
 
 CO 
 
 «£; 
 
 •I-* 
 
 
 08-i3 
 
 THUS <C «00 
 
 s 
 
 S-H 
 
 E-l 
 
 iZ-^ 
 
 MO.-IO 0<© 
 CO rH t-H 
 
 K 
 
 Is 
 
 
 »B-IZ 
 
 «Nrt«#i-i(DC 
 
 CO 
 00 
 
 § " 
 
 
 
 
 
 
 
 Id 
 
 
 IB-SI 
 
 tH00i-i os« 
 
 ss 
 
 'P o 
 
 
 8I-SI 
 
 s -^ °=ss 
 
 s 
 
 11 
 
 
 BI-Zl 
 
 iH OJt* 
 
 o 
 
 s? 
 
 
 a 
 
 
 
 
 
 <a-a 
 
 
 .fi 
 
 
 zi-e 
 
 eo 
 
 M 
 
 ^•1 
 
 
 
 6-fl 
 
 -- 
 
 - 
 
 la 
 
 
 
 
 
 
 
 
 
 
 
 d 
 
 
 
 
 
 : :a 
 
 
 
 
 
 o 
 
 
 
 
 
 ; o S 
 
 
 
 
 
 ■J3 
 
 
 
 
 
 * ^ S 
 
 
 
 
 
 U 
 
 
 i 
 
 •s 
 1 
 
 1 
 
 > 
 
 'i 
 t 
 
 
 +■ 
 
 *< 
 
 
 
 
 
 
 ^21 
 
 gains 
 sains 
 gains 
 gains 
 ;ains 
 gains 
 
 ^ 
 
 
 
 
 
 
 i- 
 
 
 
 
 
 
 
 
 
 
 
 
 88 « ca girt sa;. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 g d a S e S ^ 
 
 
 
 
 
 
 
 (otoioiBtoid § 
 
 
 
 
 
 
 
 C 
 
 >c 
 
 JOOOC 
 
 )C 
 
 I 
 
 
 138 
 
NATURE AND EXTENT OF DELINQUENCY 139 
 
 first conviction, and only one misdemeanant at the latest conviction. It 
 is also of interest to note that in neither the Probation nor the Work- 
 house group, which are limited to misdemeanants, is there any woman 
 who was convicted of a felony as the first offense. 
 
 One might expect, from a study of the differences between the first 
 and latest offense of the women in this study that there would be a 
 relationship between the age at first conviction and the nature of the 
 first offense. This we find to be true in Table 29 where the correla- 
 tion ratio between age at first conviction and nature of first offense 
 has been calculated. The ratio of .35 zt .037 would indicate that there 
 Is a genuine relationship between these two factors. Reference to the 
 mean values given in the extreme right-hand column shows that the 
 order of arrangement of kinds of offenses in accordance with in- 
 creasing age at first conviction is as follows: (1) General criminality, 
 (2) Offenses against chastity, (3) Offenses against property rights, 
 (4) Offenses against the family, (5) Offenses against regulations for 
 public health, (6) Offenses against the person, (7) Offenses against 
 the administration of government. 
 
 {c) First Sentence 
 
 After observing the similarity between the first and latest offenses, 
 we might expect to see a similarity in the sentences imposed for them. 
 This we find to be the case in Table 30. In each institutional group 
 studied it is evident that the largest proportion of cases were sent to 
 an institution of that type at the time of the first sentence. We must 
 remember that a fairly large percentage in each group are first of- 
 fenders and so in those cases the first sentence will apply to the present 
 sentence. The similarity in type of sentence is especially striking in 
 the Workhouse and Probation groups, one of which has 91.2 per cent 
 first sentenced to institutions such as the Workhouse, the Penitentiary 
 or County Jails, and the other with 92.1 per cent first put on proba- 
 tion. It would seem, particularly in the Workhouse group, that if one 
 were once sentenced to the Workhouse, she would continue to be sent 
 there after subsequent convictions, and that possibly there is some proc- 
 ess of selection in the courts which makes women of a certain type 
 more likely to be sentenced there. 
 
140 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 30 
 
 FIRST SENTENCE 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups 
 
 
 Institutional Groups 
 
 
 First Sentence 
 
 Bed- 
 ford 
 
 Auburn 
 
 Mag- 
 dalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Proba- 
 tion 
 
 Total 
 
 Probation or Suspended Sen- 
 
 22.8 
 5.0 
 5.0 
 
 58.4 
 
 8.9 
 
 7.5 
 7.5 
 1.3 
 5.0 
 
 23.8 
 55.0 
 
 26.3 
 1.3 
 
 65.8 
 6.6 
 
 10.9' 
 2.7 
 1.8 
 
 10.9 
 
 70.9 
 
 2.7 
 
 2.0 
 1.0 
 1.0 
 
 4.9 
 91.2 
 
 92.1 
 2.0 
 
 1.0 
 2.9 
 
 2.0 
 
 27 5 
 
 Fine 
 
 3 2 
 
 Juvenile Institution 
 
 1 8 
 
 Reformatory Institution 
 
 Penitentiary, Workhouse or 
 County Jail 
 
 23.3 
 36 1 
 
 State Prison 
 
 8 2 
 
 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100 
 
 
 
 Number of cases 
 
 101 
 
 80 
 
 76 
 
 110 
 
 102 
 
 102 
 
 571 
 
 
 
 
 
 (d) Juvenile Delinquents 
 
 While we are discussing the first offense, it may be well to note 
 the number of our cases who have been juvenile delinquents, in the 
 sense of being convicted in a court of law before 16 years of age. In 
 Table 21, where 16 years has been interpreted as being any age be- 
 tween 15 years, 6 months, and 16 years, 6 months, we find 26 cases or 
 4.4 per cent of our total group who appear as juvenile delinquents, 
 though all were actually convicted before the age of IS years, 6 months. 
 If we select, however, cases of all those who were convicted before 
 they had reached their 16th birthday, the upper limit of the Juvenile 
 Court age, we find 40 women or 6.8 per cent of our total group. 
 
 This small percentage of women who have been convicted as juve- 
 nile delinquents is striking in view of the general belief that, as Dr. 
 Healy states :^^ 
 
 " Healy, VVilliam. "The Individual Delinquent." Boston. Little, Brown & 
 Co., 1915, p. 10. ' 
 
NATURE AND EXTENT OF DELINQUENCY 141 
 
 "Practically all confirmed criminals begin their careers in childhood or 
 early youth. The fact of this remarkable early development of a definite 
 tendency towards criminality was soon clear to us, both through observable 
 trends in young offenders, and through the life histories of older delinquents. 
 . . . Another writer, cited by Morrison, states, 'It is an ascertained fact that 
 there is scarcely an habitual criminal in the county of Staffordshire who 
 has not been imprisoned as a child.' Even more important is the thor- 
 ough research of Matz, who investigated the prison population of the 
 province of Pommern. He found that in 70 per cent the first imprison- 
 ment had been inflicted before the 21st year, and that of the repeated 
 offenders by far the greater number had received their first punishment 
 before the 17th year. It must be remembered that here Matz is speaking 
 not of the commission of the first offense, but of receiving the first punish- 
 ment." 
 
 Lombroso writes, "All great criminals have given proof of perver- 
 sity in their youth, especially at the age of puberty and sometimes 
 even before." ^^ He goes on to cite instances of numerous French- 
 Italian penologists who have found many criminals with the "tendency 
 to theft," "little pilferings," etc., when they were very young. He 
 does not mean that these early delinquencies were necessarily brought 
 into court, but that the individual had a "tendency" toward criminality 
 and showed it in early youth by various forms of "perversity." 
 
 At present the chief criterion of assuming that a person is criminal 
 is the fact of his conviction in a court of law. To argue that all great 
 criminals, though not convicted when children, began their criminal 
 careers in childhood, would be to assume that it was not common for 
 those who grow to adult life without contact with the courts also to 
 perform the acts which the criminal may have done when he was a 
 child. Of this we have no proof. Normal children undoubtedly do a 
 certain amount of pilfering and commit other anti-social acts. Until 
 we have some way of measuring the extent of "perversity" in the 
 normal youth who does not become a confirmed criminal, we are not 
 in a position to state that the criminal who was not convicted during 
 his childhood but who did show a "definite tendency towards crimi- 
 nality" necessarily began his "criminal career in early youth." To 
 hold that view carried to the extreme would be to agree with Lom- 
 broso that'X^Precocity in crime points to the fact that criminality, much 
 more than insanity, is an inherited characteristic. This reminds us 
 
 ^'Lombroso, Cjesare. "Crime: Its Causes and Remedies." Translated by 
 Henry P. Horton. Boston. Little, Brown & Co., 1911, p. 178. 
 
142 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 that precocity is one of the distinguishing features of savage peoples, — 
 a new proof of the atavistic origin of crime." ^^ \ 
 
 We might cite here cases of several of the women in Auburn 
 Prison who were convicted of serious offenses in the eyes of the law, 
 and who were recidivists. Yet in the group of eighty women, whose 
 records were carefully investigated, there were only two cases where 
 there had been convictions under sixteen years of age. One of the 
 women was feeble-minded, epileptic and consequently very unstable. 
 Her crimes seemed more a matter of her instability than of any crimi- 
 nal intent. The other juvenile delinquent in the Auburn group was a 
 woman who would, without doubt, fit in with Dr. Healy's idea of a 
 confirmed criminal beginning his career in childhood or early youth. 
 Her first conviction occurred when she was ten years and eight months 
 of age. At that time she ran away from home, and when found was so 
 troublesome that she was committed to the Catholic Protectory as an 
 ungovernable child. When- fourteen, she was arrested as a common 
 prostitute and sent to the New York State Training School at Hudson. 
 In this institution she was so troublesome that she was thought to be 
 insane and was transferred to Matteawan, where she remained seven 
 months and was discharged as not insane. The report from Matteawan 
 showed at that time a history of prostitution, stealing and drug habit. 
 Since then she has been arrested many times for soliciting and loiter- 
 ing, but has served only two terms, one for running a disorderly house 
 and one for assault. Her present conviction was for murder in the 
 second degree. For $500 she had made the arrangement for two men, 
 belonging to a Black Hand gang, to kill an Italian whose wife wanted 
 him out of the way so she could marry another man. This one case is 
 practically alone, out of the group of the most serious offenders we 
 have, in showing marked criminal traits during childhood and adoles- 
 cence. Since the group of women studied in Auburn Prison covered 
 the commitments for two consecutive years, it would seem that the 
 sample of offenders is representative enough for us to state that among 
 the most serious women offenders in New York State during the 
 years 1915-16, there was no proof that their criminal careers, in the 
 sense of being convicted in court, began in childhood or early youth. 
 Some may object that the juvenile court age limit is set too low and 
 that the convictions which take place within the next few years are as 
 significant as those which may occur earlier. Out of 587 cases, how- 
 ever, we find that only 47.2 per cent were convicted under 24 years of 
 " Op. cit., p. 177. 
 
NATURE AND EXTENT OF DELINQUENCY 143 
 
 age; that is, that over half of our cases did not reach the courts until 
 
 they were well into the adult period. From this observation, it would 
 
 be difificult to state that the determinants of • delinquent careers are 
 
 necessarily the conditions of youth, since the first of the career in 
 
 the courts comes at a period widely separated from the conditions 
 
 of youth. 
 
 The forty women who had been juvenile delinquents are found in 
 
 the institutional groups in the following numbers : 
 
 Bedford 18 cases 
 
 Auburn 2 " 
 
 Magdalen 8 " 
 
 Probation . . . ^ 5 " 
 
 Penitentiary 5 " 
 
 Workhouse 2 • " 
 
 Two of the Bedford and one of the Magdalen cases came into this 
 study at the time of their first conviction when they were nearly but 
 not quite sixteen years of age, though their legal age was sixteen.^" 
 
 The numbers^ are too small to make any comparison with other 
 groups, but the .main factors in the criminal record of the juvenile 
 delinquents are given to show- that there is no strikingly marked dif- 
 ference from the rest of our cases.^ The present offenses of this group 
 are distributed as follows: 
 
 Offenses against the Person 2 
 
 Offenses against Chastity (all soliciting or loitering) 22 
 
 Offenses against the Family, etc 1 
 
 Offenses against Regulations for Public Health 1 
 
 Offenses against Administration of Government 1 
 
 Offenses against Property Rights 10 
 
 General Criminality 3 
 
 Only five of the forty cases were felons at the time of this study. 
 
 The first offenses of the women in this delinquent group were as 
 
 follows : 
 
 Offenses against Chastity 6 cases or IS.0% 
 
 Offenses against Property Rights 3 " " 7.5% 
 
 General Criminality 30 " " 75.0% 
 
 Unknown 1 " " 2.5% 
 
 All of the first offenses were misdemeanors with the exception of 
 one case. 
 
 The number of previous convictions of this group who were juve- 
 nile delinquents does not show that they have been recidivists to as 
 great an extent as have the women in some of our institutional groups. 
 The mean number of convictions is 1.82, which is less than the mean 
 number for either the Penitentiary or Workhouse groups. 
 
 '"Legal age is the age given by the delinquent in court at the time of her 
 conviction. 
 
144 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 No previous convictions 3 cases 
 
 1 
 2 
 3 
 
 4 
 12 
 
 conviction 22 
 
 convictions 7 
 
 3 " 
 
 " 4 " 
 
 " 1 case 
 
 6 " 
 
 " 15.0% 
 
 7 " 
 
 " 17.5% 
 
 4 " 
 
 " 10.0% 
 
 6 " 
 
 '■ 15.0% 
 
 1 case, 
 1 " 
 
 " 2.5% 
 " 2.5% 
 
 1 " 
 
 " 2.5% 
 
 The number of months served, as would be expected, is somewhat 
 higher than for the institutional groups or the total group of cases, 
 since juvenile delinquents usually receive an institutional commitment 
 which in many cases involves a long and indefinite stay in a reforma- 
 tory institution. The sentence for a similar offense among older per- 
 sons would be much shorter in most cases. The distribution of time 
 served in penal institutions follows : 
 
 Never served time 14 cases, or 34.8% 
 
 Served less than 1 year 6 "~" 
 
 1 to 2 years 7 
 
 2 " 3 " 
 
 3 " 4 " 
 
 4 " 5 " 
 
 5 " 6 " 
 
 7 " 8 " 
 
 As a group, it may be said that the juvenile delinquents do not 
 show up as noticeably different in any way from, the rest of the 
 cases. The numbers are too small for statistical comparisons, but 
 the smallness is significant as showing that among the sample of 
 women covered in our study, the problem of juvenile delinquency has 
 been a minor one, and for the most part, there is no proof that the 
 most serious female delinquents have begun their criminal careers in 
 childhood or early youth. 
 
 It should be noted that the chief insistence on the extensiveness 
 and importance of juvenile convictions in connection with the problem 
 of recidivism has come from persons concerned primarily with men 
 delinquents or with boy offenders. It is entirely possible that the situ- 
 ation is quite different as it affects women delinquents. Unfortunate- ' 
 ly, adequate data for comparison are not available. Goring's figures are 
 the most valuable for this purpose.^^ We have already called atten- 
 tion to the fact that the mean age of first conviction is lower for his 
 group than for ours. (See page 130). The difference in numbers of 
 very youthful offenders is also striking. 
 
 Referring to Goring's table (Table 285) we find that 337 (or 15.1 
 per cent) out of a total of 2,225 men, none of whom were first offend- 
 ers, had been convicted before they were fifteen years old. On the 
 " Op. cit. Table 285, p. 424. 
 
NATURE AND EXTENT OF DELINQUENCY 145 
 
 other hand, out of a total of 304 women who had had one or more 
 previous convictions, only fifteen cases (or 4.9 per cent) had been con- 
 victed prior to their fifteenth year.^^ 
 
 SUMMARY 
 
 In conclusion, we shall summarize the most important factors which 
 arise in considering the delinquency among the women in our study. 
 First, it is well to keep in mind that the New York City Police De- 
 partment classification of offenses which we have used throughout 
 is composed of seven main divisions, each of which is unevenly di- 
 vided as to felonies and misdemeanors. Since two of our groups, the 
 Workhouse and the Probation groups, are made up entirely of mis- 
 demeanants and Auburn entirely of felons, while both the more and 
 less serious offenses are found in the other three groups, it is natural 
 that we should find great irregularities in the distribution of the 
 various divisions of the New York City Police Department classifica- 
 tion among our several groups of delinquent women. While in the 
 total group the largest percentage of women fall among the offenders 
 against chastity, with offenders against proEerty_jughts having the 
 next largest percentage, this ratio holds true in no institution except 
 Bedford, though the offenders against chastity have the highest per- 
 centage in each group except Auburn and the Penitentiary. 
 
 In showing the extent of delinquency, our best measure seems to 
 be the number of wevious^coiiyictions, since the, number of terms and 
 length of time served are both open to serious criticisms. They both 
 disregard the possibility of the suspended sentence or probation, and 
 in addition to this the sentences for the same offense vary so between 
 institutions of a reformatory type and the Workhouse, for instance, 
 that a term cannot mean the same in both cases, and the different 
 length of time is no criterion. The distribution of number of convic- 
 tions among the institutional groups shows that the Workhouse and 
 Penitentiary have the largest mean number of convictions, and the 
 Probation group the smallest, while the other three groups stand fairly 
 close together on intermediate ground. Between felons and misde- 
 meanants there appears to be no demonstrable difference in the mean 
 number of convictions, but among the offenders against property 
 rights and the offenders against chastity there seems to be a valid dif- 
 
 ""The above data are easily obtained from Table 30 by subtracting from the 
 total number of SS4 the 250 cases of first offenders. The number convicted 
 before IS years of age can be counted up from the table directly. 
 
146 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 ference both in the tendency of the offenders against property rights 
 to have a larger number of convictions and in the variation of the 
 distribution of convictions, the offenders against property rights hav- 
 ing a wider scattering. 
 
 In the total group, the percentage of recidivists is 55.3. This per- 
 centage varies, however, in the institutional groups from 16.3 in the 
 Probation to 77.0 in the Workhouse. Among the felons, the percent- 
 age of recidivists is 48.7, considerably lower than the percentage of 
 male felon recidivists, in studies both of Sing Sing Prison and Auburn 
 Prison. = 
 
 The age at first conviction varies among the institutional groups, 
 from 19.58 years in Bedford to 32.1 years in the Workhouse. Among 
 the felons and misdemeanants it is interesting to note that there is ap- 
 parently a real difference between the age at first conviction with a 
 tendency for the felons to be convicted at a later age. There seems 
 to be a somewhat like difference between the age at first conviction of 
 the offenders against property rights and offenders against chastity, 
 with a tendency for the offenders against property rights to be con- 
 victed at a later age. 
 
 In considering the nature of the first offense, we find that the offend- 
 ers against chastity still have the largest percentage of cases in all 
 institutions except Auburn and the Penitentiary, where the offenders 
 against property rights have the highest percentage, of cases. There 
 is a large increase in the general criminality offenses due to the fact 
 that in this group the semi-juvenile delinquencies occur which we 
 should expect in many of the younger cases to be the first offense. 
 
 One of the most important things found in the study of the first 
 contacts with the law, is the surprisingly small number of women who 
 have been convicted as juvenile delinquents, ah^ within this small 
 group the lack of marked dissimilarity to the total number of cases 
 studied. 
 
 The basic elements of the criminal record as given in this chapter 
 will be used in the later chapters' as showing the relationship between 
 the nature and extent of delinquency, the various environmental fac- 
 tors and mentality. These relationships, as previously stated, will be 
 treated as associated with the delinquency, but not necessarily as 
 causative factors, since there are no figures for the general population 
 to prove that such factors are not there also a commonly occurring 
 condition. 
 
CilAPTER VII 
 
 MISCELLANEOUS CONSIDERATIONS 
 
 THOUGH the problems in which we are primarily interested in this 
 study are the larger ones of school, work, home, family background, 
 and sex life in their relation to delinqviency, there are various more gen- 
 eral elements to be considered which affect these more important prob- 
 lems in numerous ways. The factor of age, for instance, may be very 
 necessary in determining whether there is any significant difference in 
 the number of convictions between two groups. It may also be the 
 basic element in the difference between prevailing wages of any two 
 groups to be compared, or in the length of time prostitution has been 
 carried on, since in the one case we might expect those who are still 
 very young to have a lower prevailing wage, and in the other, a shorter 
 time in prostitution. There are various other miscellaneous points to 
 be noted, relating to the social status of the women, and these will be 
 presented -briefly so that they may be used for reference in later chap- 
 ters.- Among these, we shall consider the record of previous commit- 
 ments to hospitals for the insane, habits as shown by the amount of 
 alcoholism, drug addiction, and use of tobacco among the different in- 
 stitutional groups, and a very brief summary of the criminal records of 
 the twenty-one women in the special Intoxication Group of the Work- 
 house. Civil condition, age at marriage, and religion will also be shown 
 in their distribution among the institutional groups. Though these 
 quite tmrelated factors do not add appreciably to the understanding of 
 any one problem, they serve as a background to the later discussion of 
 comparisons of institutional groups. 
 
 AGE 
 
 Probably one of the most significant factors in a study of a some- 
 what heterogeneous group of individuals is the element of age. Age 
 may help to account for a very high or a very low number of convic- 
 tions, for a long or short history of prostitution, for a high or a low 
 
 147 
 
148 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 prevailing wage. Accordingly, it is necessary to define clearly what 
 unit is taken for measuring age, and to keep the factor of age in mind 
 in connection with each important subject. Throughout this study, age 
 is given in years, in round numbers. Each age in years includes the 
 time between six months previous to and six months following that 
 birthday. That is, sixteen years, as given in Table 31, includes offend- 
 ers who at the time of the present conviction were of any age between 
 fifteen years, six months, and sixteen years, six months.^ 
 
 TABLE 31 
 
 AGE AT PRESENT CONVICTION 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups 
 
 
 Age 
 
 iNSfelTnTIONAL GbOUPB 
 
 
 
 
 
 
 
 
 
 Total 
 
 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Probation 
 
 
 14 to 
 
 18 yrs... 
 
 22.8 
 
 
 14.5 
 
 .9 
 
 2.0 
 
 5.5 
 
 7.5 
 
 18 " 
 
 22 " .. 
 
 37.6 
 
 8.9 
 
 32.9 
 
 4.6 
 
 4.0 
 
 35.9 
 
 20 1 
 
 22 " 
 
 26 " .. 
 
 19.9 
 
 18.8 
 
 25.0 
 
 16.6 
 
 14.9 
 
 24,0 
 
 19.6 
 
 26 " 
 
 30 " .. 
 
 11.9 
 
 13.9 
 
 15.7 
 
 17.4 
 
 16.8 
 
 17.4 
 
 15 7 
 
 30 " 
 
 34 " .. 
 
 6.0 
 
 11.3 
 
 3.9 
 
 18.4 
 
 14.9 
 
 12.1 
 
 11 5 
 
 34 " 
 
 38 " .. 
 
 1.0 
 
 16.3 
 
 6.6 
 
 13.8 
 
 9.9 
 
 
 7.9 
 
 38 " 
 
 42 " .. 
 
 1.0 
 
 10.1 
 
 1.3 
 
 8.3 
 
 14.9 
 
 3.3 
 
 6.7 
 
 42 " 
 
 46 " . . 
 
 
 7.6 
 
 
 4.6 
 
 5.9 
 
 
 3.0 
 
 46 " 
 
 50 " .. 
 
 
 
 3.9 
 
 
 
 8.3 
 
 5.0 
 
 2.2 
 
 3 5 
 
 50 " 
 
 54 " .. 
 
 
 
 5.1 
 
 
 
 1.8 
 
 5.0 
 
 
 2 1 
 
 54 " 
 
 58 " .. 
 
 
 
 2.5 
 
 
 
 1.8 
 
 3.0 
 
 
 1.3 
 
 58 " 
 62 " 
 
 62 " . . 
 66 " .. 
 
 
 
 1.3 
 
 
 
 2.8 
 .9 
 
 2.0 
 1 
 
 
 1.1 
 .4 
 .2 
 .2 
 
 66 " 
 
 70 " .. 
 
 
 
 1.3 
 
 
 
 
 
 
 70 " 
 
 74 " .. 
 
 
 
 
 
 
 
 1.0 
 
 
 Total.. . 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cases 
 
 101 
 
 80 
 
 76 
 
 109 
 
 101 
 
 92 
 
 559 
 
 Mean 
 
 
 21.67 
 
 33.8 
 
 22.97 
 
 33.67 
 
 34.9 
 
 24.47 
 
 
 
 
 28.79 
 
 O^m- . 
 
 
 + .480 
 
 ±1.17 
 
 ±.603 
 
 ±.950 
 
 ±1.09 
 
 ±.647 
 
 
 
 ±.429 
 
 a 
 
 4.82 
 
 10.42 
 
 5.26 
 
 9.92 
 
 10.97 
 
 6.21 
 
 
 
 
 10.13 
 
 <Ta. . . 
 
 
 
 ± S'SO 
 
 ±.824 
 
 
 ±.672 
 
 ±.772 
 
 ±.458 
 
 
 
 
 
 
 
 
 ±.303 
 
 datI^wLre*thiT<-on?,l't!''nSr- '"^ "'""^J" *^' ^*"<ly' ^^ ^^^^^ «" verified 
 data where this could be obtained, but where we did not have any verifica- 
 
 ^^*K^,f"?.f^'''^■'*^*,^"^"* ^^ ''^^" "«ed. For 88 Bedford cales on 
 which both the subject's statement of age and the verified age were avail^ 
 able we find by calculating the correlation between these two ffcto7s thaT?he 
 coefficient of correlation, is .958, indicating that there is a very S^elation- 
 ship between the age which the subject gives as her Le at vZl Af fJ,» !.„c^nf 
 conviction, and the actual age determined by verification ^ 
 
MISCELLANEOUS CONSIDERATIONS 
 
 149 
 
 Table 31, followed by Chart 8, shows the distribution by four-year 
 intervals of the ages of delinquent women in the various institutional 
 groups with the mean age and standard deviation for each group. The 
 lower age limit of fourteen was found in only one case, that of a col- 
 ored girl committed to the Workhouse.^ This is the only case in which 
 the age is below sixteen years. The chart indicates what has been stated 
 previously concerning the range of ages within the institutional groups.^ 
 Bedford has only eight cases over the thirty-year age limit, while a 
 high percentage of 22.8 falls in the lowest age group — from fourteen to 
 eighteen years. The range of ages is comparatively short, running 
 from IS to 38 years. Auburn, on the other hand, starts with the later 
 age group of 18 to 22, and has a long range up to 70 years. Here the 
 mean,. as we should expect, is much higher. The Magdalen group is 
 more like the Bedford, with a fairly high percentage of cases in the 
 first age group, and a range up to forty-two years, though there is 
 somewhat more of a massing of cases toward the upper end of the 
 scale, resulting in a larger mean age. The Penitentiary shows a range 
 more like that in Auburn and very different from the Magdalen and 
 Bedford. The mean age in the Penitentiary is also very close to that 
 
 Per 
 
 Cent 
 
 TOTAL 
 
 Chart VIII 
 
 Age at Present Conviction 
 
 Per cent distribution by institutional groups 
 
 " The girl had given her age to the judge as sixteen years when as a mat- 
 ter of fact she was only fourteen years, four months, of age. In addition to 
 this, she had been convicted as an adult criminal four months earlier, when she 
 was only fourteen, and had been sentenced to the Workhouse for thirty days. 
 
 " See Chapter II for accounts of institutional groups. 
 
Pm 
 
 Cent 
 
 BEDFORD 
 
 :^— r 
 
 1 I I I I I I I I I I T" 
 14 18 22 26 30 34 38 42 -46 SO S4 58 
 
 I I I- 
 
 66 70 74 
 
 Per 
 Cent 
 
 Per 
 Cent 
 
 AUBURN 
 
 MAGDALEN 
 
 >4 18 22 26 30 34 
 
 ACE 
 
 ■Chart VIII (Continued) 
 Age at Present Conviction 
 Per cent distribution by institutional groups 
 150 
 
Per 
 Cent 
 
 Per 
 Cent 
 
 Per 
 Cent 
 
 PENITENTIARY 
 
 WORKHOUSE 
 
 PROBATION 
 
 JZ^ 
 
 O-R. 
 
 I I 
 
 14 18 22 26 30 34 
 
 42 46 50 54 
 AGE 
 
 62 66 70 74 
 
 vitt & 
 
 Chart VIII (Continued) 
 Age at Present Conviction 
 Per cent distribution by institutional groups 
 151 
 
152 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 of Auburn. The Workhouse, as well as having the youngest case, also 
 has the oldest case in the entire group, a woman of seventy-two years, 
 and therefore has the longest range of years of any of the institutions. 
 The mean age is slightly older than for any other group, though there 
 is a large standard deviation. The Probation women with a com- 
 paratively short range of cases have a smaller percentage in the lowest 
 age group, than have Bedford and the Magdalen, and a slightly higher 
 percentage in the higher age groups so that we should expect the 
 mean age to be higher than the average ages in the other two groups. 
 The total group, which will be used in many of the correlations to 
 follow, is based on 559 cases, and has a mean age of 28.79±.429 years, 
 with a fairly large standard deviation of 10.13 ± .303 years. As 
 a general basis of comparison, it is of interest to note that the institu- 
 tions under consideration divide quite sharply into classes with re- 
 spect to age. Bedford, the Magdalen, and the Probation group differ 
 only slightly from one another in mean age, showing a marked se- 
 lection of younger women. Auburn, the Penitentiary, and the Work- 
 house, on the other hand, while likewise differing only slightly from 
 one another, are made up of considerably older groups, as shown 
 by their mean ages given above. Wherever it has been felt that the 
 factor of age was important to consider in the relation between any 
 other two observed factors, the partial correlation coefficient between 
 the two traits for constant age has been used.* 
 
 As stated in the introduction, it was found to be impossible to ob- 
 tain a physical and medical history of the women in all of the institu- 
 tional groups. Two of the institutions, the Workhouse and the Peni- 
 tentiary, kept very meager records ; the Probation group had no physi- 
 cal examination forthe most part, and the physical records in the Mag- 
 dalen, Auburn, and Bedford groups differed so substantially in the type 
 of record kept that it seemed impracticable to try to use the medical 
 data available, except for several unconnected items. A few of the 
 general factors relating to the physical status of the women in this 
 study will be considered in other chapters. Venereal disease, for in- 
 stance, will be included in the chapter relating to sex history. The 
 number of full-term children or miscarriages will be considered in 
 connection with other family relationships and hereditary influences in 
 Chapter IX. 
 
 * See Chapter III. 
 
MISCELLANEOUS CONSIDERATIONS 
 
 153 
 
 HABITS 
 
 There is, however, the consideration of certain habits relating to 
 physical status which we will present here, as a means of helping to 
 define the total group. These habits are the use of alcohol, drugs and 
 tobacco. The material on which these figures are based is not only the 
 statement of the subject but the corroboration of other informants 
 interviewed by the field worker. We have not included as excessive 
 alcoholics or drug addicts any women concerning whom there was 
 reasonable doubt as to whether or not alcohol or drugs were used to 
 excess. 
 
 TABLE 32 
 
 USE OF ALCOHOL 
 
 Per Cent Distribution Among Delinquent Women in Institutional 
 
 Groups 
 
 
 Inbtitdtional Groups 
 
 
 Use of Alcohol 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Probation 
 
 Total 
 
 Never 
 
 Moderate 
 
 Excessive 
 
 65.3 
 16.8 
 
 17.8 
 
 35.9 
 33.3 
 30.8 
 
 79.5 
 
 11.0 
 
 9.6 
 
 55.6 
 
 26.9 
 17.6 
 
 35.6 
 33.7 
 30.7 
 
 50.0 
 34.2 
 15.9 
 
 53.2 
 26.2 
 20.6 
 
 Total .... 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cases 
 
 101 
 
 78 
 
 73 
 
 108 
 
 101 
 
 82 
 
 543 
 
 From Table Z2 it is seen that the Magdalen and Bedford have the 
 largest percentages of women who have never used alcohol, while fhe 
 Penitentiary, Probation, Auburn, and Workhouse follow in order of 
 decreasing percentages. It is of interest to note that Bedford and the 
 Magdalen which have the lowest average ages also have the largest 
 percentage of non-alcoholics. 
 
 The largest percentage of excessive drinkers, on the other hand, is 
 found in the Auburn group, and is followed by the Workhouse, Bed- 
 ford, Penitentiary, Probation, and the Magdalen. It is interesting to 
 note that the degree of alcoholism is much aHke in the Auburn and 
 
154 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 Workhouse groups where one might expect to find wide discrepancies. 
 The fact that from this group of Workhouse women all those convicted 
 of intoxication are omitted, serves to change the general trend of Work- 
 house commitments. In this particular comparison, therefore, we must 
 remember that only a part of the Workhouse is represented, and that 
 if the total intoxication cases were included the comparison would be 
 quite different. 
 
 The Special Intoxication group in the Workhouse which has been 
 described in Chapter I ^ is composed of twenty-one cases. It may be 
 well here to give some of the more important data on this small num- 
 ber of cases which, though not consecutive in admission to the Work- 
 house, were chosen from the total group of Workhouse women ob- 
 served, with no conscious bias. The number is too small to use for 
 any extensive comparison, but a few facts may serve to show how the 
 group is different from the Workhouse group which we have used in 
 this study. Though we hesitate to draw conclusions from results with 
 so few cases, we believe, on the basis of rather extensive observation, 
 that the discrepancy indicated above is genuine. 
 
 It is probably worthy of note that there are no colored women in 
 this group of intoxication cases. Twelve of the women are native 
 white, while of the nine foreign born, seven come from Ireland, one 
 from England, and one from Finland. 
 
 As stated, the nature of the present offense in each of these cases is 
 for intoxication and falls in the general classification of offenses 
 against regulations for public health, safety and policy, and also in the 
 misdemeanor group. 
 
 The extent of delinquency among the women of this group is of 
 particular interest in that the mean number of convictions is 11.0 in 
 comparison with 2.82 for the non-intoxication cases studied in the 
 Workhouse and with 1.63 for the total of the six institutional groups. 
 It is also probable that this mean number is based on only a part of the 
 convictions for many of the older women in the intoxication group. 
 Since they have been coming into the courts for years and before the 
 finger-print system was established, it is likely that many of thfe writ- 
 ten records for constant repeaters were not found by the field worker. 
 The range of convictions is from no previous convictions in 2 cases to 
 35 in one case. The first offenders were only 2, or 9.5 per cent of the 
 total, in comparison with 23.0 per cent for the rest of the Workhouse 
 group, and with 44.7 per cent first offenders for the total group. The 
 ■'See Chapter 11, p. 24. 
 
MISCELLANEOUS CONSIDERATIONS 
 
 155 
 
 length of time served runs from no time served to 94 months and 6 
 days. The mean number of months served is 20.42. 
 
 If we turn to the first offenses of the women in this intoxication 
 group we find that the age at the first conviction has a range of from 
 25 to 71 years, with a mean age of 40.7 years. This is considerably 
 older than the average age at first conviction of the larger Workhouse 
 group, 32.1 years, but very likely does not represent the actual age at 
 first conviction for several of the older women who say they have "been 
 coming to the Island ever since I can remember." The nature of the 
 first offense in ten cases or practically half of the total was in the same 
 group of offenses against regulations for public health, and nearly all 
 were intoxication cases. Nine women, or 47.4 per cent, were first con- 
 victed of general criminality offenses. If we add all of the convic- 
 tions of each of the twenty-one cases, we have 212 previous convic- 
 tions. Of this number, only four convictions were for offenses involv- 
 ing prostitution, all the others being for intoxication or disorderly con- 
 duct." 
 
 " To show the impossibility of having reliable information if the statement 
 of the woman is used when she has had a record like the following, both the 
 woman's statement and the verified record of convictions are given for two of 
 the intoxication cases in the Workhouse. 
 
 Woman's Statement 
 
 "At first insisted 1. 
 that her first con- 2. 
 viction was two 3. 
 years ago. Later 
 stated that she 
 'stayed out of 
 prison for three 5. 
 years at a time, 6. 
 two years ago.' 
 Then said she had 7. 
 had 5 and 10 day 
 sentences and 6 
 months the last 
 time. No further 
 information could 
 be obtained from 
 the woman, who 
 seemed contused 
 and said she 'nev- 
 er could remem- 
 ber dates'." 
 
 8. 
 
 9. 
 10. 
 11. 
 12. 
 13. 
 14. 
 15. 
 16. 
 17. 
 18. 
 19. 
 20. 
 21. 
 22. 
 23. 
 
 Verified Record 
 
 1/14/03, Disorderly conduct, 
 3/29/04, Intoxication, 
 4/30/04, Disorderly conduct. 
 
 4. 8/12/04, 
 
 9/14/05, 
 4/ 3/06, 
 
 5/ 7/05, 
 
 9/ 1/05, 
 
 11/18/05, 
 
 6/20/06, 
 
 10/ 7/07, 
 
 5/16/08, 
 
 11/ 2/08, 
 
 6/ 3/09, 
 
 1/24/10, 
 
 8/23/11, 
 
 10/20/11, 
 
 11/24/11, 
 
 7/ 9/12, 
 
 7/25/12, 
 
 10/14/12, 
 
 6/ 7/13, 
 
 8/ 1/14, 
 
 Intoxication, 
 Disorderly conduct. 
 
 Intoxication, 
 Disorderly conduct, 
 
 Intoxication, 
 
 it 
 
 Disorderly conduct. 
 
 Intoxication, 
 Disorderly conduct, 
 
 Intoxication, 
 Disorderly conduct. 
 
 Workhouse, 5 days. 
 20 days. 
 " 6 mos., or 
 
 $500 fine. 
 Workhouse, 1 mo., or 
 
 $500 fine. 
 Workhouse, 160 days. 
 " 1 mo., or 
 
 $300 fine. 
 Workhouse, 6 mos., or 
 
 $500 fine. 
 
 Workhouse, 5 days. 
 
 6 mos. 
 
 6 mos. 
 
 6 mos. 
 
 6 mos. 
 
 6'mos. 
 
 6 mos. 
 
 6 mos. 
 
 10 days 
 
 10 days. 
 
 180 days. 
 
 5 days. 
 
 30 days. 
 
 60 days. 
 
 10 days. 
 
 30 days. 
 
156 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 By the nature of the group, all of the special intoxication cases 
 have used alcohol to excess. None of the women in the group denied 
 this. All denied use of drugs in any form and there was no outside 
 evidence obtained to contradict their statements. It was found that 
 three of the group used tobacco to excess. 
 
 While we are considering the question of alcoholism, it will be ad- 
 visable to show the relation of the alcoholic and non-alcoholic groups 
 to the number of convictions. For this comparison, alcoholic has been 
 used to include both the excessive drinkers in the regular group of 
 587 cases, and the special intoxication cases in the Workhouse. The 
 non-alcoholic includes those who never use alcohol and the moderate 
 
 Wmnan'a Statement 
 24. 
 
 25. 
 
 26. 
 27. 
 28. 
 29. 
 30. 
 
 31. 
 32. 
 33. 
 34. 
 35. 
 
 2. 
 
 "Been coming 
 here again and 
 again for years. 
 Came out of jail 
 six months ago. 
 Was arrested the 
 nejct day and got 
 six months." 
 
 10. 
 11. 
 
 12, 
 13, 
 
 15. 
 16. 
 17. 
 18. 
 
 9/ 2/14, 
 
 9/29/14, 
 
 6/11/15, 
 6/23/15, 
 7/16/15, 
 9/16/15, 
 10/24/15, 
 
 10/31/15, 
 
 11/10/15, 
 
 12/20/15, 
 
 3/27/16, 
 
 5/22/16, 
 
 9/ 5/11, 
 11/22/11, 
 12/27/11, 
 
 8/19/13, 
 12/30/13, 
 
 3/14/14, 
 
 Verified Record 
 
 Intoxication, Workhouse, 5 days, or 
 
 $5.00 fine. 
 " House of Good Shep- 
 
 herd, 9/29/14-4/5/15. 
 Disorderly conduct. Workhouse, 10 days. 
 Intoxication, " 15 days. 
 
 30 days. 
 30 days. 
 Disorderly conduct, " 5 days, or 
 
 $5.00 fine. 
 Intoxication, Workhouse, 10 days. 
 
 30 days. 
 
 " " 3 moB. 
 
 Disorderly conduct, " 30 days. 
 
 Intoxication, " 6 mos. 
 
 Intoxication, 
 Disorderly conduct. 
 
 30 
 1 
 6 
 
 Intoxication, 
 
 30 
 
 30 
 
 1 
 
 $500 fine. 
 Workhouse, 6 
 10 
 „ , , $10 fine. 
 
 9. 2/15/15, Disorderly conduct. Workhouse, 3 
 
 $100 fine. 
 ." _ " Workhouse, 10 
 Intoxication and dis- 
 orderly conduct. " 60 
 Disorderly conduct, " 20 
 
 mo. 
 mos. 
 days. 
 
 8/ 4/14, 
 2/ 5/15, 
 
 7/22/15, 
 9/15/15, 
 
 11/14/15, 
 12/ 4/15, 
 
 $10 fine. 
 
 mo., or 
 
 mos. 
 days, or 
 
 mos., or 
 
 days. 
 
 days, 
 days, 
 days, or 
 
 14. 12/14/15, Vagrancy, violation 
 of Tenement 
 House Law, Workhouse, 5 
 
 Disorderly conduct, " 30 
 
 Intoxication, " 30 
 
 " " 4 
 
 12/19/15, 
 5/16/16, 
 6/23/16, 
 
 10/24/16, 
 
 days. 
 
 6 
 
 mos. 
 mos. 
 
MISCELLANEOUS CONSIDERATIONS 
 
 157 
 
 drinkers, A comparison of the average number of previous con- 
 victions for the alcoholic and non-alcoholic groups showed that the 
 average for the alcoholic group is 5.35 convictions, while the average 
 for the non-alcoholic group is only 2.285 convictions. That this is 
 a valid difference is shown by the ratio 5.83. A comparison of the 
 dispersions of number of convictions among the two groups indicates 
 that there is a valid difference in the dispersion of the two groups, 
 with a wider scattering among the alcoholic group. 
 
 TABLE 33 
 
 ALCOHOLIC* AND NON-ALCOHOLIC 
 
 Comparison of the Means and Standard Deviations of the Number of 
 
 Previous Convictions Among Alcoholic and Non-Alcoholic 
 
 Women of Total Group 
 
 
 Alcoholic 
 
 Non-Alcoholic 
 
 Difference 
 
 d 
 ffd 
 
 Chances that 
 
 real differ- 
 ence does not 
 exist are 1 
 in: 
 
 Mean 
 
 5.35 
 + .517 
 
 2.285 
 ±.0984 
 
 3.07 
 
 5.83 
 
 00 
 
 (Tin ■■••.-• 
 
 
 
 
 (T 
 
 5.96 
 + .365 
 
 2.044 
 + .0696 
 
 3.92 
 
 10.53 
 
 00 
 
 (Xa 
 
 
 
 
 Cases 
 
 133 
 
 431 
 
 
 
 
 
 
 *ThoaB who use alcohol to excess. Includes 21 cases from Special Intoxication Group in Workhouse. 
 
 In determining what percentage of the cases are drug-addicts, no 
 measure of "moderate" has been used, and none have been counted as 
 drug-users who did not have a habit at the time of the present convic- 
 tion or shortly before. The following table shows that the percentage 
 of drug-users varies from 6.2 per cent in the Probation group to 33.7 
 per cent in the Penitentiary, where a large percentage of the women 
 convicted of possessing or selling drugs are sentenced. The five women 
 drug-users in the Probation group were not convicted of using drugs 
 and the fact of their being drug-addicts was not discovered until after 
 the probation sentence was given. 
 
 The use of tobacco also varies considerably among the institu- 
 tional groups. The Workhouse has the lowest percentage who never 
 use tobacco and is followed by the Penitentiary, Auburn, Probation. 
 Bedford, and Magdalen in order of increasing percentages. Of those 
 
158 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 34 
 
 USE OF DRUGS 
 
 Per Cent Distribution Among Delinquent Women in Institutional 
 
 Groups 
 
 
 Institotionad Groups 
 
 
 Use of Drugs 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Probation 
 
 Total 
 
 D;ug-User3 
 
 Non-Dmg-Users 
 
 14.9 
 85.1 
 
 13.0 
 87.0 
 
 13.7 
 
 86.3 
 
 33.7 
 66.4 
 
 22.8 
 77.2 
 
 6.2 
 93.8 
 
 18.3 
 81.7 
 
 Totals. . . . 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cases 
 
 101 
 
 77 
 
 73 
 
 107 
 
 101 
 
 81 
 
 540 
 
 who use tobacco to excess, however, the order is changed from Pro- 
 bation with the smallest percentage to the Magdalen, Bedford, Peni- 
 tentiary, Auburn, and Workhouse in increasing order. 
 
 TABLE 36 
 USE OF TOBACCO 
 
 Per Cent Distribution Among Delinquent Women in Institutional 
 
 Groups 
 
 
 Institutional Groups 
 
 
 Use of Tobacco 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Probation 
 
 Total 
 
 Never 
 
 82.2 
 
 2.0 
 
 15.8 
 
 79.2 
 20.8 
 
 84.7 
 
 2.8 
 
 12.5 
 
 72.0 
 
 9.4 
 
 18.7 
 
 59.0 
 
 8.0 
 
 33.0 
 
 81.3 
 10.7 
 
 8.0 
 
 75.6 
 
 Moderate 
 
 Excessive 
 
 5.6 
 18.8 
 
 Total .... 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cases 
 
 101 
 
 77 
 
 72 
 
 107 
 
 100 
 
 75 
 
 532 
 
MISCELLANEOUS CONSIDERATIONS 159 
 
 There are many possibilities of comparison of habits with other 
 studies of delinquents. Of these only two will be noted as directly 
 comparable with certain of our institutional groups. First of all, a 
 comparison of Dr. Guibord's study of 200 Bedford cases in 1914 may 
 be of interest in showing whether there seems to be any marked change 
 in the make-up of this group in a few years .'^ 
 
 Habits of 200 Bedford Habits of Women in 
 
 Women in 1914. Present Study for: 
 
 Group A. Group B. Bedford Total Group 
 Excessive or frequent use 
 
 of alcohol 35.0% 27.0% 34.6% 46.8% 
 
 Drug-addicts 21.0% 7.0% 14.9% 18.3% 
 
 Excessive use of tobacco . 20.0% 17.0% 15.8% 18.8% 
 
 The most noteworthy discrepancy is that between the percentages 
 of drug-users in the three Bedford groups. This amount of variation 
 would indicate that the proportion of drug-users in this institution 
 varies from year to year to a considerable extent. 
 
 In a study of 200 male felons in Auburn Prison, Dr. Heacox finds 
 results quite different from those for our group of women in Auburn 
 Prison or for our total group. The latter, which includes a high per- 
 centage of misdemeanants, we should expect to differ from a group of 
 male felons. The following percentages from the above groups are 
 known to have used alcohol, drugs, or tobacco to either a moderate or 
 an excessive degree. 
 
 Auburn Men' Auburn Women Total Women 
 
 Alcohol 87.5% 64.1% 46.8% 
 
 Drugs 2.5% 13.0% 18.3% 
 
 Tobacco 95.0% 20.8% 24.4% 
 
 It is surprising to find so much larger a percentage of drug-addicts 
 in the group of Auburn women than among the Auburn men. There 
 seems to be no clear reason for the difference, since in neither group 
 are commitments possible for drug-using. 
 
 RECORD OF COMMITMENTS TO HOSPITALS FOR THE INSANE 
 
 We may note briefly, at this point, the number of women in the 
 institutional groups who have at any time been patients in a hospital 
 
 'Guibord, Alberta S. B. "Physical States of Criminal Women." Journal 
 of the American Institute of Criminal Law and Criminology, vol. VIII, no. i. 
 May 1, 1917. 
 
 'From unpublished figures on 200 consecutive admissions to Auburn Prison. 
 
160 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 for the insane. We have not included here the cases of obviously 
 psychopathic women who have not been in hospitals, since there was 
 no routine psychiatric examination in any group except Bedford, and 
 there was no equally reliable information for comparison on the other 
 groups. The numbers and percentages of women who have at some 
 time been in a hospital for the insane are as follows : 
 
 Number Per Cent 
 
 Bedford 6 S.9 
 
 Auburn 3 3.4 
 
 Magdalen 0.0 
 
 Penitentiary 5 4.5 
 
 Workhouse 8 7.J 
 
 Probation •... 2 2.0 
 
 Total 24 4.1 
 
 This small percentage of 4.1, it must be remembered, does not in- 
 clude all who might have been diagnosed as insane, had a proper exam- 
 ination been given, but merely those who had at some time during their 
 lives been diagnosed as insane and committed to a hospital for the 
 insane. 
 
 The diagnoses as we were able to obtain them from the hospitals on 
 these 24 cases are as follows : 
 
 Constitutionally inferior 5 
 
 Manic Depressive S 
 
 Manic Depressive with Alcoholism 3 
 
 Dementia Prjecox 2 
 
 Alcoholic Psychosis, Korsakoff Type 1 
 
 Stuporous Melancholia 1 
 
 Involutional Melancholia 1 
 
 Infective Exhaustion 1 
 
 Mental Defective with Psychosis 1 
 
 Not Insane 1 
 
 Diagnosis unknown 3 
 
 In Dr. Katharine B. Davis' study of 647 prostitutes at Bedford,' 
 twenty cases, or 3.1 per cent of the group, were transferred to hospitals 
 for the insane while they were at Bedford. This does not include, how- 
 ever, women who had at any time in the past been in such a hospital, 
 so that the percentage is probably lower than it should be for exact 
 comparison with our data. 
 
 'Kneeland, George J.: "Commercialized Prostitution in New York City," 
 Chapter VIII by Katharine Bement Davis. A Study of Prostitutes Committed 
 to the State Reformatory for Women at Bedford Hills. The Century Com- 
 pany, 1916, p. 186. •^ 
 
MISCELLANEOUS CONSIDERATIONS 
 
 161 
 
 CIVIL CONDITION 
 
 Turning to other general items indicating the social status of the 
 women deUnquents, we may first consider the question of civil condi- 
 tion. The following table shows that the percentage of single women 
 varies between the institutional groups to a large extent. As would 
 be expected, the highest percentages of single women are found in 
 two groups, Bedford and Magdalen, where the average age is 
 lowest. The total group shows 42.4 per cent of the women single. 
 Among those who have been married the percentage who were widowed 
 or divorced at the time of the present conviction varies between the 
 groups somewhat as the percentage of those married varies. That is, 
 those groups with the highest percentage married, namely. Auburn, 
 Penitentiary and Workhouse, have the highest percentage who are 
 divorced or widowed. If we compare the total delinquent group with 
 the general female population fifteen years of age and over in New 
 York State, we see that the delinquent group has an appreciably 
 smaller percentage of women who are married. This varies, however, 
 in a comparison with the institutional groups, those groups with the 
 highest average ages. Auburn, Penitentiary, and Workhouse, having 
 a higher percentage of married women than the general population. 
 
 TABLE 36 
 
 CIVIL CONDITION 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups 
 
 and of General Female Population 15 Years of Age and 
 
 Over in New York State, in 1910 
 
 
 Institutional Groups 
 
 Total 
 
 General 
 
 Civil Condition 
 
 Bed- 
 ford 
 
 Auburn 
 
 Mag- 
 dalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Proba- 
 tion 
 
 Female 
 Population 
 
 Single 
 
 63.4 
 36.6 
 
 16.3 
 
 83.8 
 
 69.7 
 30.3 
 
 29.6 
 70.4 
 
 27.7 
 
 72.3 
 
 50.5 
 49.5 
 
 42.4 
 
 57.6 
 
 33.8 
 
 Married 
 
 66.2 
 
 
 
 Widowed 
 
 Divorced 
 
 2.0 
 2.0 
 
 3.8 
 17.5 
 
 1.3 
 4.0 
 
 2.8 
 15.7 
 
 4.9 
 14.9 
 
 2.2 
 4.3 
 
 2.9 
 
 9.8 
 
 11.3 
 .3 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cases .... 
 
 101 
 
 80 
 
 76 
 
 108 
 
 101 
 
 ■ 93 
 
 559 
 
 3,291,714 
 
162 STUDY OF WOMEN DELINQUENTS. IN NEW YORK 
 
 The age at first marriage is presented by institutional groups to 
 show the range of ages and where the greatest massing occurs. The 
 mean age at first marriage for the total group is 20.5 years ±.268. 
 The standard deviation of 4.68±.189 years indicates that there is not 
 a particularly wide scattering of ages. There are 92 cases or 30.0. per 
 cent of the total group who were married under 18 years, which is the 
 legal age of consent in New York State. There were 30 cases or 9.8 
 per cent married under 16 years, the limit of the Juvenile Court age. 
 
 TABLE 37 
 
 AGE AT FIRST MARRIAGE 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups 
 
 
 Institutional Groups 
 
 
 Age 
 
 
 
 
 
 
 
 Total 
 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Probation 
 
 
 13 to 16 yrs.. 
 
 8.3 
 
 10.8 
 
 4.4 
 
 9.2 
 
 11.9 
 
 10.3 
 
 9.8 
 
 16 " 19 " . 
 
 44.4 
 
 30.8 
 
 13.1 
 
 27.6 
 
 22.4 
 
 43.6 
 
 30.1 
 
 19 " 22 " . 
 
 25.0 
 
 29.2 
 
 30.5 
 
 22.4 
 
 23.9 
 
 23.1 
 
 25.2 
 
 22 " 25 " . 
 
 13.9 
 
 15.4 
 
 34.8 
 
 17.1 
 
 26.9 
 
 7.7 
 
 18.6 
 
 25 " 28 " . 
 
 5.6 
 
 9.2 
 
 13.1 
 
 9.2 
 
 7.5 
 
 7.7 
 
 8.5 
 
 28 " 31 " . 
 
 2 8 
 
 3.1 
 
 4.4 
 
 7.9 
 
 4.5 
 
 5.1 
 
 4.9 
 
 31 " 34 " . 
 
 
 1.5 
 
 
 1.3 
 
 1.5 
 
 
 1.0 
 
 34 " 37 " . 
 
 
 
 
 2.6 
 
 
 
 .7 
 
 37 " 40 " . 
 
 
 
 
 
 1.5 
 
 2.6 
 
 .7 
 
 40 " 43 " . 
 
 
 
 
 1.3 
 
 
 
 .3 
 
 43 " 46 " . 
 
 
 
 
 1.3 
 
 
 
 .3 
 
 Total. . . 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of case 
 
 3 36 
 
 65 
 
 23 
 
 76 
 
 67 
 
 39 
 
 306 
 
 Mean Age at First Marriage (Total Group). . .20.50 ± .268 yiars. 
 a (Total Group) 4.68 ± .189 years. 
 
 If we compare the age of the husband at this first marriage, we 
 see from Table 38 that the distribution begins at a later age and runs 
 to a later age. Only eleven men or five per cent were married before 
 they were eighteen years of age to the women in this study, and only 
 two men or .9 per cent were married before they were sixteen. The 
 average age of the first husband at marriage is 25.27±.47S years, with 
 a standard deviation of 7.056±:.336 years. This shows that the men 
 were of a considerably older average age than the women whom they 
 married. 
 
MISCELLANEOUS CONSIDERATIONS 
 
 163 
 
 TABLE 38 
 
 AGE OF FIRST HUSBAND AT MARRIAGE 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups 
 
 
 Institutional Groups 
 
 
 Age 
 
 
 
 
 
 
 
 Total 
 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 Peni- 
 tentiary 
 
 Worlt- 
 house 
 
 Probation 
 
 
 13 to 16 yrs.. 
 
 
 2.0 
 
 
 
 
 3.5 
 
 .9 
 
 16 " 19 " . 
 
 2.8 
 
 13.7 
 
 R 
 
 3 
 
 7.0 
 
 13.0 
 
 6.9 
 
 9.1 
 
 19 " 22 " . 
 
 22.2 
 
 19.6 
 
 18 
 
 8 
 
 20.9 
 
 10.9 
 
 34.5 
 
 20.4 
 
 22 " 25 " . 
 
 27.8 
 
 15.7 
 
 37 
 
 5 
 
 27.9 
 
 30.4 
 
 31.0 
 
 26.7 
 
 25 " 28 " . 
 
 19.4 
 
 17.7 
 
 12 
 
 5 
 
 20.9 
 
 19.6 
 
 10.4 
 
 17,7 
 
 28 " 31 " . 
 
 16.7 
 
 11.8 
 
 
 
 9.3 
 
 4.4 
 
 10.4 
 
 9.5 
 
 31 " 34 " . 
 
 2.8 
 
 11.8 
 
 18 
 
 8 
 
 4.7 
 
 8.7 
 
 
 7.2 
 
 34 " 37 " . 
 
 
 
 6 
 
 3 
 
 2.3 
 
 4.4 
 
 3.5 
 
 2.3 
 
 37 " 40 " . 
 
 
 
 
 
 2.3 
 
 
 
 .5 
 
 40 " 43 " . 
 
 
 2.0 
 
 
 
 
 4.4 
 
 
 1.4 
 
 43 " 46 " . 
 
 2.8 
 
 3.9 
 
 
 
 2.3 
 
 
 
 1.8 
 
 46 " 49 " . 
 
 2.8 
 
 2.0 
 
 
 
 2.3 
 
 
 
 1.4 
 
 52 " 55 " . 
 
 2.8 
 
 .... 
 
 
 
 
 2.2 
 
 
 .9 
 
 61 " 63 " . 
 
 
 
 
 
 
 2.2 
 
 
 .5 
 
 Total. . . 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of case. 
 
 3 36 
 
 51 
 
 16 
 
 43 
 
 46 
 
 29 
 
 221 
 
 Mean Age of First Husband at Marriage (Total Group) .. 25 . 27 ± .475 y;ars 
 <r (Total Group) 7.056 ± .336 years 
 
 RELIGION 
 
 The religion of the women shows varying differences in the institu- 
 tional groups. Before discussing these differences in detail it will be 
 well to bear in mind the fact that we have no representation here of 
 institutions such as the House of the Good Shepherd where the inmates 
 are very largely Catholic. This particularly affects the percentages in 
 Bedford, Magdalen, and the Workhouse, which receive much the same 
 kind of cases that are sent to the House of the Good Shepherd. Table 
 39 shows that in these three groups there are, as we should expect, a 
 smaller percentage of Catholic than of Protestant women, but to offset 
 this we must remember that the House of the Good Shepherd is almost 
 entirely Catholic and that the percentages of women in these three 
 institutional groups classified by religion are not, therefore, representa- 
 tive of the religion of delinquent women as a whole. In general, the 
 figures for Auburn and Probation ought not to be seriously affected 
 by the existence of Catholic institutions, since these two groups tend to 
 
164 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 receive a distinct type of case, which in most instances, would not re- 
 ceive any other kind of sentence. The Penitentiary, might have been 
 affected, to a certain extent, particularly with such offenses as possess- 
 ing drugs, which are punishable by commitment to a private institution 
 like the House of the Good Shepherd. In these three groups, Auburn, 
 Penitentiary, and Probation, we see that the percentages of Catholics 
 are larger than those of Protestants, and that the difference is most 
 marked in the Probation group with 55.1 per cent of its cases Catholics 
 and 30.3 per cent Protestants. The percentages for our total group of 
 delinquent women are of little value for anything except material 
 descriptive of the group, because this total includes the three institu- 
 tional groups which are largely Protestant because the Catholic women 
 have been sent to a Catholic institution. From the three groups, there- 
 fore, which can give any accurate indication of the percentage of Cath- 
 olics, Hebrews and Protestants, it is interesting to note that the Catho- 
 lics in each case make up the highest percentage, the Protestants next 
 highest, and the Hebrews a relatively small percentage. It is impossible 
 to obtain any recent figures from the Census or from any other exten- 
 sive studies to compare the percentages from the denominational groups 
 among delinquent women with the same for the general population, and 
 it is, therefore, difficult to state whether the percentages, of Catholics, 
 Protestants and Hebrews are in excess or below the percentages for 
 these denominational groups in the general population. 
 
 TABLE 39 
 
 RELIGION 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups 
 
 
 Institutional Groups 
 
 
 Religion 
 
 Bfedford 
 
 Auburn 
 
 Magdalen 
 
 Penil 
 tentiary 
 
 Work- 
 house 
 
 Probation 
 
 Total 
 
 Catholic 
 
 36.6 
 
 46.3 
 
 22.7, 
 
 43.0 
 
 35.0 
 
 66.1 
 
 40.0 
 
 Greek Cath. . 
 Roman Cath. 
 
 Hebrew 
 
 Protestant 
 
 1.0 
 35.6 
 
 8.9 
 
 54.5 
 
 3.8 
 42.5 
 
 11.3 
 
 42.6 
 
 22!7 
 41.3 
 36.0 
 
 43!6 
 16.0 
 42.1 
 
 35!6 
 10.0 
 55.0 
 
 55] i 
 
 14.6 
 30.3 
 
 .7 
 39.3 
 
 15.9 
 
 44.0 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cases 
 
 101 
 
 80 
 
 75 
 
 107 
 
 100 
 
 89 
 
 552 
 
MISCELLANEOUS CONSIDERATIONS 165 
 
 SUMMARY 
 
 From this chapter of somewhat heterogeneous social factors in the 
 hves of the dehnquent women under consideration, we may note the 
 following points. 
 
 The average age varies from 21.7 years in the Bedford group to 
 34.9 years in the Workhouse. The institutional groups divide them- 
 selves into two groups, those of appreciably younger average ages, 
 Bedford, Magdalen and Probation, and those of an older average age. 
 Auburn, Penitentiary and Workhouse. The average age of the total 
 group which we shall use in later chapters is 28.8 years. 
 
 The habits of the women in institutional groups are of considerable 
 importance, especially in reference to the use of alcohol, drugs* and 
 tobacco. From Table 39, we found that the percentage who used alco- 
 hol moderately or to excess varied considerably between the institu- 
 tional groups, with the Magdalen having the largest percentage of non- 
 alcoholics and the Workhouse the smallest percentage. A brief survey 
 of the 21 intoxication cases from the Workhouse shows that the aver- 
 age number of convictions among this group is 11.0, almost seven 
 times as large as the average number of convictions for the total of the 
 six institutional groups. If we compare the mean number of convic- 
 tions for the non-alcoholic group, that is, those who never use alcohol 
 or do so moderately, with those who use alcohol to excess, including the 
 special intoxication group in the Workhouse, we find that the crude 
 differences of the mean convictions are large and that these differences 
 are significant. That is, the average number of convictions for the 
 alcoholic group is much larger than for the non-alcoholic group. 
 
 The ttse of drugs also varies markedly in the institutional groups, 
 with the lowest percentage of drug-addicts in the Probation group from 
 a court which has no jurisdiction over charges of possessing or using 
 drugs, and the largest percentage in the Penitentiary which receives 
 the bulk of the women who are convicted of possessing drugs. 
 
 The use of tobacco, while not of so hiuch importance as the drug- 
 addiction and use of alcohol, is of some interest, especially in compari- 
 son with other groups of delinquents. A group of male felons in Au- 
 burn Prison, for instance, has 95.0 per cent of the group using tobacco 
 to excess, while in our group of Auburn women the excessive tobacco 
 users are only 20.8 per cent. 
 
 The record of commitments to hospitals for the insane shows that 
 4.1 per cent of our total group have at some time been patients in such 
 
166 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 hospitals. This does not include those who may be mentally diseased 
 or deteriorated, but who have not been in a hospital for the insane. 
 
 If we proceed to the other social factors of civil condition, age at 
 marriage, etc., we find that the groups with the youngest average age 
 at present conviction have the largest percentage unmarried. The aver- 
 age age at first marriage for the total group of women is 20.5 years, 
 while the average age of the first husband at the time of marriage to 
 the women in this study is 25.3 years. 
 
 The distribution by religion is difKcult to interpret since the num- 
 ber of Catholics in Bedford, Magdalen, and the Workhouse is affected 
 by the fact that the House of the Good Shepherd, a Catholic institu- 
 tion, takes the cases which could be committed to these three in- 
 stitutions, and therefore unduly increases the percentages of Protes- 
 tants and Hebrews in these institutions. In the three institutional 
 groups which would be little affected by the House of the Good Shep- 
 herd, the Catholics have the highest percentage of cases, the Protes- 
 tants the next highest and the Hebrews a relatively small percentage. 
 
CHAPTER VIII 
 NATIVITY AND COLOR IN RELATION JO DELINQUENCY 
 
 THE report of the City Council Committee on Crime in Chicago 
 states that: 
 
 "Statistics furnished in the police report show that the native Americans, 
 white and colored both, have a percentage of arrests and convictions 
 considerably greater than their percentage of population. On the other 
 hand, the various foreign groups show a smaller per cent of convictions 
 than their proportion of the population entitles them to have. It is, of 
 course, popularly believed that immigration is a cause of crime. This 
 belief has largely been due to a comfortable theory that we are superior 
 to the people of Europe, and to a desire to shift the responsibility for our 
 shortcomings on to other people. No facts have ever been found to sup- 
 port this belief. The Chicago statistics in this respect agree with the 
 statistics furnished by the United States census and the Federal Commis- 
 sion on Immigration. The report of the commission states emphatically 
 that 'no satisfactory evidence has yet been produced to show that immi- 
 gration has resulted in an increase in crime disproportionate to the in- 
 crease in adult population. Such comparable statistics of crime and 
 population as it has been possible to obtain indicate that immigrants are 
 less prone to commit crime than are native Americans.' The special cen- 
 sus report on prisoners after analyzing the statistics of nativity for 
 the whole country said that it was 'evident that the popular belief that 
 the foreign born are filling the prisons has little foundation in fact.' 
 Chicago statistics furnish further confirmation of this statement.""- 
 
 Our study of 587 delinquent women, including the important 
 types of women offenders throughout New York State, adds to 
 the data already collected in the reports which have been mentioned, 
 by showing that the native born white women offenders have a 
 representation among the total group of female delinquents slightly 
 greater than their percentage of the general population in the state, 
 the native born colored a percentage considerably greater than their 
 
 ' "Report of the City Council Committee on Crime of the City of Chicago," 
 March 22, 1915, p. 59. 
 
 167 
 
168 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 percentage of the general population, and the total foreign born, on 
 the other hand, a percentage considerably less than their proportion 
 of the general female population of the same ages. In addition to a 
 comparison of the numbers of delinquent women within the nativity 
 and color groups, with the general female population in New York 
 City over fifteen years of age, several comparisons have been made of 
 the degree of delinquency between the total foreign, total native, 
 native white, and native colored. These comparisons have been worked 
 out in the order followed in Chapter VI, considering first the nature 
 of the present offense in its two classifications, the number of previous 
 convictions followed by the percentage of recidivists and first offenders, 
 and finally the age at first conviction, nature of first offense and first 
 sentence. 
 
 BIRTHPLACE AND COLOR 
 
 It is very desirable, in any comprehensive survey of the relation of 
 nativity and color to delinquency, to shov; comparisons between the 
 various foreign countries and groups of countries from which the de- 
 linquents come. Unfortunately, the total number of foreign born 
 women in our entire group is only 172, and the largest number from 
 any one foreign country is twenty-four. Even by grouping together 
 those born in western Europe (65 cases) or southern and eastern 
 Europe (79 cases), the number is too small for statistical uses, and 
 it would not be safe to draw even the most general conclusions from 
 so few observations. 
 
 There is, however, a vast amount of material which has been writ- 
 ten to show the comparisons between the total foreign and the total 
 native born in their relation to criminality. To supplement that mate- 
 rial, we have used our total foreign born group which we believe to be 
 a true sample of the foreign born women delinquents in New York 
 State, and have compared this group with both the native white and 
 native colored found in the same institutions. 
 
 The following frequency table shows the distribution of the women 
 studied, by the country of birth and by color. It will be noted that 
 there are four colored foreign born cases, one in Auburn, two in the 
 Penitentiary, and one in the Probation group. These four cases were 
 counted among the total foreign born throughout, except for purp.oses 
 of comparison with the Census figures where they were added to the 
 total negro group, leaving the foreign white for comparison with the 
 
NATIVITY IN RELATION TO DELINQUENCY 169 
 
 TABLE 40 
 
 BIRTHPLACE 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups by 
 
 Birthplace 
 
 
 Institdtional Groups 
 
 
 Birthplace 
 
 Bed- 
 ford 
 
 Auburn 
 
 Mag- 
 dalen 
 
 Peni- 
 tentiary 
 
 Work- Proba- 
 house tion 
 
 Total 
 
 Total Native Born 
 
 8S.1 
 
 51.3 
 
 68.4 
 
 57.4 
 
 70.0 77.2 
 
 69.1 
 
 
 
 Native White 
 
 68.3 
 19.8 
 
 11.9 
 
 35.0 
 16.3 
 
 48. 8 
 
 68.4 
 31.6 
 
 38.0 
 19.4 
 
 42.6 
 
 34. ( 
 36. ( 
 
 30. ( 
 
 3 68.5 
 D 8.7 
 
 ) 22.8 
 
 51.5 
 
 Native Colored 
 
 Total Foreign Born 
 
 17.6 
 30.9 
 
 
 
 
 3.0 
 
 6.3 
 
 i'.s 
 i'.s 
 
 2.5 
 
 'sis 
 
 1.3 
 
 5.0 
 
 11.3 
 
 '2^5 
 1.3* 
 
 'sis 
 'iis 
 
 1.3 
 1.3 
 
 10.5 
 
 'iis 
 
 i'.O 
 1.3 
 4.0 
 
 'iis 
 
 ■2i6 
 
 '2i6 
 1.3 
 1.3 
 
 iis 
 
 4.6 
 '2i8 
 
 "ig 
 
 1.9 
 1.9 
 5.6 
 2.8 
 5.6 
 1.9 
 
 "ig 
 
 .9 
 .9 
 
 2.8 
 .9* 
 
 '7i4 
 "i9* 
 
 2.( 
 
 '2ic 
 4.C 
 l.C 
 7.C 
 l.C 
 l.C 
 
 iic 
 
 '5ic 
 
 'iic 
 
 5.0 
 
 ) 6.5 
 1.1 
 
 'iii* 
 iii 
 
 iii 
 
 ) 1.1 
 
 1.1 
 
 1.1 
 
 iii 
 
 '2i2 
 1.1 
 1.1 
 S.3 
 
 5.2 
 
 Belgium 
 
 .2 
 
 Bermuda Islands 
 
 Canada 
 
 1.0 
 
 .2 
 
 .7 
 
 Cuba 
 
 
 .2* 
 
 Denmark 
 
 
 .2 
 
 England 
 
 
 .7 
 
 Finland 
 
 
 .5 
 
 France 
 
 
 .7 
 
 
 
 3.1 
 
 Hungary 
 
 1.0 
 1.0 
 2.0 
 
 1.4 
 
 
 4.0 
 
 Italy 
 
 2.7 
 
 New Zealand 
 
 Norwav 
 
 .2 
 .4 
 
 
 
 .2 
 
 Austrian Poland 
 
 Russian Poland 
 
 Porto Rico 
 
 1.0 
 1.0 
 
 .4 
 1.8 
 .4* 
 
 
 
 .4 
 
 
 1.0 
 1.0 
 
 4.3 
 
 Scotland 
 
 .5 
 
 Sicily 
 
 .4 
 
 
 
 1.8 
 
 Switzerland 
 
 
 .2 
 
 
 
 .2 
 
 British W. Indies 
 
 
 .2* 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100. 
 
 100.0 
 
 100.0 
 
 
 
 Number of cases 
 
 101 
 
 SO 
 
 76 ■ 
 
 108 
 
 100 
 
 92 
 
 557 
 
 ♦Colored. 
 
 
 
 
 
 
 
 
170 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 native white.^ Since the number of cases is too small to regroup satis- 
 factorily, even into English-speaking and non-English-speaking foreign 
 bom women, we have listed the countries of birth in alphabetical order, 
 with no attempt to classify as to geographical or language distribu- 
 tion. It is obvious, first of all, from this table, that the percentages 
 of native and foreign born women vary considerably between the in- 
 stitutional groups. 
 
 Bedford has the lowest percentage of foreign bom women. This 
 may be partly explained by the fact that Bedford receives commit- 
 ments not only from New York City, but also from the counties along 
 the Hudson River as far north as Rensselaer County, and that Bed- 
 ford has the smallest percentage of commitments from New York City 
 where there is a higher percentage of foreign born in the general popu- 
 lation than in the rest of New York State. On the other hand. Auburn 
 Prison has the highest percentage of foreigfn born though it takes pris- 
 oners from all over the State. A later table will partially explain this 
 by showing that a larger percentage of felons is found among the 
 foreign born than among the native white or native colored. Since 
 Auburn is distinctly a felony group, we may, then, expect a large per- 
 centage of its women to be foreign born. In all the New York City 
 groups, which are made up almost entirely from women convicted in 
 New York City, we should expect to find a higher percentage of foreign 
 born than in Bedford. It is of interest to note that the Penitentiary, 
 which has the highest percentage of felony cases, next to Auburn, 
 has the second highest percentage of foreign born women. The 
 total group of foreign born, 30.9 per cent of all women delinquents, 
 which we shall use for the comparative data to follow, is of course, 
 midway between the extremes of the Bedford and Auburn groups, 
 but gives a picture of the total foreign born in a sample drawn from 
 the whole of New York State. 
 
 If we turn to the percentage of native colored within the various 
 institutional groups, it is obvious that there are marked differences 
 here. The Magdalen does not take colored women, so that they are 
 
 ' For the most part, these four cases are not extreme enough in any way seri- 
 ously to affect the central tendency of the foreign born in any particular group 
 of facts used. The ages are as follows : 27, 28, 31 and 27. Nature of present 
 offense : offense against chastity, 1 ; offenses against property rights, 3. Num- 
 ber of previous convictions: none in two cases; 1 in one case; 10 in one case. 
 Age at first conviction: 27, 28, 28, 29. First- offense: offense against chastity, 1; 
 offenses against property rights, 3. First sentence : Probation, 1 ; Reformatory, 
 1 ; Penitentiary, 1 ; State Prison, 1. 
 
NATIVITY IN RELATION TO DELINQUENCY 171 
 
 straightway eliminated from the discussion. The probation group has 
 a small percentage of colored because of the meager facilities for 
 supervising colored girls on probati'on. Only the most promising col- 
 ored girls are considered for probation instead of an institutional sen- 
 tence because of the difficulty of looking after them without an ade- 
 quate staff of probation officers. Within the three groups, Bedford, 
 Auburn and the Penitentiary, the differences in the percentages of 
 colored women are not marked. The very high percentage of colored 
 women in the Workhouse is undoubtedly partly due to the practice of 
 giving short Workhouse terms to first or second offenders who might, 
 if they had been- white, have been given a chance on probation ; and 
 partly due to the fact that several of the private institutions in the 
 city refuse to take colored women, for whom, then, there is no alterna- 
 tive but a long Reformatory sentence or a Workhouse term. The total 
 group of colored women, 17.6 per cent of all the women in the study, 
 will be used in the comparisons to follow. 
 
 (a) Comparison of Nativity and Color Between Delinquent Women 
 and General Female Population in New York City 
 
 In comparing the delinquent women with the general female 
 population, with respect to nativity and color, we have chosen the popu- 
 lation of New York City rather than New York State for the com- 
 parison, and have used only the delinquent women convicted in New 
 York City. This plan seemed more advisable since the distribution of 
 the nativity groups varies so markedly between New York City and 
 New York State, and so few of our cases, only 11.9 per cent, were 
 convicted in the counties outside of Greater New York. From the 
 Census figures for New York City in 1910, we were able to get only 
 the number of females in the general population over fifteen years 
 of age. Since the members of this delinquent group are all supposedly 
 sixteen years of age or over, there will be a slight but unavoidable dif- 
 ference in the basis of comparison. 
 
 From chart 9, made up on the basis of the total group of delinquent 
 women convicted in New York City, it is evident that the native white 
 have a slightly larger representation in the group of delinquents than 
 they have in the general female population of the same ages. We 
 might also compare here the percentage of native white in the general 
 population with the percentages of native white delinquent women in 
 
172 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 the Magdalen, Penitentiary, Workhouse and Probation groups in 
 
 Table 40.^ 
 
 It will be seen that the percentage of native white in the Mag- 
 dalen is 68.4, or 21.6 per cent larger than the percentage of native 
 white in the general female population in New York City over fif- 
 teen years. In the Penitentiary and the Workhouse, the native white 
 run 8.8 per cent and 12.8 per cent below their representation of the 
 native white in the general population. In the Probation group the 
 native white delinquents have 21.7 per cerit more than their propor- 
 
 PER CENT 
 
 40 60 
 
 NATIVE WHITE *| 
 
 FOREIGN WHTE ^Jf 
 
 ALL OTHERS 
 
 I I GENERAL FEMALE POPULATION 
 
 ■|m DELINQUENT WOMEN IX 
 
 Chart IX 
 
 Percentage Comparison Showing Nativity and Color Among (1), General 
 Female Population over IS Years of Age in New York City in 1910, and (2), 
 481 Female Delinquents Convicted in New York City. 
 
 tionate representation among the general population. Our total de- 
 linquent group in New York City, then, while being only slightly 
 higher in the percentage of native white than the same class in the 
 population of New York City, varies within itself as to proportions, 
 two of the institutional groups having less, and two of the groups 
 more than their proportionate share of the native white. 
 
 If we turn to the group of negroes in Chart IX, we note that 
 their proportion among the deUnquent women is markedly higher 
 than we should expect to find from the number of negro women over 
 
 ' Since Bedford and Auburn each have a number of commitments from dis- 
 tricts outside of New York City and Table 40 includes all the women studied, 
 both from New York City and New York State, we shall not compare these 
 two groups with the general female population in Chart 9. 
 
NATIVITY IN RELATION TO DELINQUENCY 173 
 
 fifteen years of age in New York City. While they are a small part, 
 only 2.4 per cent of the population in question, they furnish 19.3 per 
 cent of the delinquents. Within the New York City institutional 
 groups in Table 40, it will not be possible to make many comparisons 
 since the Magdalen takes no negro women. The Penitentiary, how- 
 ever, has 21.2 per cent of its group negroes,* the Workhouse 36.0 per 
 cent and the Probation group 9.8 per cent. The Workhouse is particu- 
 larly striking in having fifteen times as many negro women as one 
 would expect from the female negro population over 15 years in New 
 York City. As noted before, the very much larger proportion of col- 
 ored in the Workhouse as compared with the Probation group is prob- 
 ably accounted for by the fact that many of these women are sent to 
 the Workhouse on their first sentence instead of being given probation. 
 
 The foreign white delinquent women have a far smaller percentage 
 in our total group than have the foreign white women in the female 
 population over fifteen years of age in New York City. This is worthy 
 of mention in view of the fact that many persons still believe that 
 the amount of crime in this country, and especially in New York City, 
 is greatly increased by immigration. New York City with a foreign 
 born white population of females over fifteen years amounting to 50.8 
 per cent of its total female population offers a ready field for those of 
 foreign birth who intend to violate the law. And yet they contribute 
 only 32.2 per cent of our total delinquent group. In none of the in- 
 stitutional groups do the foreign born white contribute as high a per- 
 centage as they have in the general female population over fifteen 
 years, and in most instances the difference is very marked. 
 
 The group of "All Others" shown on Chart 9 by the lowest bar is 
 composed of 1 case only, in the delinquent group. It seems absurd to 
 draw any coniparison based on 1 case, but as the chart shows, the per- 
 centage of delinquent women and women in the general population 
 included under "All Others" are the same. 
 
 It has been suggested that one reason why the foreign born ought 
 to have a smaller percentage in the criminal class than in the general 
 population in this country is because of the immigration law which 
 excludes from admission into the United States "persons who have 
 been convicted of or admit having committed a felony or other crime 
 or misdemeanor involving moral turpitude," and "prostitutes, or 
 women or girls coming into the United States for the purpose of pros- 
 
 *In the Penitentiary anli Probation groups add percentage of foreign born 
 colored, starred in Table 40, to native colored, to obtain total negro group. 
 
174 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 titution or for any other immoral purpose."^ Though this law is diffi- 
 cult for any Board of Review to enforce strictly since they have no 
 knowledge of the past lives of the immigrants coming before them, it 
 is probably true that the law does act as a deterrent on some criminals 
 who would not risk being returned from the United States. The num- 
 ber of aliens who are actually debarred at the time of coming to this 
 country or later deported as criminals, however, was only 912 or .8 
 per cent of the total aliens admitted to the United States from July 1, 
 1913, to June 30, 1914.° Though the number excluded as criminal 
 aliens is small, this probably has a slight effect on the percentage of 
 foreign born delinquents in New York City. 
 
 AGE CLASSIFIED BY NATIVITY AND COLOR 
 
 Before proceeding to the nature of delinquency among the nativity 
 groups in our study, we shall consider the question of age at present 
 conviction, since the age factor will to a certain extent affect the num- 
 ber of convictions to be discussed later. The following table (Table 
 41) gives the frequency distribution of age for all groups.'' At first 
 glance, it is obvious that the foreign born have the longest range of 
 ages, the native white the next in order and the native colored the 
 shortest in range. On the other hand, the massing of ages appears 
 to be in the earlier age groups, particularly in the native white group. 
 
 In orcfer to compare the central tendencies of the different nativity 
 groups, the following three tables are presented. 
 
 Table 42 indicates that the mean age of the native white is 26.92 
 years, or slightly younger than the mean age of the native colored 
 group, 27.00 years. Since the ratio of the actual difference between 
 the means to the standard deviation of the differences is only .08, the 
 chances are even that the observed differences might have occurred by 
 chance. There is, therefore, no demonstrable difference between the 
 mean age of the native colored and the native white. The comparison 
 of the standard deviations of the two groups, however, indicates a 
 wider scattering of ages in the native white than in the native colored 
 group. The ratio of this difference to the standard deviation of the 
 difference is 2.91, which indicates that there is with little doubt a 
 
 "Immigration Act of Feb. 20, 1907, as Amended by the Acts of March 26, 
 1910, and March 4, 1913. Section 2. 
 
 ""Annual Report of the Commissioner General of Immigration,'' pp. 104 
 and 108. , 
 
 'All of the following discussion in this chapter is based on the total group 
 of women delinquents committed from all of New York State. 
 
NATIVITY IN RELATION TO DELINQUENCY 175 
 
 TABLE 41 
 AGE AT PRESENT CONVICTION 
 
 Per Cent Distribution of Delinquent Women Classified by Nativity and 
 
 Color 
 Total Group 
 
 Age 
 
 14 to 
 
 18 " 
 
 22 
 
 26 
 
 30 
 
 34 
 
 38 
 
 42 
 
 46 
 
 50 
 
 54 
 
 58 
 
 62 
 
 66 
 
 70 
 
 18 years , 
 
 22 
 
 26 
 
 30 
 
 34 
 
 38 
 
 42 
 
 46 
 
 50 
 
 54 
 
 58 
 
 62 
 
 66 
 
 70 
 
 74 
 
 Total. 
 
 Number of cases . 
 
 Nativity and Colob 
 
 Total 
 Foreign Born 
 
 1.2 
 
 12.9 
 
 15.8 
 
 14.6 
 
 16.4 
 
 11.1 
 
 7.6 
 
 5.9 
 
 4.1 
 
 4.1 
 
 2.3 
 
 1.8 
 
 1.2 
 
 100.0 
 
 171 
 
 Total 
 Native Born 
 
 10.1 
 
 23.1 
 
 21.3 
 
 15.8 
 
 9.4 
 
 6.5 
 
 6.2 
 
 1.8 
 
 3.1 
 
 1.4 
 
 .8 
 
 .8 
 
 100.0 
 
 385 
 
 Native 
 White 
 
 11.2 
 
 25.8 
 18.5 
 15.7 
 9.1 
 5.2 
 5.6 
 2.1 
 3.8 
 1.1 
 1.1 
 1.1 
 
 100.0 
 
 287 
 
 Native 
 Colored 
 
 7 
 15 
 29 
 16 
 10 
 10, 
 
 8, 
 
 1. 
 
 1. 
 
 1. 
 
 100.0 
 
 Total 
 
 7.4. 
 
 20.0 
 
 19.6 
 
 15.5 
 
 11.5 
 
 7.9 
 
 6.7 
 
 3.1 
 
 3.4 
 
 2.0 
 
 1.3 
 
 1.1 
 
 .4 
 
 .2 
 
 .2 
 
 100.0 
 
 556 
 
 TABLE 42 
 
 NATIVE WHITE AND NATIVE COLORED 
 
 Comparison of the Means and Standard Deviations of Age at Present 
 Conviction for Native White and Native Colored 
 
 
 Native 
 White 
 
 Native 
 Colored 
 
 Difference 
 
 d 
 
 Chances that real 
 
 difference does 
 
 not exist are 1 
 
 in: 
 
 Mean 
 
 26.92 
 ±.560 
 
 27.00 
 ±.761 
 
 -.08 
 
 .08 
 
 2 
 
 
 
 
 
 0" 
 
 9.48 
 ±.396 
 
 7.54 
 ±.538 
 
 1.95 
 
 2.91 
 
 556 
 
 (J a 
 
 
 
 
 Cases 
 
 287 
 
 98 
 
 
 
 
 
 
176 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 valid difference in the variability of the two groups, with the native 
 white having a wider scattering. 
 
 Having indicated that there is no demonstrable difference between 
 the mean ages of the native white and native colored, we should be 
 justified in comparing the ages of the total native born and the total 
 foreign born women. But, because of the apparently valid difference 
 between the dispersion of ages in the native white and native colored 
 groups, we have calculated the comparison of the differences between 
 the mean ages of the native white and total foreign born — ^26.9 vs. 
 32.9. The ratio of the difference of the means to the standard devia- 
 tion of the differences (5.92) indicates that there is undoubtedly 
 a valid difference between the native white and total foreign born, 
 with reference to their mean age, with a tendency for the foreign born 
 delinquent women to be older than the native born. A difference, 
 though less striking, in the dispersion of the ages of the two groups is 
 also seen in the ratio of 1.89, with the foreign born having a wider 
 scattering of ages. 
 
 TABLE 43 
 NATIVE WHITE AND TOTAL FOREIGN BORN 
 
 Comparison of the Means and Standard Deviations of Age at Present 
 Conviction for Native White and Total Foreign Bom 
 
 
 Native 
 White 
 
 Total 
 Foreign Born 
 
 Difference 
 
 d 
 
 Chances that real 
 
 difference does 
 
 not exist are 1 
 
 in: 
 
 Mean 
 
 26.92 
 
 ±.560 
 
 32.85 
 ±.830 
 
 -5.92 
 
 5.92 
 
 00 
 
 
 
 
 
 0" 
 
 9.48 
 ±.396 
 
 10.82 
 
 ±.587 
 
 -1.34 
 
 1.89 
 
 34 
 
 (yff 
 
 
 
 
 
 287 
 
 170 
 
 
 
 
 
 
 Since there is a valid difference between the mean ages of the 
 native white and total foreign born, and no demonstrable difference 
 between the mean ages of native white and native colored, we should 
 expect, as shown in Table 44, to find a real difference between the 
 mean ages of the total native and the total foreign born. The ratio 
 6.22 indicates a valid difference between the mean ages of. the two 
 groups and the ratio 2.67 indicates that there is also a- valid difference 
 in the dispersion, the total foreign born group having a wider scatter- 
 ing. It will be well to keep this difference in mind for the later dis- 
 
NATIVITY IN RELATION TO DELINQUENCY 177 
 
 TABLE 44 
 
 TOTAL NATIVE BORN AND TOTAL FOREIGN BORN 
 
 Comparison of the Means and Standard Deviations of Age at Present 
 Conviction for Total Native Bora and Total Foreign Bom 
 
 
 Total 
 Native Born 
 
 Total 
 Foreign Born 
 
 Difference 
 
 d 
 0-d 
 
 Chances that real 
 
 difference does 
 
 not exist are 1 
 
 in: 
 
 Mean 
 
 26.94 
 ±.460 
 
 32.85 
 ±.830 
 
 -5.90 
 
 6.22 
 
 00 
 
 
 
 
 
 (T 
 
 9.03 
 ±.325 
 
 10.82 
 
 ±.587 
 
 -1.79 
 
 2.67 
 
 263 
 
 (ja . 
 
 
 
 
 Cases 
 
 385 
 
 170 
 
 
 
 
 
 
 cussion of the criminal record, — that the foreign born group show an 
 older mean age than either the native white or total native born, and 
 also a wider scattering than is found among the native white or total 
 native born. 
 
 Very, little has been done in gaging the nature and extent of crim- 
 inality among the foreign born. Miss Grace Abbott writes : 
 
 "The United States Immigration Commission made 'the central fea- 
 ture' of its investigation the determination of how 'the criminality of 
 the immigrant differs from that of the native born' (Vol. 36, p. 13). The 
 material available or secured by the commission was, however, very 
 meager and no attempt was made to discover to what extent apparent 
 tendencies on the part of certain races to commit certain kinds of crime 
 were the result of local conditions rather than race."* 
 
 As noted before, our numbers from individual foreign countries are 
 too small for use, but with the total foreign born group we have been 
 able to draw some conclusions as to the nature and extent of their 
 delinquency. 
 
 NATURE OF PRESENT OFFENSE, CLASSIFIED BY NATIVITY GROUPS 
 
 (a) New York City Police Department Classiftcation 
 
 Table 45, followed by Chart X, shows the nature of the present 
 offense classified by the New York City Police Department classifica- 
 
 ' Abbott, Grace. "Immigration and Crime." (Report of Committee "G" 
 of the Institute.) Journal of the American Institute of Criminal Law and 
 Criminology. Vol. VI, No. 4, Nov., 1915, p. 528. 
 
178 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 45 
 
 NATURE OF PRESENT OFFENSE 
 
 Per Cent Distribution of Delinquent Women in Nativity and Color 
 
 Groups, Classified by New York City Police Department 
 
 Classification 
 
 Nature of OfEenae 
 
 Offenses against the Person 
 
 Offenses against Chastity 
 
 Offenses against Family and 
 Children 
 
 Offenses against Regulations for 
 Public Health, Safety and 
 PoUcy 
 
 Offenses against Administration 
 of Government 
 
 Offenses against Property Rights 
 
 General Criminality 
 
 Total 
 
 Number of cases 
 
 Nativity and Colob 
 
 Total 
 
 Foreign 
 
 Bom 
 
 8.7 
 41.9 
 
 1.7 
 
 9.3 
 
 .6 
 
 34.3 
 
 3.6 
 
 100.0 
 
 Total 
 Native 
 Born 
 
 3.4 
 60.0 
 
 2.3 
 
 9.9 
 
 .5 
 
 19.7 
 
 4.2 
 
 100.0 
 
 Native 
 White 
 
 2.8 
 62.0 
 
 3.1 
 
 9.4 
 
 .7 
 
 17.4 
 
 4.5 
 
 100.0 
 
 Native 
 Colored 
 
 5.1 
 54.1 
 
 11.2 
 
 26.5 
 3.1 
 
 100.0 
 
 Total 
 
 6.0 
 54.4 
 
 2.2 
 
 9.7 
 
 .5 
 
 24.2 
 
 4.0 
 
 100.0 
 
 172 
 
 386 
 
 287 
 
 98 
 
 557 
 
 tion for the nativity and color groups. A first glance at the chart will 
 show what was evident in a comparison of the institutional groups, 
 that the offenses against chastity have the first place and offenses 
 against property rights second, in each of the three nativity groups. 
 The ratio between the two varies, however, in the groups, the foreign 
 born having the smallest percentage of offenders against chastity, and 
 the largest percentage of offenders against property rights, while the 
 native white have the largest percentage of offenders against chastity 
 and the smallest percentage of offenders against property rights. In 
 each of these two large divisions of offenses, the native colored stand 
 on middle ground. The other divisions of the classification show that 
 the foreign born have a percentage of offenses against the person three 
 times as large as that of the native white, but less than twice as large 
 
O 
 
 8 
 
 g 
 
 3 
 a" 
 
 X 
 Z 
 
 
 
 
 
 
 
 ■ 
 
 .1 1. 
 
 C8 
 U 
 
 j5 
 
 
 W 
 
 
 o 
 
 h K 
 
 n 
 
 
 z 
 
 " 
 
 o 
 
 
 
 
 
 S 
 
 K 
 
 
 O 
 
 
 s f 
 
 u 3 
 
 fL,0 
 
 
 S> o 
 
 Do 
 
 u T3 
 O 3 
 >H !« 
 
 JO 
 
 ■a 
 
 «3 
 
 U 
 
 « 
 
 3 
 
 J3 
 
 P 
 
 i s s I 
 
 H u »* S 
 
 179 
 
180 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 as that of the native colored. The offenders against the family and 
 children, and the offenders against administration of government are 
 too few to use for comparison, since there are only a total of twelve 
 cases in the one and three in the other. The offenders against regula- 
 tions for public health have their largest percentage in the native col- 
 ored group, all convicted of possessing or selling drugs. The general 
 criminality group shows no marked differences in percentages between 
 the nativity groups. Because the numbers are largest and because the 
 difference is more marked, the most important conclusion we can draw 
 from this table, is that the foreign born have a noticeably smaller per- 
 centage of offenders against chastity, and a noticeably larger percent- 
 age of offenders against property rights than have either the native 
 white or native colored. 
 
 (b) Felonies and Misdemeanors 
 
 A natural deduction from the preceding table is that the foreign 
 born probably have a higher percentage of felons than have the other 
 groups. In Chart III, it appeared that the offenders against property 
 rights are made up more largely of felons than is the group of offend- 
 ers against chastity rights. Table 46 and Chart XI show that the 
 foreign bom have a percentage of felons considerably over twice that 
 of the native white and nearly twice that of the native colored. In 
 other words, about a third of the foreign born were convicted of 
 felonies at the time we studied them, while approximately an eighth 
 of the native white and a fifth of the native colored were convicted 
 of felonies. Conversely, of course, the percentage of misdemeanants 
 or minor offenders is much larger among the native white and native 
 colored than among the foreign born. 
 
 The Special Report of the Census on Prisoners and Juvenile De- 
 linquents in Institutions states that: 
 
 "It is evident that the popular belief that the foreign born are filling 
 the prisons has little foundation in fact. It would seem, however, that they 
 are slightly more prone than the native whites to commit minor offenses." ' 
 
 Our data would not bear out this observation, so far as women are 
 concerned. On the contrary, we might say that within our group of 
 delinquents there is a very considerabfe tendency for the foreign 
 born women to commit more serious offenses than do the native born 
 within the same group. 
 
 • "Special Report of the Census Office : Prisoners and Juvenile Delinquents in 
 Institutions." 1904, pp. 18-19, 40-41. 
 
NATIVITY IN RELATION TO DELINQUENCY 181 
 
 EXTENT OF DELINQUENCY CLASSIFIED BY NATIVITY AND COLOR 
 
 We have seen so far that the foreign born group of women delin- 
 quents shows a smaller percentage among our total delinquent group 
 than their proportion of the population entitles them to have, but that 
 among the foreign born who are convicted there is a tendency for 
 them to commit the more serious oflFenses. A natural sequence to this 
 will be to measure the extent of delinquency among those of the foreign 
 born who reach the courts, and compare this with the amount of de- 
 
 TABLE 46 
 
 NATURE OF PRESENT OFFENSE 
 
 Number and Per Cent of Delinquent Women Convicted of Felonies 
 and Misdemeanors Classified by Nativity and Color 
 
 
 Nature of Pbebbkt Offense 
 
 Total 
 
 Nativity and Color 
 
 Felonies 
 
 Misdemeanors 
 
 
 Number 
 
 Per Cent 
 
 Number 
 
 Per Cent 
 
 Number 
 
 Per Cent 
 
 Total Foreign Born 
 
 Total Native Bom 
 
 58 
 59 
 
 33.7 
 15.3 
 
 114 
 326 
 
 66,3 
 84.7 
 
 172 
 385 
 
 100.0 
 100.0 
 
 Native White 
 
 Native Colored 
 
 40 
 19 
 
 13.9 
 19.4 
 
 247 
 79 
 
 86.1 
 80.6 
 
 287 
 98 
 
 100.0 
 100.0 
 
 Total 
 
 117 
 
 21.0 
 
 440 
 
 79.0 
 
 557 
 
 100 
 
 
 
 PER CENT 
 O 20 40 eo 80 100 
 
 ITOREIGN BORN 
 NATIVE WHITE 
 NATIVE COLORED 
 
 HlSOEUBANANn 
 
 Chart XI 
 
 Per cent Distribution of Felons and Misdemeanants Classified by Nativity 
 and Color. 
 
182 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 linquency among the native born. The only measure of extent of de- 
 linquency which we shall use in this connection is the number of pre- 
 vious convictions. 
 
 (o) Number of Previous Convictions 
 
 (1) Comparison of Differences Among Nativity Groups. — Table 
 47 shows the range of number of previous convictions among the 
 various nativity groups, and indicates that the range is longest in the 
 native white group, and is of equal leiigth among the foreign born 
 and native colored. This table taken in connection with the three 
 following tables is of value in enabling us to make comparisons between 
 the mean number of convictions within the various groilps. 
 
 TABLE 47 
 NUMBER OF PREVIOUS CONVICTIONS 
 
 Per Cent Distribution of Delinquent Women Classified by Nativity 
 
 and Color 
 
 
 Nativitt and Color 
 
 
 Number of Previoua Convictions 
 
 Total 
 
 Foreign 
 
 Born 
 
 Total 
 Native 
 Born 
 
 Native 
 White 
 
 Native 
 Colored 
 
 Total 
 
 
 
 52.3 
 19.2 
 11.6 
 4.7 
 4.1 
 2.9 
 1.2 
 
 "!6- 
 
 .6 
 .6 
 .6 
 
 '".6 
 .6 
 .6 
 
 41.3 
 24.9 
 
 10.7 
 
 8.1 
 
 4.4 
 
 3.1 
 
 1.6 
 
 2.1 
 
 .8 
 
 .8 
 
 .5 
 
 .3 
 
 .5 
 
 .3 
 .3 
 .3 
 
 45.0 
 
 27.2 
 
 9.8 
 
 7.3 
 
 3.5 
 
 1.7 
 
 .7 
 
 1.4 
 
 .4 
 
 .7 
 
 .7 
 
 .4 
 
 .4 
 
 "'!4 
 
 ";4 
 
 .4 
 
 30.6 
 
 18.4 
 
 13.3 
 
 10.2 
 
 7.1 
 
 7.1 
 
 4.1 
 
 4.1 
 
 2.0 
 
 1.0 
 
 'i!6 
 
 44 7 
 
 1 
 
 23 2 
 
 2 
 
 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 8 
 
 11.0 
 7.0 
 4.3 
 3.1 
 1.4 
 1.4 
 7 
 
 9 
 
 7 
 
 10 
 
 11 
 
 12 
 
 13 
 
 14 
 
 18 
 
 20 
 
 31 
 
 .5 
 .4 
 .4 
 .2 
 .4 
 .4 
 .2 
 .2 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 
 Number of cases 
 
 172 
 
 385 
 
 287 
 
 98 
 
 557 
 
 
NATIVITY IN RELATION TO DELINQUENCY 183 
 
 We shall first compare the mean number of convictions and the 
 amount of variation about these means for the native white and native 
 colored groups. Table 48 suggests that the native colored tend to 
 have been convicted more often than the native white since the mean 
 number of convictions in the one is 2.49 and in the other only 1.52. 
 Since the ratio of the difference of the means to the standard de- 
 viation of the difference is 2.81, it would seem that this apparent dif- 
 ference is almost certainly valid. From a comparison of the standard 
 deviations, however, it is clear that there is no demonstrable difference 
 in the dispersion of the two groups. 
 
 Having shown that there is a significant tendency for the native 
 colored to have a larger number of convictions than the native white, 
 it is necessary for us to compare the total foreign born with the native 
 white before extending this comparison to the total native born and 
 total foreign born. A comparison of the means alone would suggest 
 that the native white tend to have been convicted slightly more often 
 than the foreign born, but since the difference between the means is 
 only .36 times the standard deviation of the difference, we have no 
 evidence that this is more than the chance variation which might occur 
 through random sampling. (See Table 49.) 
 
 Since there is no demonstrable difference in number of convictions 
 between the native white and the total foreign born, and since it has 
 been shown (Table 48) that there is a tendency for the native colored 
 to have a larger mean number of convictions than the native white, we 
 are justified in stating that there must also be a valid difference in the 
 
 TABLE 48 
 
 NATIVE WHITE AND NATIVE COLORED 
 
 Comparison of the Means and Standard Deviations of Number of 
 Previous Convictions for Native White and Native Colored 
 
 
 Native 
 White 
 
 Native 
 Colored 
 
 Difference 
 
 d 
 
 era 
 
 Chances that real 
 
 difference does 
 
 not exist are 1 
 
 in; 
 
 Mean 
 
 1.52 
 ±.176 
 
 2.49 
 + .299 
 
 -.97 
 
 2.81 
 
 400 
 
 (Tm 
 
 
 
 
 (T 
 
 2.96 
 + .124 
 
 2.96 
 ±.211 
 
 .01 
 
 .02 
 
 2 
 
 (r<r 
 
 
 
 
 Cases 
 
 287 
 
 98 
 
 
 
 
 
 
184 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 49 
 
 TOTAL FOREIGN BORN AND NATIVE WHITE 
 
 Comparison of the Means and Standard Deviations of Number of 
 Previous Convictions for Total Foreign Bom and Native White 
 
 
 Total 
 Foreign Born 
 
 Native 
 White 
 
 Di£ferenoe 
 
 d 
 
 Chances that real 
 
 di£Ference does 
 
 not exiat are 1 
 
 in: 
 
 Mean 
 
 1.42 
 ±.201 
 
 1.52 
 ±.175 
 
 -.10 
 
 .36 
 
 3 
 
 (Tin 
 
 
 
 
 
 
 2.64 
 ±.142 
 
 2.96 
 ±.124 
 
 -.32 
 
 1.70 
 
 22 
 
 (Ja . . . 
 
 
 
 
 
 172 
 
 287 
 
 
 
 
 
 
 mean number of convictions of the native colored and the total foreign 
 born, the native colored having a larger mean number of convictions. 
 
 For comparison with other groups of data, it may be well to com- 
 bine the native white and native colored and compare the mean num- 
 ber of convictions and the standard deviation for the total native born 
 with the same constants among the foreign born. Table 50 shows that 
 in a comparison of the means, the total native born have a slightly 
 larger mean number of convictions than the total foreign born. Since 
 this difference is only 1.3S times the standard deviation of the dif- 
 ference, our figures fail to offer convincing evidence of a real 
 
 TABLE 60 
 
 TOTAL FOREIGN BORN AND TOTAL NATIVE BORN 
 
 Comparison of the Means and Standard Deviations of Number of 
 
 Previous Convictions for Total Foreign Bom and 
 
 Total Native Bom 
 
 
 Total 
 Foreign Born 
 
 Total 
 Native Born 
 
 DifEerenoe 
 
 d 
 (Td 
 
 Chances that real 
 
 difference does 
 
 not exist are 1 
 
 in; 
 
 Mean 
 
 1.42 
 ±.201 
 
 1.77 
 ±.152 
 
 -.34 
 
 1.35 
 
 11 
 
 
 
 
 
 2.64 
 ±.142 
 
 2.99 
 ±.108 
 
 -.35 
 
 1.95 
 
 39 
 
 (To- 
 
 
 
 Cases 
 
 172 
 
 385 
 
 
 
 
 
 
NATIVITY IN RELATION TO DELINQUENCY 185 
 
 difference between the native and the foreign bom with respect to this 
 factor. 
 
 In interpreting the extent of delinquency of the groups distin- 
 guished by nativity and color, therefore, it will be well to bear in mind 
 ( 1 ) that the native colored are distinguished from both the native white 
 and the foreign born by having a larger mean number of convictions, 
 and (2) that, although the foreign born show a slightly smaller mean 
 number of convictions than do either the native white or the total native 
 born, the evider^ce fails to justify our considering these differences 
 demonstrably valid. 
 
 (2) Comparison of Differences Between Felons and Misdemeanants 
 in Nativity Groups. — It will be of interest to turn from the total group 
 of delinquent women classified by nativity and color to their classifica- 
 tion into felons and misdemeanants, keeping in mind the fact that the 
 foreign born have a larger percentage of their total group felons than 
 have either the native white or native colored. 
 
 Table 51 shows the distribution of number of convictions for na- 
 tivity groups among both the felons and misdemeanants. A hasty 
 glance shows that the range of convictions for the foreign born felons 
 is shorter than for any of the other groups of felons, and that it is 
 slightly longer among the native white and total native bom misde- 
 meanants than among the foreign born or native colored misdemean- 
 ants. 
 
 In order to determine whether there is a valid difference in the 
 number of previous convictions between the total foreign and the na- 
 tive born felons, the constants of the two groups have been compared. 
 The total native born felons were used instead of the native white and 
 native colored separately since the numbers were too small for this 
 separate use. Table 52 shows that the native born felons have a 
 larger mean number of convictions than have the total foreign born 
 felons, and that this observed difference is more than three times the 
 standard deviation of the difference. 
 
 From this we are entirely safe in saying that there appears to be a 
 valid difference between the mean number of previous convictions of 
 the total native born and total foreign born felons, with the native 
 born felons having a considerably larger mean number of previous 
 convictions. The difference in the dispersion of the groups also ap- 
 pears to be valid, with the native born felons having a much wider 
 scattering. 
 

 
 s 
 
 
 
 Ci 
 
 
 
 OS 
 
 
 
 <u 
 
 
 
 a 
 
 
 
 0) 
 
 
 
 •d 
 
 
 
 CO 
 
 
 
 
 
 
 § 
 
 
 
 •d 
 
 
 
 d 
 
 
 
 c« 
 
 
 
 03 
 
 
 tn 
 
 V 
 
 
 fe 
 
 ■a 
 
 
 
 
 3 
 
 
 P 
 
 
 
 a 
 
 ^:i 
 
 
 
 
 
 
 
 u 
 
 •g oi 
 
 1-1 
 
 ^ 
 
 ,^t 
 
 lO 
 
 t3 
 
 1:4 
 
 
 
 
 f5 
 
 1 
 
 
 fo 
 
 
 II 
 
 
 PC 
 
 
 
 ' 
 
 
 
 1 
 
 a 
 o 
 
 l-l 
 
 i 
 
 i 
 
 ■3 
 1 
 
 r-l •* 10 00 ■* rH 1-1 "3 1> (N UJ N (M "O 10 ■ d 
 
 
 
 8 
 
 iH 
 
 3 
 
 CO ■* 1-1 1~^ T)i M .-1 r-l ; 
 
 
 t* C3 iH CD "3 00 -co ■ 'CO • -CO | | 
 ,-i05C0OI>t-INM -rH • -rt • -.-I • • 
 
 
 
 8 
 
 g 
 
 
 ,-1 IN ^- CO CO 00 00 ■* -* ■* ■* . .tH ; •■* 
 
 
 
 8 
 
 (N 
 
 ■* o> a> t- M N ; ; ; ; 
 
 a 
 
 '1 
 
 tH 1> t~ CO -* (M 10 M CD C<3 CO CO -COCO -CO 
 
 
 
 § 
 
 CD f 
 CO 
 
 tH CD 00 -it CO T-l --I 
 
 1 
 
 1 
 
 CO t^ (N i-H CO »C 00 -OSOi -OS -ososos - • 
 
 
 
 
 
 
 
 i-H 
 
 
 00 CD CO CO >0 CO 1-H • • • ■ • 
 ■^T-ti-H • ; ; ; 1 
 
 1 
 
 3 
 5 
 
 ocD-^iH co'i>t*cor*ait^ -d ■ ■ 'd ■ 
 
 
 
 8 
 
 i-H 
 
 1> 
 
 t-l as 05 10 C5 t-l T-l (N 1-1 i-H 
 
 IOt-1 
 
 11 
 
 
 
 
 s 
 
 i-H 
 
 
 
 (Mi-Hi-Ht-I l-l iH 
 
 
 ?5 
 II 
 
 000*010 ■ '0 -lOiO -lO ■ ■ -lO • 
 
 »Ol-l T-i • • • 
 
 
 
 1—1 
 
 5 
 
 •Bo 
 1 
 
 ■^ CC T-H 10 -^ t> -^ tH TfH> t^ '-t^ '- ' -t^ '• 
 
 
 
 8 
 
 T-t 
 
 OS 
 
 10 
 
 c<i lo d 06 CO i-H CO ui CO 1-1 T-H .^ . . .,-1 . 
 
 t:H 1-i 1-H 
 
 J 
 1 
 
 CO T-t in b- 1> t^ • • ■ -t* • • ^ ■ ■ ■ * 
 
 -^ 00 I-H 1-1 i-H • ■ * -i-l 
 
 
 
 8 
 
 I— 1 
 
 s 
 
 1 
 
 1 
 
 1 
 
 
 
 
 
 
 
 
 1 
 
 <4H 
 
 s 
 
 1? 
 
 
 
 
 
 
 
 
 
 
 OTHCMCO-^iOOt^OOOSOiHCqcOTtioOOrH 
 i-Hi-Hi-ii-li-li-HCQCO 
 
 186 
 
NATIVITY IN RELATION TO DELINQUENCY 187 
 
 TABLE 52 
 
 TOTAL NATIVE BORN FELONS AND TOTAL FOREIGN BORN 
 
 FELONS 
 
 Comparison of the Means and Standard Deviations of Number of 
 
 Previous Convictions of Total Native Bom Felons and Total 
 
 Foreign Bom Felons 
 
 
 Total 
 
 Native Born 
 
 Felona 
 
 Total 
 
 Foreign Born 
 
 Felons 
 
 Difference 
 
 d 
 <T«. 
 
 Chances that real 
 
 difEerence does 
 
 not exist are 1 
 
 in: 
 
 Mean 
 
 2.53 
 
 ±.489 
 
 .79 
 ±.209 
 
 1.73 
 
 3.26 
 
 
 (Tm 
 
 1667 
 
 
 
 tr 
 
 3.75 
 ±.345 
 
 1.60 
 
 ±.148 
 
 2.16 
 
 5.74 
 
 
 a,T 
 
 00 
 
 
 
 Cases 
 
 59 
 
 58 
 
 
 
 
 
 
 Though it would be possible, so far as actual numbers are con- 
 cerned, to make a comparison between the native white or colored and 
 the total foreign born misdemeanants, we shall use only the total na- 
 tive born with the total foreign born (Table S3) so that this ratio will 
 be comparable with that in Table 52. While a comparison of the means 
 alone in Table 53 would suggest that the foreign bom misdemeanants 
 
 TABLE 53 
 
 TOTAL NATIVE BORN MISDEMEANANTS AND TOTAL 
 FOREIGN BORN MISDEMEANANTS 
 
 Comparison of the Means and Standard Deviations of Number of 
 
 Previous Convictions of Total Native Bom Misdemeanants 
 
 and Total Foreign Bom Misdemeanants 
 
 
 Total 
 Native Born 
 Misde- 
 meanants 
 
 Total 
 Foreign Born 
 Misde- 
 meanants 
 
 Difference 
 
 d 
 
 era 
 
 Chances that real 
 
 difference does 
 
 not exist are 1 
 
 in: 
 
 Mean 
 
 1.63 
 ±.156 
 
 1.75 
 ±.280 
 
 -.12 
 
 .37 
 
 3 
 
 Cm 
 
 
 
 
 a 
 
 2.81 
 ±.110 
 
 2.99 
 ±.198 
 
 -.18 
 
 .79 
 
 5 
 
 <Ta 
 
 
 
 
 Cases 
 
 326 
 
 114 
 
 
 
 
 
 
188 'STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 have a slightly larger mean number of convictions, the ratio of the 
 difference between the two to the standard deviation of the difference 
 is only .17, indicating that there is no demonstrable difference between 
 the central tendencies of the two groups of offenders. The difference 
 between the standard deviations of the two groups is not demonstra- 
 bly vaHd. 
 
 (&) Recidivists and First Offenders Among Nativity and 
 Color Groups 
 
 If we turn to Table 54 showing the percentage of recidivists and 
 first offenders among the felons and misdemeanants, as well as among 
 the total group of delinquents classified by nativity and color, we note 
 that within each of the groups the foreign born have a smaller per- 
 centage of recidivists than have either the native colored, native white 
 or total native born. In each group the native colored have the high- 
 est percentage of recidivists, the native white and total native born 
 the next highest, and the foreign born the smallest percentage. These 
 groups differ from one another by more marked degrees, however, 
 among the felons than among the misdemeanants. In the total group, 
 including both felons and misdemeanants, 52.3 per cent of the foreign 
 born have had no previous convictions, while only 44.9 per cent of 
 the native white and 30.6 per cent of the native colored have had no 
 previous convictions. Chart XII presents in graphic form the data 
 regarding recidivism for the three nativity groups. 
 
 PER CENT 
 
 FOREIGN BORN 
 
 NATIVE WHITE 
 
 NATIVE COLORED 
 
 ^^m RECIDIVISTS 
 
 I I FIRST OFFENDERS . ^ 
 
 Chart XII 
 
 Per cent Distribution of Recidivists and First Offenders- Classified by 
 Nativity and Color. 
 
o 
 U 
 
 I 
 
 
 
 •d 
 
 
 
 V 
 
 
 V) 
 
 a 
 
 
 P^ 
 
 (0 
 
 
 M 
 
 m 
 
 
 Q 
 
 C) 
 
 
 s 
 
 
 
 p^ 
 
 D3 
 
 
 
 ^§ 
 
 
 o 
 
 a 
 
 tH 
 
 ^ 
 
 i 
 
 lO 
 
 1 
 
 Q 
 
 3 
 
 (A 
 
 
 1 
 
 
 rn 
 
 CO 
 
 
 H 
 
 H 
 
 
 r/i 
 
 
 
 
 p)k 
 
 
 a 
 
 ^ 
 
 
 
 1 
 
 o 
 
 0) 
 
 Hi 
 
 1 
 
 mox 
 
 55.3 
 
 44.7 
 
 o 
 
 o 
 o 
 
 1-i 
 
 u 
 
 pMOJOQ 
 
 oi o 
 to CO 
 
 o 
 
 s 
 
 1— 1 
 
 $ 
 
 
 
 o 
 
 8 
 
 1—1 
 
 00 
 
 niog 
 
 
 O 
 
 8 
 
 1-1 
 
 00 
 CO 
 
 UJOg 
 IB?ox 
 
 1> w 
 
 E: si 
 
 O 
 
 8 
 
 I— 1 
 
 t-H 
 
 a 
 S 
 
 IB^ox 
 
 o o 
 
 l>^ CO 
 
 O 
 
 8 
 
 1—1 
 
 s 
 
 ■* 
 
 ■paTOpo 
 
 8AI»BN 
 
 tH CO 
 
 i c3 
 
 O 
 
 8 
 
 1— ( 
 
 1> 
 
 8AT^«^ 
 
 OS ,-1 
 
 o 
 
 o 
 o 
 
 
 niog 
 
 
 o 
 
 8 
 
 1—i 
 
 CD 
 IM 
 CO 
 
 ujog 
 
 Tl3l9IOJ 
 
 I^»ox 
 
 00 IM 
 
 3 ^ 
 
 o 
 
 8 
 
 1— I 
 
 1— 1 
 
 
 Io»ox 
 
 t> CO 
 00 i-i 
 
 o 
 
 o 
 
 o 
 
 1— ( 
 
 1—1 
 
 paioioo 
 
 SAIJOM 
 
 r- CO 
 
 f2 i 
 
 o 
 
 8 
 
 1-1 
 
 05 
 1-H 
 
 9AIJBii 
 
 o o 
 
 o 
 
 8 
 
 i-H 
 
 ? 
 
 niog 
 
 BAI^BN 
 
 "3 ^ 
 
 o 
 
 o 
 
 o 
 
 .-H 
 
 05 
 
 niog 
 
 a3t3io^ 
 
 IB?ox 
 
 t- CO 
 
 oi O 
 CO to 
 
 o 
 
 8 
 
 1—1 
 
 g 
 
 
 
 1 
 
 J 
 
 1 
 O 
 
 1 
 
 
 1 
 
 
 O 
 
 u 
 
 (D 
 
 a 
 
 189 
 
190 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 (1) Comparison With Male Felons in Sing Sing. — If we use 
 only the felons in Table 54, we have material comparable to the 
 study made by Dr. Glueck at Sing Sing of 608 male felons." This 
 is illustrated in Chart XIII which makes four comparisons. The 
 first of these, between the female foreign born felons and the male 
 foreign born felons, shows that there is a much smaller percentage of 
 recidivists among the female foreign born felons than among the males, 
 
 FOREIGN BORN 
 
 PER Cent 
 
 20 40 60 80 
 
 FEMALE FELONS 
 
 MALE FELONS 
 
 FEMALE FELONS 
 If ALE FELONS 
 
 NATIVE BORN 
 
 RECIDIVISTS 
 
 FIRST OFFENDERS 
 
 Chart XIII 
 
 Comparison of the Percentages of Recidivists and First Offenders Among 
 the Foreign and Native-Born of 117 Female Felons and 608 Male Felons in 
 Sing Sing Prison. 
 
 a diiiference between 39.7 per cent of recidivists in the one and 49.8 per 
 cent in the other. A second comparison shows that the native born 
 female felons have a smaller percentage of recidivists than the native 
 born male felons, a difference between 57.6 per cent in the one case 
 and 75.9 per cent in the other. The third comparison has been made 
 in Table 54, showing that the foreign born female felons have a much 
 smaller percentage of recidivists than the native born female felons. 
 The fourth comparison between foreign born male felons and native 
 born male felons shows that among the male felons also, the foreign 
 born have a percentage of recidivists very much smaller than the per- 
 
 " Op. cit., p. 140. 
 
NATIVITY IN RELATION TO DELINQUENCY 191 
 
 centage of recidivists among the native bom. In a comparison between 
 the nativity groups, then, in each case the foreign born show a smaller 
 percentage of recidivists, and in a comparison between felons classified 
 by sex, the foreign bom female felons show the smaller percentage of 
 recidivists.^^ 
 
 (c) Relationship Between Length of Time in this Country and 
 Number of Previous Convictions 
 
 (1) Number of Previous Convictions and Years in United States. 
 — Before concluding the discussion of the extent of criminality among 
 the foreign and native born, we shall consider the objection that the 
 foreign born delinquents are not represented in proportion to their 
 quota in the general population and have a smaller percentage of 
 the repeated offenders because they may have been in this country 
 too short a time to have come into conflict with the law and to have had 
 many convictions. In order to find out whether there seems to be any 
 relation between the number of convictions of the foreign born and 
 the length of time in the United States or the age at coming to the 
 United States, the two following correlations are presented. 
 
 Table 55 shows that the range of number of years in the United 
 States is from 1 to 46 years, with the mean number of years 14.04± 
 9.17. The coefficient of correlation (.33 + .069) would indicate that 
 there is a small, but significant, relationship between the length of time 
 in this country and the number of convictions. The correlation ratios 
 are slightly larger, .39±.065 and .36+ .067. The tendency toward a 
 higher record of convictions with a longer residence in this country is 
 very evident from the table. 
 
 (2) Number of Previous Convictions and Age at Coming to the 
 United States. — Table 56 correlates the number of previous convictions 
 with the age at coming to the United States instead of with the number 
 of years in the United States as in Table 55. The mean age at coming 
 to this country is 18.88±9.29 years. The small negative coeffi- 
 cient of correlation, —.17, only slightly more than twice its stand- 
 ard deviation of .074, would indicate that the relationship is small, but 
 that there is a slight tendency for the foreign born who come to the 
 United States when young tp have a larger number of convictions 
 
 "As stated in Chapter 6, p. 127, Dr. Glueck has used the term "recidivist" 
 to mean an individual who has served a previous sentence m prison, while our 
 use of recidivist includes all who have had a previous conviction. His per- 
 centage of recidivists would, therefore, be proportionately larger if put on a 
 basis exactly comparable with ours. 
 

 
 'o? 
 
 1 
 
 
 
 
 
 
 
 
 
 
 
 
 1 
 
 1 
 
 
 U3 CQOSOCOIOC 
 
 = ^ 
 
 
 h 
 
 CO W (N <N ^ d C 
 
 5 p^ 
 
 
 
 □9 
 
 E3 
 
 
 JTo (N CO OJ ■* a 
 
 J i-H 
 
 °. 
 
 tHCO-^ 
 
 o 
 
 .-iT-icqMccP 
 
 ' 5:: 
 
 ■^ 
 
 H 
 
 
 i-C 
 
 
 
 00 
 
 ^ 
 
 ^ 
 
 
 
 
 
 
 
 
 t- 
 
 
 
 
 "^ 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 *' 
 
 
 
 
 lO 
 
 
 
 
 iH 
 
 
 
 
 
 •^ 
 
 rH 
 
 ^ 
 
 
 
 
 
 
 
 CO 
 
 tH 
 
 ^ 
 
 
 
 
 
 
 DD 
 
 
 
 
 
 a 
 
 (N 
 
 
 
 
 o 
 
 
 
 
 t^ 
 
 B 
 
 
 
 
 
 'co 
 
 
 
 
 > 
 
 rH 
 
 "-t 
 
 '-I 
 
 M 
 
 § 
 
 
 
 
 
 
 
 
 
 o 
 
 O 
 
 r-l 
 
 rt 
 
 
 (0 
 
 
 
 
 
 p 
 
 
 
 
 
 OS 
 
 -^ 
 
 ^ 
 
 
 H 
 
 
 
 
 
 
 
 
 
 Ah 
 
 00 
 
 rH 
 
 iH 
 
 
 fe 
 
 
 
 
 
 O 
 
 
 
 
 
 a 
 
 I-- 
 
 
 
 
 H 
 
 
 
 
 
 «D 
 
 >-t rH 
 
 C^ 
 
 
 1 
 
 
 
 
 
 
 lO 
 
 rt ^,-1 (M 
 
 ir 
 
 ■* 
 
 rnCfl^lMrH 
 
 f-1 
 
 
 
 
 
 
 
 
 
 
 
 
 eo 
 
 M IMrtrH.- 
 
 I> 
 
 ^ 
 
 N 
 
 r-iCCrt^MN.- 
 
 O 
 
 OS 
 
 
 
 
 <N 
 
 «5 
 
 ^ 
 
 rH <Ncqt-cor-«: 
 
 CO 
 
 00 
 
 
 
 
 CO 
 
 CO 
 
 o 
 
 .-1 cfl to rti oi i>- CO oe 
 
 o 
 
 to 
 
 
 
 rHNOJ 
 
 OJ 
 
 1-1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 (U 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 •tJ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 03 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -*3 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 QQ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 V 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 -»^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 •a 
 
 
 
 D3 
 
 
 
 
 & 
 
 
 
 S- , 
 
 
 
 
 g 
 
 
 
 
 
 
 
 
 
 
 s. 
 
 
 
 
 »ri 
 
 
 
 COrttOrHSD^tDrHCC 
 
 
 
 
 (S 
 
 
 
 -* ^ CO CO (MM rH r-( 
 
 
 
 
 
 
 
 O; ^ - - . - 
 
 
 
 
 
 
 
 
 ,3 
 
 § = 
 
 
 
 i-<«OiHi©i-tCDi-(«D.- 
 
 
 
 
 
 Tt< CO CO N IN i-H rH 
 
 - I 
 
 <U V 
 
 
 
 
 S| 
 
 
 
 
 eaiB^g pa^iuQ ui 
 
 
 3 
 
 
 
 EU 
 
 39 
 
 A. 
 
 JO 
 
 J 
 
 aq 
 
 m 
 
 nji 
 
 •I 
 
 
 
 
 Z 
 
 II II 
 so 
 •§•§ 
 |.^ 
 oo 
 
 a a 
 
 cS es 
 
 nS 
 
 coo 
 
 qq 
 4HI 
 
 gS 
 o d 
 
 •s>. 
 
 S!0 
 
 tNO» 
 II II 
 b b 
 
 
 II ll=!>S 
 
 g a II "So 
 ■■SCD t. S B 
 
 Sl§^^2 
 
 g.Sgs 
 "P S S.2 
 S s "-S 
 fe Si*- S 
 
 V'«-<M d 
 o d 2 
 
 S S"3.2 
 
 3 P O O 
 
 !z;ZOO 
 
 19a 
 
CO 
 
 I 
 
 CO 
 
 I 
 
 bo 
 
 
 •d 
 
 s 
 
 CO 
 
 d 
 
 •I 
 
 I 
 
 o 
 U 
 
 CO 
 
 o 
 
 •p 
 
 O 
 
 
 n 
 
 
 B 
 
 d 
 
 o 
 
 •^ 
 
 o 
 O 
 
 Means 
 (Number of 
 
 Previous 
 Convictions) 
 
 ■OS t* i-t TiH TjH O *H 
 O O tH tH ^ (N CO 
 
 1-1 
 
 
 1 
 
 
 o 
 
 00 
 
 1-< 
 
 i-H CS lO CO i-l 1-H 
 
 1 
 I 
 
 o 
 
 g 
 
 O 
 
 Ah 
 
 O 
 M 
 
 a 
 m 
 S 
 
 D 
 
 12; 
 
 2 
 
 1—1 
 
 i-H 
 
 T-l 
 
 iH 
 
 1-1 
 
 1-1 
 
 CO 
 
 1-1 
 
 1—1 
 
 « 
 
 
 
 ^ 
 
 rH 
 
 tH 
 
 2 
 
 1—1 
 
 1—1 
 
 OS 
 
 iH 
 
 rH 
 
 00 
 
 1— ( 
 
 1— I 
 
 ^- 
 
 
 
 CO 
 
 1— t 1—1 
 
 (M 
 
 »o 
 
 I— 1 1—1 1—1 (N 
 
 lO 
 
 •<*< 
 
 lO 1—1 1—1 
 
 '-^ 
 
 to 
 
 1—1 
 
 CO 
 
 I— 1 1—1 CO 1—1 I— < 
 
 i> 
 
 <N 
 
 ^ 1-1 rH i-H CD -^ (M Tj< 
 
 o 
 
 l> 
 
 o 
 
 - 
 
 i-ii-HCOcoOir^ioco 
 
 i-H 
 
 n 
 
 CD 
 
 o 
 
 C^Ni-ICDOC0r-HM(N-^ 
 rH 1—1 CO 1—1 
 
 en 
 
 00 
 
 
 
 
 CO 
 
 3 
 
 Means 
 
 (Age at coming to 
 United States) 
 
 
 COt-fCDiHCDi-HCOi-ICOi-lCO 
 lOuO-^-^COCOCSKNi-lT-l 
 
 i-HCOi-tCOi-tCDi-HCOT-HCOrH 
 
 s»tB»g pajnifi o% Sonnoo ^e aSy 
 
 
 
 a 
 
 03 
 
 c3 
 
 cS 
 
 s 
 
 ^ 
 
 g 
 
 t-t 
 
 
 
 
 
 
 -tJ 
 
 ^ 
 
 ^ 
 
 N 
 
 CQ 
 
 PQ 
 
 
 g 
 
 o 
 
 a 
 g 
 
 ■H 
 
 -H 
 
 ^ 
 
 o 
 
 ■* 
 
 1 
 
 (N 
 
 Cd 
 
 
 
 OS 
 
 
 
 
 II 
 
 II 
 
 t/j 
 
 B- 
 
 B- 
 
 -s 
 
 i I 
 
 
 1 
 
 > 
 
 o 
 
 
 ;^ 
 
 3 
 
 .a 
 
 1 
 
 s^g 
 
 -B 
 
 
 IN05 
 
 6(1 
 
 a 
 
 1^ 
 
 II II 
 b b 
 
 .a 
 
 tt 
 
 a 
 
 s 
 
 -s 
 
 » 
 
 
 
 1 
 
 1 
 
 ^§8 
 
 
 
 s 
 
 •HOO 
 
 fl 
 
 .3 
 
 
 
 
 -(^ 
 
 
 g. 
 
 
 
 
 II II 
 
 O 
 
 i> 
 
 ^^ 
 
 S 
 
 1 
 
 
 
 0) CD 
 
 .2 
 
 iS 
 
 
 §S 
 
 -4^ 
 
 s. 
 
 3 
 o 
 
 
 i^ 
 
 •s 
 
 ■p 
 
 S'pa 
 
 193 
 
194 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 than those who come later in life. The correlation ratios are slightly 
 larger, .20±:.074 and .24 ± .072. Inspection of the table shows clearly 
 that those having fewer convictions were older when they came to this 
 country than those having many, and that the number of convictions 
 decreases with increase in age of coming to this country. 
 
 Since, however, both the variables in the above correlation — ^num- 
 ber of previous convictions and age at coming to this country — are 
 more or less closely related to the factor of present age, we have com- 
 puted the partial correlation coefficient for the number of previous con- 
 victions with the age at coming to the United States, for constant 
 present age.^^ This coefficient we find to be — .377 ±.066. In other 
 words, when we eliminate the effect of actual age a higher degree of 
 relationship is observed between frequency of conviction and an early 
 age of immigration. 
 
 Accordingly, from both the correlation between number of con- 
 victions and years in the United States, and that between number of 
 convictions and age at coming to this country, actual age being kept 
 constant, it would seem that there is a certain amount of relationship 
 between the length of time in this country or the age at coming to 
 this country and the degree of recidivism, with a tendency for those 
 who have been here longer or who came here younger to have more 
 convictions. 
 
 Several explanations for this relationship might be sought. In the 
 first place there is the very obvious possibility of inadequacy of in- 
 formation concerning convictions in other countries since these can not 
 be verified as can those of this country. In the second place there is 
 the probability that the same act might not constitute an offense against 
 the law in another country. This is particularly true of the offense of 
 prostitution which will not lead to arrest in most of the European 
 countries if carried on under more or less exactly defined conditions. 
 Of course, as has already been stated, a woman known to be a prosti- 
 tute will not be admitted to this country, but there is no reason to be- 
 lieve that this law would ever be enforced with complete success. 
 
 Another way of stating the facts might be, that the longer the immi- 
 grant stays in this country, the more likely she becomes to be con- 
 victed of offenses against the law and the nearer the percentage of de- 
 linquents among the foreign born approaches the much higher per- 
 
 "The data required for this computation are the following: 
 
 r (number convictions with age coming U. S.) = — .173 (170 cases) 
 r (number convictions with age) = .197 (169 cases) 
 
 r (age at coming to this country with age) = .610 (170 cases) 
 
NATIVITY IN RELATION TO DELINQUENCY 195 
 
 centage of delinquents among the native born. It is time that the most 
 of the foreign born delinquent women ;n this study have never lived 
 outside of New York City and that the effect of the crowded urban 
 conditions under which they are forced to live may serve to increase 
 the number of their convictions the longer they live here. 
 
 Whatever the explanation of the relation may be, however, the 
 fact of existence of correlations of this size between recidivism and 
 time in this country would be sufficient to interfere with our stressing 
 greatly the lesser degree of recidivism of the foreign group. The fact 
 that, they have not spent their whole lives in this country tends to 
 counteract the allowance to be made for their higher mean age. Ac- 
 cordingly we are left uncertain as to whether there is any valid differ- 
 ence between the\ native born delinquents and the foreign born with 
 respect to "criminalistic tendency" as indicated by number of con- 
 victions. ■-, The fact that there is a smaller percentage of the total for- 
 eign born population than of the native born who become delinquents 
 at all, as indicated by our group, would not be affected by the above 
 relationship between time in this country and number of convictions. 
 Accordingly it seems possible to say that, while there are fewer, rela- 
 tively, of the foreign born women than of the native born who become 
 delinquents, the fact of a significantly lower degree of recidivism 
 on the part of the foreign born delinquents can not be established by 
 our data. 
 
 It is of interest to note that in the group of foreign born women 
 studied, very few cases were found where the "crimes or misde- 
 meanors were committed in entire ignorance of the law because of 
 adherence to national customs which, innocent in a rural district, are 
 dangerous in the city and have therefore been prohibited."^* This 
 may be because we have not included in our study the cases where 
 fines were imposed, or the suspended sentences in the city magistrates' 
 day courts for violation of city ordinances, etc. 
 
 A further opinion often voiced is that much of the crime among the 
 foreign born is due to the different ethical standards in other countries 
 which make it difficult for an immigrant witfi ideals unlike ours to 
 adjust himself to our legal ways of thinking. In the group of 172 
 foreign born delincjuents whose cases we studied there were very few 
 where the ethical standards brought from another country seemed to 
 
 "Abbott, Grace. "Immigration and Crime" (Report of Committee "G" of 
 the Institute). Journal of the American Institute of Criminal Law and Crimi- 
 nology. Vol. VI, No. 4. Nov., 1915, p. 529. 
 
196 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 have any effect in the commission of the crimes for which the women 
 were convicted at the time of our study. There are a few striking 
 exceptions to this, particularly in the Auburn group, and therefore in 
 connection with the most serious offenses. "-^ 
 
 FIRST CONTACTS WITH THE LAW 
 
 The problem of the first contacts of the foreign born with the law 
 is to many people of as much or more importance than the subsequent 
 convictions. Some of the main factors of this first contact, the age 
 at first conviction, nature and seriousness of first offense and first 
 sentence will now be considered. 
 
 (a) Age at First Conviction 
 
 Table 57 shows that the range of age at first conviction is much 
 longer among the foreign than among either of the native born groups. 
 The percentage convicted under sixteen years of age is only 2.4 among 
 the foreign born group, while it is 6.3 in the native white and 4.2 in the 
 native colored. This smaller percentage of foreign bom juvenile de- 
 linquents is undoubtedly affected by the fact that 64.7 per cent of the 
 women did not come to this country until after they were over the 
 juvenile court age, and the possibility of convictions at early ages in 
 many foreign countries is not so great as in the United States, where 
 the Juvenile Court has been more fully developed. The age at first 
 conviction among the native groups will be more valuable if we 
 compare the mean ages at first conviction. Table 58 indicates that 
 there is very little difference between the means of the native white 
 and native colored, 24.5 as against 24.4, a difference which does not 
 appear demonstrably valid statistically. A comparison of the stand- 
 
 "A notable exception to this is the case of a Syrian girl, twenty-four years 
 
 of age, who was seduced by her lover, M , under promise of marriage. 
 
 When she became pregnant he refused to marry her. The subject's brother-in- 
 law, R , with whom she was living, offered to give M the furniture of 
 
 his own fiat and help him financially in other ways if he would marry the girl, 
 but he refused to do so. She became unable to stand the taunts and insults 
 
 that M subjected her to and so one morning after she had met him to make 
 
 a last appeal and he had refused her, she followed him into a restaurant and, 
 according to her own statement, shot him. The story is complicated by the 
 
 fact that the brother-in-law, R , was supposed to h^ve shot M , was 
 
 convicted of murder in the first degree and sentenced to the chair, while the 
 girl was given a short term in Auburn. On account of her insistence, how- 
 ever, that she had committed the act, because M had "taken her honoj-," 
 
 R 's sentence was commuted to life imprisonment. Though the girl had 
 
 lived in this country for eight years she followed the standard of the small 
 Syrian community in which she had grown up, that if a man "takes a girl's 
 honor and refuses to "make it good" she may take his life. 
 
NATIVITY IN RELATION TO DELINQUENCY 197 
 
 TABLE 67 
 
 AGE AT FIRST CONVICTION 
 
 Per Cent Distribution of Delinquent Women Classified By Nativity 
 
 and Color 
 
 Total Group 
 
 Age at First Conviction 
 
 8 to 
 
 Nattvitt and Color 
 
 12 " 
 
 16 " 
 
 16 " 
 
 20 " 
 
 20 " 
 
 24 " 
 
 ?4 " 
 
 28 " 
 
 28 " 
 
 32 " 
 
 3?, " 
 
 36 " 
 
 36 " 
 
 40 " 
 
 40 " 
 
 44 " 
 
 44 " 
 
 48 " 
 
 48 " 
 
 52 " 
 
 52 " 
 
 66 " 
 
 56 " 
 
 60 " 
 
 60 " 
 
 64 " 
 
 64 " 
 
 68 " 
 
 68 " 
 
 72 " 
 
 
 
 
 Total 
 
 
 
 Number of cases 
 
 Total 
 
 Foreign 
 
 Born 
 
 1.8 
 
 8.9 
 
 16.6 
 
 16.0 
 
 14.2 
 
 14.8 
 
 8.3 
 
 5.9 
 
 3.0 
 
 2.4 
 
 100.0 
 
 169 
 
 Total 
 Native 
 Born 
 
 . .8 
 
 5.0 
 
 27.0 
 
 22.8 
 
 17.5 
 
 9.2 
 
 6.5 
 
 3.9 
 
 4.7 
 
 1.6 
 
 1.3 
 
 .6 
 
 .3 
 
 100.0 
 
 382 
 
 Native 
 White 
 
 1.1 
 
 5.2 
 
 28.0 
 
 22.0 
 
 16.4 
 
 10.1 
 
 3.9 
 
 3.5 
 
 5.2 
 
 1.8 
 
 1.8 
 
 .7 
 
 .4 
 
 100.0 
 
 286 
 
 Native 
 Colored 
 
 4.2 
 
 24.0 
 
 26.0 
 
 20.8 
 
 6.3 
 
 10.4 
 
 6.2 
 
 3.1 
 
 1.0 
 
 100.0 
 
 96 
 
 Total 
 
 .7 
 
 4.0 
 
 21.4 
 
 20.9 
 
 17.1 
 
 10.7 
 
 8.4 
 
 5.3 
 
 6.1 
 
 2.0 
 
 1.6 
 
 1.1 
 
 .9 
 
 .6 
 
 .2 
 
 .2 
 
 100.0 
 
 561 
 
 TABLE 68 
 
 NATIVE WHITE AND NATIVE COLORED 
 
 Comparison of the Means and Standard Deviations of Age at First 
 Conviction For Native White and Native Colored 
 
 
 Native 
 White 
 
 Native 
 Colored 
 
 Diflference 
 
 d 
 
 Chances that real 
 
 diiTerence does 
 
 not exist are 1 
 
 in: 
 
 Mean 
 
 24.51 
 + .623 
 
 24.44 
 
 + .727 
 
 .08 
 
 .08 
 
 2 
 
 
 
 
 
 (X 
 
 8.84 
 + .370 
 
 7.12 
 ±.514 
 
 1.72 
 
 2.72 
 
 303 
 
 (Ta 
 
 
 
 
 Cases 
 
 286 
 
 96 
 
 
 
 
 
 
198 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 ard deviations, however, indicates that there is probably a valid differ- 
 ence in the dispersion of these two groups, with the native white having 
 a wider scattering. 
 
 Since there seems to be no valid difference in the central tendencies 
 of the age at first conviction between the native white and native 
 colored, we are justified in comparing the total native born and the 
 total foreign born, instead of comparing the native white and native 
 colored separately with the foreign born. 
 
 Table 59 shows that there is a valid difference in the mean age 
 at first conviction between the total native and total foreign born, 
 with a tendency for the foreign born to be convicted at a considerably 
 later age. A comparison of the standard deviations shows that there 
 is also a real difference between the dispersion of the two groups with 
 a wider scattering of cases in the total foreign bom group. 
 
 TABLE 59 
 
 TOTAL NATIVE BORN AND TOTAL FOREIGN BORN 
 
 Comparison of the Means and Standard Deviations of Age at First 
 Conviction For Total Native Bom and Total Foreign Bom 
 
 
 Total 
 Native Born 
 
 Total 
 Foreign Bom 
 
 Difference 
 
 d 
 
 Chances that real 
 
 difference doea 
 
 not ejdat are 
 
 in: 
 
 Mean 
 
 24.50 
 + .432 
 
 31.18 
 
 + .874 
 
 -6.68 
 
 6.85 
 
 8 
 
 
 
 
 
 g 
 
 8.44 
 + .305 
 
 11.37 
 + .618 
 
 -2.93 
 
 4.24 
 
 89,286 
 
 a<T 
 
 Cases 
 
 382 
 
 169 
 
 
 
 
 
 
 (&) Nature of First Offense 
 
 The nature of the first offense will be considered as the offenses 
 for the institutional groups were, in Chapter VI, first by the New York 
 City Police Department classification, and second, by felonies and mis- 
 demeanors. We must bear in mind here that a considerable number of 
 our group of delinquent women are first offenders and therefore that 
 in those cases the first offense applies to the present offense. 
 
 Table 60 shows, as all the other classifications have, that within 
 each group the highest percentage of first offenses are offenses against 
 chastity. The foreign born, however, show a much smaller percentage 
 
NATIVITY IN RELATION TO DELINQUENCY 199 
 
 TABLE 60 
 
 NATURE OF FIRST OFFENSE 
 
 Per Cent Distribution of Delinquent Women in Nativity and Color 
 
 Groups, Classified by New York City Police Department 
 
 Classification 
 
 
 
 Natovitt J 
 
 IND COLOB 
 
 
 
 Nature of Offense 
 
 Total 
 
 Foreign 
 
 Bom 
 
 Total 
 Native 
 Born 
 
 Native 
 White 
 
 Native 
 Colored 
 
 Total 
 
 Offenses against the Person 
 
 Offenses against Chastity 
 
 Offenses against Family and Chil- 
 dren 
 
 7.0 
 33.3 
 
 2.3 
 
 10.5 
 
 1.8 
 31.0 
 14.0 
 
 2.4 
 49.2 
 
 1.6 
 
 7.3 
 
 .3 
 13.4 
 25.9 
 
 1.8 
 50.4 
 
 2.1 
 
 8.8 
 
 .4 
 12.7 
 23.9 
 
 4.1 
 46.9 
 
 3.1 
 
 16.3 
 31.6 
 
 3.8 
 44.3 
 
 1 8 
 
 Offenses against Regulations for 
 PubHc Health, Safety and Pol- 
 icy 
 
 8 3 
 
 Offenses against Administration 
 
 of rrnVP-TTlTTlRnt , . 
 
 7 
 
 Offenses against Property Eights. 
 General CriminaHty 
 
 18.8 
 22 2 
 
 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 
 
 Number of cases 
 
 171 
 
 382 
 
 284 
 
 98 
 
 553 
 
 
 
 of offenses against chastity than do either the native white or native 
 colored. The second highest percentage of offenses falls in the offenses 
 against property rights for the foreign born group and in the general 
 criminality group for the native white and native colored. The other 
 divisions of offenses show no particularly striking differences, except in 
 offenses against the person where the foreign born have a slightly 
 higher percentage of cases. The important thing shown in the table is 
 that the. foreign born tend to commit for their first offenses a much 
 smaller percentage of offenses against chastity than do the other na- 
 tivity groups, and that they have nearly as large a percentage of their 
 first offenders committing offenses against property rights, while the 
 native white and native colored have their second largest percentage 
 of offenders in the general criminality group. This may partly be 
 
200 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 explained by the fact of the later age at first conviction among the 
 foreign bom. The general criminality offenses are very largely com- 
 posed of semi-juvenile offenses such as ungovernable or incorrigible 
 child, which apply only to those convicted at an early age. Since the 
 difference in the age at first conviction between the total native and 
 total foreign born seems to be valid, it is reasonable to suppose that 
 the later age at first conviction of the foreign born is one reason for the 
 few general criminality offenses within that group. 
 
 TABLE 61 
 
 NATURE OF FIRST OFFENSE 
 
 Per Cent Distribution of Delinquent Women in Nativity and Color 
 
 Groups Classified as Felons and Misdemeanants for the 
 
 First Offense 
 
 
 Nattvitt and Colob 
 
 
 Nature of Offense 
 
 Total 
 
 Foreign 
 
 Boin 
 
 Total 
 
 Native 
 Bom 
 
 Native 
 White 
 
 Native 
 Colored 
 
 Total 
 
 Felons 
 
 28.1 
 71.9 
 
 10.0 
 90.1 
 
 9.5 
 90.5 
 
 11.2 
 
 88.8 
 
 15.6 
 
 
 84.5 
 
 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 
 
 Number of cases 
 
 171 
 
 382 
 
 284 
 
 98 
 
 553 
 
 
 
 A further classification of the first offenses of the native and for- 
 eign born shows that the total foreign born have almost three times 
 as high a percentage of felons as the native white and about two and 
 a half times as many as the native colored. We should have expected 
 this from the high percentage of offenders against property rights in 
 Table 60, since this group of offenders has 54.9 per cent of its total, 
 felons.^^ 
 
 Again, this is contrary to the report of the Census on Prisoners and 
 Juvenile Delinquents in Institutions, that the foreign born "are slightly 
 more prone than the native white to commit minor offenses."" 
 
 " See Chapter VI. 
 
 "Op. cit., pp. 18-19, 40-41. 
 
NATIVITY IN RELATION TO DELINQUENCY 201 
 
 (c) First Sentence 
 
 Since the first offenses of the foreign born are in general more 
 serious than the first offenses of the native white or colored, we might 
 expect that the first sentence would be more severe. This we find to 
 be the case. 
 
 TABLK 62 
 
 FIRST SENTENCE 
 
 Per Cent Distribution of Delinquent Women Classified by Nativity and 
 
 Color 
 
 II 
 
 Nativity and Coloh 
 
 
 First Sentence 
 
 Total 
 
 Foreign 
 
 Born 
 
 Total 
 Native 
 Born 
 
 Native 
 White 
 
 Native 
 Colored 
 
 Total 
 
 Probation or Suspended Sentence . 
 Fine 
 
 19.2 
 1.2 
 
 1.7 
 20.3 
 
 42.4 
 15.1 
 
 29.6 
 4.2 
 
 1.8 
 26.5 
 
 33.5 
 6.6 
 
 34.8 
 
 4.6 
 
 2.1 
 30.7 
 
 21.6 , 
 6.3 
 
 14.3 
 3.1 
 1.0 
 
 11.2 
 
 67.3 
 3.1 
 
 26.4 
 3.2 
 
 Juvenile Institution 
 
 1.8 
 
 Reformatory Institution 
 
 23.9 
 
 Penitentiary, Workhouse or 
 County Jail 
 
 36.3 
 
 State Prison 
 
 8.4 
 
 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 
 
 
 172 
 
 385 
 
 287 
 
 98 
 
 557 
 
 
 
 In a comparison between the groups it is obvious first of all that 
 the foreign born have a much larger percentage of women whose first 
 sentence was state prison than have either of the other groups. The 
 total number of those committed to institutions for the first sentence is 
 79.5 per cent among the foreign born, while among the native white this 
 number is 60.7 per cent, and among the native colored 82.6 per cent. 
 The high percentage of colored committed to the Penitentiary, Work- 
 house group is, as has been explained previously, probably due to the 
 fact that so few colored women are put on probation and instead are 
 given a short Workhouse term. It seems obvious that the native white 
 
202 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 are more likely to be given a chance on probation or suspended sen- 
 tence than the foreign born. Though 71.9 per cent of the foreign born 
 were first convicted of misdemeanors, most of which might have been 
 given a probation sentence, only 19.2 per cent were given such a 
 sentence. 
 
 SUMMARY 
 
 In summarizing the relationship between nativity and delinquency, 
 there are several points which stand out clearly. The first of these is 
 that the foreign white group of delinquents studied from New York 
 City is in much smaller proportion to the total group of delinquents 
 than are the foreign white females over fifteen in New York City to 
 the general female population over fifteen. That is, one might con- 
 clude, the foreign white in New York City are less likely to come into 
 conflict with the law than either the native white or native colored. 
 
 Secondly, a consideration of present age shows that the foreign born 
 are markedly older than the native white or native colored. 
 
 The nature of the present offense is more serious among the 
 foreign born than among the native white or native colored, since the 
 one has 33.7 per cent of its offenders felons, while the other two have 
 13.9 per cent and 19.4 per cent felons, respectively. Though the 
 foreign born have their highest percentage in the offenses against chas- 
 tity group, as have the native white and native colored, this percentage 
 among the foreign born is considerably less than within the other 
 groups. On the other hand, the offenders against property rights have 
 their largest proportion in the foreign born group, though in each 
 nativity group this class of delinquents has second place. , 
 
 A comparison of the extent of delinquency among the three 
 nativity groups shows that there is a tendency for the native colored 
 to have more convictions than the native white, but that there is no 
 demonstrable difference between the mean number of convictions for 
 the native white and total foreign born groups. Due to the weight 
 of the native colored in the total native born group, there is a possible 
 slight tendency for the total native born to have a larger mean number 
 of convictions than the total foreign born. When allowance is made 
 for both present age and time in this country it seems clear that we 
 can not consider that any significant difference in recidivism between 
 these groups is established. 
 
 If we turn to the comparison of felons and misdemeanants within 
 the nativity groups, we find that there seems to be a valid difference 
 
NATIVITY IN RELATION TO DELINQUENCY 203 
 
 in the tendency for the total native born felons to have a larger mean 
 number of convictions than the total foreign born felons. There 
 seems to be no demonstrable difference in a like comparison between 
 the misdemeanants in these two nativity groups. 
 
 Passing from the number of convictions to the more general view 
 of the percentage of recidivists and first offenders, we find that only 
 47.7 per cent of the total foreign born are recidivists, while 55.1 per 
 cent of the native white arid 69.4 per cent of the native colored have 
 had previous convictions. That is, in addition to being less likely to 
 come into conflict with the law at all, there seems to be a slight ten- 
 dency for the foreign born who get into the courts not to repeat 
 their offenses, though lack of information concerning the early crimi- 
 nal record of the foreign group may operate in this connection also. 
 
 The partial correlation coefficient for the number of previous con- 
 victions with the age at coming to the United States, for constant 
 present age shows that there is a certain amount of relationship be- 
 tween the age at coming to this country and the degree of recidivism, 
 with a tendency for those who came here when younger to have more 
 convictions. This must, of course, be given weight in any compari- 
 son of number of convictions among the native and foreign born, 
 which shows that the foreign born have a smaller average number of 
 convictions than the native born. 
 
 It is also of interest to note again that the foreign born show a 
 valid difference in their tendency to be first convicted at a mean later 
 age than the total native born group, and that there seems to be no 
 demonstrable difference between the mean age at first conviction of the 
 native white and native colored. While the foreign born are first 
 convicted at a later age, they are also first convicted of more serious 
 offenses than the native white or native colored. The sentences fol- 
 lowing these more serious first offenses are, as we should expect, more 
 severe than in the groups of native white and native colored women 
 where the percentage of misdemeanants is much higher. It seems 
 true, however, even making allowance for the difference in serious- 
 ness of offense, that the native white are more likely to be given a 
 chance on probation or suspended sentence than the foreign born. 
 The figures would also indicate what is considered a well established 
 fact in New York City, that the negroes are seldom considered for 
 probation or suspended sentence because of the meager facilities in 
 the probation system for supervising negro women. 
 
 The interest in the nativity and color groups in any such study as 
 
204 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 this is not, of course, confined solely to the consideration of the crimi- 
 nal record. The other factors of the educational background, eco- 
 nomic efficiency, and history of sex irregularities of the foreign and 
 native born will be considered in later chapters dealing especially with 
 these subjects. 
 
CHAPTER IX 
 
 SIGNIFICANT FACTORS IN EARLY HOME CONDITIONS 
 AND FAMILY STATUS 
 
 TN attempting to understand any delinquent, and the possible factors 
 ^ which have been associated with her delinquency, one of the first 
 queries is about the kind of home in which she has grown up, whether 
 she has seen there such offenses as that which she has committed, or 
 whether the lack of training in the home has permitted petty offenses 
 to pass unnoticed which have finally culminated into contact with the 
 law. Closely associated with the more external survey of the home 
 conditions is the consideration of the woman's heredity, including the 
 defective physical and mental strains and other defects, such as ex- 
 cessive alcoholism and criminal records of the parents. Other factors 
 which determine to a certain extent the relative advantages or disad- 
 vantages of the early home surroundings are the number of children 
 in the family, particularly in the poor family; the subject's order in 
 her fraternity, that is, whether she was oldest or youngest in a large 
 family ; the parents' ages in relation to the ages of the women in whom 
 we are interested; the parents' nativity; and the ages of the women at 
 the time of either parent's death. These elements, of course, enter into 
 the estimate of the home conditions, but are presented separately for 
 comparison with similar data on delinquent groups. Since there are 
 no figures comparable to ours on the home conditions of the general 
 population, this most important comparison will have to be omitted 
 for the most part. 
 
 METHOD OF ESTIMATING HOME CONDITIONS DURING CHILDHOOD AND 
 
 ADOLESCENCE 
 
 In considering home conditions during childhood and adolescence, 
 the factor of age again arises to make the situation difficult. In many 
 of the younger cases, the home visited by the field worker was the 
 home in which the woman had lived for years, and was with little 
 change the home during childhood and adolescence. In these cases 
 
 205 
 
206 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 we have first-hand and reliable information concerning the home. 
 When a woman is forty or fifty years old, however, it becomes much 
 more difficult even to estimate what the earlier home conditions were. 
 It has been necessary, therefore, in the cases of these older women, 
 to obtain every possible record of social and charitable organizations, 
 and to learn all that we could from older friends and relatives in 
 order to have sufficient material on which to base an estimate. In many 
 cases, however, we have had only the woman's statement to go by. 
 We have thrown out of each group all of the subject's statements 
 which we felt were too unreliable for use. In this respect, the Work- 
 house is, again, least satisfactory. The real data of interest concern- 
 ing early background can not be obtained from a person unless there 
 is a spirit of cooperation and friendliness. As previously stated, the 
 heterogeneity of the group at the Workhouse, the hurry and con- 
 fusion, made a satisfactory interview impossible in many cases, and 
 the more personal details such as information regarding family and 
 the status of the home suffered most as a result of this. The reasons for 
 making this difficult are two. In the first place, many of the women 
 felt that it was unnecessary for them to be accurate in giving informa- 
 tion which they thought would be difficult to verify; also, many of 
 the older women were too confused and their memories were too poor 
 to give anything more than the most inaccurate and obviously incor- 
 rect statements. We have, therefore, omitted the estimate of home 
 conditions for the Workhouse group, throughout, except in Table 67 
 which shows how much of this information was based on the sub- 
 ject's statement only and how marked the tendency is for the Work- 
 house women to place themselves in the higher estimates of the home 
 conditions. We are convinced from our contact with the group of 
 women in the Workhouse and from the material we have been able 
 to verify, that such a percentage as 94.1 per cent in the three higher 
 classes of home conditions is absurd, and that this high percentage is 
 largely due to the tendency noted in other groups of facts about 
 the Workhouse women — the tendency to make out their past as much 
 better than it really was. 
 
 The material on home conditions, after the field worker's visits 
 and interviews had been recorded and summarized, was carefully 
 gone over by two persons, and with some of the institutional groups, a 
 third person, each one making an estimate of the three factors con- 
 sidered under home conditions, and a total estimate of home condi- 
 tions. The work was done by these persons independently of one 
 
FACTORS IN EARLY HOME CONDITIONS 207 
 
 another and finally re-checked by one person who treated in a uniform 
 manner all of the cases in which a difference had been noted. We 
 shall first take up the factors considered in making up the total estimate 
 of the home during childhood and adolescence, namely, (1) the eco- 
 nomic status, (2) the moral standards, (3) parental supervision. 
 Throughout, if a child spent the most of her time until adolescence in a 
 Home or Orphanage, the weight was given in making the estimate to 
 the conditions in the Orphanage. If she spent the most of her life in 
 her own home with a few years only in the Orphanage, the estimate 
 was based on conditions in her own home. 
 
 (a) Estimate of Economic Status in the Home 
 
 The estimate of the economic status in the home is divided into 
 five classes, which do not attempt to include all classes of society, but 
 those from a very low economic status through those of considerable 
 independent means. In general, the two lowest classes include those 
 who are, for the most part, behind their income and unable to get 
 along on what they earn. The third class which we have called medi- 
 ocre or fair is made up of the poor, but the normally self-sustaining, 
 while the two upper classes are those who are ahead of their income, 
 and comfortably situated. In making this estimate, the following 
 points were noted, not as arbitrary standards for determining the 
 classification, but merely as suggestions : 
 
 1. Very poor. Includes families which are dependents all or a 
 large part of the time and willing to accept aid repeatedly. May 
 also include families which have accepted less financial aid, but 
 have not had enough money to eke out a decent existence and 
 have lived in primitive fashion. 
 
 2. Poor. Includes families able to get along with difficulty, but re- 
 quiring only occasional financial aid; those with incomes obviously 
 below a living wage, though some may have unusual ability in 
 the management of the income. 
 
 3. Mediocre or Fair. Includes those who are normally self-sup- 
 porting but without saving or surplus. An occasional free treat- 
 ment in hospitals need not place a family in class 2, while an 
 industrial insurance policy, for example, need not raise it to 
 class 4. 
 
 4. Good. Includes those able to live comfortably on their incomes, — - 
 those who have money saved in the bank, or insurance. Those 
 
208 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 who have property heavily mortgaged so that keeping up the 
 interest is a great financial drain might lower the estimate to 
 class 3 or 2, as the case might be. 
 5. Very good. Includes families of considerable means, so that 
 luxuries are possible, such as extensive traveling, etc. 
 
 The distribution of these delinquent women by the estimate of the 
 economic status is shown in Table 63. From this table it is clear 
 that the same trend runs through each group, i.e., a small percent- 
 age both in the very poor and the very good classes, and a concen- 
 tration in the mediocre and poor classes. The largest percentage in 
 each group, except Bedford, falls in the fair or mediocre class with 
 the second largest percentage in the poor class. Bedford has the 
 largest per cent of any of the groups in class !■, and the largest per- 
 centage of her total in class 2. This institution undoubtedly is af- 
 fected by the element of having so many cases sent from the country 
 districts. Out of the nine cases, for instance, classed as very poor in 
 the economic status, seven cases or 77.7 per cent, were committed from 
 country districts, while only 11.9 per cent of the total group were 
 committed from courts outside of New York City. It is of interest 
 to note that Bedford, in addition to having the largest percentage of 
 women in the poorest classes, also has the only two cases where the 
 early economic conditions were considered very good. In one of these 
 cases, the girl was an only child. Her father, a wealthy man, was a 
 member of a large publishing firm, a member of the Chamber of' 
 Commerce and a prominent man in the community. They lived in the 
 most fashionable residence section in Brooklyn. The girl was given 
 every opportunity, was sent to a private school, given music lessons, 
 etc. The other case is that of a girl whose father was a forei^ cor- 
 respondent for a large newspaper, a mining expert, and a well-known 
 politician. She and her sister traveled all over the world with the 
 father, had private tutors for years, and finally went to an expensive 
 finishing school. In both of these cases the economic standards were 
 very good. The Magdalen has the largest percentage in the two best 
 classes, while the Penitentiary comes second. The Penitentiary, on 
 the other hand, has the smallest percentage in the two poorest classes, 
 with the Probation group having the next smallest percentage in these 
 two classes. For the total group, the table shows that 41.4 percent of 
 the cases come from homes where the economic status during child- 
 hood and adolescence was poor, and where the family was nearly al- 
 ways behind its income. Forty-five per cent of the cases were from 
 
FACTORS IN EARLY HOME CONDITIONS 
 
 209 
 
 families perhaps normally self-supporting but poor, while only 13.6 
 per cent come from comfortable homes. There are no figures which 
 will make a comparison of our group with the total population pos- 
 sible, so that in our discussion of the home conditions of this group 
 we can hope to give little more than a picture of the conditions as 
 we found them. 
 
 TABLE 63 
 
 ESTIMATE OF ECONOMIC STATUS IN THE HOME DURING 
 CHILDHOOD AND ADOLESCENCE 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups 
 
 (Workhouse Omitted) 
 
 
 iNSTITnTIONAL GROUPS 
 
 
 Estimate of Economic Status 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 Peni- 
 tentiary 
 
 Probation 
 
 Total 
 
 Very poor 
 
 Poor 
 
 Fair 
 
 Good 
 
 Very good 
 
 9.8 
 
 50.1 
 
 31.6 
 
 6.5 
 
 2.2 
 
 9.0 
 38.4 
 43.5 
 
 9.0 
 
 5.7 
 32.9 
 42.9 
 18.6 
 
 2.2 
 22.7 
 
 57.2 
 18.4 
 
 2.3 
 34.5 
 49.5 
 13.8 
 
 5.7 
 
 35.7 
 
 45.0 
 
 13.1 
 
 5 
 
 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cases 
 
 92 
 
 78 
 
 70 
 
 93 
 
 87 
 
 420 
 
 There are, however, certain comparisons from other studies of 
 delinquency which may be of importance here. In the consideration 
 of the economic status we may note particularly the study of children in 
 the Juvenile Court in Chicago, made by Miss Breckinridge and Miss 
 Abbott in 1912.^ In discusing the "Poor Child" they have made four 
 large economic groups indicating the status in the homes of the chil- 
 dren. Their classes are: 
 
 Group I. Very poor families. Not normally self-sustaining. 
 Group II. Poor, but normally self-sustaining. 
 Group III. Fairly comfortable. 
 Group IV. Comfortable. 
 
 'Breckinridge, Sophonisba P., and Abbott, Edith. "The Delinquent Child 
 and the Home." Russell Sage Foundation, 1912, p. 72. 
 
210 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 Roughly, the cases they have included in Group I would corre- 
 spond to those we have included in classes 1 and 2. Those in Group II 
 would include our class 3, Group III would include class 4, and Group 
 IV would include class 5. If we rearrange our classes in this way, 
 we find that the rough percentage comparison is as follows : 
 
 Children in Juvenile Our Delinquent Group 
 Court in Chicago 
 Boys Girls 
 
 Group I 38.2 68.8 41.4 
 
 Group II 37.9 21.0 45.0 
 
 Group III 21.2 7.6 13.1 
 
 Group IV 1.7 1.3 .5 
 
 No Home 1.0 1.3 
 
 It is interesting to note that the boys in the Juvenile Court appear 
 to come from better homes than the girls and that our group is be- 
 tween the two. Though no exact comparison is possible, we may ob- 
 serve that in each group the preponderance of cases is in the two lower 
 groups of the very poor, and the poor but normally self-supporting. 
 The study of the children in the Juvenile Court, of course, gave the 
 worker an opportunity to visit the childhood home and make an esti- 
 mate on what was actually observed at that time. Because our esti- 
 mate necessarily had to be based on a matter of record and second- 
 hand information for most of the older women, it is interesting to 
 see that the same tendency is shown, for the delinquents in both groups 
 to come from the poorer economic classes. 
 
 (&) Estimate of Moral Standards in the Home 
 A second general factor which was observed in a study of the 
 home conditions was the moral standards in the home during child- 
 hood and adolescence. This again has been estimated, and is grouped 
 into five classes, as follows : 
 
 1. Very poor. Here there were included the most flagrant cases of 
 low moral standards, such as excessive alcoholism, sex offenses, 
 criminal records, etc., in the immediate family in which the girl 
 was brought up. Extreme overcrowding, with consequent lack 
 of privacy, an extremely bad neighborhood and environment should 
 be considered in connection with the standard of the immediate 
 family. 
 
 2. Poor. Includes families with low moral standards but appreciably 
 better than class 1. This class may include cases where there may 
 not be definite active immorality of any kind in the immediate 
 family, but where there is absence of standards in parents or where 
 
FACTORS IN EARLY HOME CONDITIONS 211 
 
 parents have so little force that they are not able to establish decent 
 standards. An example of this occurs in one of the Bedford 
 cases where the parents did not themselves steal, drink or commit 
 illegal acts, and yet they countenanced petty stealing in the girl, 
 and were willing to accept the small things she had stolen and 
 brought home. 
 
 3. Fair or mediocre. Includes cases where, for the most part, the 
 members of the immediate family have a reputable standing in 
 the community. A criminal record for one member of the family, 
 if it does not directly affect the girl, or moderate alcoholism need 
 not necessarily lower a case to class 2. 
 
 4. Good. Includes cases where all of family have good standards, and 
 where there is every chance that the girl should have good moral 
 standards, in so far as the character of her surroundings has 
 affected her. 
 
 5. Very Good. Includes cases where there are unusually high stand- 
 ards in the parents and in the immediate environment. 
 
 With this classification, we see in Table 64 that the distribution 
 of the estimate of moral standards is, in general, very like that of the 
 previous estimate. In each institutional group there is a larger per- 
 centage of the cases in the lowest class than in the lowest economic 
 class, but there are only two cases, and those in the Penitentiary, in the 
 highest class. Bedford has the highest percentage in the poor class, 
 while the other institutional groups have the highest percentage of 
 their cases in the fair or mediocre group. The Penitentiary has the 
 smallest percentage below mediocre, and is followed in order of in- 
 creasing percentages for this group by Probation, Auburn, Magdalen 
 and Bedford. The total group shows a larger percentage in the two 
 highest classes of moral standards than is found in the two highest 
 estimates of economic status. The factor of the number of country 
 cases committed to Bedford for petty offenses undoubtedly helps, 
 again, to swell th^ percentage of cases falling in the class of very 
 low moral standards. Our observation has been that more flagrant 
 offenses against moral standards will be tolerated in rural than in 
 urban communities. Very likely several of the Bedford cases, which 
 will be cited later, might have been spared contact with the law if they 
 had been removed from the extremely immoral surroundings when 
 they were young. Another factor which may account for Bedford's 
 larger representation in these lower groups is that we know the women 
 in Bedford much more thoroughly than the women in any other group, 
 and it is true, we are convinced from observation, that the more 
 
212 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 64 
 
 ESTIMATE OF MORAL STANDARDS IN THE HOME DURING 
 CHILDHOOD AND ADOLESCENCE 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups 
 (Workhouse Omitted) 
 
 
 iNBTITtrTIONAL GkOUPS 
 
 
 Estimate of-Moral Standards 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 Peni- 
 tentiary 
 
 Probation 
 
 Total 
 
 Very poor 
 
 14.2 
 
 14.1 
 
 11.4 
 
 7.6 
 
 6.8 
 
 lO.S 
 
 Poor 
 
 44.7 
 
 16.6 
 
 20.0 
 
 8.6 
 
 21.9 
 
 22.6 
 
 Fair 
 
 31.6 
 
 46.1 
 
 47.2 
 
 51.8 
 
 57.5 
 
 46.7 
 
 Qood 
 
 9.8 
 
 23.0 
 
 21.5 
 
 30.1 
 
 15.0 
 
 20.0 
 
 Very good 
 
 
 
 
 2.2 
 
 .... 
 
 .2 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cases 
 
 92 
 
 ■ 78 
 
 70 
 
 93 
 
 87 
 
 420 
 
 thorough the investigation, the less advantageous to the subject is the 
 trend away from the information which she has given. 
 
 (c) Estimate of Parental Supervision in the Home 
 
 The third factor, and one of the most important in making an 
 estimate of the home conditions is the element of parental supervision 
 in the home. A home may be economically very comfortable and the 
 moral standards may be fair, yet the parents may be so harsh, or so 
 lacking in force, understanding and sympathetic insight, that their 
 effect on the children is very bad. We have not considered separately 
 in this connection the markedly psychopathic and rnentally abnormal 
 girls who may have been brought up in a home where the parents 
 had a fairly intelligent understanding, but were unable to manage the 
 girl because she required the supervision of some one who had much 
 more skill and experience in handling abnormal cases. In such a case, 
 the parental supervision would be classed as fair, because our interest 
 in this connection is in the status of the home, not in the idiosyncrasies 
 of the individual. If the difficulty of adjustment lies in the individual 
 
FACTORS IN EARLY HOME CONDITIONS 213 
 
 herself, that will be considered in another connection. For determining 
 this estimate which has been very carefully made, we have considered 
 the following suggestions : 
 
 1. Veryypoor. Includes cases where one or both parents are dead, 
 and the guardian makes no effort to look after the child. Extreme 
 cases where one or both parents are living but do not attempt 
 supervision of any kind fall here, and also cases where parents are 
 very unintelligent in the control of the child. 
 
 2. Poor. Includes less extreme cases than 1. May include cases 
 where the mother has to go out to work and no one is left at home 
 to look after the girl, or where there may be very strict super- 
 vision, but so little understanding that the supervision does not 
 count for anything. 
 
 3. Fair. Includes cases where the supervision and intelligent control 
 in the home are sufficiently good that a normal child stands a 
 fair chance of being guided by it. As noted before, parents who 
 are fairly intelligent in their control of a normal child but who can 
 not manage an abnormal child will be placed in -this group, since 
 we are desirous of showing in this place the kind of home from 
 which the most of our delinquents come, and not the make-up of 
 the child. 
 
 4. Good. Includes cases where children have been carefully and in- 
 telligently supervised, and ought certainly, if normal, to have been 
 guided by their supervision. 
 
 5. Very good. Includes cases where supervision has been excellent 
 and shows unusual insight into the nature of the child. 
 
 Table 65 shows how the women in institutional groups are cHs- 
 tributed by this classification. Except for the Penitentiary, the per- 
 centage in the very poor class is considerably larger than in either the 
 estimate of economic status or moral standards. In each institutional 
 group the percentage in the poor class is also larger than this same 
 class in any of the other estimates considered. Again, there are only 
 2 cases in the very good class, and those from the Penitentiary. 
 These 2 cases which were also in the highest class of moral standards 
 will be considered later. As a whole, it is interesting to note that in 
 the bulk of the cases the estimate of parental supervision falls into 
 much lower classes than either the moral or economic factors, and 
 that only 3.1 per cent had good or unusually good supervision at 
 home, while only 35.0 per cent were intelligently enough controlled 
 so that they stood a fair chance of developing normally. 
 
214 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 {d) Total Estimate of the Home 
 
 If we compare these three general factors which have gone into 
 making an estimate of the home conditions, we may see that the ele- 
 ment of very poor or poor parental supervision has affected the largest 
 number of cases, while very poor or poor economic conditions have 
 affected the next largest number. In considering the childhood home, 
 as a whole, these three elements vary in weight in different cases. A 
 child well supervised and living in a home with good moral standards 
 may still be too poor economically to be thought of as growing up in 
 a fair home, while a home with good moral standards and a fair 
 amount of money, may be so lacking in control of the children as to 
 lose all the effect of the other two elements. 
 
 In making up the total estimate of the home conditions during 
 childhood and adolescence, the three estimates just discussed, the 
 economic status, moral standards and parental supervision, were con- 
 sidered, and from a careful survey of these three factors and their 
 
 TABLE 66 
 
 ESTIMATE OF PARENTAL SUPERVISION IN THE HOME 
 DURING CHILDHOOD AND ADOLESCENCE 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups 
 (Workhouse Omitted) 
 
 
 Institutional Ghoups 
 
 
 Estimate of 
 Parental Supervision 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 Peni- 
 tentiary 
 
 Probation 
 
 Total 
 
 Very poor 
 
 Poor 
 
 Fair 
 
 Good 
 
 Very good 
 
 17.4 
 
 66.7 
 
 25.1 
 
 1.1 
 
 24.3 
 
 39.7 
 
 33.3 
 
 2.6 
 
 15.7 
 
 52.9 
 
 28.6 
 
 2.9 
 
 6.4 
 
 36.7 
 
 61.8 
 
 4.3 
 
 2.2 
 
 18.4 
 
 44.9 
 
 34.5 
 
 2.3 
 
 16.0 
 
 46.0 
 
 35.0 
 
 2.9 
 
 2 
 
 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cases 
 
 92 
 
 78 
 
 70 
 
 93 
 
 87 
 
 420 
 
FACTORS IN EARLY HOME CONDITIONS 215 
 
 relative weights in the home, a classification of 5 groups was made, 
 which we have called : 
 
 1. Very poor. Includes cases where all of the elements entering into 
 the home life are bad, and where there is every chance that the 
 child has been subjected to surroundings too bad to make her a 
 good citizen unless she is a very unusual child, and later able to 
 get away from her early associations. 
 
 2. Poor. Includes homes which, in general, are better than those 
 in the previous class. A home, fair in certain respects, may be 
 so pulled down by one very bad element that the estimate falls 
 here instead of in class 3. 
 
 3. Fair. Includes homes which are, for the most part, not markedly 
 bad or markedly good in any one respect, though there may be 
 cases here where the moral standards, for instance, are excellent, 
 while the other factors are only mediocre and therefore keep the 
 estimate in class 3. 
 
 4. Good. Includes homes which can offer good economic and moral 
 standards, and where there is intelligent enough understanding 
 of the child to make the other factors count for what they should. 
 
 5. Very good. Homes of unusual standards and offering excellent 
 opportunities. It is not necessary that each of the previous 3 
 estimates should have been in the highest group in order to have 
 a case fall here. 
 
 Table 66 shows the distribution of our cases by this method of 
 estimation, and makes it clear, first of all, that the total estimate seems 
 to approximate the estimate of economic status more than any other 
 factor. From this it would seem that the economic factor determines 
 to a very large extent whether or not the home seems a desirable 
 place to bring up children. The table shows in each institutional group 
 the concentration of cases in the poor and mediocre groups, with very 
 few cases in the bettar classes. The Penitentiary group shows the 
 smallest percentage falling in the two poorest classes while the Mag- 
 dalen has the next larger, and is followed by Probation, Auburn and 
 Bedford in order of increasing percentages. Of the total group of 
 women, 8.3 per cent came from homes which were very bad in every 
 way and which were not fit places for children to live, 38.1 per cent 
 came from poor homes which had at least one and perhaps more bad 
 elements, 47.1 per cent came from fair or mediocre homes, while only 
 6.5 per cent came from homes which were good or excellent. 
 
 Chart XIV shows graphically the figure given in Table 66, and 
 indicates that the two upper classes contribute the smallest percentages 
 
216 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 66 
 
 ESTIMATE OF HOME CONDITIONS DURING CHILDHOOD AND 
 
 ADOLESCENCE 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups 
 (Workhouse Omitted) 
 
 
 Institutional Gboups 
 
 
 Estimate of Home Conditions 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 Peni- 
 .tentiary 
 
 Probation 
 
 Total 
 
 Very poor 
 
 Poor 
 
 Fair 
 
 Good 
 
 10.9 
 51.1 
 31.5 
 
 6.5 
 
 12.8 
 42.3 
 37.2 
 
 7.7 
 
 10.0 
 
 38.5 
 
 47.1 
 
 4.3 
 
 5.4 
 
 22.6 
 
 63.4 
 
 6.4 
 
 2.1 
 
 3.4 
 36.8 
 
 55.2 
 4.6 
 
 8.3 
 38.1 
 47.1 
 
 6.0 
 
 Very good 
 
 .5 
 
 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cases 
 
 92 
 
 78 
 
 70 
 
 93 
 
 87 
 
 420 
 
 of delinquent women; while in each chart the second and third classes 
 show the largest numbers. 
 
 (e) Comparison Between Verified and Unverified Data on Home 
 
 Conditions 
 
 It may be advisable to give here the figures which show how the 
 estimates were made up on the various groups, and how the subject's 
 statement differs from the verified data. The estimates previously 
 presented are on combined data ; that is, in addition to all of the veri- 
 fied statements which enabled us to determine the kind of home, the 
 subject's statement also was used where we felt such information was 
 reliable. Table 67 shows that the percentage of verified material on 
 home conditions varies in the different institutional groups. In Bedford 
 none of the subject's statements alone have been used and in Auburn 
 and the Magdalen the estimates were based on verified data in all 
 but 15.4 per cent and 13.9 per cent of the cases, respectively. The 
 Penitentiary estimate is based on the subject's statement only in 
 31.1 per cent of the cases, while the Probation group has 36.8 per 
 

 1 — ■ 5 
 
 
 
 8 
 
 
 9 
 
 8 
 
 9 -' 
 
 il 
 
 8 
 
 a 
 
 
 
 
 
 
 
 
 
 
 L 
 
 1 
 
 
 ■ 
 
 
 • 
 
 
 
 8 
 
 g i 
 § i 
 
 00 9 
 
 8 
 
 
 s 
 
 8 
 > 
 
 t i 
 
 B 
 
 
 
 ■ 
 
 
 
 
 
 
 1 
 
 i 
 
 . 
 
 .1 
 
 ■ - 
 
 
 i 
 
 
 
 1 " 
 
 n 
 
 < 
 
 
 B 
 8 
 
 it 
 
 8 
 
 
 
 
 
 
 t}7 
 
 i\t 
 
 
 i i 
 i i ! i ? 
 
 
 hill 
 
 •a 
 a 
 o 
 
 U 
 
 6 
 o 
 
 K 
 
 3 
 O 
 
 
 (U ^ 
 
 W s 
 
 
 217 
 
218 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 67 
 
 ESTIMATE OF HOME CONDITIONS DURING CHILDHOOD AND 
 
 ADOLESCENCE 
 
 Per Cent Distribution by Institutional Groups Showing Subject's 
 Statement Only, Verified Data Only, and Combined Data 
 
 
 INSTIT0TIONAL GsOUP 
 
 
 Estimate of Home Conditions 
 
 Bed- 
 ford 
 
 Auburn 
 
 Mag- 
 dalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Proba^ 
 tion 
 
 Total 
 
 Subject's Statement Only 
 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Very poor and poor 
 
 Fair, good and very good. . . . 
 
 
 50.0 
 60.0 
 
 30.0 
 70.0 
 
 17.2 
 
 82.8 
 
 6.9 
 94.1 
 
 25.0 
 75.0 
 
 17.2 
 82.8 
 
 Number of cases 
 
 
 12 
 
 10 
 
 29 
 
 68 
 
 32 
 
 151 
 
 
 
 
 Verified Data Only 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100 
 
 
 
 Very poor and poor 
 
 Fair, good and very good. . . . 
 
 62.0 
 38.0 
 
 56.1 
 43.9 
 
 51.7 
 48.3 
 
 32,8 
 67.2 
 
 44.0 
 56.0 
 
 49.1 
 50.9 
 
 50.8 
 49.2 
 
 
 92 
 
 66 
 
 60 
 
 64 
 
 25 
 
 66 
 
 362 
 
 
 
 Combined Data 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100 
 
 
 
 
 62.0 
 38.0 
 
 56.1 
 44.9 
 
 48.6 
 61.4 
 
 28.0 
 72.0 
 
 16.1 
 83.9 
 
 40.2 
 69.8 
 
 40 9 
 
 Fair, good and very good. . . . 
 
 59.1 
 
 Number of ca^es 
 
 92 
 
 78 
 
 70 
 
 93 
 
 93 
 
 87 
 
 513 
 
 
 
 cent and the Workhouse 73.1 per cent of women whose statements of 
 early home conditions could not be verified. To show how the per- 
 centages of cases in the two poorest classes and those in the three best 
 classes vary between the subject's statement and the verified data, we 
 may note in Table 67 that the two lowest classes have only 5.9 per cent 
 in the Workhouse by the subject's statement alone, based on 68 cases, 
 while there are 44.0 per cent in these two classes if we consider the 25 
 cases in the Workhouse group whose early home conditions we were 
 able to verify. By combining the 68 statements and 25 verified cases, 
 the percentage in the two lowest classes becomes 16.1 per cent, which is 
 obviously much too low in comparison with the trend shown in the 
 verified Workhouse data. For this reason the Workhouse group 
 
FACTORS IN EARLY HOME CONDITIONS 219 
 
 has been omitted in the discussion of home conditions. In each 
 institutional group the percentage of the poorer classes is smaller when 
 we take the subject's statement alone, than wh^ we use the verified 
 data alone, but in no group except the Workhouse is the difference be- 
 tween the combined data and the verified data so striking as to rnake 
 the combined data of little value. 
 
 (/) Presentation of Cases Representing Various Classes of Home 
 
 Conditions 
 
 Because the group of women in Bedford has been most carefully 
 studied and because we have come to know these families most thor- 
 oughly, we shall present a number of the cases in this group and two 
 cases in the Penitentiary, showing the various types of homes included 
 in the five classes of home conditions. 
 
 The two cases we shall give, representing class 1, are in many 
 ways alike, but show varying degrees of immorality, poor economic 
 standards, and parental supervision. The method of treatment has 
 also been different in these two cases. 
 
 1. Mary M , 16 years and 2 months old at the time of her com- 
 mitment to Bedford, was born in a small country town. Her 
 father died one month before Mary was born; her mother, who 
 had a questionable reputation, went to live with another man soon 
 after the child's birth, and Mary and her sister were taken by the 
 maternal grandmother, with whom Mary has lived nearly all of 
 her life. The home conditions there have been very poor. The 
 grandfather for years kept the "Penny Bridge" for which he re- 
 ceived a very small pittance. Living in the same house were two 
 of Mary's uncles, both of them shiftless, good-for-nothing and never 
 contributing anything to the family support. As a result, the 
 family often had to be aided by the town, in order to eke out 
 the most miserable kind of an existence. The school record 
 shows that Mary and her sister came to school very poorly clad 
 and having had nothing to eat. Through the school, a complaint 
 was made to a charitable organization who gave the family 
 financial aid, but did not remove the children from the house. The 
 house which the field worker visited was a small frame house con- 
 sisting of 3 rooms with a rental of $3 a month. In this house, which 
 was filthy and malodorous at the time of the visit, the grandmother 
 and two uncles lived, but occasionally an aunt was there with 
 one of her children. The grandmother said that there was a 
 curtain in the upstairs room between the bed in which she and 
 
220 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 Mary slept and her sons' bed. The moral standards have been, 
 as the judge said, "as low as possible." The grandmother is 
 notorious as an immoral woman, and at one time left home to 
 live with a negro workman. Mary herself is supposed to have 
 been immoral with her uncle who is undoubtedly mentally defec- 
 tive, and who has a long jail record for minor offenses. In 
 this home, each of the three factors of economic status, moral 
 standards and parental supervision was placed in the lowest 
 class. These poor home conditions were known by the girl from 
 shortly after her birth until the time of commitment to Bedford. She 
 was convicted of prostitution but had, with little doubt, been promis- 
 cuous sexually long before, and is thought to have prostituted" for 
 money. This she had seen in her home, and in addition, the fact that 
 no moral standards had been inculcated in her, her need of money 
 and the willingness on the part of her family to have her prosti- 
 tute, were all factors making this easy method of earning money 
 the most desirable and profitable thing for her to do. 
 
 2. Another case which we grouped in the lowest class is that of 
 
 Jane H . At the time of her commitment to Bedford for 
 
 "Exposure of person" she was twenty years and eight months of age. 
 She was born in a small village in New York State and lived 
 there with her people until she was about eight years old when she 
 was placed in an Orphanage. Three years later she was returned 
 to her people, but after a time was again brought back to the 
 Orphanage where she remained until she was eighteen. Our infor- 
 mation is obtained from the very complete records of the Orphan- 
 age, from the Humane Society which handled the case when the 
 children were taken from the home, and from the older children 
 in the family who were able to remember something of the 
 early home conditions. There were eight children in the family and 
 
 they all lived in the depths of the backwoods of R County, 
 
 New York. Both the mother and father were very heavy drinkers 
 and used to "carouse and act shameful." Three of the children in 
 the family were illegitimate, and the father as well as the mother 
 was known to be grossly immoral. When the children were 
 brought to the Orphanage they seemed more like wild animals 
 than human beings; they would run and hide when they saw 
 any one coming. They had. had no training of any kind. The 
 house was "little better than a pig-pen"; the children were often 
 left there at night with no fire, and even by huddling together 
 they could not keep warm. There was never enough to eat- 
 usually only dry bread in the house. The children were not sent 
 to school and "ran wild the most of the time." In this family 
 
FACTORS IN EARLY HOME CONDITIONS 221 
 
 the conditions of living were probably even lower than in the 
 case just cited, but the children were found before they had spent 
 their entire childhood in the home, and were given fair ad- 
 vantages during the time they were in the Orphanage. The 
 effect on the girl under consideration, however, since she spent 
 altogether over nine years with her parents, seemed important 
 enough to estimate the home conditions as very poor. 
 
 For Class 2 we shall present two cases, one from New York and 
 one from a small village. They both represent conditions appre- 
 ciably better than those just given. 
 
 1. Alice B was committed to Bedford at the age of twenty years, 
 
 seven months for contracting an infectious disease in the practise of 
 debauchery. She was born in Brooklyn and lived there the most 
 of her life. She lived with her parents until she was nine, when 
 her mother died, and she was taken by her maternal grandmother. 
 While the mother was living, the family had a hard time to get 
 along because the father spent so much money for drink. He 
 was not faithful to his wife and treated her very badly. The 
 grandmother's home, to which Alice went when she was nine, is in a 
 fairly good residential district and is a good house which the 
 grandmother owns. The father gave no money for the children's 
 support, however, and the grandmother would not spend her money 
 in caring for them. Various complaints were made to the Children's 
 Society concerning this home and alleging that the children did 
 not have proper guardianship. The school reported that "the 
 grandmother is a miser. She goes around picking up wood, etc. 
 The children used to sleep on the floor, on piles of filthy rags in- 
 fested with vermin with only newspapers over them in winter. 
 Their clothing was filthy and ragged when they came to school, 
 which was only irregularly." The Children's Society also had 
 reports of Alice having to "get beer" for the grandmother, which 
 necessitated her being on the streets entirely too much. Upon 
 investigation by the Society, it seemed that these reports had 
 been exaggerated but that there undoubtedly was some truth in 
 them, and that there had not been proper supervision. These con- 
 ditions were improved under the supervision of the Children's 
 Society. In this home, the economic standards, except for the 
 time before the intervention of the Children's Society, were not 
 so low as in the other cases, and in the grandmother's home 
 there was a certain amount of ready money which she later con- 
 sented to use for the children. Under the supervision of the 
 Children's Society, Alice and her sister were looked after more 
 
222 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 carefully. The moral standards while low in the father's home 
 were not so essentially a part of the home as in the cases of 
 Mary M and Jane H . 
 
 2. Jennie B , committed to Bedford for violating the liquor tax 
 
 law, was born in a small village 25 years and 6 months before her 
 conviction. Her mother died when she was three years old and 
 she went to live with her maternal grandparents where she re- 
 mained for three years. She then went to live with her stepmother 
 who did the best she could to provide for the family and was 
 a very fine woman. The father was so alcoholic and abusive, how- 
 ever, that the stppmother left him when Jennie was eleven. After 
 this, Jennie kept house for her grandfather, father, uncle and 
 brother. The home conditions were bad for her during this 
 period because she had little supervision and her father and uncle 
 were drinking men. She used to go to the saloon for her father 
 at that time, and the men who frequented this place would give 
 her nickels to sit on the bar and sing for them. When she was 
 fourteen she went out to do housework. This girl, aside from 
 her father's drinking, had a decent home until she was 3 and from 
 the time she was 6 until she was 11. After that, and at her most 
 impressionable years, she was thrown on her own resources, had 
 no one to look out for her, and there was no woman in the 
 house. As a whole, we feel that each of the factors in the home, — 
 the economic and' moral standards, and the parental supervision 
 belong in the poor class, and that the total estimate of the home 
 would fall in this class. 
 
 For Class 3, which represents cases in appreciably better home 
 surroundings, we shall give two cases. 
 
 1. Sarah S was committed to Bedford when seventeen years, six 
 
 months of age for associating with dissolute persons and being in 
 danger of becoming morally depraved. She was born in New York 
 City and lived there all of her life. Her father has for years worked 
 in a shoe factory and in the rush seasons earns from $14 to $18 
 a week though he only earns about $6 a week during the slack 
 seasons. The oldest boy, who is not married, earns $14 a week 
 and helps out with the family expenses. There are five children in 
 the fraternity including Sarah. On the whole, because of the 
 mother's good managing ability and the fact that the father has 
 always worked steadily, though at times he earned little, the 
 economic standards may be classed as mediocre. The family 
 has never needed to accept charitable aid, and the father has 
 
FACTORS IN EARLY HOME CONDITIONS 223 
 
 been able to keep up his membership in a Jewish benefit lodge. 
 There seems to be a good family spirit and the family have 
 good moral standards. Sarah is the only one in the family who 
 has ever been in court. The parents are quite foreign and have 
 not quickly adapted themselves to the customs in this country. 
 As a result, a large part of the supervision has fallen on the 
 oldest boy who has exerted his influence rather unpleasantly at 
 times. On the whole, however, the home has been fair and the 
 girl had a chance of making good so far as the influence of her 
 home conditions was concerned. In our estimates of the factors 
 in the home, we have classed the estimate of the moral stand- 
 ards as good and all of the other estimates as fair. 
 
 2. Kate M was sent to Bedford when eighteen years and five 
 
 months old for prostitution. She was born in a village in Massachu- 
 setts and lived there and in a small city in New York state most of 
 her life. Her father died shortly before her birth. He was an 
 erratic, unstable sort of man, shiftless and quite lazy, so that 
 there was no money left after his death, and his wife and children 
 had to live with his father. A few years after his death, the 
 mother remarried and took her children to New York state with 
 her. The home with the Stepfather was fairly good. He worked 
 steadily and tried to provide well. When Kate was nine, however, 
 her grandparents in Massachusetts offered to take care of her 
 and so she went back to them, remaining until she was fourteen. 
 The grandfather owned several houses and provided well for the 
 child. She went to school regularly, finishing grammar school, 
 and* the grandfather was anxious to send her through college. 
 They were very respectable old people and had a good standing 
 in the community, but were undoubtedly too old to exert proper 
 authority, so that Kate was allowed to have her own way in 
 everything. When fourteen years old, she decided that she wanted 
 to go back to her mother, and despite her grandparents' protesta- 
 tions, she did so. After reaching home, she found that she did not 
 get on well with her mother, after having been away from home for 
 five years. She resented her mother's affection for the children by 
 the second husband and imagined that her mother slighted her in 
 every possible way. Finally, she left home the day after Christmas 
 because she thought her mother had not given her as many gifts 
 as the other children had. Without any supervision on the out- 
 side, she steadily went from bad to worse and within a few months 
 had been arrested and committed to Bedford. In estimating this 
 girl's case, the economic status and moral standards were classed 
 as good, parental supervision as poor, and the total estimate as 
 
224 STUDY OF WOMEN DELINQUENTS IN NEW YUKK 
 
 fair. It would seem that this girl with her good mentality, her 
 responsiveness and desire to please, might have been managed so 
 that she would not have come into contact with the law. She was 
 always a difficult child, high-strung, nervous and irritable, but these 
 qualities were only exaggerated by the lax supervision in her grand- 
 parents' home and the more strict and less sympathetic atmosphere 
 in her mother's home when she was of the age where she needed 
 very careful guidance. 
 
 Class 4, which includes good homes, comfortable economically, 
 with good moral standards and parental supervision, comprises only 
 6.0 per cent of the cases in the total group. (Table 66.) The fol- 
 lowing case is typical of a large part of these cases. 
 
 1. Carrie C was committed to Bedford when 21 years, 5 months 
 
 of age for petit larceny. She was born in a village on Long 
 Island and lived there, always in the same house which her 
 mother owned, until she was married at the age of twenty. The 
 field worker in this case had an excellent opportunity to see for 
 herself the house and neighborhood in which the girl had grown 
 up. Carrie's father was a mason and earned from $30 to $40 a 
 week. Since he worked steadily and had only three children to bring 
 up, they were always comfortable economically. They bought a 
 two-story and basement house in which Carrie was born, and 
 when the father died in 1914 he left a $3,000 life insurance. Both 
 the father and mother are well thought of in the community 
 and are both spoken of as having good moral standards. The 
 mother has meant to do well for her children, and since her 
 husband's death has spent a great deal of money to make them 
 comfortable. She is of a nagging, dictatorial disposition, how- 
 ever, and at times has undoubtedly been disagreeable. There 
 was a certain amount of friction between Carrie and her mother, 
 particularly over the man the girl wanted to marry. In her an- 
 tagonism to him and her way of handling the situation, the mother 
 showed poor judgment and little understanding. There does not 
 seem to be evidence, here, of a very important connection betweeii 
 the girl's offense and her home conditions. Her family were honest 
 and law-abiding and had taught her to be so. Her thefts from 
 the department store in which she was working over the holidays 
 seem to have been prompted by the suggestions of the floor-walker 
 who wanted to share the proceeds with her and by her own economic 
 need at that time, since her husband was ill and she was too proud 
 to ask her mother for money. 
 
FACTORS IN EARLY HOME CONDITIONS 225 
 
 2. A quite different case for this class is that of a colored girl, Sarah 
 
 E , 23 years, 9 months of age at the time of her commitment to 
 
 Bedford for prostitution. She was an illegitimate' child and was 
 boarded by her mother with her foster mother from the time she 
 was two weeks old until she was eighteen months old. Then, be- 
 cause her mother was not -paying for the child's board, Mrs. E 
 
 took Sarah to her mother and told her she could not keep her. The 
 
 mother later abandoned Sarah and when Mrs. E heard of this 
 
 she went to the Society which was caring for the child and took her 
 home with her. Sarah has lived with her foster parents since then 
 and has had a good home with them. They own their own home in 
 Brooklyn, a two-story and basement frame house with a yard around 
 it. The house is well furnished and in good condition. All of the 
 family are insured and the foster father has earned good wages as a 
 porter. The foster parents have good moral standards and are 
 well spoken of by every one in the community. Undoubtedly, 
 Sarah has been given excellent training. The only objection 
 which could be raised in regard to the amount of supervision is 
 that the foster mother left the home to work out quite often when 
 Sarah was younger and this left her at home with no one to look 
 out for her after school hours. On the whole, however, the home 
 ranks high, — way above that for any other colored girl in our group. 
 We have estimated the home as good in economic status and moral 
 standards and as fair in parental supervision. The girl in question 
 is very low grade mentally and probably no home training could 
 have kept her from doing certain things if the temptation or oppor- 
 tunity to do so were put in her path. Certainly, in a survey of her 
 actual home surroundings there is no obvious association with her 
 delinquency. 
 
 The fifth class, those with excellent homes, has only two representa- 
 tives in our total group, and they both come in the Penitentiary. In 
 considering the estimates of certain factors in the home, only two other 
 cases were found where any element in the home conditions came up 
 to the highest class. These two cases, both in Bedford and previously 
 mentioned in connection with the economic status, reached the highest 
 class only in the economic status, however. 
 
 1. The first of the Penitentiary cases whose early home surroundings 
 
 were very good is Nina R , who at the time of her commitment 
 
 to the Penitentiary for petit larceny was 49 years and 5 months of 
 age. The field worker was able to visit the childhood home and 
 obtain information both from members of the immediate family and 
 
226 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 from friends, so that our information is probably reliable. Nina 
 
 R was born in a small town in the East and lived there until 
 
 she went to boarding' school, when she was about fifteen. Her father, 
 a wheelwright by trade, was a steady worker and "plain, simple and 
 upright in every way." He owned his home and they were in com- 
 fortable circumstances, though never wealthy. The mother, also, 
 was a woman of fine character. The grandfather was a Quaker 
 preacher and all of his children were Quakers of strict standards. 
 There were five children in Nina's fraternity. The oldest studied to 
 be a doctor and one other child, after teaching school for fourteen 
 years, took a nurse's training course. All who knew the parents felt 
 that they were intelligent and made a very good home for their chil- 
 dren. Nina herself was sent to boarding school from which she 
 graduated and then taught school for twelve years. In moral stand- 
 ards and parental supervision, this family was rated as being very 
 good, and the economic status was classed as good. 
 
 2. The other woman who came from a very good home is Hannah 
 
 M who at 58 years, 4 months was committed to the Penitentiary 
 
 for arson. She was born in Germany and lived there until she was 
 nineteen years old, when she came to this country with her husband. 
 We were able to interview the woman's older brother who seemed 
 sincere and anxious to give any information which might be of bene- 
 fit. He says that the family in Germany were comfortably well-to- 
 do, very good people and well educated. The father was an in- 
 terior decorator, not wealthy, but determined to give his children 
 the best possible education. Hannah was educated for a teacher 
 and was in school until she was nineteen, when she was married to a 
 physician. So far as we know, this home offered very good ad- 
 vantages. We have estimated the economic status as good only, 
 since there seemed to be no great surplus of money, but have con- 
 sidered the other factors as very good. 
 
 In neither of these two cases were the home conditions in any 
 sense involved in the difficulty with the law which came much later in 
 life. 
 
 As a whole, the home conditions of the bulk of our cases are repre- 
 sented by such homes as those of Alice B and Sarah S , i.e., 
 
 (1) cases where the home has been poor, and where only a child of 
 unusual initiative would reach out beyond what the home had offered 
 her, and (2) cases where the home was suflficiently good so that there 
 seemed a fair chance for the child to develop into a good instead of 
 an anti-social person. 
 
as' 
 
 s 
 
 09 
 
 d 
 
 •I 
 •3 
 ti 
 o 
 O 
 
 a 
 
 o 
 
 69-99 
 
 99-89 
 
 £9-09 
 
 09-19 
 
 l9-f9 
 
 W-IS 
 
 TS-8» 
 
 8^-S* 
 
 S^6E 
 
 o 
 O 
 
 S 
 
 6S-9S 
 
 9S-SS 
 
 EE-OS 
 
 §11 
 
 "-J 
 
 OE-iZ 
 
 ZS-^ 
 
 1-2 
 
 o g3 
 
 ?«-IZ 
 
 IZ-8T 
 
 8I-SI 
 
 Bl-Zt 
 
 Zt-6 
 6 -9 
 
 v o cd o <D 
 
 BUOI^TpnOQ 
 
 |0 ajBTniijsg 
 
 227 
 
228 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 (g) Relationship Between Estimate of Home Conditions and Age at 
 
 First Conviction 
 
 In various later chapters we have considered the relationship of 
 given social factors to the estimate of the home conditions. At this 
 point we shall consider the relationship between the age at first con- 
 viction and the home conditions during childhood and adolescence. 
 Table 68 presents these data in a correlation ratio. The ratio of .31 
 with a relatively small standard deviation of .044 would indicate that 
 there is a significant relationship between the age at first conviction and 
 the estimate of home conditions, with a tendency for those who come 
 from the worst homes to be convicted earliest. This is shown by the 
 mean age at first conviction for the various estimate groups, with the 
 mean age at first conviction in the poorest class of home conditions at 
 22.1 years while there is a progression to a mean age at first conviction 
 of 53.5 years for the two cases which fall in the best group of home 
 conditions. 
 
 (h) Relationship Between Estimate of Home Conditions and Number 
 of Previous Convictions 
 
 If we turn to Table 69, showing the correlation between the num- 
 ber of previous convictions and the estimate of home conditions, we 
 find that the correlation ratio is .05 with a standard deviation of .049. 
 This seems to indicate that there is no relationship between the kind 
 of home in which a child grows up and the extent of criminality which 
 develops later. 
 
 FAMILY STATUS 
 
 So closely associated with any survey of home conditions as to 
 be almost inseparable from it are certain other factors, such as the 
 nativity of the parents, their ages in relation to the age of the subject, 
 the number of children in the family and the subject's order in the 
 fraternity. Though these factors are taken into consideration, in gen- 
 eral, in making the estimate of the home conditions, they seem import- 
 ant enough to repeat in detail at this point. 
 
 (o) Nativity of Parents 
 
 Table 70 gives the nativity of the parents by institutional groups, 
 and shows that the percentage of women with both parents native born 
 varies among the groups. Bedford has a high percentage, 57.3 with 
 both parents native born, Auburn 30.7 per cent. Probation 28.4 per 
 
10 
 
 I 
 
 O 
 
 O 
 
 I 
 
 
 t3 
 
 
 i 
 
 1 
 
 •43 
 
 d) 
 
 ij 
 
 M 
 
 •? 
 
 3 
 
 S 
 
 
 O 
 
 ^ 
 
 CO 
 
 ^ 
 
 S 
 
 
 o 
 
 p< 
 
 V 
 
 3 
 
 0) 
 
 h 
 
 £ 
 
 O 
 
 «4H 
 
 '^ 
 
 O 
 
 O 
 
 u 
 
 H 
 
 « 
 
 
 ^ 
 
 
 
 
 ^ 
 
 
 ^ 
 
 
 B 
 
 o 
 
 o 
 O 
 
 «S!3'? 
 
 
 
 S-a.S.9 
 
 
 O 
 
 
 
 
 Me 
 Nui 
 Pre 
 nvic 
 
 OfHr^FHrt 
 
 
 
 
 
 oo 
 
 
 
 ^ 
 
 
 
 
 
 
 
 r^j-i 
 
 
 B 
 
 
 
 
 S 
 
 - 
 
 -• 
 
 
 
 
 
 
 
 
 
 (N 
 
 
 
 
 
 
 t^ 
 
 
 
 
 r-( 
 
 
 
 CO 
 
 
 
 
 
 
 
 ui 
 
 
 
 
 iH 
 
 
 
 ■^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 g 
 
 
 
 
 
 
 
 i 
 
 cq 
 
 tM ^ 
 
 M 
 
 
 
 
 
 O 
 
 ^ 
 
 i-H 
 
 ^ 
 
 
 i-H 
 
 
 
 13 
 
 2 
 
 
 
 
 o 
 
 N W 
 
 CO 
 
 
 
 
 
 (S 
 
 
 
 
 
 rt^ 
 
 « 
 
 fa 
 
 o 
 
 
 
 
 
 
 
 Pt 
 
 00 
 
 <N^ 
 
 CO 
 
 
 
 
 
 
 
 
 
 t* 
 
 i-Hi-ieo 
 
 U3 
 
 :? 
 
 
 
 
 CD 
 
 Ttt^ 
 
 lO 
 
 >o 
 
 .-HCOM 
 
 t* 
 
 ■* 
 
 iHUSCO 
 
 N 
 
 eo 
 
 N-HOCS 
 
 lO 
 
 
 
 
 (M 
 
 "SS^ 
 
 1-1 
 
 
 
 
 
 
 
 o 
 
 O 
 
 
 O) 
 
 
 
 
 
 
 
 iM 
 
 
 ' 
 
 -a 
 
 
 
 
 »- 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 u 
 
 
 
 l_ 
 
 l> j^ 
 
 
 
 bl-b i t 
 
 1 
 
 
 
 
 
 
 
 !>OfHM> 
 
 Fh 
 
 
 snompuoo ai 
 
 nOH 
 
 
 
 
 JO 
 
 a» 
 
 Bl 
 
 ai 
 
 *M 
 
 a 
 
 
 "O 
 
 |o 
 
 ax 
 <« a 
 
 11 
 
 229 
 
230 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 70 
 
 NATIVITY OF PARENTS 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups 
 
 
 Institutional Groups 
 
 
 Nativity 
 
 Bed- 
 ford 
 
 Auburn 
 
 Mag- 
 dalen 
 
 Peni- 
 tentiary 
 
 Worlt- 
 house 
 
 Proba- 
 tion 
 
 Total 
 
 Both Native 
 
 57.3 
 
 30.7 
 
 18.8 
 
 26.9 
 
 40.5 
 
 28.4 
 
 34.8 
 
 
 
 White 
 
 40.6 
 16.7 
 
 35.4 
 
 17.3 
 13.3 
 
 61.3 
 
 18.8. 
 66.7 
 
 12.9 
 14.0 
 
 65.6 
 
 14.6 
 
 25.8 
 
 50.6 
 
 27.2 
 1.2 
 
 55.6 
 
 22.3 
 
 Colored 
 
 12.6 
 
 Both Foreign 
 
 55.1 
 
 
 
 White 
 
 36.4 
 
 60.0 
 1.3 
 
 8.0 
 
 66.7 
 14.5 
 
 62.4 
 3.2 
 
 7.5 
 
 60.6 
 9.0 
 
 53,1 
 2.5 
 
 16.1 
 
 53.9 
 
 
 1.2 
 
 One Foreign: one Native 
 
 7.3 
 
 10.1 
 
 White 
 
 7.3 
 
 8.0 
 
 14.6 
 
 7.5 
 
 5.6 
 3.4 
 
 13.6 
 2.5 
 
 9.1 
 
 Colored 
 
 1.0 
 
 
 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 
 
 Number of cases 
 
 96 
 
 75 
 
 69 
 
 93 
 
 89 
 
 81 
 
 503 
 
 
 
 cent, the Penitentiary 26.9 per cent and the Magdalen 18.8 per cent. 
 In view of the fact that Auburn and the Penitentiary have the highest 
 percentage of foreign born women, it is of interest to observe that the 
 highest percentage of foreign or mixed parentage occurs in the Mag- 
 dalen. The Probation group also has a higher percentage of women 
 with foreign or mixed parentage than Auburn, though the percentage 
 of foreign born women in the Probation group is smaller than that in 
 Auburn. 
 
 If we compare the nativity and parentage of our total group with 
 the nativity and parentage of the general female population in New 
 York state in 1910 ^ we find the following results : 
 
 Nativity and Parentage 
 
 Native White 
 
 Native Parentage 
 
 Foreign or Mixed Parentage ., 
 
 Foreign White 
 
 Negro 
 
 All Others , 
 
 General Female Population Delinquent 
 
 New York State Women in 
 this Study 
 
 61.7 S0.2 
 
 34.1 20.4 
 
 27.6 29.8 
 
 36.5 30.9 
 
 1.7 18.8 
 
 .1 2 
 
 'Thirteenth Census of the United States. 1910. Vol. III. 
 
FACTORS IN EARLY HOME CONDITIONS 
 
 231 
 
 In view of the fact that the foreign born delinquent women in this 
 study are represented in the total group of delinquents by much less 
 than the representation we might expect from their numbers in the 
 general population (See Chapter VIII), it is interesting to find that 
 among the native white of foreign or mixed parentage the delinquent 
 group has a larger percentage than the general female population^ 
 The native white of native parentage, on the other hand, are 34.1 per 
 cent of the general population but only 20.4 per cent of the women in 
 this study. It would seem from this, that the element of foreign par- 
 entage is significant as a factor in the home conditions which may be 
 associated with delinquency. In many of the families studied the in- 
 ability of the foreign parents to adjust themselves to American cus- 
 toms is striking, particularly when the children become quickly 
 adapted to American customs and are ashamed of the parents because 
 of their foreignness. 
 
 (b) Age of Parents at Time of Subject's Birth 
 
 A further factor of importance in the home is the relationship 
 between the ages of the parents and the children. Table 71 shows the 
 
 TABLE 71 
 
 AGE OF MOTHER AT SUBJECT'S BIRTH 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups 
 
 
 Age of Mother 
 
 Institutional Gboups 
 
 
 
 Bed- 
 ford 
 
 Auburn 
 
 Mag- 
 dalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Proba- 
 tion 
 
 Total 
 
 11 to 
 
 
 
 3.2 
 17.5 
 22.2 
 23.7 
 11.1 
 14.3 
 4.8 
 3.2 
 
 1.8 
 10.7 
 26.8 
 23.2 
 14.3 
 16.1 
 5.4 
 1.8 
 
 1.9 
 13.5 
 19.2 
 
 34.6 
 
 11.5 
 
 11.5 
 
 3.9 
 
 1.9 
 
 1.9 
 
 4.3 
 
 27.7 
 
 19.2 
 
 14.9 
 
 19.2 
 
 4.3 
 
 8.5 
 
 2.1 
 
 4.8 
 11.3 
 
 27.4 
 
 24.2 
 
 21.0 
 
 4.8 
 
 4.8 
 
 1.6 
 
 2 4 
 
 16 " 
 
 21 " 
 
 22.2 
 
 33.3 
 
 17.8 
 
 18.9 
 
 3.3 
 
 2.2 
 
 2.2 
 
 17.3 
 
 21 " 
 
 26 " 
 
 •25.7 
 
 26 " 
 
 31 '■ 
 
 22.7 
 
 31 " 
 
 36 " 
 
 16.2 
 
 36 " 
 
 41 " 
 
 8.7 
 
 41 " 
 
 46 " 
 
 4.6 
 
 46 " 
 
 51 " 
 
 2.2 
 
 61 " 
 
 56 " 
 
 .3 
 
 
 
 
 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 
 
 
 Number of nasfis ... 
 
 90 
 
 63 
 
 56 
 
 52 
 
 47 
 
 62 
 
 370 
 
 
 
 
 Mean Age of Mother at Subject's Birth (Total Group) 27.66 + .401 years 
 
 0- 7.71 +.284 
 
232 STUDY "OF WOMEN DELINQUENTS IN NEW YORK 
 
 mother's age at the time of the subject's birth. The youngest mother 
 was eleven years of age, and the oldest, 51 years of age when the sub- 
 ject was born. In the total group, 2.4 per cent of the cases or nine of 
 the delinquent women were born while their mothers were between 
 eleven and sixteen years of age. The largest number of cases occurs 
 in the age group of 21 to 26 years. For the total group the mean age 
 of the mother at time of the subject's birth was 27.66 years with a 
 standard deviation of 7.71 years. 
 
 About the same trend as that noted above is shown with regard to 
 the age of the father at the time of the subject's birth. Table, 72 
 shows that the range of ages starts at 13 years and runs to 83 years. 
 In this latter group there are only two cases and in both instances 
 the father was 81 years old at the time the subject was born. The 
 earlier age groups do not show as high a percentage for the father 
 as they do for the mother. One woman in the Workhouse was born 
 when her father was thirteen, and the four other women included in 
 this group were bom when their fathers were seventeen and eighteen 
 years old. For the total group the average age of the father at the ' 
 time of the subject's birth was 32.95 years with a standard deviation 
 of 10.50 years. This average age, while somewhat higher than that 
 of the mother at time of the subject's birth, is not so much higher as 
 to make the extreme age of the father an important factor in the life 
 of the delinquent. 
 
 (c) Age of Subject at Time of Parenfs Death 
 
 Another factor often advanced as causative of delinquency, is the 
 breaking up of the home, through death, divorce, etc., while the child 
 is young. It is difficult for us to present in detail a table showing the 
 exact status of homes regarding the parents who are separated, dead, 
 imprisoned, etc., since our group of women is of varying ages, many of 
 them so old that we should naturally expect their parents to be dead, 
 and the fact of their being dead not significant as indicating any ab- 
 normal home condition. It seems of interest, however, to know what 
 percentage of these women were very young when either parent died. 
 Table 73 has been made to show the percentage of delinquent women 
 who were under five, ten, fifteen and twenty years of age at the time 
 of the father's and of the mother's death. The table shows that there 
 are varying percentages of women in each of these age groups. The 
 percentage of women who were under five at the time of the father's 
 death runs from 8.3 per cent in Auburn and the Magdalen to 21.4 per 
 
FACTORS IN EARLY HOME CONDITIONS 233 
 
 TABLE 72 
 
 AGE OF FATHER AT SUBJECT'S BIRTH 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups 
 
 
 Institutional Ghohps 
 
 
 Age of Father 
 
 Bed- 
 ford 
 
 Auburn 
 
 Mag- 
 dalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Proba- 
 tion 
 
 Total 
 
 13 to 18 years 
 
 
 ii!9 
 
 20.3 
 
 23.7 
 
 13.6 
 
 15.3 
 
 5.1 
 
 5.1 
 
 3.4 
 
 ■i!7 
 
 's.z 
 
 26.3 
 21.1 
 12.3 
 
 7.0 
 17.5 
 
 5.3 
 
 'i!8 
 1.8 
 
 3.9 
 
 13.5 
 
 21.2 
 
 17.3 
 
 13.5 
 
 15.4 
 
 7.7 
 
 1.9 
 
 3.9 
 
 1.9 
 
 4.6 
 
 11.4 
 
 9.1 
 
 34.1 
 
 15.9 
 
 6.8 
 
 9.1 
 
 4.6 
 
 2:3 
 '2;3 
 
 1.9 
 
 11.5 
 
 23.1 
 
 23.1 
 
 13.5 
 
 13.5 
 
 1.9 
 
 5.8 
 
 3.9 
 
 1.9 
 
 1 4 
 
 18" 23 " 
 
 8.3 
 
 28.6 
 
 23.8 
 
 15.5 
 
 14.3 
 
 2.4 
 
 4.8 
 
 1.2 
 
 10 1 
 
 23 " 28 " 
 
 22 4 
 
 28 " 33 " 
 
 24 1 
 
 33 " 38 " 
 
 13 5 
 
 38 " 43 " 
 
 12 4 
 
 43 " 48 " 
 
 6 9 
 
 48 " 53 " 
 
 4 6 
 
 53 " 58 " 
 
 2 
 
 58 " 63 " 
 
 1.2 
 
 63 " 68 " 
 
 
 .3 
 
 68 " 73 " 
 
 1.2 
 
 3 
 
 73 " 78 " 
 
 .3 
 
 78 " 83 " 
 
 
 .6 
 
 
 
 
 Total . . . 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 
 
 Number of cases 
 
 84 
 
 59 
 
 57 
 
 52 
 
 44 
 
 52 
 
 348 
 
 
 
 Mean Age of Father at Subject's Birth (Total Group) 32.95 + .562 
 
 a 10 . 50 ± . 397 
 
 cent in the Workhouse. The total group shows 11.8 per cent who were 
 under five at the time of the father's death. Probably the next two 
 age groups are of more importance, however, because they include 
 not only those who were very young when the father died but also those 
 who were of the ages where the absence of a father in the home would 
 be of the greatest importance. For those children who were under ten 
 at the time of the father's death there were 18.6 per cent for the total 
 group, with varying percentages in the institutional groups. The per- 
 centage of children under fifteen at time of the father's death, 24.9 
 per cent for the total group, is probably large enough to indicate a 
 serious lack of supervision in most of these childhood homes. The 
 next age group, while of less importance, because many of the women 
 affected had left home before their twentieth year, shows that 34.7 
 of the women were under twenty years at the time of the father's 
 death. We have not presented any further figures sirice we are most 
 interested in the home during childhood and adolescence. 
 
234 STUDY OF WOMEN DELINQUENTS IN NEW YUKK 
 
 TABLE 73 
 
 AGE AT TIME OF PARENTS' DEATH 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups 
 
 
 
 
 Age 
 
 A.T Time 
 
 OF Death of; 
 
 
 
 
 Father 
 
 Mother 
 
 
 Under 
 5 
 
 years 
 
 Under 
 10 
 
 years 
 
 Under 
 
 15 
 years 
 
 Uhder 
 
 20 
 years 
 
 Under 
 
 5 
 years 
 
 Under 
 
 10 
 years 
 
 Under 
 15 
 
 years 
 
 Under 
 
 20 
 years 
 
 Bediford 
 
 10.4 
 
 8.3 
 
 8.3 
 
 9.3 
 
 21.4 
 
 12.5 
 
 17.7 
 15.3 
 15.3 
 16.3 
 27.4 
 18.8 
 
 22.9 
 20.8 
 20.8 
 26.7 
 34.5 
 22.5 
 
 35.4 
 26.4 
 26.4 
 34.9 
 42.9 
 40.0 
 
 5.1 
 10.4 
 10.8 
 11.0 
 12.8 
 
 7.1 
 
 10.2 
 14.3 
 16.2 
 14.3 
 17.4 
 9.5 
 
 20.4 
 20.8 
 23.0 
 20.9 
 18.6 
 13.1 
 
 28.6 
 
 Auburn 
 
 23.4 
 
 Magdalen 
 
 28.4 
 
 Penitentiary 
 
 31.9 
 
 Workhouse 
 
 25.6 
 
 Probation 
 
 19.1 
 
 
 
 Total 
 
 11.8 
 
 !8.6 
 
 24.9 
 
 34.7 
 
 9.4 
 
 13.5 
 
 19.4 
 
 26.3 
 
 
 
 Number of cases 
 
 68 
 
 91 
 
 122 
 
 170 
 
 48 
 
 69 
 
 99 
 
 134 
 
 If we turn to the child's age at time of the death of the mother we 
 see that, for the most part, the percentages of delinquent women who 
 were young at time of the mother's death are smaller than the per- 
 centages at the time of the father's death. There are 9.4 per cent of 
 our total group of women who were under five at the time of the 
 mother's death, 13.5 per cent who were under ten, 19.4 per cent under 
 fifteen, and 26.3 per cent under twenty years of age. ^ 
 
 At this point we may compare with our data Goring's ^ figures for 
 the ages of male criminals at the time of the mother's death. The 
 percentages are as follows : 
 
 Under S years 
 " 10 " 
 " IS " 
 " 20 " 
 
 This shows a smaller percentage of male criminals in the youngest 
 age group, but a larger percentage in the other groups, than we find 
 among the delinquent women of this study. 
 
 Goring's Study 
 
 Women in this Study 
 
 1:1% 
 
 9.4% 
 
 15.4% 
 
 13.5% 
 
 25.2% 
 
 19.4% 
 
 35.5% 
 
 26.3% 
 
 ^Of. cit., p. 422. 
 
FACTORS IN EARLY HOME CONDITIONS 235 
 
 (d) Number of Children in Fraternity 
 
 Another element entering into the consideration of home and fam- 
 ily conditions is the size of the family. Many contend that most crim- 
 inals come from large families, especially those of poorer economic 
 status where the large number of children makes decent living condi- 
 tions impossible and leads to prostitution or stealing as a means of 
 livelihood. Others feel that an only child is more likely to "go wrong" 
 because she has had less training in the appreciation of property rights 
 and in adapting herself to others. An instance of this occurs in the 
 case we have mentioned previously in this chapter of the only child 
 in a wealthy family. She -was pampered and spoiled until she was 
 eighteen when she insisted, despite her parents' protest, upon marrying 
 a man who was of little account. Her life from that time on, — her 
 unwillingness to take the responsibility of looking after her children, 
 her selfishness at the time of her mother's illness and death, and her 
 gradual drifting into prostitution — seems in part an outgrowth of her 
 early training. 
 
 To show the range of number of children in the fraternities of the 
 delinquents, we present Table 74. This indicates that the range is 
 very long in Bedford, Auburn and the Penitentiary groups and pro- 
 gressively shorter in the Workhouse, Probation and Magdalen. The 
 total group has a range of one to eighteen children in the family, 
 with the greatest concentration between three and seven children. The 
 mean number of children for the total group is 5.76±.148 with a stand- 
 ard deviation of 3.8±.10S. 
 
 The Census figures for 1910 give the number of persons per fam- 
 ily in New York State as 4.5.* The Census figures do not include the 
 average number of children per family, but we are safe in saying that 
 this number must be appreciably less than^S, the average number of 
 persons per family. This would indicate' thaf the women in our group 
 come from families of appreciably larger average size than the fan;- 
 ilies in the general population. We might also compare here the aver- 
 age number in the families of 647 prostitutes studied in Bedford and 
 reported on by Dr. Katharine Bement Davis in Mr. Kneeland's 
 book on "Commercialized Prostitution in New York City." = Dr. 
 Davis found that the average number of children in the families was 
 3.99, a number somewhat less than the average number for our group. 
 
 •Thirteenth Census of the United States. 1910. Vol. III. 
 ' Op. cit.. p. 170. 
 
236 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 74 
 NUMBER OF CHILDREN IN FRATERNITY 
 Per Cent Distribution by Institutional Groups 
 
 
 Institutional Groups 
 
 
 Children in 
 Fraternity 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 Peni- 
 tentiary 
 
 Work- 
 houae 
 
 Probation 
 
 Total 
 
 1 
 
 9.9 
 11.9 
 14.9 
 11.9 
 7.9 
 7.9 
 7.9 
 5.9 
 3.0 
 4.0 
 2.0 
 4.0 
 5.9 
 1.0 
 2.0 
 
 3.8 
 
 6.3 
 
 8.8 
 
 12.5 
 
 8.8 
 
 16.3 
 
 12.5 
 
 3.8 
 
 7.5 
 
 7.5 
 
 3.8 
 
 2.5 
 
 2.5 
 
 2.5 
 
 lis 
 
 5.3 
 
 8.0 
 
 14.7 
 
 13.3 
 
 17.3 
 
 9.3 
 
 17.3 
 
 5.3 
 
 4.0 
 
 5.3 
 
 11.3 
 
 4.7 
 
 10.4 
 
 12.3 
 
 12.3 
 
 8.5 
 
 5.7 
 
 8.5 
 
 7.6 
 
 3.8 
 
 2.8 
 
 .9 
 
 7.6 
 
 2.8 
 
 .9 
 
 12.0 
 4.0 
 
 11.0 
 
 16.0 
 8.0 
 
 14.0 
 5.0 
 9.0 
 7.0 
 2.0 
 2.0 
 4.0 
 3.0 
 3.0 
 
 14.6 
 
 14.6 
 
 13.5 
 
 6.7 
 
 10.1 
 
 9.0 
 
 13.5 
 
 3.4 
 
 4.5 
 
 2.3 
 
 4.5 
 
 2.3 
 
 1.1 
 
 9.8 
 
 2 
 
 8.2 
 
 3 
 
 12.2 
 
 4 
 
 12.2 
 
 5 
 
 10.5 
 
 6 : 
 
 10.7 
 
 7 
 
 9.8 
 
 8 
 
 6.2 
 
 9 
 
 5.6 
 
 10 
 
 4.0 
 
 11 
 
 2.5 
 
 12 
 
 2.4 
 
 13 
 
 3.6 
 
 14 
 
 1.6 
 
 17 
 
 .5 
 
 18... 
 
 .2 
 
 
 
 Total . . . 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of oases 
 
 101 
 
 80 
 
 75 
 
 106 
 
 100 
 
 89 
 
 551 
 
 Mean Number of Children in Fraternity (Total Group). 
 a 
 
 .5.76±.148 
 .3. 48+. 105 
 
 This again we may compare with Goring's' figures for the families , 
 of male criminals. He finds that the mean number of children in the 
 family is 6.89 with a standard deviation of 3.85. This would indicate 
 that the men in his group came from slightly larger families than do the 
 women in our group. He has computed from this the correlation 
 ratio of nature of crime with the number of children in the family 
 and finds that the ratio is .1 + .04. From the values of his means, he 
 finds that those convicted of fraud tend to be drawn from smaller 
 families than those convicted of other offenses. He explains this by 
 showing that the social class from which the fraudulent criminals are 
 mainly drawn tends to be restricted in size of family. 
 
 ° Op. cit., p. 422. 
 
FACTORS IN EARLY HOME CONDITIONS 2Z7 
 
 {e) Order in Fraternity, of Delinquent Women 
 
 The order in the fraternity, which is closely connected with the 
 number of children in the family, is of interest because of the belief 
 that the oldest in a large family may have much of the responsibility 
 of looking after the others and may be neglected if there are many 
 ^hers who are not able to look out for themselves. Table 75 shows 
 ^hat there is a large percentage in our group, 28.9 per cent, who were 
 the oldest in the family^Since Table 74 shows that there are only 9.8 
 per cent of the cases where the subject was the only child, this letvii 
 19.1 per cent of the women who were not the only child and who were 
 the oldest in the family. In the total group, and by comparison with 
 Table 74, we find that the delinquent women, in large part, come from 
 the older members Jn.the_fraternity. 
 
 (/) Number of Full-term Children 
 
 Though not logically related to the subject's home as a child or 
 her own family conditions, we shall consider here, in connection with 
 
 TABLE 75 
 
 ORDER IN FRATERNITY 
 
 Per Cent Distribution by Institutional Groups 
 
 
 iNSTITnTIONAL GbODPS 
 
 
 Order in 
 Fraternity 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Probation 
 
 Total 
 
 1 
 
 29.7 
 
 20.8 
 
 16.8 
 
 15.8 
 
 6.9 
 
 5.0 
 
 2.0 
 
 1.0 
 
 i'.o 
 
 1.0 
 
 22,8 
 
 15.2 
 
 10.1 
 
 17.7 
 
 8.9 
 
 7.6 
 
 3.8 
 
 2.5 
 
 5.1 
 
 1.3 
 
 2.5 
 
 1.3 
 
 i'.i 
 
 23.3 
 
 20.6 
 
 11.0 
 
 9.6 
 
 12.3 
 
 15.1 
 
 4.1 
 
 1.4 
 
 2.7 
 
 28.6 
 26.4 
 
 16.5 
 
 14.3 
 
 5.6 
 
 4.4 
 
 '2^2 
 
 i.'i 
 
 29.8 
 10.6 
 19.2 
 13.8 
 
 8.5 
 
 8.5 
 
 6.4 
 
 i.'i 
 
 'i!i 
 'i!i " 
 
 37.5 
 17.1 
 18.2 
 8.0 
 3.4 
 5.7 
 6.8 
 1.2 
 1.2 
 
 'i;2 
 
 28.9 
 
 2 
 
 18.4 
 
 3 
 
 15.6 
 
 4 
 
 13.3 
 
 5 
 
 7.4 
 
 6 
 
 7.4 
 
 7 
 
 3.8 
 
 8 
 
 1.1 
 
 9 
 
 1.5 
 
 10 
 
 .8 
 
 11 
 
 .8 
 
 12 
 
 .6 
 
 13 
 
 
 14 
 
 .4 
 
 
 
 Total . . . 
 
 100.0 
 
 100.0.- 
 
 100.0 
 
 100.0, 
 
 100.0 , 
 
 100.0 ■ 
 
 100.0 
 
 Number of cases 
 
 101 
 
 79 
 
 73 
 
 91 
 
 94 
 
 88 
 
 526 
 
238 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 76 
 
 NUMBER OF FULL-TERM CHILDREN 
 
 Per Cent Distribution by Institutional Groups 
 
 
 Institutional Groups 
 
 
 Full-Term 
 Children 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Probation 
 
 Total 
 
 
 
 59.4 
 24.8 
 7.9 
 4.0 
 3.0 
 1.0 
 
 25.6 
 33.3 
 9.0 
 11.5 
 5.1 
 2.6 
 1.3 
 5.1 
 1.3 
 2.6 
 1.3 
 1.3 
 
 62.9 
 
 24.3 
 
 7.1 
 
 2.9 
 
 'i!4 
 1.4 
 
 47.7 
 19.6 
 8.4 
 8.4 
 3.7 
 5.6 
 2.8 
 
 i'.Q 
 
 .9 
 
 52.5 
 
 16.8 
 
 14.9 
 
 7.9 
 
 5.0 
 
 '2;6 
 
 63.0 
 
 26.1 
 
 6.5 
 
 1.1 
 
 3.3 
 
 52.1 
 
 1 
 
 23.7 
 
 2 
 
 9.1 
 
 3 
 
 6.0 
 
 4 
 
 3.5 
 
 5 
 
 1.8 
 
 6 
 
 
 
 .9 
 
 7 
 
 
 
 .9 
 
 8 
 
 .2 
 
 9 
 
 .7 
 
 10 
 
 .4 
 
 11 
 
 .6 
 
 18 
 
 .2 
 
 
 
 
 
 Total . . . 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cases 
 
 101 
 
 78 
 
 70 
 
 107 
 
 101 
 
 92 
 
 549 
 
 the size of her own fraternity, the number of full-term children, total 
 number of pregnancies and number of illegitimate pregnancies she 
 has had. The number of full-term children, which we shall present 
 first, includes both legitimate and illegitimate children. Table 76 
 shows that the percentage of women who have had no children is 52.1 
 per cent for the total group, but that the percentage varies between 
 institutional groups, the three groups where the average age is highest 
 and where there are the highest percentages of married women, hav- 
 ing the smallest percentage without children. The range of number of 
 children is also longest in the three groups first mentioned, i.e., Au- 
 burn, the Penitentiary and the Workhouse. 
 
 (g) Number of Pregnancies 
 
 If we compare with the number of full-term children the number 
 of times the delinquent women in this study have been pregnant we 
 find a somewhat different distribution.- Table 77 shows that while 
 the range for the total group remains the same as in the previous 
 table (Table 76), the percentage who have never been pregnant is 
 
FACTORS IN EARLY HOME CONDITIONS 
 
 239 
 
 much smaller in each institutional group, indicating that there is, 
 throughout, a considerable number of women who have had miscar- 
 riages or abortions. For the total group, there are 9.8 per cent of 
 the women who have been pregnant but who have never had a full- 
 term child. 
 
 The number of illegitimate pregnancies is difficult to obtain accu- 
 rately, particularly when the subject herself is unwilling to talk freely 
 about it. With the best data that could be obtained, however, we find 
 as shown in Table 78 that a very high percentage in each institutional 
 group have had no illegitimate children, the highest percentage being 
 in the three groups where there are the largest numbers of married 
 women and in the Probation group. Bedford and the Magdalen which 
 have the largest number of illegitimate pregnancies have a range of 
 none to six illegitimate pregnancies in the one case and none to three 
 illegitimate pregnancies in the other. The Bedford woman who had 
 six illegitimate pregnancies is an Austrian woman who has lived with 
 
 TABLE 77 
 
 NUMBER OF PREGNANCIES 
 
 Per Cent Distribution by Institutional Groups 
 
 
 Institutional Groups 
 
 
 Pregnancies 
 
 Bedfo 
 
 rd Auburn 
 
 Magdalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Prob£ 
 
 Total 
 ition 
 
 
 
 49. 
 
 23. 
 
 10. 
 8.< 
 4.( 
 2.( 
 l.( 
 
 5 16.9 
 
 i 23.4 
 
 3 16.9 
 
 3 15.6 
 
 ) 3.9 
 
 3 6.5 
 
 3 2.6 
 
 1.3 
 
 1.3 
 
 3.9 
 
 2.6 
 
 1.3 
 
 1.3 
 
 1.3 
 
 1.3 
 
 44.3 
 
 32.9 
 
 17.1 
 
 1.4 
 
 '2;9 
 
 37.4 
 22.4 
 
 10.3 
 
 10.3 
 
 4.7 
 
 8.4 
 
 2.8 
 
 i'.Q 
 .9 
 
 .... 
 
 "■]9 
 
 43.6 
 16.8 
 14.9 
 13.9 
 5.9 
 2.0 
 
 'i!6 
 'i;6 
 'i!6 
 
 58 
 
 22 
 
 12 
 
 1 
 
 3 
 
 2 
 
 7 42 3 
 
 1 
 
 8 -23 2 
 
 2 
 
 13 3 
 
 3 
 
 1 8.8 
 
 4 
 
 3 3.8 
 
 5 
 
 2 4.0 
 
 6 
 
 1.1 
 
 7 
 
 
 
 .6 
 
 8 
 
 
 
 .2 
 
 9 
 
 .9 
 
 10 
 
 .6 
 
 11 
 
 .4 
 
 12 
 
 .2 
 
 13 
 
 .2 
 
 14 
 
 .2 
 
 17 
 
 .2 
 
 18 
 
 .2 
 
 
 
 
 
 Total . . . 
 
 100. C 
 
 ) 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100 
 
 100.0 
 
 Number of cases 
 
 101 
 
 77 
 
 70 
 
 107 
 
 101 
 
 92 
 
 ! 548 
 
240 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 78 
 
 NUMBER OF ILLEGITIMATE PREGNANCIES 
 
 Per Cent Distribution by Institutional Groups 
 
 
 Institutional Groups 
 
 
 Illegitimate 
 Pregnancies 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 Peni- 
 tentiary 
 
 Work- 
 hotise 
 
 Probation 
 
 Total 
 
 ■ 
 
 64.4 
 
 18.7 
 
 6.9 
 
 6.9 
 
 2.0 
 
 ■i!6 
 
 84.4 
 
 11.7 
 
 1.3 
 
 '2.6 
 
 64.3 
 
 27.1 
 
 7.1 
 
 1.4 
 
 83.0 
 
 12.3 
 
 .9 
 
 .9 
 .9 
 .9 
 .9 
 
 86.1 
 5.9 
 3.0 
 2.0 
 1.0 
 2.0 
 
 89.1 
 
 7.9 
 3.3 
 
 79.0 
 
 1 
 
 13.4 
 
 2 
 
 3.7 
 
 3 
 
 2.0 
 
 4 
 
 .7 
 
 5 
 
 .9 
 
 6 
 
 .4 
 
 
 
 Total... 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cases 
 
 101 
 
 77 
 
 70 
 
 106 
 
 101 
 
 92 
 
 547 
 
 two consorts. She was probably never sexually promiscuous and did 
 not go with any other man while she was living steadily with one con- 
 sort. In this total group there are 21 per cent of the women who have 
 had one or more illegitimate pregnancies. 
 
 HEREDITARY FACTORS 
 
 The more important part of the family history which requires a 
 much further study than we can give it at this point is the matter of 
 the heredity and the details of certain defective strains in the family. 
 Because this is such a large subject and needs careful presentation of 
 family histories to make the material worth anything, we shall give 
 here only the summarized data which we were able to obtain through 
 the field workers' visits. This material shows the occurrence of cer- 
 tain specified defective strains in the mother, father, sisters or broth- 
 ers of the women in this study. The distribution of delinquent women 
 by institutional groups, for specified defects, gives in each case the 
 number of women in whose immediate families (mother, father or 
 sibs) the given defect occurs. We have selected for observation cer- 
 tain significant hereditary factors, particularly those stressed by mod- 
 ern criminologists. 
 
 "Alcoholic" as we have used it in this connection includes only 
 those who are excessively alcoholic. "Criminal Record" includes 
 
FACTORS IN EARLY HOME CONDITIONS 241 
 
 those who have had a conviction in any court for any offense. "Epilep- 
 tic" includes, wherever possible, those who have been so diagnosed 
 by the family physician; where this diagnosis was not possible the 
 statements of reliable informants have been used. "Feeble-minded" 
 probably more than any other defect noted here has a much smaller 
 percentage of cases than there should be because of the caution of the 
 field workers in calling any one feeble-minded who was not obviously 
 unable to look out for himself. "Insane" includes only those who 
 have been confined in hospitals for the insane or who have been diag- 
 nosed as insane. "Neurotic" has been used to include persons of de- 
 cidedly high-strung, irritable, unstable temperament, difficult to live 
 with. "Sexually irregular" includes those concerning whom we have 
 been able to get definite evidence of sexual promiscuity. "Suicide" 
 and "Tubercular" are self-explanatory. "Venereal disease" is inclu- 
 sive of both syphilis and gonorrhea; the information on this is prob- 
 ably very unreliable because of the difficulty of obtaining this informa- 
 tion either from the subject or other members of the family, except 
 in Bedford, where the immediate families of nearly all of the girls were 
 very well known. "Wanderer," which is a term used in the eugenics 
 studies made under Dr. Charles B. Davenport, includes those con- 
 cerning whom we had definite evidence of marked nomadic tendencies. 
 
 Table 79 shows the frequency of occurrence of these defective 
 strains in the immediate families of the women in this study.'' There 
 are, it is seen, 21.8 per cent of the women in whose families some 
 member has been excessively alcoholic. This is a much higher per- 
 centage than Dr. Davis found in a study of prostitutes at Bedford,' 
 where 35 girls or 5.4 per cent of her total group came from families 
 in which there was alcoholism. However, as she states, her figures 
 are largely based on the girls' own statements and are probably much 
 lower than they should be. 
 
 The percentage of women in our group who come from homes in 
 which some member of the family has had a conviction is 15.9. 
 Though we have no way of measuring the percentage of families in 
 the general population in which there has been a conviction, this per- 
 centage would seem to be so high as to be of great significance in 
 
 ''The distribution does not show the amount of overlapping which occurs. 
 There may be alcoholism, epilepsy and insanity all within the same family. In 
 this event, each defective strain is counted separately so that the percentage 
 of women having alcoholism in the family represents all cases where alcoholism 
 occurs either as the only defective strain or in combination with others. 
 
 ' Op. cit., p. 172. 
 
242 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 79 
 
 DEFECTIVE STRAINS IN FATHER, MOTHER AND MEMBERS OF 
 
 FRATERNITY 
 
 Distribution by Institutional Groups and Per Cent Distribution for 
 
 Total Group of Delinquent Women in Whose Immediate Families 
 
 Specified Defective Strains Occur 
 
 
 ' 
 
 Total 
 
 Defective Strains 
 
 Bed- 
 ford 
 
 Aubxirn 
 
 Mag- 
 dalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Prdba-- 
 tion 
 
 Number 
 
 Per 
 
 Cent 
 
 Alcoholic 
 
 32 
 
 32 
 
 10 
 
 8 
 
 2 
 
 9 
 
 32 
 
 1 
 
 20 
 
 5 
 
 6 
 
 21 
 
 13 
 
 3 
 
 1 
 
 4 
 2 
 8 
 1 
 16 
 
 5 
 
 18 
 15 
 
 4 
 
 2 
 
 3 
 
 13 
 
 2 
 8 
 
 2 
 
 12 
 6 
 2 
 1 
 5 
 6 
 7 
 3 
 
 20 
 
 12 
 7 
 1 
 2 
 4 
 1 
 8 
 2 
 
 18 
 
 i 
 
 19 
 
 10 
 
 1 
 
 6 
 
 'i 
 11 
 
 ie 
 
 2 
 2 
 
 114 
 83 
 17 
 22 
 17 
 25 
 79 
 
 9 
 98 
 
 7 
 16 
 
 21.8 
 
 Criminal Record 
 
 15.9 
 3.3 
 
 
 4.2 
 
 Insane 
 
 3.3 
 
 Neurotic 
 
 4.8 
 
 Sexually Irregular 
 
 Suicide 
 
 15.1 
 1.7 
 
 Tubercular 
 
 18.8 
 
 Venereal Disease 
 
 1.3 
 3.1 
 
 
 
 Number of Women with 
 Specified Defective 
 Strains in Family 
 
 67 
 
 38 
 
 35 
 
 39 
 
 35 
 
 42 
 
 256 
 
 49.0 
 
 Number of Women not 
 Having Specified De- 
 fective Strains in 
 Familv 
 
 34 
 
 38 
 
 40 
 
 65 
 
 51 
 
 38 
 
 266 
 
 61 
 
 
 
 Total Number of Women 
 
 101 
 
 76 
 
 75 
 
 104 
 
 86 
 
 80 
 
 522 
 
 100.0 
 
 affecting home conditions. Closely associated with the criminal record 
 is the amount of sexual irregularity. In our group we find that there 
 was sexual irregularity in the homes of 15.1 per cent of the women. 
 Tuberculosis is found in the immediate families of 18.8 per cent 
 of the cases. Each of the other defective strains noted occurs in less 
 than 5 per cent of the families. We feel that the information concern- 
 ing the defects which show the lower percentages is probably inade- 
 quate and that exact comparisons are not justified. As a whole, we 
 find that 49.0 per cent of the women in our group had at least one 
 of the defective strains we have observed, in one or more members of 
 their families. Many of these families had several of the defects in 
 
FACTORS IN EARLY HOME CONDITIONS 243 
 
 both the mother and father, and the members of the fraternity.^ Fifty- 
 one per cent of the women in the gpoup came from families in which 
 we were unable to find any of the defects specified. 
 
 It is difficult to say whether any of these defects of poor heredity 
 are directly responsible for the women getting into difficulties with 
 the law, since we do not know whether such defects are present in the 
 same proportion in the general population. The most we can say is 
 that, from this brief survey of hereditary strains which are believed 
 to be significant, the immediate families of at least 49.0 per cent of 
 the women in our study are affected. 
 
 Dr. Davis found that 20.09 per cent of the prostitutes at Bedford 
 had "degenerate strains"^" very closely corresponding to the defective 
 strains we have just discussed. As stated before, she felt that this did 
 not include nearly all of the defects which should have been included 
 had it been possible to make more thorough investigations. Dr. Glueck, 
 in his study of men at Sing Sing," found that "alcohol and tuberculo- 
 sis seem to have been rather significant hereditary factors." 
 
 SUMMARY 
 
 In summarizing the more important factors which relate to the 
 early home or the family status of the women whom we are studying, 
 there stands out most clearly the fact that the large part of the women 
 have come from poor homes, and that a large percentage come from 
 families with defective strains. That we might treat the informa- 
 tion on all of our cases alike, an estimate was made, uniformly for all 
 cases, of the early home conditions and the three factors which we felt 
 were most important in the home, i.e., the economic status, moral 
 standards and parental supervision. By each of these estimates, the 
 majority of the cases fall in the very poor, poor or mediocre classes. 
 In the total estimate of the home conditions during childhood and 
 adolescence, 46.4 per cent of the cases fall in the two poorest groups, 
 
 "A girl of 17 committed to Bedford came from a family in which the father 
 was excessively alcoholic, going off on "sprees" periodically; he was feeble- 
 minded, and had never been able to learn to read or write ; he was sexually 
 promiscuous and syphilitic; he had had innumerable convictions for intoxica- 
 tion and stealing; and he was a "wanderer," leaving his family every few 
 months and moving about from place to place, always thinking he could "do 
 better" in another town. The mother of the girl was also feeble-minded, epilep- 
 tic and sexually promiscuous. Two of the younger boys in the family, though 
 under 16 years, had been brought into court several times for petty stealing, 
 and were finally sentenced to a juvenile institution on larceny charges. 
 
 "O/-. cit., p. 173. 
 
 ^^Op. cit., p. 107. 
 
244 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 47.1 per cent in the mediocre class, and 6.5 per cent in the class of 
 homes which are good or excellent. 
 
 How much of a factor the element of poor home conditions is, 
 as causative of delinquency, it is difficult to judge, since we have no 
 similar estimates for the general population. Within our. own group, 
 however, we are able to show that there is a significant relationship 
 between the age at first conviction and the estimate of home condi- 
 tions, with a tendency for those who were brought up in the poorest' 
 homes to be convicted at an earlier age than those who were brought . 
 up in better homes. , 
 
 Certain of the details of family status have been separately con- " 
 sidered, though to a certain extent they enter into the study of homef ■ 
 conditions. Among these details we find that for the most part the age 
 of the parents was not extreme in relation to the age of the subjects-' 
 in this study. There is a fairly high percentage, 24.9 per cent, of these ■ 
 women who were under fifteen at the time of the father's death, and 
 19.4 per cent who were under fifteen at the time of the mother's death. 
 A similar observation made by Goring shows that he found 25.2 per 
 cent of male criminals under fifteen at the time of their mother's 
 death. The number of children in the family shows that the general 
 population for New York state has fewer children in the family ,than 
 there are in the families of the delinquent women. 
 
 In any study of the conditions found in homes, we can not hope 
 to do more than give a picture of the kind of homes from which the 
 most of the persons in whom we are interested come. Very few at- 
 tempts have been made to use uniform schemes for estimates or for 
 grading homes, and up to this time the descriptive method has been 
 felt to be of more value. Because of this trend, there is little available 
 data for comparison, and we are able to say little more regarding 
 the home conditions of our group than that these women come from 
 the poorer classes, to a very large extent, and that, in addition to poor 
 economic opportunities, there are low moral standards and poor super- 
 vision of the children. 
 
CHAPTER X 
 EDUCATIONAL BACKGROUND 
 
 VIEWS REGARDING RELATION OF EDUCATION TO CRIME 
 
 THE significance of education as a factor in relation to crime has 
 been an important moot point in criminology. There have been all 
 yarieties of contentions, including such diverse views as the assertion 
 that education is a definitely harmful factor, increasing any latent 
 "'tendency toward crime; the insistence that lack of education is an im- 
 '^portant cause of criminality; and the belief, at variance with both of 
 the above, that education as such is not an important factor, but ap- 
 pears to figure merely because it is symptomatic of other genuinely 
 significant influences, such as mentality and social background. Ade- 
 quate data have not been oiJered as yet for establishing any of these 
 views. Aschaflfenburg^ goes so far as to insist that it will inevitably 
 remain lacking. He says : 
 
 "No statistical proof of the influence of education, on criminality can 
 ever be brought. . . . But — and this too, unfortunately, needs to be ex- 
 pressly mentioned — there can also be no proof to the contrary. The simple 
 childlike view, that the degree of education obtained in the lower schools 
 menaces the harmless, deeply moral, mode of thought of the people, this 
 sentimental glorification of the people in their primitive state, is based 
 on entirely vague and unfounded prejudices. It is not increasing education 
 that causes the growth of crime, but the changes that have taken place in 
 all external conditions in the course of the last centuries.'' 
 
 We are not willing to agree that it will be impossible ever to 
 analyze the complex social situation sufficiently to reach any conclu- 
 sions regarding the influence of the educational factor, since modern 
 statistical methods can go far toward solving just such difficulties. 
 Statistics, however, must have an adequate basis of facts on which to 
 build, and that, we must admit, is not available at present. Such a 
 
 ^ Aschaffenburg, G. "Crime and Its Repression,'' p. 137. Trans, by A. Al- 
 brecht. Little, Brown & Co. 1913. 
 
 245 
 
246 STUDY OF WOMEN DELINQUENTS IN NH-W yukK 
 
 mode of approach as that of Lombroso,^ in spite of its apparent foun- 
 dation in fact, is worse than futile, since it may lead to the most erro- 
 neous conclusions. The general line of argument, with its almost in- 
 conceivable fallacies, is well illustrated by the following passage from 
 Lombroso : 
 
 "Tocqueville shows that in Connecticut criminality has increased with 
 the increase in instruction. In the United States the maximum figures 
 for criminality . . . were noted in Wyoming, California, and Nevada, 
 which gave the minimum number of illiterates . . . ; and the minimum 
 figures for criminality were found in New Mexico . . . South Carolina 
 . . ., Alabama, Mississippi, Georgia, and Louisiana, which had the high- 
 est number of illiterates. Nebraska, Iowa, Maine, and Dakota were ex- 
 ceptional, having a small number of criminals and illiterates both, as a 
 result of other causes which we shall see presently. In England the coun- 
 ties of Surrey, Kent, Gloucester, and Middlesex, where there is a higher 
 degree of education, gave the maximum degree of criminality, while the 
 minimum was shown by the more illiterate districts. North Wales, Essex, 
 and Cornwall."^ 
 
 It is obvious that no argument can carry weight if it is based 
 simply on a crude comparison of dififerent decades or different sec- 
 tions of the country with reference to the amount of illiteracy on the 
 one hand and the amount of criminality on the other. The whole 
 social situation underlying such differences in illiteracy may be re- 
 sponsible for the differences in criminality, the variations in illiteracy 
 being merely side-products. Specifically, there are two conspicuous 
 ways that this changing social situation might operate, both of which 
 would tend to explain the apparent relationship, mentioned above. In' 
 the first place, the same conditions which lead to better school oppor- 
 tunities are likely to involve a more compHcated form of life, with 
 more tensions, restrictions, and interferences of one individual with 
 another, leading thus to a greater probability of at least the minor 
 types of offenses. In the second place, the general social change is 
 likely to be accompanied by the establishment of more rigid and 
 more strictly enforced standards of conduct, so that what may be 
 only deplored in one community ma^ be a legal offense in another. 
 These varying legal standards may well account for such differences 
 in "criminality" as Lombroso cites between various sections of this 
 
 'See Lombroso, C. "Crime: Its Causes and Remedies." Chap. VIII. 
 Trans, by H. P. Horton. Little, Brown & Co. 1911. 
 " Op. cit., p. 106. 
 
EDUCATIONAL BACKGROUND 247 
 
 country, and which he naively accepts as proof that one section has 
 more delin'quency than another. The United States Census has 
 pointed out very convincingly in such statements as the following 
 that such an inference is untenable, and that the difference in ap- 
 parent criminality must first be sought in a comparison of the laws 
 and the provisions for law-enforcement of the several communities. 
 
 "In general it must be borne in mind that the number of commitments 
 represents simply the number of violations of law which have been 
 punished by imprisonment. A high ratio of commitments to population may, 
 therefore, be indicative not so much of exceptional criminality and law- 
 lessness as of exceptional efficiency on the part of the local authorities, 
 the police, and the courts in following up infractions of the law, by 
 detecting criminals and offenders and bringing them to trial. The ratio 
 may thus reflect, in some degree, the public sentiment of the community as 
 regards the punishment of crime and minor offenses."* 
 
 We mayj therefore, disregard as quite unfounded Lombroso's fears 
 regarding, the dangers of education for criminals or prospective crimi- 
 nals — feairSv, which he does not himself entirely dismiss with his "com- 
 forting asjcirance that education is not so fatal as it appears at first 
 to be."* "'^ foreign to the current acceptance in this country of the 
 value of education is the point of view that is fearful of the ^ influence 
 of school training, that it seems a superfluity even to consider the prop- 
 osition that education may tend to increase the amount of crime. We 
 fear, however, that the present point of view on the matter is built on 
 very Httle more adequate factual material than is its predecessor, but 
 that this, like its predecessor, is mainly a matter of conviction. Reli- 
 ance upon education as a deterrent from crime is in accord with the 
 spirit of the times, and is doubtless the safest hypothesis to act upon, 
 since, in the absence of evidence to the contrary, we believe that educa- 
 tion is a desirable asset, whether it is a preventive of crime or not. 
 
 Goring's discussion of the problem,"^ though very cursory, has the 
 merit of limiting itself to the actual data available. Considering first 
 ■the relation to recidivism of what he designates as "formal" education, 
 meaning thereby opportunity for instruction as represented by the 
 kind of school attended and the age at leaving school, he reaches the 
 following conclusions : 
 
 * United States Census, 1910, Volume on "Prisoners and Juvenile Delin- 
 quents," p. 19. 
 " Op. cit, p. 108. 
 ° Op. cit, pp. 274-277. 
 
248 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 "From the above differences of mean in relation to their probable 
 errors, it would appear, firstly, that there is no significant relation between 
 a convict's formal education, when a child, and the frequency of his subse- 
 quent convictions for crime, or that, if any relation there be, it is those 
 who have received no school education who are the least frequently con- 
 victed; and, secondly, that convicts receiving the longest terms of im- 
 prisonment are those who have been industrial and reformatory school- 
 boys, and that those receiving the shortest terms are those who have not 
 been educated at any orthodox school. Our conclusion is that the kind of 
 school-education they may have received has no traceable influence upon 
 the subsequent careers of convicts; but that, since industrial and reforma- 
 tory school-boys must be the pick of those with the greatest law-breaking 
 proclivities, this accounts for the fact that convicts with the worst penal 
 records consist of those who have passed through industrial and reforma- 
 tory schools. 
 
 "With regard to age at leaving school, our conclusion is similar, '/and 
 is based upon the' partial correlation coefficients of age with penal record 
 for constant intelligence. . . . This conclusion is that their age, on leaving 
 school, has no appreciable influence upon the subsequent penal career of 
 convicts."' 
 
 His conclusions regarding the significance of wh'at he calls "effec- 
 tive" education are practically identical with the above. By effective.' 
 education he means "the apparent profit derived from whatever formal 
 education has been received," which he measures by " ( 1 ) the standard 
 or form reached by the subjects on leaving school, and (2) the grade 
 of education apportioned to each convict on his reception into prison 
 by the school-master." With the second of these measures we shall 
 not concern ourselves, since, as Goring points out, the effect of prison 
 schools undoubtedly enters in here. Accordingly the prison ex- 
 perience itself is being measured to some extent. With regard to 
 the relation of recidivism to the first-mentioned measure— grade at- 
 tained on leaving school — he concludes, on the basis of his data {op. 
 at., p. 276) that "with increasing number of convictions recorded 
 against convicts, there is no significant change in the' mean standard 
 attained by them on leaving school." Before considering this point 
 decisively established, however, he proceeds to eliminate the factor of 
 intelligence, through the medium of partial correlations, making intel- 
 ligence the constant factor. The partial correlation coefficient of 
 school education with number of convictions per year, for constant 
 
 ■ Op. cit.. p. 27S. 
 
EDUCATIONAL BACKGROUND 249 
 
 intelligence, is .0565 dz .0297; that of school education with fractions 
 of year imprisoned, for constant intelligence, is .0632 zt .0297. It is 
 evident that these small coefficients cannot be thought of as signifi- 
 cant. Accordingly Goring concludes (loc. cit., p. 276) "that the 
 profit derived from school education exerts no influence upon the 
 subsequent penal records of convicts." 
 
 LACK OF DATA ON EDUCATION FOR GENERAL POPULATION 
 
 The data offered by Goring are entirely convincing, as justifying 
 such conclusions as he permits himself to draw. He leaves completely 
 untouched, however, the question of how the "convict" group com- 
 pares with the general adult population with regard to either "formal" 
 or "effective" education. His reasons are doubtless the same as 
 those that we shall have to offer : viz., the lack of inf orfriatjon regard- 
 ing any random ' sample of the general population. ' Until we have 
 such data regarding the educational background of -adults, we shall 
 not be able to make a pretense of answering the question as to whether 
 the delinquent is -differentiated from the average adult by either a 
 deficit 'of schooling or an excess, or whether no difference at all can 
 be demonstrated. \ 
 
 The fact remains, unfortunately, that for us, as for Goring and 
 earlier investigators, the supplementary data are lacking which might 
 make possible a solution of the problem. To any one familiar with 
 the manifold recent studies of school systems and communities with 
 special reference to problems of retardation and elimination, this may 
 seem at first thought an unnecessarily pessimistic statement. Refer- 
 ence to such studies, however, will make clear the fact that they have 
 been concerned entirely either with children actually in school, or with 
 those leaving to go to work. These furnish no basis for estimating 
 the educational status of the general adult population. Accord- 
 ingly we shall be forced again into the position which we have had 
 to take so frequently during the course of the present study, of 
 finding it necessary to offer the facts regarding our group, without 
 being able to draw any final conclusions because of the absence of the 
 supplementary information about the total population. We shall fol- 
 low our usual practice, however, of presenting the inforrnation which 
 we have secured, in the hope that, at some future time, it may be of 
 service in carrying through the necessary comparisons. Its present 
 usefulness is largely confined to that of helping to describe our groups, 
 and of furnishing a background for the interpretation of the educa- 
 
250 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 tional status of an individual in the light of that of the whole delin- 
 quent group of which she is a part. 
 
 DISTINCTION BETWEEN EDUCATIONAL OPPORTUNITIES AND 
 EDUCATIONAL ATTAINMENTS 
 
 We shall approach the educational problem from two quite differ- 
 ent points of view, according as we think primarily of the educational 
 opportunities, or of the educational attainments of these individuals. 
 If we understand by educational opportunities genuine and not merely 
 potential opportunities, it would appear that, if we could isolate this 
 aspect of the total, we should have the best indication of the effect of 
 "schooling" per se. This would be the angle of approach, for ex- 
 ample, to the problem as to whether mere presence in school beyond 
 the age of fourteen tends to be a deterrent of delinquency — a conten- 
 tion which may be urged entirely without reference to the question 
 as to whether there is any important gain in educational attainments 
 during these years. Any measvre of educational achievement is com- 
 plicated by the fact that achievement is to a marked degree a com- 
 posite result of educational opportunity and the further factor of in- 
 nate mental capacity. Accordingly, in measuring an individual's at- 
 tainments, we may be measuring his mental capacity more than we are 
 measuring the effect of general school training or the lack of it. In 
 this chapter, therefore, we are quite certain to anticipate to some ex- 
 tent the discussion of the mental capacities of the women under con- 
 sideration. 
 
 As we shall see, it will be impossible to distinguish satisfactorily 
 between measures of opportunity and measures of attainment, and 
 equally impossible to eliminate from either of these measures the 
 effects of differences in mental capacity, in personality, and in emo- 
 tional trends. We can, however, make a rough distinction on the 
 basis of certain obvious implications of the data at hand. 
 
 Meaning by "opportunity" the amount and kind of schooling to 
 which one was actually "exposed," and not simply the chance for such 
 schooling, we should naturally consider under this head such facts 
 as the kind of school attended, the regularity of attendance, the age 
 at leaving school, and the total number of years in school. Only the 
 first of the above factors, however, — kind of school attended— is in 
 any thorough-going way independent of the characteristics of the per- 
 son concerned. We should expect that the other three factors would 
 be influenced by differences in mentality and in such traits as ambi- 
 
EDUCATIONAL BACKGROUND 251 
 
 tion, energy, and alertness, though probably not to the same degree that 
 one's attainments are affected. 
 
 The best measure of school attainment for our purposes appears 
 to be the school grade completed. Though comparisons with averages 
 of the general community are out of the question, we can compare 
 our data with the age-grade requirements set up by various school 
 systems, taking into account the age of leaving school as the age which 
 should determine the grade in the case of the delinquent women. We 
 can also interpret the grade attainment with reference to the number 
 of years actually spent in school, assuming that a year in school should 
 net one grade completed. 
 
 Even without reference to these determining factors of age at 
 leaving school, or of number of years in school, the grade attained 
 has some descriptive value as an absolute measure of school accom- 
 plishment. In a general way, at least, the difference between having 
 reached third grade, eighth grade, high school or college is significant 
 of a difference in degree of academic education, entirely apart from 
 the question as to whether the possibility of attaining one or another 
 of these may not have been determined by the degree of mentality. 
 
 A still further means of measuring educational accomplishment is 
 through a determination of the amount of the educational equipment 
 which one has retained from her school experience. For this pur- 
 pose we have made use of several varieties of educational tests which 
 have recently been developed. The type of information furnished 
 by these differs in two important ways from that afforded by school 
 grade as an index of educational achievement. In the first place, it 
 is more objective, since it is in no sense dependent on the subject's 
 own statement of her attainment, as is the school grade for a large 
 number of our cases where verification has proved impossible. In 
 the second place, it affords an insight into the present educational 
 working equipment of the individual, which may not be indicated 
 adequately by the statement concerning the grade reached in school. 
 Either of two influences may have been operative, tending in quite 
 different directions. That is, the subject may have either dropped 
 from the degree of ability represented by her last school grade, be- 
 cause of entire absence of practice, or she may have proceeded to reach 
 a much higher level of accomplishment by virtue of her own initiative 
 and ability stimulated by special demands of her environment. Which- 
 ever alteration occurs, the measure of present attainment is of con- 
 siderable interest since it is the present individual that we are con- 
 
252 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 cerned with, at least in so far as we are planning for her institutional 
 treatment or her later adjustments. 
 
 It is important, however, to recognize that the present status of 
 educational attainments is likely to be even more closely associated 
 with mental capacity than is the grade actually reached in school. 
 The latter is subject to the various artificial factors operative to move 
 all children through the schools at the same pace, provided differ- 
 ences in ability' be not too striking. The amount retained from the 
 school training and, still more, the additional amount built upon this 
 should tend to differentiate more exactly the bright individual from 
 the dull. In so far as this is true we are obviously measuring men- 
 tality by these educational tests, quite as much as we are measuring 
 educational factors per se. 
 
 VERIFICATION OF SCHOOL RECORD 
 
 The general process of verification of records was explained in 
 Chapter V. The most satisfactory means of verifying school records 
 was by writing to the school and obtaining the individual's record card. 
 To accomplish this necessitated, in the first place, having precise in- 
 formation regarding the school. It was not enough to know, for ex- 
 ample, that the woman went to school in the Bronx or on the East 
 Side, New York. The number and location of the school must also 
 be obtained. In many cases, a woman was inaccurate in recalling these 
 details, either purposely or because she had not thought of her school 
 for many years. Accordingly it frequently required repeated efforts to 
 obtain the desired information. In the second place, it was essential 
 that the school in question should have the facts. In many smaller 
 towns and country places there is no record-keeping system. Unless, 
 then, the investigator could find some teacher who had known the 
 woman when she was in school there was no way of verifying her 
 statement. Even in New York City there was no systematic record- 
 keeping until 1909. Accordingly, the record cards were usually in- 
 complete for the earlier school years of our cases. With the women 
 who had had their schooling in foreign countries there was practically 
 no possibility of verification except through the statements of relatives. 
 
 With these conditions in mind, it is not surprising to find that veri- 
 fication was obtained for only about half of our total group. The 
 wide differences in degree of verification of the various institutional 
 groups may be indicated by the following figures showing percentages 
 of verification : 
 
EDUCATIONAL BACKGROUND 253 
 
 Grade Finished: Percentage of Records Verified 
 
 Bedford Auburn Magdalen Penitentiary Workhouse Probation Total 
 
 n 38 72 27 14 42 46 
 
 It is evident that a particularly high degree of reliability attaches to 
 the figures on Bedford and the Magdalen and that, on the other hand, 
 the Workhouse figures amount to little more than the women's own 
 unverified statements. 
 
 With a view to determining what difference, if any, was likely to 
 exist between verified data and data based solely on the women's own 
 statements, we computed the correlation of verified with unverified 
 data on school grade for 207 cases and found the coefficient to be .85. 
 The mean grade based on verified records was found to be 4.6, whereas 
 the mean as given by the women's statements alone is 5.3. It is evi- 
 dent that the tendency is toward exaggeration, toward making a better 
 impression. It must be remembered that the above figures are neces- 
 sarily based simply on comparison of verified data with those original 
 statements on •mhich additional information was later obtained. Ob- 
 viously, we can not compare the original statements with verified rec" 
 ords where the latter were never obtained. It is, therefore, impossible 
 to state whether the tendency to exaggerate holds to the same degree for 
 the whole group. It might be assumed that those statements on the 
 basis of which it was impossible ever to obtain a verified record would 
 have shown an even greater discrepancy from the actual facts. An- 
 other factor, however, enters in since an appreciable portion of the 
 unverifiable statements were those reporting little or no schooling, 
 especially as given by the foreign group. To take this factor into ac- 
 count, we have compared the means, based on the original statements 
 of all cases on whom we had any records, with the means, based on 
 verified data of all cases on whom we had verified records, with the 
 following results: 
 
 Mean School Grade Based on 
 
 Women's Statements Verified Records 
 
 4.88 ± .135 4.63 ± .178 
 
 (437 cases) (211 cases) 
 
 It is apparent, therefore, that the mean school grade as obtained from 
 the women's statements alone is lower when we consider the entire 
 group than when we consider that portion of the group whose records 
 it was possible to verify (4.9 as against 5.3). In other words, the 
 mean is actually lowered, as we assumed it might be, by the addition 
 of the cases whose records were unverifiable. Even so, the mean 
 
254 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 based on women's statements alone is slightly higher than the mean 
 based on verified records (4.9 as against 4.6). 
 
 By combining the verified data for all cases on whom this was 
 available with the data based on original statements for the remainder 
 of the group, we have the best available information on schooling for 
 our group as a whole. This is the type of data used, therefore, 
 throughout this chapter except in the few instances where we have 
 confined the discussion to the verified data. From Table 87 we find 
 that the mean grade based on the best available records for 447 cases 
 is 4.58±.127, which is practically identical with the mean given by 
 verified records alone on the 211 cases (4.63 + . 178). If we may as- 
 sume that there was some degree of exaggeration in the additional 
 236 records which were based only on women's statements, it is evi- 
 dent that the true mean for the group would be slightly lower still. 
 In other words, our data may be interpreted as probably giving, in 
 all cases, a slightly higher mean value for school grade than the facts 
 would warrant. On the other hand, they are lower by a slight but 
 appreciable degree than data based on the women's statements alone. 
 The latter point should be borne in mind when our data are compared 
 with other data where there has been no verification. 
 
 EDUCATIONAL OPPORTUNITIES 
 
 (a) Kind of School Attended 
 
 Our first consideration under the head of educational opportuni- 
 ties is that of kind of school attended. (See Table 80.) While the 
 various types of school doubtless represent, to some extent, differences 
 in kind of opportunity, we have not attempted to evaluate these differ- 
 ences. The largest percentage in each group have had public school 
 education. This is to be expected, since by far the largest portion 
 of the population has been trained in the public schools. Bedford has 
 the largest percentage of public school cases and Auburn the smallest, 
 the other groups showing percentages almost identical with one an- 
 other. The per cents from the parochial schools vary from 2.2 per 
 cent in the Penitentiary to 16.2 per cent in Auburn. We see no ex- 
 planation for so marked a variation of these two extremes, since 
 both institutions are necessarily non-sectarian, both receive relatively 
 older cases for the most part, and, moreover, in the two institutions 
 the range of offenses is fairly similar. There is still more striking 
 variation among the institutions regarding numbers educated en- 
 
EDUCATIONAL BACKGROUND 
 
 255 
 
 tirely in foreign schools, Bedford having much the lowest per cent 
 — 3.1 — and Auburn the largest — 33.8. This difference appears mainly 
 explicable in terms of the relative numbers of foreign cases in the 
 various groups. (See Chapter VIII.) The data concerning foreign 
 schools are of significance as showing that an appreciable number of 
 women, in all groups except Bedford, have had all their schooling 
 under markedly different social circumstances from those of the school 
 systems of this country. Moreover, in the cases where attendance in 
 foreign schools is reported, it is particularly likely to have been 
 so brief and interrupted as to amount to little. The footnote to the 
 table shows the numbers who have never attended school. It is no- 
 ticeable that the institutions having the younger women have the few- 
 est of such cases — Probation none, Magdalen 3, Bedford 4 — indicating 
 that, in this section of the country at least, complete absence of school 
 training is not likely to be an important matter in the future. 
 
 TABLE 80 
 
 KIND OF SCHOOL ATTENDED* 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups 
 
 
 Institutional Groups 
 
 
 Kind of School 
 
 Bed- 
 ford 
 
 Auburn 
 
 Mag- 
 dalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Proba- 
 tion 
 
 Total 
 
 
 7.2 
 6.2 
 2.0 
 
 77.3 
 
 3.1 
 4.1 
 
 1.5 
 16.2 
 
 47! i 
 
 33.8 
 
 1.5 
 
 5.8 
 4.3 
 
 63]8 
 26.1 
 
 2.2 
 
 2.2 
 
 2.2 
 
 64.0 
 
 27.0 
 
 2.2 
 
 3.3 
 
 8.7 
 
 4.3 
 
 62.0 
 
 21.7 
 
 3.5 
 
 11.8 
 
 3.5 
 
 62.4 
 
 10.6 
 
 8.2 
 
 
 Orphauage or Home 
 
 Parochial 
 
 4.0 
 8.0 
 
 Private 
 
 2.2 
 
 Public 
 
 63.6 
 
 Schools in Foreign Countries 
 
 19.4 
 
 Schools in Foreign Countries 
 and Public Schools in U. S. 
 
 2.8 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 
 
 
 97 
 
 68 
 
 69 
 
 89 
 
 92 
 
 85 
 
 500 
 
 
 
 *The following numbers in the various groups never attended school: Bedford, 4; Auburn, 12; 
 Magdalen, 3; Penitentiary, 15; Workhouse, 7; Total, 41. 
 
 (b) Regularity of Attendance 
 
 Of importance also in estimating the influence of schooling is in- 
 formation concerning regularity of attendance. Such information is 
 
256 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 peculiarly difficult to secure in the case of these older women, many 
 of whom have only the vaguest of recollections as to their behavior 
 with reference to the school situation. However, we have summa- 
 rized in Table 81 the information available through school reports or 
 parents' statements on this point. It is needless to point out that 
 standards of judgment are far from uniform and that, accordingly, no 
 careful comparisons can be made. We note that in each institutional 
 group the largest percentage are described as having been "fair" in 
 attendance, though a goodly number are referred to as "poor." T\k 
 groups vary considerably in the per cents described as "very poor" 
 but in general these numbers are low. Much lower still are the per 
 cents to whom a record of "very good" attendance is attributed. It 
 is of interest to note that the group is not so predominantly charac- 
 terized by very poor school attendance as we might have anticipated. 
 How they compare in this respect with an average group of adults 
 we have no means of saying. 
 
 TABLE 81 
 
 ATTENDANCE IN SCHOOL 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups 
 
 
 Institutional Gkoufs 
 
 
 Attendance 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Probation 
 
 Total 
 
 Very poor 
 
 Poor 
 
 14.4 
 
 29.9 
 
 62.6 
 
 3.1 
 
 13.8 
 32.3 
 53.8 
 
 3.1 
 23.1 
 69.2 
 
 4.6 
 
 1.3 
 24.1 
 72.2 
 
 2.5 
 
 7.0 
 15.1 
 
 77.9 
 
 4.9 
 28.4 
 66.7 
 
 7.6 
 25.3 
 
 Fair 
 
 65.4 
 
 Very good 
 
 1.7 
 
 Total. .. 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cases 
 
 97 
 
 65 
 
 65 
 
 79 
 
 86 
 
 81 
 
 473 
 
 (c) Age at Leaving School 
 
 We have indicated that both age at leaving school and number of 
 years in school may be considered indices of amount of school oppor- 
 tunity. For comparisons of homogeneous communities, with definite 
 and well-enforced standards regarding age at entering school and 
 
EDUCATIONAL BACKGROUND 257 
 
 without influx of large numbers of children — especially foreign cases 
 — who enter school late, the two measures may be used fairly inter- 
 changeably as measures of school contacts. The child who leaves 
 school at fourteen years may ipso facto be thought of as having had 
 less schooling than the one who leaves at sixteen, and more than the 
 one who leaves at twelve. Where, however, the stationary conditions 
 suggested above have not existed in a community the age of leaving 
 school becomes a far more ambiguous measure of school opportunity 
 than the figure which gives the actual time spent in school. 
 
 Nevertheless data on the age of leaving school can not be neg- 
 lected, in the light of the importance attached to this factor in con- 
 nection with the problems of compulsory education laws and of 
 retardation. As a result of the interest in these directions during 
 recent years we have before us numerous statements as to what should 
 be the minimal standards of a modern community. These are higher 
 than the actual records of even the more advanced communities, for the 
 most part, and so can not be used as a means of comparing the de- 
 linquent group with the general population. 
 
 In the development of compulsory education and child labor laws 
 there has been almost universal acceptance of the view that the age 
 of fourteen is the lowest age at which dropping out of school should be 
 tolerated. By most of those who deal directly with children leaving 
 school to go to work this is considered too low a standard. We hear 
 of the "wasted years" from fourteen to sixteen. In certain states the 
 minimum age for leaving has been moved on to fifteen or sixteen, 
 either absolutely or with some accompanying grade requirement. Thus 
 in New York State, which concerns us most closely, the Compulsory 
 Educational Law has required, since 1917, that in order to leave at 
 fourteen years of age a child must have completed eighth grade; in 
 order to leave at fifteen he must have completed sixth grade; if he 
 has not completed sixth grade he may not leave school until he is six- 
 teen. 
 
 (1) Data on Institutional Groups and on Total. — Table 82 shows 
 the percentages leaving school at various ages, for the separate insti- 
 tutional groups and for the total. Women who never attended school 
 are not included in this table, since it is obviously impossible to state 
 their age at leaving school. The percentages of cases who never at- 
 tended school may be obtained from Table 84. The age of leaving 
 school as given is understood to be the nearest age. For example, any 
 age falling between nine and a half and ten and a half is given as ten. 
 
258 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 This custom has been followed at least where the age has been figured 
 precisely from records. In cases where verbal statement has been 
 relied upon, the informant may be assumed to have performed a 
 rougher process of the same sort in turning the age into round num- 
 bers. In comparing the groups it is well to remember the discrepancies 
 in chronological age between the Bedford, Magdalen and Probation 
 groups on the one hand, and the Auburn, Penitentiary and Workhouse 
 groups on the other. In each of the latter three there are considerable 
 numbers of women who had their schooling at a time when com- 
 pulsory education requirements concerning age at leaving school were 
 less rigid than at present. (For the range of ages of the various 
 groups see Table 31.) 
 
 TABLE 82* 
 
 AGE AT LEAVING SCHOOL 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups 
 
 Age at 
 
 Institutional Groups 
 
 
 Leaving 
 School 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Probation 
 
 Total 
 
 7 
 
 'i!6 
 'i.x 
 
 7.3 
 
 7.3 
 
 27.1 
 
 29.2 
 
 20.8 
 
 2.1 
 
 2.1 
 
 1.0 
 
 3.1 
 
 "■i.i 
 
 9.4 
 
 7.8 
 
 10.9 
 
 18.8 
 
 20.3 
 
 10.9 
 
 7.8 
 
 4.7 
 
 1.6 
 
 1.6 
 
 'i!5 
 
 2.9 
 
 2.9 
 
 4.4 
 
 7.3 
 
 11.6 
 
 30.4 
 
 24.6 
 
 13.0 
 
 i'.'h 
 
 "\.2 
 3.7 
 3.7 
 
 8.7 
 
 J.8.5 , 
 
 25.9 
 
 16.1 
 
 11.1 
 
 6.2 
 
 i'.b. 
 
 2A 
 
 3.5 
 
 17.6 
 
 . 11.8 
 
 15.3 
 
 28.2 
 
 8.2 
 
 5.9 
 
 3.5 
 
 1.2 
 
 1.2 
 
 ■2!4 
 
 2.4 
 
 18.1 
 
 35.0 
 
 26.5 
 
 7.2 
 
 6.0 
 
 2.4 
 
 4 
 
 8 
 
 2 
 
 9 
 
 1 5 
 
 10 
 
 2 7 
 
 11 
 
 3 8 
 
 12 ,. 
 
 13 
 
 9.0 
 14 
 
 14 
 
 25 7 
 
 15 
 
 23 2 
 
 16 
 
 11 7 
 
 17 
 
 4 2 
 
 18 
 
 1 9 
 
 19 
 
 1 5 
 
 20 
 
 2 
 
 
 
 Total . . . 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cases 
 
 96 
 
 64 
 
 69 
 
 81 
 
 85 
 
 83 
 
 478 
 
 Mean 
 
 14.57 
 ±.158 
 
 13.16 
 ±.304 
 
 13.83 
 ±.224 
 
 14.14 
 ±.226 
 
 14.17 
 ±.223 
 
 14.39 
 ±.148 
 
 14.10 
 ±.088 
 
 
 
 ^ 
 
 1.55 
 ±.112 
 
 2.43 
 ±.215 
 
 1.87 
 ±.159 
 
 2.04 
 ±.160 
 
 2.05 
 ±.157 
 
 1.35 
 ±.105 
 
 1.93 
 ±.062 
 
 (ja. \ 
 
 
 ♦This table includes only those cases who have attended school (or at least a part of one year. 
 
EDUCATIONAL BACKGROUND 259 
 
 Referring to the table it may be noted that all the groups show an 
 appreciable number who left school before they were fourteen; in 
 other words, before the age which has now practically universal ac- 
 ceptance in this country as the minimum age at which dropping out 
 of school should be allowed. For the total group the percentage 
 who left before fourteen is 31.6, or approximately one-third of the 
 group who actually attended school. (Reference to Table 84 shows 
 that 8.8 per cent of the total group had never attended school at all.) 
 For the three groups made up of younger women the percentages 
 leaving school before fourteen are as follows: Bedford 17.7 per cent, 
 Magdalen 30.6 per cent, and Probation 22.9 per cent. The relatively 
 large percentage which appears for the Magdalen is shown by refer- 
 ence to the data to be due mainly to the large percentage of foreign 
 cases in this group. The three groups which contain considerable 
 numbers of older women give the following percentages: Auburn 
 53.1, Penitentiary 35.8, Workhouse 36.5. With reference to the very 
 high percentage shown for Auburn it may be noted that there are 
 not only many older women here but also large numbers of foreigners 
 and a fair number of women from isolated country districts with poor 
 school facilities. Table 82 shows the mean age at leaving school for 
 the total group to be 14.1 years: for the various institutional groups 
 the means range from 13.16 for Auburn to 14.57 years for Bedford. 
 
 (2) Comparatwe Data on Non-Delinquent Groups. — The fore- 
 going figures are sufficient to show that the group of delinquent 
 women fail to meet the demands of present-day public opinion 
 regarding minimum age for leaving school. Whether they fall 
 below the actual standards of the adult community, as represented 
 by the facts of their school records, we have no means of knowing. 
 The establishing of the facts regarding the ages of ehmination of 
 children dropping out of school within recent periods of time has not 
 proved a simple problem. With the merits of the controversy center- 
 ing about this point we can not here concern ourselves.* There has 
 been agreement as to the fact that the source of the difficulty lies in the 
 lack of facts regarding individual school histories and the necessity 
 of making inferences from records of school attendance, after cor- 
 recting for numerous changing conditions, such as death-rate and 
 increase of population. No one has had the temerity, apparently, to 
 
 ' See especially Thorndike, E. L. "The Elimination of Pupils from School." 
 Bulletin No. 4, 1907. Whole No. 379. Department of the Interior, Bureau of 
 Education. 
 
 Ayres, L. P. "Laggards in Our Schools, ' 1909. 
 
260 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 attempt to estimate the ages of elimination for the general adult popu- 
 lation. Accordingly our only available comparison is that with the 
 disputed figures representing the estimates of the ages of elimination 
 of children during fairly recent periods. It will be remembered that 
 these figures vary not only with the method of estimating, but with the 
 character of the communities considered as well. 
 
 We may consider the following figures representative of estimates 
 which have received serious consideration: 
 
 (1) Thorndike's estimate of the numbers remaining in school at 
 the various ages, for every hundred children who were in school at 
 eight years of age, for cities of 25,000 or over in 1900. (Op. cit., 
 p. 23.) 
 
 Thorndike's figures, reversed to show percentages leaving at given 
 ages, rather than those remaining,^ are as follows: 
 
 Age at leaving school . .8 9 10 11 12 13 14 IS 16 17 18 or over. 
 Percentages leaving 2 1 9 18 23 17 13.5 7.9 8.6 
 
 (2) Ayres' figures on decline in attendance, summarized for 58 
 cities, assuming a stationary population. (Op. cit., Table 11, p. 28 and 
 Table 12, p. 30.) 
 
 Ayre's figures taken from Table 12 and expressed as percentages 
 leaving school at given ages are as follows : 
 
 Age at leaving school 11 12 13 14 IS 16 or over 
 
 Percentages leaving 2 9.8 30 3S 17 8 
 
 (3) Ayres' figures on the estimated ages of elimination for Cleve- 
 land, Ohio.^° While these are more sectional in their interest than are 
 the figures from the other groups, they have an advantage for that 
 reason, in representing a definite community and not a composite of 
 several. It should be remembered, however, that it represents a de- 
 cidedly high-grade standard, as compared with the country as a whole, 
 since the data are recent and should show the influence of the present 
 compulsory education law of Ohio, which requires that boys remain 
 in school until fifteen and girls until sixteen. (We are not concerned 
 
 ° In reversing the figures, offered by Thorndike and by Ayres, our procedure 
 has been as follows r If a given per cent, are indicated as remaining in school 
 tilt a given age, it is assumed that the difference between this and 100 per cent 
 had left before this age. For example, if 100 per cent were retained until ten 
 years and 98 per cent until eleven, we infer that 2 per cent left school at ten 
 years, i. e., during the period that they were counted ten years old^'" 
 
 Ayres, L. P. "Child Accounting in the Public Schools." Cleveland Edu- 
 cation Survey, 1915, p. 28 and p. 34, table 4. 
 
EDUCATIONAL BACKGROUND 261 
 
 here with the evidence which the figures oflEer to the effect that the 
 law is by no means adequately enforced.) 
 
 The figures for Cleveland as offered by Ayres are as follows : 
 
 Age at leaving school 12 13 14 15 16 17 18 19 20 or over. 
 
 Percentages leaving 3 14 30 20 12 9 8 3 1 
 
 The three estimates presented above evidently offer very different 
 standards of comparison. Thus it appears by Thorndike's estimate 
 that 30 per cent leave school before fourteen, by Ayres' estimate for 
 fifty-eight cities that 40 per cent leave before this age, and -by Ayres' 
 estimate for Cleveland that 17 per cent drop out before fourteen. 
 In the light of the discrepancies and disagreements we can not feel 
 that the figures are very convincing. If we take them at their face 
 value our delinquent groups, with the exception of Auburn, do not 
 show up badly, except in comparison with the Cleveland data. Even 
 in this comparison the Bedford group appears favorably, with only- 
 17.7 per cent leaving before fourteen. 
 
 In the absence of more conclusive data we are forced to consider 
 ihe relationship of the delinquent women to the remainder of the 
 population, with respect to age at leaving school, an open question. 
 The implications of the data before us are to the effect that, while 
 the delinquent women have not met the standards of present opinion 
 regarding the minimal age for leaving school, they are probably not 
 seriously behind the population as a whole in this respect. All the 
 arguments which apply to the population as a whole, to show the 
 harmfulness and loss involved in a high rate of elimination in the 
 earlier years, apply also to the delinquent group. But in the absence 
 of a clear distinction between the delinquent women and the general 
 population we lack evidence to show that an early age of leaving school 
 is one of the determinants of delinquency. 
 
 (3) Relationship Between Age at Leaving School and Age at First 
 Conviction. — Turning from comparison of the delinquent group with 
 other groups to consideration of relationships within the delinquent 
 group, special interest attaches to the relationship between age at leav- 
 ing school and age of first conviction. It would appear quite possible 
 that an early age of leaving school might be associated with an early 
 age of first conviction. This comparison is shown in Table 83 for 
 those of our total group who have actually attended school at some 
 time. Inspection of the table fails to show any striking relationship. 
 The coefficient of correlation is —. 03 ±.046, indicative of entire absence 
 
262 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 of relationship. The correlation ratios were computed and found to be 
 .2Sih.043 and .21±.044 respectively. Even this slight evidence of rela- 
 tionship can not, howeer, be taken at its face value because of the lack 
 of consistent trend in the lines of means, at least in case of the ratio 
 of age at leaving school on age of first conviction. In the case of the 
 ratio of age at first conviction on age at leaving school, reference to 
 the means given in the extreme right-hand column of the table shows 
 the suggestion of a curious relationship, to the effect that those who 
 left school at ages not far from the average — especially the ages from 
 13 to 16 — tend to have been convicted at slightly earlier ages than 
 either those who left school very young or who remained in school 
 to ages considerably above the average. It is not surprising to find 
 the tendency toward later first conviction with later age at leaving 
 school. Aside from any question of increased advantage to the in- 
 dividual, accruing from longer school contacts, there is a tendency 
 toward setting up a lower limit here. That is, except in case of the 
 early juvenile convictions, a conviction is very likely to interrupt and 
 thereby terminate the school career. Accordingly, the continuance in 
 school beyond the age of sixteen or seventeen years is likely to be 
 somewhat symptomatic of the fact that such interruption has not 
 occurred. Examination of the table suggests the operation of this 
 tendency. Reasons for the delayed age of first conviction of those who 
 left school very early are not apparent from the table. Consideration 
 of the facts regarding the foreign members of the group affords, how- 
 ever, a clue. It has been shown (see pp. 196-197) that these cases had a 
 later average age of first conviction than the native-born. It is also 
 known that their educational opportunities have been much slighter 
 (see p. 265). Reference to our records shows that the foreign-born 
 form the large majority of the group who left school earlier than thir- 
 teen. This factor, therefore, evidently accounts for the negative trend 
 in the lower portion of the curve. In view of the slightness of the 
 relationship indicated between age at leaving school and age at first 
 conviction, and on consideration of the complication of the situation 
 by other factors, it is apparent that we can offer no evidence to show 
 that age at leaving school has operated, either positively or negatively, 
 as a determinant of the age at first conviction. 
 
 We shall defer until our consideration of grade attainments the 
 discussion of age at leaving school in relation to grade, which involves 
 the problem of retardation. 
 
s 
 
 I 
 
 ■s 
 
 o 
 <! 
 
 ■i 
 
 ■s 
 
 89 0*09 
 
 09 0% 13 
 
 i9 o» f9 
 
 »SO» TB 
 
 19 o* 8* 
 
 8» o* S* 
 
 9» o» Z» 
 
 Z* o* 68 
 
 98 o* E£ 
 
 88 0% 08 
 
 OS o» IZ 
 
 lZO%fz 
 
 ?ZO*TZ 
 
 IE ot 8t 
 
 81 o% St 
 
 SI oi z\ 
 
 ZI0»6 
 
 6 o»9 
 
 OS t^ CO :d eo iH 00 ^ t^ lO 
 
 -it»Osg«DT-*Mt*MM«t*iH« 
 W lO fH M CO ^ rt T-l 
 iHtH 
 
 r-l tH tH d i-l W iH 
 
 I-* « CO W Tjt ^ iH rH 
 
 iH tHMfHeONCOrHeC- 
 
 •HrHCOCDlONOaeONi-H 
 
 N i-H Ttl CO (D a CO »H 
 
 NCCr^OOrtCO-^COINrHrH 
 
 O01«t*«DiOTj<«(N^O01Mt- 
 
 II II 
 b b 
 
 II II 
 
 £ * 41 
 
 la's o 
 
 II II 
 
 a a 
 ^1 
 
 MM 
 
 is, 
 
 •II fl i 
 
 ^•-•§ 
 II 11^ 
 
 U 
 
 1-2 « 
 ^ .S •" 
 " P s 
 
 a O F— < 
 
 -«£ 
 
 w o a Tj 
 
 IS 
 
 <!<! 
 
 %a J 
 
 §^-«! a 
 
 Is ■§ 
 
 a be 
 
 ]Ooqog 3aiAB3^ !(6 d3y 
 263 
 
264 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 (d) Number of Years in School 
 
 (1) Data on Total Delinquent Group and on Institutional Sub- 
 Groups. — Table 84 presents data regarding number of years in school 
 by institutional groups, the distribution of which are shown graphi- 
 cally in Chart XV. These distributions include, in all cases, those 
 women who never attended school as well as those who were in school 
 for varying periods. 
 
 TABLE 84 
 
 NUMBER OF YEARS IN SCHOOL 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups 
 
 Number 
 
 Institutional Groups 
 
 
 of 
 Years 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Probation 
 
 Total 
 
 
 
 4.0 
 
 2.0 
 
 1.0 
 
 3,0 
 
 3.0 
 
 5.1 
 
 7.1 
 
 16.2 
 
 28.3 
 
 15.2 
 
 9.1 
 
 2.0 
 
 3.0 
 
 1.0 
 
 16.0 
 5.3 
 
 2.7 
 5.3 
 8.0 
 4.0 
 10.7 
 26.7 
 6.3 
 5.3 
 6.7 
 2.7 
 1.3 
 
 4.2 
 
 4.2 
 
 1.4 
 
 1.4 
 
 2.8 
 
 9.9 
 
 11.3 
 
 16.9 
 
 25.4 
 
 12.7 
 
 4.2 
 
 5.6 
 
 17.6 
 
 '4!4 
 
 4.4. 
 
 5.5 
 
 3.3 
 
 11.0 
 
 13.2 
 
 16.5 
 
 11.0 
 
 5.5 
 
 4.4 
 
 2.2 
 
 10.3 
 
 3.5 
 
 3.5 
 
 2.3 
 
 5.8 
 
 6.9 
 
 13.8 
 
 12.6 
 
 14.9 
 
 13.8 
 
 5.8 
 
 3.5 
 
 3.5 
 
 ■2!6 
 
 2.6 
 
 5.1 
 
 5.1 
 
 12.8 
 
 25.6 
 
 24.4 
 
 11.5 
 
 3.9 
 
 6.4 
 
 8 8 
 
 1 
 
 2 i 
 
 2 
 
 2 6 
 
 3 
 
 3 2 
 
 4 
 
 5 
 
 6 
 
 5 6 
 
 6 
 
 11 
 
 7 
 
 18 2 
 
 8 
 
 19 4 
 
 9 
 
 11 8 
 
 10 
 
 6 
 
 11 
 
 4 
 
 12 
 
 1 8 
 
 13 
 
 2 
 
 14 
 
 2 
 
 
 
 Total.... 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cases 
 
 99 
 
 75 
 
 71 
 
 91 
 
 87 
 
 78 
 
 501 
 
 Mean 
 
 7.31 
 + .267 
 
 5.32 
 ±.384 
 
 6.79 
 ±.312 
 
 5.95 
 
 ±.377 
 
 6.26 
 ±.348 
 
 7.23 
 ±.223 
 
 6.50 
 ±.136 
 
 (Tm 
 
 
 a 
 
 2.65 
 ±.189 
 
 3.33 
 
 ±.272 
 
 2.63 
 ±.221 
 
 3.60 
 ±.267 
 
 3.25 
 
 ±.246 
 
 1.97 
 ±.158 
 
 3.052 
 ±.096 
 
 a<T 
 
 
 All groups show a wide range of years spent in school. In the 
 total group this extends from no years to fourteen years. The dis- 
 persion as measured by the standard deviation is appreciably higher for 
 the three groups containing older women. The average time spent in 
 
EDUCATIONAL BACKGROUND 
 
 265 
 
 school for the total group is 6.5 years. The three groups containing 
 on the whole younger women show higher average periods of attend- 
 ance than do the other three, though the dividing lines are not sharp. 
 Bedford and Probation show the highest average time spent in school 
 (7.31 and 7.23 years respectively) and Auburn the lowest (5.32 years). 
 The factors cited as explanatory of the low average age of leaving 
 school of the Auburn women apply also here. 
 
 The numbers who have never been to school vary from per cent 
 for the Probation group to 17.6 per cent for the Penitentiary women. 
 Of the total group 8.8 per cent had never attended school. Again 
 the three groups of younger women make a much better showing than 
 do the other three. 
 
 (2) Data on Groups Classified by Nativity and Color. — Com- 
 parison by nativity and color, rendered possible by Table 85, shows 
 that the foreign bom portion of the group contributes by far the 
 largest number of those very deficient in school opportunity as meas- 
 ured by number of years in school. Practically one-fourth (24.5 per 
 cent) have had no schooling, and the mean number of years in school 
 is only 4.92. The group which falls next with respect to lack of 
 schooling, the native colored, is separated from the foreign born by 
 a wide gap. They show 4.7 per cent with no schooling, and a mean 
 number of years in school of 6.42. When these two elements— the 
 foreign-born and the negro — are eliminated from the total, we find 
 
 Pet 
 
 Cent 
 
 TOTAL 
 
 YEARS 
 
 Chart XV 
 
 Number of Years in School 
 
 Per cent distribution by institutional groups. 
 
Per 
 
 Cent 
 
 BEDFORD 
 
 ' . ' J ' 3 ' , ' 5 ' 6 ■ 7 ■ 8 ■ 9 -.0 " il • « ■ .3 ■ .4 ■ 
 
 xvr E 
 
 Per 
 Cent 
 
 AUBURN 
 
 xvi^ 
 
 Per 
 Cent 
 
 MAGDALEN 
 
 12 3 4 S e 7 8 9 10 11 12 13 14 
 
 YEARS 
 
 Chart XV (Continued) 
 
 Number of Years in School 
 
 Per cent distribution by institutional groups. 
 
 366 
 
Per 
 
 Cent 
 
 PENITENTIARY 
 
 Per 
 
 Cent 
 
 XVI^ 
 
 WORKHOUSE 
 
 Per 
 Cent 
 
 PROBATION 
 
 -T— P— I— ^i— T^-!«— -P— » I I ^— ^— ^ 
 
 12 3 4 5 8 7 8 g 10 11 12 13 14 
 
 YEARS XVl-( 
 
 Chart XV (Continued) 
 
 Number of Years in School 
 
 Per cent distribution by institutional groups. 
 
 267 
 
268 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 the native-white making a far more satisfactory showing. Only 0.8 
 per cent show no record of school attendance, and the mean number 
 of years in school is 7.47— a mean which is almost high enough to 
 permit of an average record of graduation from elementary school, 
 assuming normal progress. We shall see however that the record falls 
 decidedly short of this possibility. 
 
 TABLE 85 
 
 NUMBER OF YEARS IN SCHOOL 
 
 Per Cent Distribution of Delinquent Women Classified by Nativity 
 
 and Color 
 
 
 NATtVITT AND COLOB 
 
 
 Years in School 
 
 Total 
 Foreign Born 
 
 Total 
 Native Bom 
 
 Native 
 White 
 
 Native 
 Colored 
 
 Total 
 
 
 
 1 
 
 24.5 
 5.2 
 1.9 
 5.8 
 5.2 
 6.5 
 5.8 
 14.8 
 13.5 
 8.4 
 4.5 
 1.9 
 1.9 
 
 1.7 
 
 1.2 
 
 2.9 
 
 2.0 
 
 4.9 
 
 5.2 
 
 13.3 
 
 19.7 
 
 22.0 
 
 13.3 
 
 6.6 
 
 4.9 
 
 1.7 
 
 .3 
 
 .3 
 
 .8 
 
 .4 
 
 .8 
 
 1.9 
 
 4.2 
 
 3.8 
 
 14.2 
 
 21.2 
 
 25.0 
 
 15.8 
 
 6.8 
 
 4.6 
 
 1.2 
 
 .4 
 
 4.7 
 
 3.5 
 
 9.3 
 
 2.3 
 
 7.0 
 
 9.3 
 
 10.5 
 
 15.1 
 
 12.8 
 
 6.8 
 
 9.3 
 
 5.8 
 
 3.5 
 
 "\.b. 
 
 8.8 
 2.4 
 
 2 
 
 2 6 
 
 3 
 
 3.2 
 
 4 
 
 5.0 
 
 5 
 
 5.6 
 
 6; 
 
 11.0 
 
 7 
 
 18.2 
 
 8 
 
 19 4 
 
 9^ 
 
 11.8 
 
 10 
 
 6.0 
 
 11 
 
 4 
 
 12 
 
 1 8 
 
 13 
 
 .2 
 
 14 
 
 2 
 
 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100 
 
 
 
 Number of cases 
 
 155 
 
 346 
 
 260 
 
 86 
 
 501 
 
 Mean 
 
 4.92 
 ±.294 
 
 7.21 
 ±.130 
 
 7.47 
 ±.125 
 
 6.42 
 ±.351 
 
 6 50 
 
 (Tm 
 
 + 136 
 
 
 
 a 
 
 3.66 
 
 ±.208 
 
 2.42 
 ±.092 
 
 2.01 
 
 ±.088 
 
 3.25 
 
 ±.248 
 
 3.05 
 + .096 
 
 (Ja ; 
 
 (3) Impossibility of Comparison with General Population.— In 
 spite of the fact that we attach more importance to figures showing 
 years in school than to those showing age at leaving, as measures 
 of amount of schooling, the former are even less serviceable than the 
 latter for purposes of comparison with other groups. Even estimates 
 are lacking on this point. We are therefore forced to refrain from 
 
5 ^ i 9 
 
 iiii 
 
 I 
 
 m 
 
 
 
 ca 
 
 I 
 
 s.a 
 
 H Q 
 
 
 a 
 
 a> 
 m 
 
 I 
 I 
 
 4) 
 g 
 
 I 
 
 CO 1-1 03 GO to «D C4 00 '^ Oi t* to 
 
 I 
 
 ■a 
 
 M ««« 1-1 i-HCq rH 
 
 oieo»Oi-iN 
 
 .H 1-1 CC tP t* ■* CO COMrHCCeo 
 
 i-i i-( iH CO "5 00 Cq t^ W5 ^ »H N CO 'il 
 
 WCOWMt*U3pCO00t*N«i-i00 
 
 Ot*U3p 
 1-iMW 
 
 ^C0Mi-<O0S00t-O"3-*C0NrHO 
 
 269 
 
 «co 
 II II 
 b b 
 
 SCO 
 to 
 
 II II 
 
 Si 
 
 II II 
 
 o o 
 
 II 
 
 o in 
 
 nn 
 
 
 fl-H 
 
 i-t 1H 
 fHCO 
 
 II II 
 
 II 
 
 i-9 
 
 ■s-s 
 
 ■54; •a a 
 
 li !i 
 
 II il 
 
 •"■" .3 S 
 loo 
 
 Ziz; 
 
 ■■'a'a 
 
 ■■Sziz; 
 
 I" 
 S.S 
 
 ^1 
 
 as 
 00 
 
 03^ 
 
 §■8 
 
 
270 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 any attempt to compare our groups with the general population with 
 respect to this factor. We suspect that the general population would 
 not show an appreciably better record in this respect. 
 
 Consideration of the relationship between years in school and 
 grade attainment will be deferred until we have presented the data with 
 regard to grade completed. 
 
 (4) Relationship between Years in School and Number of Convic- 
 tions. — We are, however, interested to know whether we can discover 
 any relationship between educational opportunity as indicated by years 
 in school and extent of criminal record as shown by the number pf 
 convictions. Table 86 presents the data on this relationship, dje 
 coefiScient of correlation is practically zero.J^he ratio of number of 
 years in school on number of previous colivictions (.11±.044) would 
 indicate only the slightest of relationships at best. The ratio of number 
 of previous convictions on years in school is, however, large enough 
 (.31 ± ;041) to call for careful inspection of the table to discover 
 whether there is actually an observable tendency here. Referring to 
 the values of the mean number of convictions for various periods of 
 school attendance shown in the extreme right-hand column of the 
 table, we note such oscillation of the various means about 'the mean 
 of the whole group that it is impossible to discover the presence of 
 any one tendency, other than an apparent trend toward a lower average 
 number of convictions for those in the highest groups for length of 
 schooling. 
 
 EDUCATIONAL ATTAINMENTS 
 
 (a) Grade Finished 
 
 In Tables 87 to 91 we present the main facts regarding grade 
 attainment for the groups, classified by institution and by nativity and 
 race. In considering these tables it is important to note that our 
 figures refer always to grade jmished. We have thus avoided the 
 necessity of stating that one was iii^ fifth grade, for exariiple, merely 
 because she had finished fourth, when possibly she had not returned 
 the following year. Our usage, however, fails to take complete 
 account of the facts, since in the reverse case we still refer to the 
 woman as having completed only fourth grade, even though she may 
 have spent part of the year in fifth. The distinction is not important 
 and should be borne in mind merely to avoid misinterpretation in com- 
 parison of our data with other figures which are usually expressed as 
 grade reached. For this comparison an additional year of credit 
 
EDUCATIONAL BACKGROUND 
 
 271 
 
 TABLE 87 
 
 GRADE FINISHED IN SCHOOL 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups 
 
 
 Institutional Groups 
 
 
 Grade Finished 
 
 
 
 
 
 
 
 Total 
 
 
 Bed- 
 
 Auburn 
 
 Mag- 
 
 Peni- 
 
 Work- 
 
 Proba- 
 
 
 
 ford 
 
 dalen 
 
 tentiary 
 
 house 
 
 tion 
 
 
 None* 
 
 9.2 
 
 25.0 
 
 6.7 
 
 15.1 
 
 12.7 
 
 
 11.4 
 
 1 
 
 4.1 
 
 5.9 
 
 1.7 
 
 3.5 
 
 
 
 2.7 
 
 2 
 
 12.2 
 
 10.3 
 
 6.7 
 
 2.3 
 
 11.1 
 
 5.6 
 
 8.1 
 
 3 
 
 19.4 
 
 8.8 
 
 3.3 
 
 14.0 
 
 7.9 
 
 9.7 
 
 11.4 
 
 4 
 
 14.3 
 
 8.8 
 
 23.3 
 
 17.4 
 
 14.3 
 
 9.7 
 
 14.5 
 
 5 
 
 15.3 
 
 13.2 
 
 18.3 
 
 9.3 
 
 11.1 
 
 18.1 
 
 14.1 
 
 6 
 
 10.2 
 2.0 
 
 5.9 
 8.8 
 
 18.3 
 8.3 
 
 12.8 
 3.5 
 
 7.9 
 22.2 
 
 23.6 
 18.1 
 
 13.0 
 
 7 
 
 9.6 
 
 8 
 
 11.2 
 1.0 
 
 8.8 
 2.9 
 
 10.0 
 1.7 
 
 14.0 
 3.5 
 
 12.7 
 
 9.7 
 1.4 
 
 11.2 
 
 1st Year High School 
 
 1.8 
 
 2nd " " " 
 
 
 
 
 1.2 
 
 
 2.8 
 
 .7 
 
 3rd ' 
 
 
 
 
 
 
 
 
 4th " " " 
 
 1.0 
 
 i.5 
 
 1.7 
 
 3.5 
 
 
 i.4 
 
 1.6 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cases 
 
 98 
 
 6^ 
 
 60 
 
 86 
 
 63 
 
 72 
 
 447 
 
 Mean 
 
 4.07 
 
 3.66 
 
 4.93 
 
 4.64 
 
 4.64 
 
 5.72 
 
 4. 58 
 
 
 ±.247 
 
 ±.367 
 
 ±.300 
 
 ±.328 
 
 ±.324 
 
 ±.233 
 
 ±.127 
 
 
 2.44 
 ±.174 
 
 3.03 
 ±.260 
 
 2.32 
 ±.212 
 
 3.04 
 ±.232 
 
 2.57 
 ±.229 
 
 1.97 
 ±.165 
 
 2.68 
 
 an 
 
 ±.090 
 
 ^Includes those who have never been in school and those who have attended but never finished 
 first grade. 
 
 should be given to our groups. Thus, for example, a mean grade 
 finished of 4.58 would signify a mean grade reached of 5.58. 
 
 (1) Data on Institutional Groups and on Total. — Chart XVI pre- 
 sents graphically the distributions shown in Table 87. Marked differ- 
 ences in distribution for the various groups appear, which it would 
 be unprofitable to discuss in detail. Numerous factors such as the 
 percentage of foreign cases, the range of age of the group, the men- 
 tality of the group and the operation of the compulsory education law, 
 have doubtless operated together to produce certain curious irregu- 
 larities in form. 
 
 The relatively high percentages, in Auburn, the Penitentiary, and 
 the Workhouse, who never completed even the first grade reflect 
 primarily the situation regarding time spent in school, thus becoming 
 
272 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 mainly measures of school opportunity. Such cases we note predomi- 
 nate in the groups having the largest proportion of foreign cases and 
 of older women. 
 
 At the other extreme we find in all the groups an extremely 
 small proportion who even entered high school and none who entered 
 college. Occasionally a woman claimed to have been a college 
 graduate but investigation never confirmed her statement on this 
 point. The complete absence of women who had had college and 
 professional training is at least interesting in view of the fact that 
 our group includes all types of offenders^ — those guilty of such crimes 
 as forgery, swindling and homicide, as well as of the minor social 
 offenses. One wishing to hold a brief for the higher education of 
 women might be inclined to point to these figures with some satis- 
 faction as indicating that advanced education afforded a woman 
 such resources in the way of training and personality that she was 
 less subject to the allurements of a criminal career than her less 
 educated sister, or that, at the very least, the figures should be in- 
 terpreted as showing that the college and professional training 
 fails to attract women of the delinquent or criminal "types." A 
 more cynical view would represent the absence of college and pro- 
 fessional women from our groups as indicative of nothing more 
 than their probable greater skill in escaping detection when they 
 have committed crimes or minor offenses. Obviously we lack the 
 
 Per 
 Cent 
 
 TOTAL 
 
 GRADE 
 
 Chart XVI 
 
 Grade Finished 
 
 Per cent distribution by institutional groups. 
 
Cent 
 
 Per 
 
 Cent 
 
 Per 
 
 Cent 
 
 BEDFORD 
 
 AUBURN 
 
 LP 
 
 oil 2 3 4 s e 7 
 
 4 . » 
 
 TT 
 
 I ■ n 
 
 2 3 4 
 
 MAGDALEN 
 
 n . n 
 
 I I I I I I I I 
 
 oil 2 3 4 5 e 7 8ll 2 3 4 
 
 ^ • ► HlEh School 
 
 GRADE XV o 
 
 Chart XVI (Continued) 
 Grade Finished 
 Per cent distribution by institutional groups 
 273 
 
Per 
 Cent 
 
 PENITENTIARY 
 
 Per 
 
 Cent 
 
 Per 
 
 Cent 
 
 WORKHOUSI 
 
 s 
 
 1 
 
 1 
 
 
 
 
 
 
 .„ 
 
 
 1 
 
 
 
 
 
 1 
 
 2 3 
 
 4 
 
 5 
 
 6 
 
 7 
 
 8 
 
 1 2 
 
 3 
 
 4 
 
 HiEh SchopI vy F* 
 
 PROBATION 
 
 .|:=^ 
 
 I 1 I I I 1 
 
 oil 2 3 4 5 6 7 S'l 2 3 4 
 
 ^ _, ^ Hieh School jiVtr. 
 
 GRADE 
 
 Chart XVI (Continued) 
 Grade Finished 
 Per cent distribution by institutional groups. 
 274 
 
EDUCATIONAL BACKGROUND 275 
 
 data which would enable us to pass judgment on the merits of 
 these divergent interpretations. 
 
 In so far as comparison of institutional groups with respect to 
 school grade is attempted it must be borne in mind that the reliability 
 of the information on this point is particularly affected by the 
 differences in completeness of verification among the various groups. 
 We have seen that the percentage of cases on whom we have verified 
 records regarding grade varies from 14 per cent for the Workhouse 
 to 77 per cent for Bedford, and that the tendency of the women 
 is, as one would expect, to better the record in their own accounts. 
 With this qualification in mind it seems unsafe to give weight to 
 the relatively slight differences in the constants for these groups 
 as shown in Table 87. The Workhouse figures should undoubtedly 
 be taken with several grains of salt. The relatively high rating for 
 the Probation group, however, would seem sufficient to indicate a 
 genuine superiority of this group. On the other hand, there seems no 
 reason to question the significance of the low figure for Auburn as 
 indicating an especially poor record of school attainment. 
 
 It is noteworthy that no group shows a record of an average at- 
 tainment equal to completion of sixth grade, which we have seen 
 is now the minimum attainment which permits a child to leave school 
 before sixteen in New York State. The average attainment for the 
 group as a whole is 4.6 grades. 
 
 (2) Data on Groups Classified by Nativity and Color. — The com- 
 parison of grade distributions for the different nativity and color 
 groups has more reliability than the foregoing comparison by insti- 
 tutional groups, since any correction for lack of verification would 
 tend to exaggerate rather than to reduce the difference. Referring 
 to the means for these groups — Table 88 — we note that the native 
 born have a decidedly higher average than the foreign born and that 
 among the native born the native white are appreciably higher than 
 the colored. Tables 89 to 91 show that we are justified in considering 
 all these differences valid. Even the native white group, however, 
 shows only an average of 5.2 grades completed. 
 
 (3) Relationship of Grade to Age Distinctions. — We have seen that 
 for both age at leaving school and number of years in school there 
 was a sharp distinction between the three institutional groups having 
 a higher average age and the three having a lower average, in favor of 
 the latter. We should expect to find the same sort of thing with regard 
 to grade. It does not appear, however, with any degree of definiteness. 
 
276 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 88 
 GRADE FINISHED 
 
 Per Cent Distribution of Delinquent Women Classified by Nativity 
 
 and Color 
 
 
 Iativity and Colob 
 
 
 Grade Finished 
 
 Total 
 Foreign Bom 
 
 Total 
 Native Bom 
 
 Native 
 White 
 
 Native 
 Colored 
 
 Total 
 
 None* 
 
 34.2 
 3.4 
 4.3 
 8.5 
 
 10.3 
 
 11.1 
 6.8 
 5.1 
 
 10.3 
 3.4 
 
 '2!6 
 
 3.3 
 
 2.4 
 
 9.4 
 
 12.4 
 
 16.1 
 
 15.2 
 
 15,2 
 
 11.2 
 
 11.5 
 
 1.2 
 
 .9 
 
 1.2 
 
 2.4 
 
 2.0 
 
 8.1 
 
 11.7 
 
 16.2 
 
 15.4 
 
 16.2 
 
 10.5 
 
 13.8 
 
 .8 
 
 1.2 
 
 1.6 
 
 6.0 
 
 3.6 
 
 13.3 
 
 14.5 
 
 15.7 
 
 14.5 
 
 12.0 
 
 13.3 
 
 4.8 
 
 2.4 
 
 11.4 
 
 1 
 
 2.7 
 
 2 
 
 8.1 
 
 3 
 
 11.4 
 
 4 
 
 14.5 
 
 5 
 
 14.1 
 
 6 
 
 13.0 
 
 7 
 
 9.6 
 
 8 
 
 11.2 
 
 1st Year High School . 
 2nd " " " .. 
 4th " " " .. 
 
 1.8 
 
 .7 
 
 1.6 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of Cases 
 
 117 
 
 330 
 
 247 
 
 83 
 
 447 
 
 Mean 
 
 3.55 
 + .304 
 
 4.95 
 + .128 
 
 5.15 
 ±.147 
 
 4.34 
 ±.247 
 
 4.58 
 + 127 
 
 ™ 
 
 
 a 
 
 3.29 
 + .215 
 
 2.32 
 ±.090 
 
 2.31 
 ±.104 
 
 2.25 
 ±.174 
 
 2.68 
 + .090 
 
 ♦Includes those who have never been in school and those who have attended but never finished' 
 first grade. 
 
 Bedford, in fact, shows a lower mean than the Penitentiary and the 
 Workhouse, and the latter institutions are not greatly inferior to the 
 Magdalen. Because of the differences in the extent of verification, 
 as we have seen, these distinctions can not be stressed as important. 
 We have, therefore, examined the relationship between grade and 
 age directly. The correlation coefficient for age with grade for 449 
 cases was found to be —.124 ± .046, indicating a slight tendency 
 toward lower school grade with increasing age. In order to discover 
 whether there is a clear-cut tendency for the older women, for 
 example, the women over 35, to have a poorer record of school at- 
 tainment than the younger women, we have obtained the means for 
 grade completed for the women over 35 and for the women under 
 35. Since the amount of verification is slight in the case of the 
 older women, we have used the data based on the women's state- 
 
EDUCATIONAL BACKGROUND 277 
 
 TABLE 89 
 
 TOTAL FOREIGN BORN AND TOTAL NATIVE BORN 
 
 Comparison of the Means and Standard Deviations of the Grade 
 
 Finished for Total Foreign Bom and Total Native Bom 
 
 of Total Group 
 
 
 Total 
 Foreign Born 
 
 Total 
 Native Born 
 
 Difference 
 
 d 
 
 Chances that real 
 
 difference does 
 
 not exist are 1 
 
 in: 
 
 Mean 
 
 3.55 
 ±.304 
 
 4.95 
 ±.128 
 
 -1.40 
 
 4.24 
 
 89,286 
 
 CFja 
 
 
 
 (7 
 
 3.29 
 ±.215 
 
 2.32 
 ±.090 
 
 .97 
 
 4.16 
 
 62,893 
 
 aa 
 
 
 
 Cases 
 
 117 
 
 330 
 
 
 
 
 
 
 ments entirely for this comparison. The mean grade for the women 
 over 35 was found to be 5.11 ± .409 and that for the women under 
 35, 6.04 ± .137. There is, therefore, an obvious difference, amount- 
 ing almost to one grade, between the averages for the older and the 
 younger women. We are justified in assuming the validity of this 
 difference since it amounts to more than twice the standard deviation 
 of the difference. 
 
 Since so genuine a difference appears to exist between these 
 groups distinguished in point of age, it has seemed desirable to 
 
 TABLE 90 
 
 NATIVE WHITE AND NATIVE COLORED 
 
 Comparison of the Means and Standard Deviations of the Grade 
 
 Finished for Native White and Native Colored of 
 
 Total Group 
 
 
 Native 
 White 
 
 Native 
 Colored 
 
 Difference 
 
 d 
 (Td 
 
 Chances that real 
 
 difference doea 
 
 not exist are 1 
 
 in: 
 
 Mean 
 
 5.15 
 
 ±.147 
 
 4.34 
 
 ±.247 
 
 .81 
 
 2.83 
 
 435 
 
 
 
 
 
 <r 
 
 2.31 
 
 ±.104 
 
 2.25 
 
 ±.174 • 
 
 .06 
 
 .30 
 
 3 
 
 ffo- 
 
 
 
 
 Cases 
 
 247 
 
 83 
 
 
 
 
 
 
278 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 91 
 
 TOTAL FOREIGN BORN AND NATIVE WHITE 
 
 Comparison of the Means and Standard Deviations of the Grade 
 
 Finished for Total Foreign Bom and Native White of 
 
 Total Group 
 
 
 Total 
 Foreign Born 
 
 Native 
 White 
 
 Difference 
 
 d 
 (Td 
 
 Chances that real 
 
 difference does 
 
 not exist are 1 
 
 in: 
 
 Mean 
 
 3.55 
 + .304 
 
 5.15 
 
 + .147 
 
 -1.60 
 
 4.75 
 
 8 
 
 (Tin 
 
 
 
 
 (T 
 
 3.29 
 + .215 
 
 2.31 
 ±.104 
 
 .98 
 
 4.11 
 
 50,505 
 
 a<r 
 
 
 
 Cases 
 
 117 
 
 247 
 
 
 
 
 compute the constants for the group under 35 for use in comparison 
 with any outside groups whose range of ages does not extend beyond 
 this age. We have in mind, for example, especially the data which 
 may be forthcoming regarding army recruits whose ages are predomi- 
 nantly lower than this. The figures for the delinquent women under 
 35 years of age are as follows : 
 
 Grade Finished 
 
 Based on Women's Statements Verified Data 
 
 Mean 6.04 ± .137 5.59 ± .174 
 
 (T 2.60 ± .097 3.56 ± .289 
 
 Number of Cases 361 197 
 
 (4) Relationship of Grade to Age at Leaving School. — In view 
 of the dearth of information regarding the educational status of the 
 general adult population, it is particularly important to search our own 
 data for internal evidences as to the normality of school progress. 
 Two such lines of evidence are of particular interest, namely, the com- 
 parison of grade attainment with age at leaving school and with the 
 number of years in school. 
 
 The former of these comparisons is closely related to the age- 
 grade tables with which we have become so familiar in all recent 
 discussions of retardation. Since, however, our women are all past 
 the school period, it becomes necessary to substitute age at leaving 
 school for actual age. Each school report in which such figures are 
 discussed sets up some standard to be met and offers as statistics 
 of retardation the percentages which fall behind this standard. That 
 
s 
 ^ 
 
 o 
 o 
 
 u 
 
 03 
 
 CS 
 0) 
 
 bo 
 <J 
 
 •a 
 
 EO 
 
 o 
 
 d 
 
 a 
 
 ■I 
 
 o 
 U 
 
 o 
 
 •a 
 
 o 
 H 
 
 
 1 
 
 
 CO 
 
 CO 
 
 1* 
 
 
 00 
 
 § 
 
 S 
 
 § 
 
 T-l 
 
 CO 
 
 rH 
 
 1-H 
 
 1— 1 
 
 
 1 
 
 CO 
 
 s 
 
 O 
 
 CM 
 
 
 - 
 
 T-H 1-1 1-* 1-1 
 
 ta 
 
 1—1 
 
 iH 
 
 ■*- 
 
 
 
 
 o 
 
 o 
 
 
 -H 
 
 i-l 1-1 
 
 CO 
 
 & 
 
 T-l 
 
 CO 
 
 1-1 tH 
 
 CO 
 
 00 
 
 1— * tH t* 
 
 T— 1 
 
 05 CD rri t— 1 
 
 00 
 CO 
 
 t* 
 
 (N CO CO <N 
 
 T— 1 
 
 CO 
 
 I— 1 
 
 in 
 
 ^ 
 
 «D 
 
 co.«20 
 
 CO 
 
 .-H i-H 
 
 s 
 
 lO 
 
 CO 05 05 IC lO 
 
 iH .-I 
 
 
 
 s 
 
 -* 
 
 T-H CD (N C3 OC CD 
 
 T-l 1-1 
 
 lO 
 
 tH 
 
 s 
 
 CO 
 
 •* OJ CO 00 00 r-fl 
 
 T— t 
 
 
 
 CO 
 
 T»l 
 
 M 
 
 rH 'Jtl 00 "O CD CO "H 
 
 -< 
 
 
 OS 
 IM 
 
 - 
 
 
 1—1 
 
 
 00 
 
 * 
 
 o 
 
 >-l 1-1 T-l (N (N >-l T-l 
 
 - 
 
 
 o 
 
 T-l 
 
 
 
 O500I:^CD>OT)le0<N'-HOC300l> 
 
 
 4 
 
 o 
 
 
 
 
 I 
 
 aoq 
 
 as 
 
 SUL 
 
 \eg 
 
 !» 
 
 B3 
 
 3v 
 
 
 
 
 •« 3 2- 
 
 CON^^g 
 
 «' 
 
 II S a S^ 
 
 53 o ^ «* 
 
 S.a o «'i 
 
 Oj Oj 03 f^(M 
 t, UDO "-I 
 
 279 
 
280 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 this standard represents an ideal, rather than an average, even for 
 children who have all the advantages of present school opportunities, 
 is evident from the fact that the proportion of "over-age" pupils 
 is practically always considerably larger than the proportion of ad- 
 vanced, and that most school systems are struggling^ with a large 
 problem of retardation. 
 
 We shall not attempt to consider the various age-grade stand- 
 ards which have been proposed, but shall consider especially in- 
 teresting for our purposes the present New York City standards, the 
 so-called "up-to-15" standard by which it is expected that a pupil 
 will have graduated from elementary school by the time he is fifteen. 
 The requirements as stated for each grade are as follows : " 
 
 
 
 
 Grade Completed 
 
 
 
 
 1 
 
 2 
 
 3 
 
 4 5 
 
 6 
 
 7 
 
 8 
 
 Age 7-8 
 
 8-9 
 
 9-10 
 
 10-11 11-12 
 
 12-13 
 
 13-14 
 
 14-15 
 
 This means that one should have completed the first grade before 
 his eighth birthday, the second grade before his ninth, etc. 
 
 Table 92 shows the age-grade relationships for our group. We 
 have enclosed within heavy lines the spaces within which would fall 
 those making normal progress in approximate accordance with the 
 New York requirements.^^ Numbers in the columns above these 
 lines represent "retarded" cases and numbers below represent "ad- 
 vanced." The enormous preponderance of retarded cases is evi- 
 dent to the most cursory inspection. 
 
 The correlation between grade finished and age at leaving school 
 is found to be .44 ± .043. While this is sufficient to show that the 
 grade attainment is related to the age at which one leaves school, 
 it is far too low to suggest that the latter factor is mainly responsible 
 for the amount of school progress. 
 
 (5) Relationship of Grade to Number of Years in School. — Of 
 greater significance than the foregoing comparison, for determining 
 whether normal progress has or has not been made, is a comparison 
 of grade and number of years that the individual has actually been 
 in school. Such a comparison is offered in Table 93. For this com- 
 
 " "Eighteenth Annual Report of the City Superintendent of Schools to the 
 Board of Education of the City of New York for the Year Ending July, 1916." 
 Appendix F, pp. 175 ff. 
 
 "A slight deviation, amounting to a half year's greater leniency, was made 
 because of the fact that our round numbers made it impossible to follow the 
 New York standards exactly. This is due to the fact that in our figures for 
 age, 7 equals 6^ to 7j4, 8 equals 7^ to 8^, etc. 
 
o 
 o 
 Jd 
 u 
 
 tS 
 
 .a -s 
 
 w 4ij 
 
 
 0) 
 
 o 
 
 I 
 
 •a 
 
 o 
 
 
 1 
 
 .-HINl>-*(NINT-ltDO!.-410l>l>eOW 
 r-l (M ■* 00 50 CO N rt (N 
 
 1-H 
 
 «n 
 
 CO 
 
 
 
 1-H tH r-l rH »H . 
 
 lO 
 
 -t- 
 
 
 o 
 
 o 
 
 IN .-1 
 
 CO 
 
 
 tH 
 
 1-H 
 
 00 
 
 .-i(NNmit>>-ioOT-i--ii-ii-i 
 
 ^ 
 
 t> 
 
 (N(NMt~or~e<j .-H 
 
 tH 
 
 CO 
 
 <D 
 
 1-1 CD O t>- COM 1-H 
 I— 1 1—1 
 
 o 
 
 lO 
 
 IN M lO (N m CO t^ (NW 
 
 T— 1 1— 1 
 
 1-1 
 
 "3 
 
 ■<** 
 
 iNiNcocomot-"-ii-iw 
 
 1-H ^ »-( 
 
 00 
 U3 
 
 CO 
 
 .-H IN lO 00 l> 00 I> CO (N 
 
 ^ 
 
 •N 
 
 1-H 1-H CO »0 t>- iO ^ CO 1-H 
 
 S5 
 
 - 
 
 i-H i-H CO i-H iH 1-H 
 
 00 
 
 8. 
 
 rHi-H CO IN-*'-! 
 
 IN 
 CO 
 
 
 TtHeo<Ni-H005oot-co>o-*coM>-io 
 
 1-H T-l I-H 1-H tH 
 
 
 jootjog aj sasai jo joqnmN 
 
 1-HO 
 
 •ooo„ 
 
 . CO 
 
 II II ' 
 
 b b ■N 
 
 TlHCO u ^J 
 
 N II a 
 
 3 c8-t^ 
 g^ Q} a 
 
 • - o 
 
 ■o ° 
 
 o cj ^ 
 •S SSfl 
 
 .^ « s 
 
 t! TC. 
 
 2« 
 .35 
 
 ^ o 
 
 
 13 O 
 3T3 
 
 281 
 
282 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 parison we have mcluded only women now 35 years of age or less 
 in order to reduce, to some extent, the heterogeneity of the group. 
 The correlation coefficient of .62 ± .033 shows that there is an un- 
 questionable relationship between school attainment, as indicated by 
 grade completed, and school opportunity, as represented by the length 
 of school attejjdance. . . 
 
 On first consideration the above figures would seem to mdicate 
 that grade is much more directly determined by the time spent in 
 school than by the age at leaving, since the coefficient for the 
 latter relationship is only .44. It must be remembered, however, 
 that those who had never attended school were necessarily omitted 
 from the figures on age at leaving school. Referring to Table 93 
 we note that there are 21 such cases and that they, naturally, show 
 perfect correspondence between opportunity and accomplishment, 
 both being nil. This would, of course, contribute materially to the 
 size of the correlation. Accordingly, we have computed the cor- 
 relation between grade and years in school, dropping out those 
 cases who had had no schooling. This gives a coefficient of .48 ± 
 .043, only slightly higher than the correlation between grade and 
 age at leaving school. 
 
 Because of the importance of this relationship we have also 
 computed the correlation using only those cases on whom we had 
 verified data for both grade and years in school (159 cases). This 
 gives a still lower correlation, .39 dz .067. We interpret this, not as 
 meaning that the relationship is less when we have reliable data, 
 but rather as due to the restriction of variability for the group 
 whose records were verified and especially to the loss of those 
 approaching the limiting value of no schooling. That these factors 
 are operative is shown by the difference in the standard deviations 
 for grade finished. For the total group under 35 years of age a 
 equals 2.5, whereas for the group on whom we have completely veri- 
 fied data for both grade and years in school a is only 1.9. 
 
 We are interested further in considering whether the observed 
 relationsjiip between grade and years in school may be due mainly 
 to a basic relationship of each of these factors with intelligence. 
 Accordingly, we have figured the correlation of grade with years 
 in school for constant intelligence, the latter being represented by 
 the score on Test Aggregate.^' The partial coefficient is found to 
 
 "The relationship of intelligence with schooling will be discussed in detail 
 in Chapter XV, pp. 493-8. At this point we are merely interested in consid- 
 
EDUCATIONAL BACKGROUND 283 
 
 be .402. This indicates that the factor of intelligence has not con- 
 tributed largely to produce the relationship noted between grade 
 and years in school. On the other hand the closeness of this relation- 
 ship is not raised when we control the factor of intelligence as in the 
 above partial. It might have been expected that when the influence of 
 intelligence was eliminated grade attainment would appear to be 
 determined almost exclusively by the length of time in school. 
 
 In addition to the mere question of relationship between these 
 factors we are also concerned to know whether the grade attain- 
 ment approximates closely to that which should be expected in view 
 of the number of years in school. That it fails to reach this level 
 is apparent at once by comparison of the constants of Tables 84 
 and 85 with those of Tables 87 and 88, respectively. Discrepancies 
 are striking whatever groups we compare. It is sufficient to note, 
 for example, the figures on the Bedford group and on the total. 
 The Bedford women show an average number of years in school 
 of 7.3, whereas the average grade completed is only 4.1. The total 
 group with an average number of years in school of 6.5 completed 
 only 4.6 grades. The discrepancy between opportunity and accom- 
 plishment is also indicated graphically by Chart XVII, which shows 
 the actual grade distribution for women under 35 in comparison 
 with the distribution which would have been found had they pro- 
 gressed at the normal rate of one grade a year. The failure to ac- 
 complish anything approximating a reasonable rate of progress sug-» 
 gests at once the probability of a generally low level of intelligence 
 in this group. That this factor is operative will be apparent from 
 our discussion in Chapter XV. The evidence, therefore, tends to 
 confirm the assumption made earlier in this chapter to the effect 
 that the record of grade completion tends of itself to be at least a 
 rough measure of mentality. In this connection, however, we should 
 not overlook the probable influence of other environmental factors 
 tending against normal progress at the rate of a grade a year. We 
 have in mind especially poor home conditions, economic stress, etc., 
 which are likely to produce irregularity of school attendance, so 
 
 ering the effect of this relationship on the correlation between the two meas- 
 ures of schooling under consideration. The correlation coefficients used in 
 computing the partial are as follows : 
 
 r,8 =.S96, rnr = .289, r„=.481. N = 383 
 (I equals intelligence as measured by Test Aggregate, G equals grade, Y equals 
 years in school, and N equals the number of cases.) 
 
284 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 that a year in school has not its full value as a measure of school 
 opportunity. 
 
 (6) Relationship of Grade to Nature of Present Offense. — Turn- 
 ing to the question of the relation of grade attained to the vari- 
 ous aspects of the criminal career of these women, we con- 
 sider first its relationship to, the nature of the present offense. 
 This is shown in Table 94. By reference to this table we note the 
 absence of any striking distinctions, although the presence of a slight 
 
 Per 
 Cent 
 
 Grade Plniahed 
 
 ....... 'Grade that ahould have 
 
 been Rnlehed 
 
 Chart XVII 
 
 Comparison of Grade Finished with Grade that Should Have Been Finished, 
 Assuming One Grade Completed in Each Year in School. Based on .l.il 
 Delinquent Women under 35 Years of Age at Present Conviction. 
 
 degree of relationship is shown by the ratio of .23 ±: .045. With 
 the rough grouping here used, however, no single type of offense 
 stands out as characteristic of the women of superior education. 
 Two groups of offenses, those against the family and those against 
 the administration of government, are conspicuous for the fact that 
 they contain no women who had finished more than the fourth grade. 
 In view of the extreme smallness of these groups no general sig- 
 nificance can be attached to this point as characterizing these types 
 of delinquencies. The fact merely serves to describe further the 
 individuals of our group who fall in these two classes. We have 
 noted previously that the instances with which we are dealing of 
 

 (b 
 
 »e 
 
 
 
 
 
 
 
 
 M 
 
 
 
 
 ill 
 
 o 
 
 oc 
 
 CO 
 
 »o 
 
 o 
 
 CO 
 
 c 
 
 K 
 
 
 
 
 CO 
 
 ■* 
 
 .-H 
 
 ■<* 
 
 IM 
 
 •* 
 
 CO 
 
 ■<* 
 
 
 
 
 1 
 
 (N 
 
 ?5 
 
 00 
 
 4 
 
 CO 
 
 o> 
 o 
 
 «: 
 
 
 
 
 ^ 
 
 H 
 
 
 
 
 
 
 
 
 
 
 
 
 4- 
 
 
 
 
 
 
 
 
 
 
 
 13 
 
 
 9 
 
 tH 
 
 e^ 
 
 
 1- 
 
 
 CO 
 
 
 N 
 
 
 
 <0 
 
 s 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 OQ 
 
 
 ■»- 
 
 
 
 
 
 
 
 
 
 
 
 ua 
 
 
 ;H 
 
 
 
 
 
 
 
 
 C 
 
 
 
 cti 
 
 
 
 
 
 
 
 
 
 
 
 
 
 w^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 O 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 o 
 
 
 (N 
 
 
 
 iH 
 
 
 c^ 
 
 
 
 0) 
 
 
 
 
 
 
 
 
 
 
 
 
 
 (0 
 
 
 
 
 
 
 
 
 
 
 
 
 
 fl ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 S d 
 
 
 
 
 
 
 
 
 
 
 
 
 
 iU 9 
 
 
 s- 
 
 
 
 
 (N 
 
 
 U3 
 
 
 i> 
 
 
 o 
 
 
 
 
 
 
 
 
 
 
 
 
 
 « 
 
 
 
 
 
 
 
 
 
 
 
 u 
 
 §1 
 
 Q 
 
 00 
 
 ^H 
 
 ?5 
 
 to 
 
 CO 
 
 c^ 
 
 ^ 
 
 
 1 
 
 P^U 
 
 t^ 
 
 1—1 
 
 IN 
 
 CO 
 
 1> 
 
 c^ 
 
 
 
 1 
 ■a 
 
 o^ 
 
 
 
 
 
 
 
 
 
 
 
 >o 
 
 
 
 
 
 
 
 
 
 
 
 h 
 
 Si 
 
 1 
 
 1 
 
 «) 
 
 (N 
 
 00 
 
 lO 
 
 fH 
 
 
 CC 
 
 
 i 
 
 Bi 
 
 fJ4 
 
 a 
 
 
 
 CO 
 
 
 
 T-H 
 
 
 ir 
 
 
 C 
 
 it 
 
 U3 
 
 (N 
 
 1-H 
 
 "3 
 
 (M 
 
 c^ 
 
 <M 
 
 
 1 
 
 O 
 
 
 
 ■* 
 
 
 
 T-H 
 
 
 tc 
 
 
 1 
 
 §« 
 
 
 
 
 
 
 
 
 
 
 
 s 
 
 _; o 
 
 
 ■* 
 
 •* 
 
 to ,-1 
 
 CO 
 
 ■* 
 
 CS 
 
 cr 
 
 o 
 
 rt 
 
 ■oa 
 
 
 
 
 CO 
 
 
 
 iH 
 
 
 CC 
 
 t^ 
 
 P 
 
 aj o 
 
 
 
 
 
 
 
 
 
 
 
 
 S 
 
 1-^ 
 
 CO 
 
 T-t 
 
 s ^ 
 
 •* (^ 
 
 05 
 
 c^ 
 
 
 b 
 
 1^ 
 
 pc(0 
 
 
 
 
 
 
 
 
 
 
 
 
 S"© 
 
 ^M 
 
 
 (N 
 
 CO 
 
 00 CO 
 
 IS 
 
 
 Ol 
 
 (M 
 
 1> 
 
 lO 
 
 o o 
 
 •9-S 
 
 •dT. 
 
 
 
 
 1-1 
 
 
 
 
 
 CC 
 
 ^ 
 
 -S 
 
 0^ o 
 
 
 
 
 
 
 
 
 
 
 
 o 
 
 T3 ' 
 
 
 rH 
 
 tH 
 
 <N 
 
 
 Tt 
 
 
 "3 
 
 
 <N 
 
 ~ s^' 
 
 51 
 
 
 
 
 
 
 
 
 
 
 
 
 O O 
 
 
 
 
 
 
 
 
 
 
 II ■ 
 
 
 1 
 
 
 s 
 
 o> 
 
 Ti 
 
 CO 
 
 CO 1- 
 
 
 0- 
 
 ci: 
 
 
 .s>. 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 §•• 
 
 d 
 
 
 
 
 
 S 
 
 fl 
 
 i 
 
 
 
 2 
 
 s5 
 
 S 
 
 
 
 
 
 £ 
 
 .S 
 
 
 
 
 •b'^S 
 
 g o 
 
 1 
 
 1 
 
 g 
 
 
 ^ ^ 
 
 i 
 
 i 
 
 (1 
 
 
 
 
 gi 
 
 
 i 
 
 1 
 
 1 
 Ph 
 
 2 
 
 > 
 
 ^ 
 
 
 .a 
 
 o 
 
 > 
 
 ■> 
 
 
 5| 
 
 
 1 
 
 ■s 
 
 4^ 
 
 
 -*s 
 
 
 if 
 
 'ca 
 
 'o 
 
 1 
 
 D 
 
 "ei 
 
 -4.9 
 
 55 
 
 (DCq 
 
 Is 
 
 
 1 
 
 
 S 
 
 
 bC c 
 
 bfl > 
 
 u> 
 
 ■E 
 
 
 11 
 
 
 03 
 
 
 
 o3 
 
 o3a 
 
 ^,9 
 
 ff 
 
 c 
 
 
 
 1 
 
 
 
 S 
 
 s-§ 
 
 gO 
 
 
 "o 
 
 ^ 
 
 
 &0 
 
 
 3 
 
 g 
 
 § 
 
 §^ 
 
 go 
 
 3 
 
 £ 
 
 
 
 * 
 
 
 
 
 
 
 ffl 
 
 
 ."" 
 
 c 
 
 
 
 
 
 
 Jtj 
 
 jj; 
 
 SH 
 
 
 so 
 
 Sd 
 
 £ 
 
 
 
 
 
 
 
 o 
 
 c 
 
 o 
 
 o 
 
 o 
 
 O 
 
 C 
 
 
 
 
 285 
 
286 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 offenses against the administration of government are not the im- 
 pressive crimes that the name may suggest. Rather, they are ex- 
 amples of stupid attempts at evading the law through unskilful 
 perjury or bribery. We are interested to note that there is^o ap- 
 preciable distinction in degree of education of those women guilty 
 of offenses against chastity as compared with those guilty of offenses 
 against property. This is of special interest since, as we shall see 
 later, the latter group shows an appreciable superiority in general 
 intelligence over the former. 
 
 (7) Relationship of Grade to Number of Previous Convictions. 
 — Considering next the relationship between grade and the num- 
 ber of convictions, we obtain the correlation presented in Table 95. 
 The coefficient, — .06 ± .047, is too small to have any significance. 
 The ratios, though slightly larger, would indicate at most a barely 
 appreciable relationship. Referring to the means, shown in the 
 column at the extreme right and in the bottom row of the table, 
 we find such irregularity that it is impossible to formulate the 
 variations as indicative of any consistent trend. It appears, there- 
 fore, that the data of this table fail to provide any evidence to 
 show that the amount of education of our delinquent women has 
 influenced the extent of their criminal records to any degree or in 
 any direction. 
 
 Since both the record of number of convictions and that of grade 
 attended have shown appreciable, though slight, relationship with 
 the factors of age and intelligence, it has seemed important to take 
 account of these two factors, because of the possibility that their 
 influence might have obscured whatever relationship between grade 
 and number of convictions existed. Computing the correlation of 
 grade with number of convictions for constant age, we obtain the 
 coefficient —.027.^* It is evident that allowance for age does not 
 serve to show up a relationship which had been obscured by this 
 factor. 
 
 The outcome is practically the same when we make allowance 
 for the possible effect of intelligence in this connection. In order 
 to utilize the whole group and not merely the English-speaking por- 
 tion, we have used the score for the non-language tests as our 
 measure of intelligence. The correlation coefficient of grade with 
 
 "The data from which this partial was computed are as follows: 
 roN = — .OSS, fsA = — .124, r^i =.232 
 (N equals number of convictions, G equals grade, A equals age.) 
 

 s 
 
 
 o 
 
 
 •43 
 
 
 o 
 
 ■o 
 
 <7> 
 
 ■g 
 
 
 o 
 
 w 
 
 O 
 
 I-) 
 
 
 § 
 
 1 
 
 
 0) 
 
 u 
 
 I 
 
 SI 
 
 d 
 
 .a 
 
 a 
 
 o 
 
 I 
 
 O 
 
 o 
 
 o 
 
 u 
 
 v 
 
 m 
 
 
 o 
 
 o 
 
 One 
 
 S 03.2+3 
 g-SP.2 
 
 6 u 
 
 
 
 HN-ijI tHCO 
 
 .-<«<NtJ< Tft CO CO 
 
 t-iHfOTti"*.-iC<INC0 
 
 Cfl iH rH »-( Th •>** Ol Oa CD t» M lO 
 
 ^W Cq ^ 00 1^ to W ■* M IN rt o 
 
 paqsrai^ apBJQ 
 287 
 
 j'3 
 
 is 
 
 CD 
 
 o o 
 
 'C'C 
 
 0) d) 
 
 s s ■ 
 
 a a V 
 
 ill 
 
 fflWg 
 
 <a 
 
 coco— ( 
 
 in 
 "I 
 
 - -s 
 "■as 
 
 ea a 
 
 ■g«'S 
 8-S'S 
 
 " oS <l> 
 
 ° a 
 
 .a-* 
 §a 
 
 |s s-gf 
 
 r:"Sl.s 
 
 •sill 
 
 
 g.2 t 
 •SiS s 
 
 op ff, 
 
288 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 number of convictions, for constant intelligence, is found to be .053." 
 In view of the absence of correlation shown in these two partials 
 we have not thought it necessary to compute the correlation of grade 
 with number of convictions for both intelligence and age constant. 
 
 (8) Relationship of Grade to Age at First Conviction. — As in- 
 dicative of any possible relationship between grade completed and 
 the beginnings of the criminal career, we have computed the corre- 
 lation between grade and age at first conviction, shown in Table 
 96. The coefficient is —.08 ± .047 which fails obviously to 
 afford evidence of the existence of any relationship. The ratio 
 of grade on age at first conviction, 17 ± .046, would show the 
 presence of at most a very slight relationship. Since we are unable 
 to discover any consistent trend in the variations in mean grade 
 with changes in age at first conviction, we do not consider that 
 the existence of even a slight relationship is established. The ratio 
 of age at first conviction on grade is appreciably larger, .35 ± .041. 
 This is sufficient to call for a careful examination of the table to dis- 
 cover what trend of relationship is indicated by these figures. 
 Reference to the means for age at first conviction in the extreme 
 right-hand column shows a tendency toward earlier first conviction 
 for those falling nearest to the average for grade finished. The 
 tendency toward delayed age at first conviction for those who had 
 no schooling and also for those who advanced beyond the elementary 
 school is striking. It will be remembered that a similar tendency ap- 
 peared for the relation between age at leaving school and age at first 
 conviction. The factors, discussed in that connection as explanatory, 
 doubtless apply also here. 
 
 (&) Amount of Illiteracy 
 
 As a further index of the school attainment of our group, from 
 the negative side, we may consider the figures showing the extent 
 of illiteracy. For comparison there are available certain figures 
 on the general population, although it is quite certain that the stand- 
 ard of what constitutes illiteracy can not be considered entirely com- 
 parable with ours. We had very accurate knowledge with reference 
 to the degree of literacy of each individual, so that we were not 
 dependent on indirect information. The census figures, on the 
 
 "The data from which this partial was computed are as follows: 
 tqn = — .055, rax =.543, rHi= .127 
 (I equals intelligence, N and G have same significance as in Note 14.) 
 
o 
 
 o 
 xi 
 u 
 03 
 
 ID 
 
 
 » I 
 
 o 
 O 
 
 -^ 
 
 .a 
 
 0) 
 
 < 
 
 d 
 
 (U 
 V 
 
 B 
 
 d 
 
 0) 
 
 o 
 O 
 
 ^ 
 
 o 
 
 O 
 
 o 
 H 
 
 .go 
 
 S6 
 
 lOOtOUSOt-tcOWi-l 
 
 ^NC 
 
 Zi-69 
 
 69-99 
 
 S9-09 
 
 09-/S 
 
 iS-»S 
 
 '^S-IB 
 
 IS-8» 
 
 8^S* 
 
 Sf-Zf' 
 
 S^-6S 
 
 (N iH i-l >^ M tH rH fH CO 
 
 68-9E 
 
 lO N(N»H Ol CO 
 
 9e-ES 
 
 CO C^ 1-^ IN ■«*< Tji ■* 
 
 EE-OS 
 
 t* CO CO IN TJH U3 
 
 DE-iZ 
 
 rHt-.^Til^C0Q0CO(O 
 
 Nwosot-cqioeOiHM 
 
 ^e-iE 
 
 coiNooooooswocqeoco 
 
 i-HO'CIN-^MCOO^'O 
 
 8T-ei 
 
 -I CO c^ CO »o lo t- o cq 
 
 ai-zt 
 
 ZI-6 
 
 6-9 
 
 ^'H CmCCb-cDiO-^COINiHO 
 
 TjICOINi-H 
 
 289 
 
 m ■* 
 
 
 (N »0 
 
 
 II II 
 
 
 ci a 
 
 
 
 
 •C 'C 
 
 
 ?. H 
 
 
 
 
 
 
 CJU 
 
 
 
 
 s§ 
 
 
 H u 
 
 be 
 
 
 
 
 
 
 <S 
 
 s PQ 
 
 -d 
 
 II 
 
 ■fl i 
 
 g 
 
 ^1 
 
 JhS 
 
 ■3 
 
 in ^1 
 
 
 J3 
 
 ■* II 
 
 B- P 
 
 
 iO 
 
 
 
 
 11 
 
 1 
 
 gfe 
 
 viction 
 Mean 
 .047 
 
 
 cm 
 
 g 
 
 fl 
 
 ■fl 
 3 
 
 fl •■ „ 
 
 
 
 
 T? -H 
 
 
 
 
 first c 
 finishe 
 -.08 
 
 g 
 
 i 
 
 es 
 
 ge at 
 rade 
 
 in 
 
 
 
 
 
 i 
 
 S 
 
 a 
 
 V 
 
 T? 
 
 "^ 
 
 .s 
 
 10<! 
 
 is 
 
 ■s £ 
 
 00 
 
290 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 other hand, are largely dependent on mere statement and on 
 indirect statement at that — information given by one member of the 
 family for the whole family. This is clearly pointed out in the 
 discussion of this subject in the census reports. Such dependence 
 on hearsay information for the general population is inevitable, but 
 doubtless results in giving figures for illiteracy vi^hich are far too 
 low. For this reason we cannot attach much importance to the 
 comparison of our figures with those of the census report, but we offer 
 this comparison as giving the best available clue to the facts. 
 
 We have used the figures for the general female population over 
 ten years of age in New York State obtained in the census enumera- 
 tion for 1910.^® We have adopted the same definition of illiteracy, 
 as signifying inability of a person ten years of age or over to write, 
 regardless of whether he can read or not. For practical purposes 
 this has meant that an individual must be able both to read and 
 write, since we have found no cases who were able to write but not 
 to read. Our standard of what constitutes reading and writing has 
 been exceedingly rudimentary. Ability to get the sense of even the 
 simplest sort of reading matter or to convey one's meaning by 
 writing, even though the spelling be very grotesque, has been con- 
 sidered adequate. 
 
 Table 97 presents the data on the delinquent women classified 
 both by institutions and by nativity and color, in comparison with the 
 data on the general female population, as regards illiteracy in the 
 sense of inability to read and write in any language. The chief 
 interest attaches to the comparison of the delinquent group classi- 
 fied by nativity and color, with the figures for the general popula- 
 tion by the same classification. We note that for each group thus 
 classified the percentage of illiterates is appreciably higher among 
 the delinquents than in the general population, according to these 
 figures. In the delinquent group as in the general population the 
 smallest percentage of illiteracy is found in the native whites (4.2), 
 and the largest percentage among the foreign whites (24.0), the per- 
 centage among the negroes (10.6) falling between the other two 
 figures. The intercomparison of these sub-groups among the de- 
 linquents — i. e., of native-white with foreign-white, etc. — may be con- 
 sidered entirely trustworthy, since a uniform standard has been 
 applied here. Some interest attaches also to consideration of the 
 percentages of illiterates in the various institutional groups. Tak- 
 
 '° Abstract "Thirteenth United States Census Report," 1910, pp. 237 ff. 
 

 
 
 li 
 
 
 
 
 
 
 
 
 
 
 
 fl 2 
 
 0« -Mb-C 
 
 cc 
 
 »o 
 
 
 
 9.S 
 
 00 -iflCO 
 
 c 
 
 U3 
 
 
 
 S3 
 
 • iHiH 
 
 
 
 
 
 vpj 
 
 
 
 
 
 o 
 
 PM 
 
 
 
 
 M <u 
 
 
 
 
 
 M 
 
 S-*^ 
 
 
 
 
 
 
 S d 
 
 
 
 U3 
 
 
 iS 
 
 |l 
 
 0.-H -MCOC 
 
 c 
 
 s? 
 
 
 Ig 
 
 
 
 CO 
 
 ^ 
 
 
 
 
 
 
 
 OIMO03CDI> 
 
 ■<1 
 
 
 
 
 <N^ rHCO 
 
 c 
 
 
 
 Eh !3 
 
 
 
 o 
 
 
 
 iz; 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 bO a 
 
 
 
 
 
 
 
 
 
 
 
 * 
 
 a »5 
 
 cciocooocc»r 
 
 c 
 
 OS 
 
 S 
 
 6 
 
 g 
 
 ^ 
 
 p' 
 
 sjssssi'' 
 
 g 
 
 ui 
 
 
 
 
 
 c 
 
 1 
 
 II 
 
 Tj<iO>OOl>i- 
 
 
 i 
 
 g 
 
 s 
 « 
 
 11 
 
 
 
 s 
 
 1 
 
 & 
 
 
 
 
 
 M 
 
 
 
 X 
 
 <! 
 
 
 ^ " 
 
 
 
 •* 
 
 Iz; 
 
 
 "a-Q 
 
 
 
 l^ 
 
 
 H 
 
 tH CONOCO Cs 
 
 If 
 
 Os" 
 
 8 
 
 
 
 g 
 
 
 t- 
 
 
 "1 
 
 
 
 
 i 
 
 
 
 
 
 
 o 
 
 
 "a 
 
 •"I 
 
 
 
 <1 
 
 Si 
 
 Vkm 
 
 
 
 
 S 
 
 t< 
 
 Ph 
 
 
 
 
 1 
 
 II 
 
 
 
 o 
 
 i 
 3 
 
 1 
 
 5 
 
 M<NOtH'<*<Cs 
 
 cs 
 
 CD 
 
 iz; 
 
 
 
 
 
 h 
 
 
 
 « 
 
 o 
 
 
 1-. ® 
 
 
 
 o 
 
 
 
 
 OS X iH r-( rt* T- 
 
 
 g 
 
 
 
 (ONW'^coa: 
 
 ^ 
 
 
 
 !z; 
 
 
 
 Csf 
 
 
 <u 
 
 
 
 
 
 
 bD V 
 
 
 
 
 
 
 (S-S 
 
 
 
 
 
 
 ■Si 
 
 0100(>00OC^ 
 
 oc 
 
 OS 
 
 
 
 «ONcDeoi><r 
 
 ^ 
 
 lO 
 
 
 
 
 <N rHi-H 
 
 -•- 
 
 
 
 i 
 
 
 
 
 
 
 
 
 
 
 <§ 
 
 
 
 
 CO 
 
 
 
 t-oo>o»oi>p- 
 
 IT 
 
 '=^- 
 
 
 
 
 <c 
 
 X 
 
 
 Ig 
 
 
 
 
 
 H 
 
 
 
 
 
 (h 
 
 
 
 >-( 
 
 
 
 ^ « 
 
 
 
 o 
 
 
 
 ^-Q 
 
 
 
 to 
 
 
 
 £i 
 
 iHffliOOJOC 
 
 ly 
 
 CO 
 
 
 
 ot^i>ooo- 
 
 
 
 
 E-*^ 
 
 r-1 tHi-1 
 
 ur 
 
 
 
 ^ 
 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 ■1— 
 
 
 
 
 
 
 
 
 
 
 
 a 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 .2 
 
 
 
 
 
 
 
 
 
 
 
 ■3 
 
 
 
 
 
 
 
 
 
 
 
 
 
 c 
 
 
 
 
 
 
 
 
 
 3 
 
 
 C 
 
 
 
 
 
 
 
 
 
 §* 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 JJ 
 
 
 
 
 
 
 
 
 
 
 cd 
 
 
 "S 
 
 
 
 
 
 b„ 
 
 " 
 
 
 
 
 
 
 fl.S S a 
 
 ■£ 
 
 •2 
 
 
 1 
 
 
 mil 
 
 h 
 
 1 
 
 
 
 
 a 
 
 
 ■s 
 
 ^ 
 
 ^ 
 
 9-i 
 
 
 o 
 
 AIM 
 
 a a 
 
 .S|g- 
 
 ■So 
 
 III 
 
 o a 
 
 V 
 
 J" " 
 
 »- CJCs S 
 
 3 -Sa 
 
 3"" go' 
 gcoHqa 
 
 291 
 
292 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 ing into account the total population of each institutional group, we 
 note that the percentages vary from 3.3 per cent for the probation 
 group to 22.8 per cent for Auburn, the latter group showing a de- 
 cidedly larger percentage than any of the others. By reference to 
 the comparisons by nativity and color, we note that this high per- 
 centage for Auburn is largely due to its foreign element. 
 
 In Table 98 we offer data showing the percentages in the various 
 groups who were unable to read and write English. Although 
 we have no comparable data for the general population, these figures 
 have some importance as showing the perceritages within the de- 
 linquent group, who are handicapped by inability to read and write 
 the language of the country in which they live. As we should expect, 
 the percentages of this table are identical with the preceding for 
 the group of native white, but are many times larger for the foreign 
 white. Half of the group of our foreign delinquents are unable 
 to read even the simplest subject matter in English. The extent 
 to which this fact keeps them from assimilating the standards of this 
 country and from becoming informed regarding its laws is quite 
 impossible to estimate. Its significance, however, should not be 
 overlooked. 
 
 (c) Results of Educational Tests 
 
 We have already commented on the importance of the educa- 
 tional tests as furnishing another type_ of measure of educational 
 attainment, the special advantage of which consists in the fact that 
 it furnishes an objective measure and is thus free from the un- 
 certainty which attaches to the women's own statements regarding 
 educational progress. In the following section we offer data on cer- 
 tain of these tests. (For general account of the tests used see Chapter 
 IV). Our selection of tests to report at this time has been deter- 
 mined, partly by the desire to have tests representative of a variety 
 of educational processes, and partly by the desire to have tests which 
 had been reasonably well standardized with school groups. With 
 these objects in mind the following tests have been chosen: Trabue 
 scale A for reading; Buckingham scale for spelling (Thorndike selec- 
 tion) ; Ayres scale for handwriting ; Courtis— Series B — for arith- 
 metic. Since we have no data on the general population based on these 
 measures, we present our figures mainly for their value for later refer- 
 ence. We have indicated such standards as are available showing what 
 
g 
 
 I 
 
 3 
 
 o 
 
 ^ 
 
 
 Sal J 
 
 
 SJ.g 
 III 
 
 ".S 
 
 ^g 
 
 
 fc5l 
 
 Z3a 
 
 S « fl 
 *n d 
 
 ^1 
 
 gj <u c 
 
 is s 
 
 ^1 
 
 OeO "rHt^O 
 
 O-H •■*too 
 
 OINOOcOt- 
 
 t> (N 00 135 so t- 
 
 ■*t-IO-^00C<3 
 
 MIMOi-H^IN 
 
 OlOt^OlOOO 
 
 T-H 05 1C 0500 
 
 Ot^t-OOOi 
 
 
 293 
 
 
 00 
 
 
 o 
 
 H 
 
294 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 score is to be expected from children now in school in any given grade. 
 Formulation of the norms which we have used may be found in the 
 appropriate references cited in Chapter IV. 
 
 (1) Reading — Trabue S cole r^:-Tah\e 99 shows the numbers and 
 percentages ofThe-variaas"institutional groups and of the total who 
 reached the median score for each grade. It shows also the median 
 for each group with the grade standard to which this corresponds. 
 English-speaking cases only were considered. From inspection of the 
 table it is apparent that all groups except the Magdalen show appre- 
 ciable percentages who are unable to do better than children in the 
 second grade. 
 
 The median score of the total group reaches the standard achieved 
 by children in fifth grade. This is interesting in comparison with 
 data regarding grade completed. We see by Table 88 that the 
 native born women had completed, on the average, 5.0 grades. Their 
 accomplishment along the lines of reading are obviously quite closely 
 in accord with these figures on grade. Among the institutional groups 
 the Magdalen, the Penitentiary and, strangely enough. Auburn, rank 
 highest, rising to the standard for sixth grade. Bedford and Proba- 
 tion are, however, on approximately the same level as these, as shown 
 by the median scores, even though they fall below the dividing line 
 between the standard for sixth grade and that for fifth. The Work- 
 house is a distinct step, lower, rising only to the level of fourth 
 grade. 
 
 (2) Spelling — Buckingham. — Table 100 shows data in similar form 
 to that given for the Trabue scale in Table 99. The grade standing 
 for the group as a whole, as well as for most of the sub-groups, is 
 higher by this measure than by the reading scale. The total reaches 
 the 6th grade standard : the Magdalen group rises to the 8th grade, the 
 Penitentiary and Probation to the 7th, Bedford and Auburn to the 
 6th, the Workhouse again drops as low as the 4th. 
 
 (3) Handwriting — Ayres Scale. — Handwriting has been measured 
 from two points of view, quality and speed. To get the former meas- 
 ure, samples were compared with Ayres' samples of adult handwriting. 
 The distributions of results are shown in Table 101. The lowness of 
 the grade levels of our groups, measured by these standards, is very 
 striking. In no case does the median of one of our groups exceed the 
 median for the 2nd grade. For three of the groups, and for the total, 
 the median falls below that for 2nd grade. 
 
s 
 
 o 
 •a 
 
 2 
 o 
 
 I 
 
 d 
 o 
 
 A 
 ca 
 
 o 
 O 
 
 d 
 cd 
 
 m 
 
 s t 
 
 S o 
 CO «, 
 
 (1) bo 
 
 S 
 
 ^ I 
 
 d "3) 
 
 <1> 
 Ph 
 
 ■a 
 d 
 
 c« 
 
 ca 
 u 
 (U 
 
 •i 
 
 d 
 
 o 
 •■C 
 
 
 i i' 
 
 t» 00 "5 OS t~ M (O "3 
 
 o 
 
 
 
 3 
 
 4 
 
 6 ^ c<i U5 >n O) •* 00 
 
 d 
 
 
 
 (N .1 rH rH r-l 
 
 o 
 
 
 
 1 
 
 " 
 
 
 iH 
 
 
 
 , 
 
 
 
 00 
 
 
 
 §S 
 
 O) (N to OS CO OS ■* lO 
 
 t- 
 
 d 
 
 ^ 
 
 
 1-° 
 
 O to CD 00 00 ■* <N •* 
 iH 
 
 s 
 
 IN 
 
 lO 
 
 
 ■1 
 
 OS b- OS 00 O M 00 00 
 
 o 
 
 
 
 § 
 
 
 00 uj 00 1^ --H M M 
 
 d 
 
 
 
 § 
 
 rH 1-H C^ rH tH 
 
 o 
 
 
 
 t> 
 
 
 tH 
 
 
 
 1 
 
 
 
 
 
 
 1 
 
 
 
 OS 
 
 is 
 
 OCOOi-iOstOC^KM 
 
 CO 
 
 d 
 
 ■5 
 
 
 1-^ 
 
 T-H T-H i-H 
 
 •o 
 
 IM 
 
 s 
 
 EQ 
 
 h d 
 
 00-*OOIMM(Nr-l(N 
 
 o 
 
 
 
 C 
 
 «l 
 
 O "3 O to ■* (N 00 (N 
 
 d 
 
 
 
 2 
 
 rH »H iH N I— I i-H 
 
 o 
 
 
 
 H 
 
 n 
 
 o 
 
 
 ^^ 
 
 
 
 
 
 
 o 
 
 
 ^ 
 
 1^ 
 
 00 ■* 00 IN 00 OS to OS 
 
 ;* 
 
 IV 
 
 :g 
 
 
 tH 1-H 
 
 t> 
 
 I-H 
 
 ■* 
 
 a 
 
 a 
 
 t~ i~ CO o o eo 50 1> 
 
 o 
 
 
 
 § 
 
 ^1 
 
 ocbeocdiNinTH-* 
 
 d 
 
 
 
 
 CO »H i-H T-H 1-1 
 
 o 
 
 1-H 
 
 
 
 1 
 
 
 
 
 
 
 
 
 
 CO 
 
 i 
 
 1^ 
 
 CO U3 O (N OS ■* rH i-< 
 
 >o 
 
 IM 
 
 1 
 
 ftfl 
 
 (N rt i-( 1-1 
 
 1> 
 
 (N 
 
 s 
 
 
 .-l(N m N 11^ CD • • 
 
 CO OS r-i 1-H lO OS • - 
 
 o 
 d 
 
 
 
 C^iHi-HCqi-H • • 
 
 o 
 
 
 
 1 
 
 
 
 iH 
 
 
 
 
 
 
 CO 
 
 
 S 
 
 a a) 
 
 (NOCDi-l00>O • • 
 
 ly 
 
 K 
 
 ^ 
 
 
 1^ 
 
 1— I t-H I— I • • 
 
 no 
 
 IN 
 
 s 
 
 
 '1 
 
 >0 i-H ■<}( CO (N (N CO <N 
 
 o 
 
 
 
 
 ■ ^ fS. 
 
 
 
 
 
 ^ 
 
 
 di-lo6>OCOCO«3CO 
 
 d 
 
 
 
 
 i-t(Ni-t iH T-l 1-1 
 
 o 
 
 
 
 
 
 
 iH 
 
 
 
 1 
 
 
 
 
 
 
 , 
 
 
 
 IN 
 
 
 
 S ID 
 
 1-° 
 
 Th 00 1> N >o lo <N m 
 
 00 
 CO 
 
 i^ 
 
 1 
 
 
 <l 
 
 ■H CD CD -* U3 m m t> 
 
 o 
 
 
 
 
 A ea 
 
 
 
 
 
 fl 
 
 ^"1 
 
 (NC001v-*odu3t> 
 
 d 
 
 
 
 
 C<J »H 1-H 1-* iH 
 
 o 
 
 
 
 s 
 
 V 
 
 
 1-H 
 
 
 
 1 
 
 
 
 
 
 
 1 
 
 
 
 (>» 
 
 is 
 
 (NIN«3i-H-*OC000 
 
 lij 
 
 tH 
 
 5 
 
 
 lOC0C<l-*C0(Ni-Hi-H 
 
 ?5 
 
 <N 
 
 lo 
 
 
 * 
 
 ^ i 
 
 
 
 
 
 : 
 
 
 i 
 
 -^ 
 
 
 
 1: 
 
 
 
 
 
 t> 
 
 3 
 
 
 
 art 
 Ob 
 
 -§ ; 
 
 
 
 
 
 ^ 
 
 
 J 
 
 
 S 
 ^ 
 
 03 g 
 
 
 1 
 
 lls" 
 
 
 :g 
 
 -^ -(^ -»^ 
 
 Q01>CC 
 
 iJ: 
 
 -4^ 
 
 ^ 
 
 ^ 
 
 ffl 
 
 
 § 
 
 o 
 
 295 
 
s 
 ^ 
 
 a to 
 
 if 
 
 CLi fi 
 
 ft fl 
 
 03+3 
 
 
 
 Pi >3 
 
 IS 
 
 oooit-^ooio-*o 
 
 lOCDO5t*00t^O3O5 
 
 Cv|C<l »H rH 
 
 Mi-HiMr^KSt-coeo 
 
 ooicit~tDi-i<Nt~'n 
 
 OOt-INCOtOOSOSO 
 
 l>OCOCOt~1^000 
 
 t^OO-^t-'O'OOSO 
 
 ooo -ooo 
 
 >Oi-ltO •OOCD-* 
 
 "3 O ■* to 1> 00 ■* U5 
 
 rot>u5T-icqtJ"3co 
 
 >-l(N d T-H >H 
 
 "5 O IN 00 rH ■* (N "O 
 
 THU3t*rMiOt*l>t* 
 
 OOOOSOOOOSOi-l 
 
 O to ■* O CO tH 05 05 
 ■!H-*(N(N(N<MCOIN 
 
 ■< 00 1> to iffl -ai CO W 
 
 
 ^ 
 
 £ 
 
 O 
 
 I 
 
 2g6 
 
 =3 
 I 
 
 a 
 
 a 
 
 a 
 
 3 
 
o t-t .3 
 
 S 3 ;3 
 
 
 .1 
 
 U3 O 00 ■* O 1> tt> 05 
 
 O 
 
 
 1 ^ 
 
 s 
 
 ^1 
 
 T-l U5 00 1> >-i CD (N to 
 ■-i(N(N>-i 
 
 8 
 
 
 ^ 
 
 g 
 
 " 
 
 
 1—1 
 
 
 & 
 
 & 
 
 
 
 
 ■* 
 
 -§ 
 
 1 
 
 
 
 
 1^ 
 
 '^Sl^SSSJ^g?? 
 
 CD 
 
 CO 
 
 m 
 
 
 1> 
 
 CO 
 
 
 
 T-l IH 
 
 ■* 
 
 
 
 o 
 
 
 . CO 00 CO iH lO 0> 00 
 • (N CD CO OS lO O CD 
 
 O 
 
 d 
 
 
 1 
 
 % 
 
 1-4 -^ i-H 
 
 o 
 
 
 n 
 
 
 
 
 
 1 
 
 1^ 
 
 ■ 1-1 OT CD ■* O 1> CO 
 
 -* 
 
 00 
 
 d 
 
 
 (N 
 
 ■* 
 
 CO 
 
 
 ^1 
 
 • ■ t^ ■* IN T-lin r-l 
 ■ • im' lO IN irj CD 00 
 
 O 
 
 
 ^ 
 
 • • »H coco 
 
 tH 
 
 
 1 
 
 
 
 
 IN 
 
 CD 
 
 i 
 
 a- 
 
 • • (N ■* OS CO t> CO 
 
 1* 
 
 1^ 
 
 ■ ■ IN(N 
 
 t~ 
 
 CVI 
 
 h 
 
 ^1 
 
 8 
 
 ■ ■•* 00 1-1 CO CD 00 
 
 O 
 
 
 
 5 
 
 ■ ■ t> 00 (N lii t^ 00 
 
 d 
 
 
 
 F4 
 
 • • INCOrt 
 
 o 
 1— ( 
 
 
 § 
 
 
 
 
 "* 
 
 1 
 
 
 
 
 s 
 
 H 
 
 H f^ 
 
 • • lO CD ICI tH <N CD 
 
 00 
 
 
 "oS 
 
 ll< 
 
 1- 
 
 1-llNi-l 
 
 CD 
 
 s 
 
 ffl 
 
 
 a 
 
 OOiOTflOiHOCOOJ 
 
 o 
 
 
 
 1 
 
 ^1 
 
 COt^CSt^CNt^i-H^ 
 
 o 
 
 
 
 i-< COi-l 1-1 
 
 o 
 
 
 
 «<! 
 
 
 tH 
 
 
 
 
 
 
 
 
 
 
 §s 
 
 (N^iOOit-OJCOiH 
 
 CO 
 
 o 
 1-1 
 
 ■a 
 
 
 
 
 1^ 
 
 iH 
 
 lO 
 
 Tj< 
 
 IN 
 
 
 . 1 
 
 03CDINTfl(N-;t(IN<N 
 
 o 
 
 
 
 
 i a 
 
 
 
 
 
 * 
 
 ^1 
 
 i>cNcoodcoodcoco 
 
 ^ 
 
 
 
 s 
 
 T-H 1-1 lH 1-H i-H i-t 
 
 
 
 
 § 
 
 o 
 
 
 1-t 
 
 
 
 g 
 
 
 
 
 ' lO 
 
 ■d 
 
 ^s 
 
 
 
 
 C0i-<^t^lOt^iO»O 
 
 00 
 
 d 
 
 d 
 
 
 1- 
 
 
 CO 
 
 CO 
 
 (N 
 
 
 ,a 
 
 OlOCDi-HCDCOOlO 
 
 o 
 
 
 
 
 h d 
 
 
 
 
 
 o 
 
 S 
 
 i-lCOOi-<OcOI>CO 
 
 d 
 
 
 
 
 i-l(N(N(N 
 
 o 
 1-i 
 
 
 
 § 
 
 
 
 
 ■* 
 
 
 
 
 
 m 
 
 1^ 
 
 NCO-HINi-lCOTtICO 
 
 05 
 
 d 
 
 "§ 
 
 
 1-1 <N -^ -^ iCi-t 1-1 
 
 OJ 
 
 CO 
 
 IN 
 
 
 l?'^ 
 
 
 1-1 
 
 
 
 
 
 
 
 
 
 
 
 
 
 £ 
 
 
 
 p 
 
 
 
 
 
 
 
 i 
 
 & 
 ^ 
 
 1 
 
 o 
 
 & 
 
 J 
 
 
 
 oooooooje 1 
 
 
 § 
 
 O 
 
 
 
 00 1> 
 
 CO 
 
 IQ 
 
 ■* 
 
 00 
 
 (M 
 
 H 1 
 
 
 297 
 
298 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 When we turn to figures on rate of writing given in Table 102, 
 we find a very different result. All of our groups but one— the 
 Workhouse — achieve the level of 8th grade accomplishment. Even 
 the Woirkhouse median is as high as that for 7th grade. It appears, 
 therefore, that the members of our group have an admirable speed but 
 a very inferior quality of penmanship. 
 
 (4) Ar ithmetic — Courtis. — As representative of accomplishment in 
 arithmetic we offer data on the two fundamental processes of addition 
 and division, as measured by the Courtis scale. In certain ways we 
 preferred the Woody scale for our purposes since it does not exceed 
 the limitations of the ability of our women so strikingly as does the 
 Courtis scale. We have chosen to report on the latter, however, be- 
 cause of the importance of the grade norms established for this scale. 
 
 Table 103 shows the distribution of scores for the problems in 
 addition. The superiority of the Magdalen is again apparent, both 
 from inspection of the distribution and from consideration of the 
 grade norm (5th) to which its median score corresponds. Bedford, 
 the Workhouse and Probation reach only 4th grade standard, while 
 Auburn and the Penitentiary fall below the 4th grade norm. 
 
 Table 104 shows the results obtained with the Courtis scale for 
 division and makes apparent how markedly this scale overshot the 
 ability of our group. Over half of the total group were unable to 
 complete a single problem correctly. In the Workhouse as many 
 as 86 per cent were unable to get a single problem, and even in the 
 Magdalen 31 per cent failed entirely. Observation of the individuals 
 at work indicated that a certain proportion of these failures were 
 due to an inabihty to recall the method of long division, though 
 this had at one time been learned. The attempt to recall this strange 
 and never-used mechanism showed a marked similarity to the per- 
 formance of the average college student v/hen suddenly confronted 
 with a problem in cube root. The level for the group as a whole 
 was below the median for 4th grade. Only Bedford and the Mag- 
 dalen rose above this, the former reaching the 4th grade standard 
 and the latter the 5th grade. 
 
 It is apparent from the foregoing account that the educational 
 tests do not offer a more favorable impression of the school attain- 
 ments of these women than do the data regarding school grade. In 
 only isolated instances do the groups rise above the level for 5th 
 grade and in numerous instances they fall to 4th grade standards or 
 below. The highest scores, from point of view of grade norms, were 
 
i^ I 
 
 s 
 ^ 
 
 y W 
 
 
 a) 
 
 
 
 
 
 
 
 
 
 
 
 
 
 i§' 
 
 lo ^ ^ CO m ■* -^ t> 
 
 o 
 
 
 
 3 
 
 S| 
 
 in CD iri Tj( U5 CO M «d 
 
 CO 
 
 S 
 
 ?: 
 
 
 ■< 
 
 o 
 
 
 1-^ 
 
 
 
 ^ 
 
 
 
 
 1 
 
 tS 
 
 
 --ICTITIINIOCOCOIN 
 
 in 
 
 
 ^ (N IM (N IM -H -H TO 
 
 t> 
 
 < 
 
 00 
 
 
 CO 
 
 ■* 
 
 
 
 
 iL rf' 
 
 t»ooeooocoeoco>o 
 
 o 
 
 
 
 o 
 
 o 
 
 <N CD C<i CO d N IN ■* 
 
 8 
 
 g 
 
 
 § 
 
 
 
 
 ^ 
 
 
 
 
 
 £ 
 
 n '-' 
 
 iNeoiHeoi-ii-H>-i<N 
 
 •* 
 
 4 
 
 :S 
 
 
 1-^ 
 
 CO 
 
 "* 
 
 00 
 
 § 
 
 il 
 
 t> t>- lO ■* CD 00 00 CO 
 
 o 
 
 
 
 GQ 
 
 P 
 
 
 O5C0«3-*oicDCDN 
 
 d 
 
 
 
 O 
 
 CO-H 1-H 
 
 o 
 
 o> 
 
 
 a 
 
 u 
 
 
 l-H 
 
 CO 
 
 
 
 
 
 •1 
 
 Is 
 
 o>o-*^i>>o>oo5 
 
 CO 
 
 
 ■s 
 
 
 1-^ 
 
 IN w 
 
 i> 
 
 
 t~ 
 
 s 
 
 1 y 
 
 (N ■ OJ 0> Tl< OS ■* CO 
 
 o 
 
 
 
 fi; 
 
 fe-3 
 
 
 
 
 
 a 
 
 
 CO ■ in lo i>^ (N ■* d 
 
 CD 
 
 0:1 
 
 
 g 
 
 CO • iH 
 
 O 
 
 I> 
 
 
 ^ 
 
 o 
 
 
 T-t 
 
 
 
 ^ 
 
 
 
 
 f 
 
 
 
 
 
 
 
 » 
 
 9fe 
 
 CO ■ ■* ■* >n (N CO t^ 
 
 00 
 
 
 
 ■s 
 
 Ph 
 
 1-" 
 
 ■* ■ 
 
 CO 
 
 00 
 
 
 1 ^ 
 
 .-1 00 t^ OJ C: 05 00 ■ 
 
 q 
 
 
 
 1 
 
 
 1-H CO in T-H 1-H tH CO • 
 00 
 
 d 
 
 o 
 
 g 
 
 
 ■< 
 
 o 
 
 
 rH 
 
 
 
 Q 
 
 
 
 
 V 
 
 
 
 
 
 
 
 
 
 h 
 
 CO IN CO i-H i-H .-l(N • 
 
 CO 
 
 :S 
 
 
 1^ 
 
 •* 
 
 m 
 
 00 
 
 
 1 3* 
 
 O5C0C0O5 -m -IN 
 
 q 
 
 
 
 
 iH d 
 
 
 
 
 
 ^ 
 
 S| 
 
 t^in >nt>. -o -co 
 
 d 
 
 05 
 
 
 g 
 
 in -T-t • i-H 
 
 o 
 
 T-l 
 
 t> 
 
 
 n 
 
 
 
 
 g 
 
 
 p 
 
 
 
 
 
 <! 
 
 1^ 
 
 <N(N <NCO •■* -in 
 
 (N 
 
 00 
 CO 
 
 
 
 1 
 
 
 g, 
 
 ■^oominminin 
 
 o 
 
 
 
 
 M 03 
 
 
 
 
 
 § 
 
 ^1 
 
 T-icDincoiniHc<»'^ 
 
 8 
 
 
 
 § 
 
 o 
 
 
 1-H 
 
 1 
 
 
 
 
 
 cq 
 
 is 
 
 IN IN O t^ --H CO lO 02 
 
 OS 
 
 4 
 
 1 
 
 
 k5-0 
 
 ■^ .— I t—f 1-H 
 
 Oi 
 
 QO 
 
 
 12; 
 
 T-4 
 
 1-1 
 
 
 
 
 d 
 
 
 
 
 
 
 
 
 
 
 a; M 
 
 -e 
 
 
 3„ 
 
 
 
 
 
 
 
 
 
 
 ftH 0) 
 
 Fl 
 
 
 1^ 
 
 .£ 
 
 ^ 
 
 ,£ 
 
 ^ 
 
 X 
 
 -^ 
 
 x 
 ^ 
 
 1 
 1 
 
 1 
 
 ft 
 
 S 
 1 
 
 
 
 3c 
 
 S 
 
 4J 
 
 CO 
 
 iS 
 
 ^ 
 
 ci; 
 
 m 
 
 
 s 
 
 
 
 
 299 
 
el 
 
 <4< C4 <«14 C4 ^ O O 00 "# « 01 C4 CO 64 CO O ra •<}< C4 -^^l '^ b- 
 
 m 
 
 I 
 
 I 
 o 
 
 I 
 
 
 ^.^^ 
 
 a 
 
 
 m 
 
 o 
 a. 
 
 
 CO 
 
 to 
 
 i 
 
 
 (I) 
 
 u 
 
 
 en 
 
 1 
 
 
 ai 
 
 09 
 
 
 M 
 
 1h 
 
 
 H 
 
 o 
 
 
 34 
 P 
 
 u 
 
 CO 
 
 s 
 
 O 
 
 J 
 
 ▼^ 
 
 
 '« 
 
 S 
 
 fe 
 
 s 
 
 pq 
 
 i 
 
 1 
 
 
 Q 
 
 
 
 1 
 
 1 
 
 
 o 
 
 IS 
 
 
 
 pj 
 
 
 P^ 
 
 ■o 
 
 
 g 
 
 S 
 
 
 W 
 
 
 
 h 
 
 bj 
 
 >> 
 
 •a 
 o 
 
 ^• 
 
 
 i!. oa 
 
 
 4. od 
 
 4au 
 cS 
 
 Cl, c 
 
 •o . -oo -oooooooo 
 ■ 01 * 'csiT}< •cDiNoocqodoo-^'o 
 
 . . . i-H 1-H iHM 
 
 • rH - -iHIN •CO<D'^(0^'^t»0 
 • • • I-H 
 
 . • ■ -^ iH t1< I-H '^ t- lO W O 00 CO U) 
 
 . . . i-Hi-i i-ieo 
 
 . . .«HC0i-HCCi-lC<I"*MMiOOiC0 
 
 H 04 «' 00 i-H CO U3 ci 03 .H 00 ^ 
 
 HMOflCO^WJ-^JiOlC^OOeOiO 
 
 C4 -0404 
 
 ■ « -04 ■ ■^ ■* N 04 --I OS ^* 00 "* i-i ■^ w 
 
 ■04 -04 ■■^■*04«i-<00eD;t*^r-l-^.-H 
 
 .,H .iH . C4 04 rH rH lO TJH CO 00 04 U3 04 "3 
 
 4 011 
 OS 
 
 
 ii! 
 
 15 -S 
 
 t-'^t^t* . -^QOlOt^ 
 
 • ■ • I I ! r I ^ ! : r rcQi>ww - -b^-^Qod 
 
 • • r-lrH-* 
 
 rHC4rHrH . .C4'^»OrH 
 
 ■Tf aTi<o»^TiiThcfioo-*cQcoa«rHQ004 
 
 CQ04COO}t>Offi01rHO>0) 
 
 •y-t 04 rt 04 iH .-H 00 so OS 04 b- tH CO « (O M in 
 
 M r-i 04 04 « 04 04 "* 
 
 SE3SO****^*OiO'^«C4»H003XI>(0»0^eoC4i-10 
 C4 C4 C4 "H rH rH 1-i i-l rH rH rH iH iH ■ 
 
 300 
 
 IS 
 
 a 
 
N • rH to ■* O ■* 00 IM (N (N ■* IN IM O (N 00 
 
 O 
 
 o 
 
 3 1, ^ 
 
 ■C<100NlOC<l-<*.-li-ltO<Ni-l,-iOiHO 
 
 rH rH 1-H CO C<I CO Ttl ro 03 
 
 (N 
 
 o 
 
 
 S>. 
 
 • •.... -O |00000000 
 
 (N -Tti-^NtOtbofflO 
 
 T-t r-l U3 
 
 "H • N M to CO CO "3 CO >0 
 
 <N 
 
 ; -*Tt<.-i -ooeo 
 
 • ■ tH 1-H -^ • CO CO 
 
 -00 
 
 1-1,-lCO -lOCO 
 
 -co 
 
 ■*■•••■*• •■*I>-*>OMt>00iOO 
 
 r-t ■ • • -T-t • ■ T-H Cfl 1-H »0 00 W to U3 c4 
 
 to 
 
 
 
 o 
 o 
 
 
 3 (-. ,_| 
 
 !■" 
 
 ■IMCM 
 ■INN 
 
 •iNai05t>-*t-co»>-*i-H 
 
 •,-lTi(-*COC<lCOI>COIN-* 
 
 to 
 
 ' CO 
 
 (^S 
 
 t>oO"-i-* 
 
 ! ! i i : ; ! ! . !co ■■* -tho 
 
 • T-t • tH t* 
 
 !-!•■* -COOl 
 
 • r-i 
 
 •Tfosait-ojoitococot-TtiTHiooto 
 
 i-l IN 1-1 -^ l> ec lO "-I to (N 
 
 o 
 o 
 
 •.-HCv|IN-*ININtOCOOOOOOINt>T)(CO 
 rH i-t T-4 N T-t C^ 
 
 
 eO«3'*COINi-i00500t^tOiO"*COC<li-lO 
 
 ^H iH T-t rH I-l tH i-t 
 
 301 
 
 o 
 Eh 
 
 ID 
 
 > 
 
 1, 
 
302 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 made on spelling and on rate of handwriting; the lowest on long divi- 
 sion and on quality of handwriting. 
 
 The consistently low record of the Workhouse by these measures 
 is of interest as tending to confirm our suspicion that they had 
 overstated the facts regarding grade attainments. 
 
 SUMMARY 
 
 From the data presented in this chapter it appears : 
 
 1. That the delinquent women fall conspicuously below the 
 standards of present-day opinion as to what should be expected 
 regarding both the age at which one should be allowed to leave 
 school and the minimum grade which he should have reached. Never- 
 theless we are unable to state whether our women fall below the 
 level of the general public in these respects. 
 
 2. That their school attainment, as indicated by grade com- 
 pleted, has not kept pace with their opportunities as represented 
 by the number of years in school. 
 
 3. That there is a distinct tendency toward poorer educational 
 opportunity and attainments among the older women of our group 
 as compared with the younger women. 
 
 4. That the native-white are superior to both the foreign born 
 and the colored with respect to both school opportunity and school 
 attainment: and that the foreign born are inferior to the colored 
 in both respects. 
 
 5. That the more striking institutional differences appear to 
 be due to one or the other of the two factors just mentioned — age 
 and nativity diffei"ences — except in so far as they are occasioned by 
 differences in degree of verification of facts. 
 
 6. That the amount of illiteracy among the delinquent groups 
 is relatively high, varying from 4 per cent for the native white to 
 24 per cent for the foreign white: that these percentages are de- 
 cidedly higher than those for similar groups in the general population, 
 though the significance of this comparison is weakened by the fact 
 that the two types of groups have not been measured by equally severe 
 standards. 
 
 7. That the relationships between education and the various as- 
 pects of the delinquent career are at most slight. 
 
 Some tendency appeared for those with least schooling and those 
 with most schooling to have been convicted first at a later age than 
 those falling in the intermediate range regarding schooling. The 
 
EDUCATIONAL BACKGROUND 303 
 
 influence of other factors than schoohng was suspected as producing 
 this relationship. 
 
 A sHght relationship was found between grade and nature of 
 offense, with those guilty of oifenses against property, against chastity, 
 and against regulations* for public health having reached slightly 
 higher grades than those guilty of other offenses. 
 
 Unambiguous evidence of relationship either between years in 
 school and number of convictions, or between grade and number of 
 convictions, was not found. Even allowance for the effect of age, 
 and of intelligence, did not disclose the presence of such a rela- 
 tionship. 
 
CHAPTER XI ■ 
 OCCUPATIONAL HISTORY AND ECONOMIC EFFICIENCY 
 
 THE history of criminal science is rife with various contentions as to 
 the essential factors most closely related to criminality. One of the 
 most fruitful fields of discussion has been, and still is, the importance 
 of stressing the economic system of recent years as entirely unassoci- 
 ated with, or, on the other hard, directly inherent in the etiology of 
 crime. The two most divergent schools in this respect are probably the 
 Italian School, fathered by Lombroso, and the Socialists, known as the 
 "Third School." The doctrine of the Italian School, as has previously 
 been stated, is based on the theory that there is a definite criminal type, 
 distinguished from the non-criminal element of the population by 
 physical and mental stigmata. Though these criminal anthropologists 
 do not deny that adverse economic conditions may often be somewhat 
 responsible for certain types of crime, they contend that the real cause 
 of crime is innate. The Socialists, on the other hand, go to the extreme 
 of emphasizing the economic factor over all others, and assuming that 
 with a more socialistic regime of the economic system the amount of 
 crime would gradually diminish. This theory was advanced by Karl 
 Marx : 
 
 "In the social production which men carry on they enter into definite 
 relations that are indispensable and independent of their will; these rela- 
 tions of production correspond to a definite stage of development of their 
 material powers of production. The sum total of these relations of produc- 
 tion constitutes the economic structure of society — the real foundation, on 
 which rise legal and political superstructures and to which correspond 
 definite forms of social consciousness. The mode of production in material 
 life determines the general character of the social, political and spiritual 
 processes of life. It is not the consciousness of men that determines their 
 existence, but, on the contrary, their social existence determines their 
 consciousness." ' 
 
 ' Marx, Karl. "A Contribution to the Critique of Political Economy," p. 11. 
 Translated from the Second German Edition by N. I. Stone. New York: The 
 International Library Publishing Co., 1904. 
 
 304 
 
OCCUPATIONAL HISTORY AND EFFICIENCY 305 
 
 Many of the tenets of the criminal anthropologists have been proved 
 unsound, particularly those relating to the presence of physical stig- 
 mata in the criminal. The work of the later criminologists has also 
 discredited the extreme views of the Socialists, that all the relations of 
 mankind, associated with criminality, are exclusively determined by 
 economic relations. 
 
 Our study of 587 women, though comparatively small in numbers, 
 includes women from all over New York state with the largest num-> 
 ber from New York city. This city of infinite economic resources 
 expending themselves in occupations of almost every description offers 
 us an admirable opportunity for observation of the industrial situation 
 as it has affected these women convicted of breaking the law. At the 
 'outset we must disagree with the extremists of the Socialist School in 
 their lack of emphasis on the individual mental defect or abnormality 
 which may be the determining factor in making one member of a family 
 an offender against the law, while all the other members — under the 
 same economic system — do not become anti-social in their actions. 
 That the present economic system is undoubtedly responsible, to a cer- 
 tain extent, for many of the factors associated with delinquency, such 
 as the economic status of home conditions discussed in Chapter IX, no 
 one will deny. But that this alone is the prime cause of crime is as 
 futile to assert as that the development of modem industry has had 
 no effect on the complications which are associated with crime. 
 
 We have attempted in this study to give a bird's-eye survey of the 
 work histories of these women from the time of starting work to the 
 present conviction, and including not only the prevailing kind of work 
 done, with the average wage for this work, but also the kind of work 
 and the wage in the first job, at time of the first and present conviction 
 and in the latest job.'^ An effort to summarize the various factors of 
 the work record has been made in the estimate of regularity of the 
 work, and the estimate of the entire work history. 
 
 It is not easy from such a summary to draw definite conclusions 
 because of the fact that for the general population no adequate wage 
 studies have recently been made for all occupations in a given com- 
 munity, and there are no other studies comparable to ours in the con- 
 sideration of wage. Even studies of delinquency are notably lacking 
 in more than brief statements of the average wage or the highest and 
 
 'Throughout, "work" has been used to signify work in which one was 
 gainfully employed, and does not include the work of women who were doing 
 their own housework. 
 
306 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 lowest wage, with no effort to determine what kind of work was 
 being done at any crucial time in the delinquent's life, how much of the 
 time he had been idle, and what his earning capacity was at any given 
 significant time. Wherever possible, comparisons will be made with 
 isolated studies of wage and kind of work for the general population, 
 but these are necessarily limited to few. That our group shows a low 
 earning capacity and irregular work might not be significant if com- 
 pared with equally reliable data for all women in New York state. 
 But since the other part of the comparison is not available at this 
 time, we shall present ours as indicative of the type of record that 
 occurs among delinquent women, and leave the rest of the comparisdh 
 to follow later. 
 
 AGE AT STARTING WORK 
 
 The Child Labor Law in New York State provides that no children 
 may be legally employed under 14 years, and that children from 14 to 
 16 must secure certificates before they can be employed legally.^ 
 Although this law has been in effect for many years, it is interesting to 
 note, from Table 105, that 80 women, or 14.8 per cent of the total 
 group, began to work before they were fourteen years of age. Of this 
 number, 21 women were foreign born so that this law would not apply 
 to them, and 26, though native born, lived during their childhood in 
 states other than New York, in most of which, however, there were 
 restrictions against child labor under fourteen years. At the least con- 
 sideration, there are between 7 and 13 per cent of the total group who 
 began to work before the age at which they were legally allowed to, and 
 which New York state has set as the minimum age possible for children 
 to work and retain their good health. Each institutional group is 
 represented among those who began to work before fourteen years, but 
 Auburn has by far the largest percentage, probably due to the fact that 
 there is in this institution a large percentage of foreign women and a 
 preponderance of the older women who began to work years ago when 
 the restrictions against child labor were few and not well enforced. 
 If we compare the percentage of women starting to work between ten 
 and fourteen years in our total group of delinquents, with the percent- 
 age of the general female population between ten and fourteen years 
 of age in New York state who were working in 1910, we find that in 
 the one case the percentage is 14.8, while in the other it is only .1 of a 
 
 » Consolidated Laws 1909, vol. 3, chapter 31, article 6, section 70, as amended 
 by Chapter 529 of the laws of 1913. 
 
OCCUPATIONAL HISTORY AND EFFICIENCY 307 
 
 per cent.* These figures are not exactly comparable since the age at 
 starting work may extend over many different years while the census 
 figures apply to those of given ages working during one specified year. 
 They do, however, indicate a trend, which it is important to observe, 
 for the delinquent women to go to work earlier than the women of the 
 general population. 
 
 TABLE 105 
 
 AGE AT STARTING WORK 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups 
 
 
 Institutional GKonPS 
 
 
 Age at Starting Work 
 
 Bed- 
 ford 
 
 Auburn 
 
 Mag 
 dalei 
 
 Peni- 
 1 tentiary 
 
 Work- 
 house 
 
 Pro 
 tio 
 
 Total 
 aa- 
 
 n 
 
 6 to 10 years 
 
 4.1 
 14.3 
 69.4 
 10.2 
 
 5.3 
 
 24.6 
 45.6 
 
 5.3 
 10.5 
 
 1.8 
 
 'i!8 
 1.8 
 1.8 
 
 'i'.s 
 
 1.^ 
 
 14.5 
 
 56.? 
 
 21. ( 
 
 2.' 
 
 2.' 
 
 I .... 
 
 ) 15.0 
 i 34.0 
 5 21.0 
 
 r 8.0 
 r 8.0 
 
 7.0 
 1.0 
 3.0 
 1.0 
 2.0 
 
 1.4 
 
 20.0 
 
 32.9 
 
 18.6 
 
 10.0 
 
 2.9 
 
 2.9 
 
 4.3 
 
 1.4 
 
 2.9 
 
 1.4 
 
 2 
 65 
 17 
 
 6 
 
 5 
 
 1 
 
 1.9 
 
 10 " 14 " 
 
 7 14.8 
 
 14 " 18 " 
 
 8 51.1 
 
 18 " 22 " 
 
 8 16.1 
 
 22 " 26 " 
 
 9 5.9 
 
 26 " 30 " 
 
 2.0 
 
 5 4.2 
 
 30 " 34 " 
 
 4 2.1 
 
 34 " 38 " 
 
 
 1.1 
 
 38 " 42 " 
 
 
 1.1 
 
 42 " 46 " 
 
 
 .9 
 
 46 " 50 " 
 
 
 .6 
 
 50 " 64 " 
 
 
 .2 
 
 54 " 58 " 
 
 
 
 58 " 62 " 
 
 
 .2 
 
 
 
 
 Total 
 
 100.0 
 
 100.0 
 
 100. ( 
 
 ) 100.0 
 
 100.0 
 
 100 
 
 100.0 
 
 
 
 Number of cases 
 
 98 
 
 57 
 
 74 
 
 100 
 
 70 
 
 7i 
 
 J 472 
 
 
 
 The range of age at starting work, it will be seen, divides itself 
 into two distinct groups — ^those who have started work before 30 to 35 
 years, represented by the Bedford, Magdalen and Probation groups, in 
 which the average age at present conviction is lowest, and those who 
 have started work at a wide range of ages, represented by the Auburn, 
 Penitentiary and Workhouse groups. In the latter three institutions 
 it may be of interest to note briefly the women who were not gainfully 
 employed until they were 42 years of age or over. The two Auburn 
 women who fall in this group are both foreign, one born in Italy and 
 
 * Thirteenth Census of the United States, 1910. 
 p. 46. 
 
 "Occupational Statistics," 
 
308 STUDY OF WOMEN DELINQUENTS IN NEW YORK ' 
 
 one in Russia. The Italian woman was married to a man who drank 
 and finally began to go with other women. When she was 44 she left 
 him, was forced to support her family and began to practise midwifery. 
 She had done no work previous to this and was not trained to do any- 
 thing. Her present conviction was for abortion performed on one of 
 her patients. The Russian woman had never done any 'work outside 
 of her home until six years before the present conviction when her hus- 
 band died and she was obliged to support her children. At that time, 
 at the age of 53, she bought a dry-goods store. The store did not 
 prove to be very successful and she attempted to burn it. She was 
 convicted of attempted arson and evidence was brought at the trial that 
 she had made various previous atternpts to dispose of the store in. a like 
 manner. 
 
 The three Penitentiary women represent somewhat difiEerent types 
 than those just mentioned. One woman, now 47 years old, was mar- 
 ried when 32. She never did any work before or after her marriage 
 until three years before her conviction, when she was short of money. 
 Previous to this her husband had left her because she was a drug user. 
 He gave her a large sum of money, but she spent so much for drugs 
 that it did not last long and she was forced to earn more. Another 
 Penitentiary woman, now 63 years, born in Austria, was married when 
 young and did no work except the housework and work on her fa- 
 ther's and husband's farm. Years after, when she had been in this 
 country for a short time, she worked three or four years as a coat 
 finisher in a factory. This was only to tide over a critical period in 
 the family finances. The third Penitentiary woman is SO years old,' 
 colored, born in the South. She has probably worked very little, 
 except for short and irregular times the year before the present con- 
 viction when her consort was in a hospital. She admitted that she had 
 prostituted many years, had used heroin steadily, and that the work 
 was only an incidental thing in her life. 
 
 The four Workhouse women who did no work outside of their 
 homes until after they were 42 are all white, two native and two for- 
 eign. The first woman is now 43. She was born in Sweden and came 
 to this country when 23. Two years later she married a man whom 
 she later divorced, and after leaving him she worked for a short time 
 at general housework. She has had a long record of prostitution and 
 the only work she did was probably sporadic. Another woman, now 
 46 years old, did no work until the year before her present arrest, 
 when she worked as waitress in a hotel. Her husband had died several 
 
OCCUPATIONAL HISTORY AND EFFICIENCY 309 
 
 ', years before. She was a most erratic, unstable person and had been in 
 a hospital for the insane for some time after her husband's death. The 
 third Workhouse case in question is a woman SL years old who has 
 been in this country for three years. Soon after arriving here she be- 
 gan to work as cook. Her husband had died fourteen years before her 
 coming to this country and during that time she had a "lover," a promi- 
 nent lawyer in Berlin, who supported her. The fourth Workhouse case 
 is a woman who has been married twice, her second husband dying just 
 one year previous to the time she began to work. She says that when 
 she was a girl her family was in comfortable circumstances. "I never 
 knew what it meant to soil my hands, we had so many Chinese ser- 
 vants." Her present conviction was for violation of the Tenement 
 House Law, and it seems likely that she had been prostituting for some 
 time. Her short work record was very irregular. 
 
 As a whole, the group of older women who did not begin to work 
 until late in life is small and represents women who have not found 
 it necessary to work before, either because they were married and being 
 supported by their husbands or because they were obtaining money in 
 an illegal manner. The work records of these women are, for the most 
 part, very brief and the various jobs are so isolated that they are 
 scarcely comparable to the records of most of the younger women. 
 
 A frequent contention is that starting to work at an early age 
 is a common cause of getting into the courts through meeting bad 
 acquaintances, losing respect for the head of the family through in- 
 creased independence, as well as for other reasons intrinsic in the 
 occupational situation, notably the tendency to get into "blind alley" 
 jobs. Table 106 shows the relationship which exists for our group 
 of women between age at first conviction and age at starting work. 
 There is clear evidence of a relationship here as shown by the coefficient 
 of .correlation of .5 lit .034. However, we can not accept these data 
 as evidence that there is a real tendency for those who start work young 
 to be convicted early in life and those who start work later not to be 
 convicted until later years, until we have taken into account the effect 
 of present age. This obviously affects both age at first conviction and 
 age at starting work, since one who is now of a given age must both 
 have been convicted before that age and have started work before then 
 if at all. Accordingly, the correlation coefficient may measure simply 
 the degree of dependence of each of fhese variables on the factor of 
 age. We have, therefore, determined the partial correlation coefficient 
 
•a 
 
 .o 
 
 
 to 
 •3 
 
 d 
 o 
 
 o 
 U 
 
 
 §* 
 
 g 
 
 a 
 
 S 
 
 o 
 
 rS 
 
 •a 
 
 
 
 
 o 
 
 
 H 
 
 £ 
 
 <0 
 
 a 
 
 4> 
 
 a 
 o 
 
 i: 
 
 o 
 
 U flS-H 
 
 99 o» 89 
 
 89 o* 09 
 
 09 o* iS 
 
 igojf9 
 
 fS0*T9 
 
 IS o» 8* 
 
 8^ °% Sf 
 
 Zf o» 68 
 
 68 o* 98 
 
 88 o* 08 
 
 08 °% IZ 
 
 fZ o» IZ 
 
 IS 0% 81 
 
 SI o» EI 
 
 ZX°% 6 
 
 »H 0>«05N^^ 
 
 I-* rH M t* ^ t^ 
 
 ^ ^ rH <N IN -d^ TtllN r-( 
 
 rtiHCO Ot-t-(M 
 
 f-< N t* ^ ^ O -^ IN 
 
 ?QU3;0COt- 
 
 W00'^O(D«C0^OC0NMt*iO 
 
 OO^QCDNQO-^OCOMCOTttOeO 
 
 310 
 
 II II 
 b b 
 
 II II 
 
 a a 
 
 V CD 
 
 as 
 
 II no 
 
 fl Bji 
 O OTl 
 
 IIS 
 
 •a -SO 
 oo 
 
 a flo 'I 
 
 §g| 
 
 ^ ea-w 
 «3| 
 
 |S| 
 
 ^|^l£ 
 M|-^ 
 
 2m 
 ^ O-P 
 
 hD'g * 
 
 a g 01 
 
 ■■SOM 
 
 5- o « 
 
 flff.S 
 a mo.Sj) " 
 
 latSio gf^ 
 
 ■2-s '-tf'-e © 
 
 II ^ s ^ 
 
 h ED 
 
 8.S 
 °ta 
 
 tag 
 
 o o 
 
OCCUPATIONAL HISTORY AND EFFICIENCY 311 
 
 for age at first conviction with age at starting work, making present 
 age the constant factor.** The correlation coefficient, .004± .046, would 
 indicate that the apparent relationship between age at first conviction 
 and age at starting work is entirely due to the relationship of each of 
 these to age. /^Accordingly, we find no confirmation for the view which 
 regards an early age of starting work as one of the important deter- 
 minants of early criminality. \ 
 
 KIND OF WORK DONE 
 
 (a) Classification of Kind of Work 
 
 Before discussing in detail the nature of work done at various times 
 during the working careers of these women, the method of classifying 
 kinds of work will be observed briefly. It is difficult to classify accu- 
 rately kinds of work done by women of varying ages for different 
 periods of time. To make this difficulty as slight as possible, therefore, 
 questions relating to kind of work done were asked very carefully with 
 a view to finding out what was meant by "factory work," "office girl," 
 "nurse," etc., both from the woman and the employer. The results 
 were codified by the main divisions of the Census Index to Occupa- 
 tions® and were later regrouped into eleven classes which seemed to fit 
 the needs of a selected group of women better than did the entire 
 census scheme which was made up from returns of occupations for all 
 wage-earners in the country, both male and female. The eleven classes 
 used were based on the conditions surrounding given kinds of work 
 as well as on degree of skill involved in the work, and include the fol- 
 lowing occupations : 
 
 1. Domestic Service includes women working at general housework, 
 nurse-girls, waitresses or cooks in private homes, lady's maids, and 
 housekeepers. This group includes all cases where living is in 
 addition to wage, and where the nature of the employment demands 
 that the women live at the place where they work. 
 
 2. Factory Work includes women working in a factory where there 
 is a group of people. Laundry workers in a laundry have been 
 put in this group also, since the numbers seemed too small for a 
 
 "The necessary data for computing the partial are the following: 
 r (age at first conviction with age at starting work) = .508 
 r (age at first conviction with age) = .964 
 
 r (age at starting work with age) = .526 
 
 " Bureau of the Census : "Index to Occupations," 1915. 
 
312 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 separate division, and the general conditions of work are similar 
 to those in factories. 
 
 3. Home Work includes workers who are able to earn their living by 
 remaining in their own homes and who work alone for the most 
 part. This is a somewhat heterogeneous group of dressmakers at 
 home, lodging-house-keepers, janitresses and those who bring work 
 home from a factory. This class is intended to show especially the 
 contrast in conditions of work from those of the previous group. 
 
 4. Restaurant and Hotel Work includes waitresses in restaurants and 
 
 hotels, chambermaids in hotels, laundry workers in hotels, a very 
 few restaurant keepers, and one manager of a small hotel. 
 
 5. Work in Stores includes clerks, salesladies, models, cash girls, mes- 
 
 sengers, errand girls, demonstrators, milliners, and five women who 
 themselves kept small stores. 
 
 6. Vaudeville Performers include show girls, acrobats, one- dancing 
 
 teacher who occasionally gave exhibition performances in theaters, 
 and one girl who sang in a cabaret in a saloon. 
 
 7. Clerical Work includes bookkeepers, two cashiers, stenographers and 
 
 typewriters, shipping clerks and one office girl for a doctor. 
 
 8. Professional Service includes two school teachers, two authors, and 
 
 one piano teacher. 
 
 9. Personal Service includes manicurists, hair-dressers, one masseuse, 
 
 and ten practical nurses and midwives. 
 
 10. Charwomen include ten women who do cleaning by the day in office 
 
 buildings, etc. 
 
 11. Miscellaneous includes seventeen occupations which fit into no gen- 
 
 eral scheme and each of which is too small in number to be given 
 a separate class. There are in this group two dog-fanciers, one for- 
 tune-teller, one peddler, one newspaper distributor, one helper in a 
 greenhouse, one commission agent, six unskilled farm workers, two 
 ushers in a theater and two telephone operators. 
 
 The discussion of kind of work done at various times in the work 
 histories of these women will be based on comparisons of these eleven 
 classes of work. We have selected as the most important specific jobs 
 in a woman's work history the kind of work (1) in the first job, (2) 
 in the latest job, (3) at the time of first conviction, (4) at the time of 
 latest conviction, and (5) prevailing work. The method of determin- 
 ing the prevailing work will be discussed later. The kind of work at 
 these given times will be considered separately by institutional groups, 
 summarized for the total group and compared with percentages for the 
 general female population when possible. 
 
OCCUPATIONAL HISTORY AND EFFICIENCY 313 
 
 (b) Kind of Work Done at Specified Times 
 
 (1) Kind of Work in First Job. — The kind of work in the first 
 employment is particularly important to consider in connection with 
 prevailing work and the estimate of efficiency of a work record. It is 
 difficult to draw any definite conclusions from the kind of work in the 
 first job, per se, because of the fact that this first work represents in 
 one case perhaps the beginning of a long and efficient work record 
 while in another case it may be only the casual job performed in later 
 life to tide over a financial embarrassment. As we have seen in Table 
 105, the age at starting work runs from the illegitimate employments 
 under fourteen years up to the Workhouse woman who started to 
 work at 58 years, after her husband's death. With these older women 
 who have had no training and who take whatever work they can most 
 easily get to tide them over any given financial stress, the first employ- 
 ment does not represent at it does with the younger women the begin- 
 ning of a work history, but merely an isolated job. Since the number 
 who started work late in life is comparatively small, however, we shall 
 present Table 107, keeping in mind that there is this constant small 
 element of women starting work late, and having no conscious se- 
 lection of kind of work since the work itself is so casual. 
 
 From Table 107 it is observed, first of all, that the large bulk of 
 cases in each institutional group, except Probation, falls in the domestic 
 service class at the time of their first job. This varies, however, be- 
 tween the institutional groups, the Workhouse having the largest per- 
 centage of domestic service workers and Probation the smallest per- 
 centage. The factory workers who have the second highest percentage 
 in all groups except Probation, where they have first place, do not 
 vary markedly between the institutional groups except in the Work- 
 house which has a very much smaller percentage of factory workers 
 than the other groups have. Home work is not of particular signifi- 
 cance, apparently, in any of the groups, and is chiefly conspicuous be- 
 cause Bedford has no representative in this group. The restaurant and 
 hotel workers vary to quite an extent and indicate that a larger per- 
 centage of women in the Workhouse than in any other group went first 
 into this work. Workers in stores furnish 16.3 per cent of the Pro- 
 bation group, and are represented to a somewhat smaller degree in the 
 other groups. Vaudeville performers have the highest percentages in 
 the Probation and Workhouse groups, with no representative in Au- 
 burn. Clerical work, on the other hand, while it has 'a percentage of 
 
314 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 107 
 
 KIND OF WORK IN FIRST JOB 
 
 Per Cent Distribution of Delinquent Women by Institutional Groups 
 
 
 Institdtional Groups 
 
 
 Kind of Work 
 
 Bed- 
 ford 
 
 Auburn 
 
 Mag- 
 dalen 
 
 Peni- 
 tentiary 
 
 Work- 
 houae 
 
 Proba^ 
 tion 
 
 Total 
 
 Dnnripstiio Service 
 
 44.6 
 29.7 
 
 46.3 
 
 25.0 
 
 6.3 
 
 2.5 
 
 7.5 
 
 'i!3 
 3.8 
 
 ■2!5 
 
 37.1 
 32.9 
 4.3 
 2,9 
 10.0 
 1.3 
 4.3 
 1.3 
 4.3 
 
 34.6 
 30.8 
 .9 
 4.7 
 7.5 
 1.9 
 5.6 
 2.8 
 2.8 
 1.9 
 2.8 
 
 47.5 
 
 12.1 
 
 5.1 
 
 10.1 
 
 11.1 
 
 6.1 
 2.0 
 
 ilo 
 iio 
 
 19.8 
 32.6 
 3.5 
 7.0 
 16.3 
 5.8 
 5.8 
 
 2.3 
 
 38.5 
 
 
 2fi9 
 
 
 3 1 
 
 Restaurant and Hotel Work . . . 
 
 5.0 
 7.9 
 2.0 
 4.0 
 1.0 
 1.0 
 
 5.5 
 99 
 
 VnnHpvillp Performers 
 
 2.8 
 
 Clerinal Work 
 
 3.7 
 
 Prnfessional Service 
 
 1.1 
 
 
 2.6 
 
 Oharwomen 
 
 A 
 
 
 2.0 
 
 2.0 
 
 
 
 Never Worked 
 
 3.0 
 
 5.0 
 
 
 3.7 
 
 5.1 
 
 3.5 
 
 3.5 
 
 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 
 
 Number of cases 
 
 101 
 
 80 
 
 70 
 
 107 
 
 99 
 
 86 
 
 643 
 
 5.8 in the Probation group, has only 2.0 per cent of the Workhouse 
 women and none in Auburn. The Penitentiary, Magdalen and Bedford 
 follow Probation in the percentage of clerical workers. Professional 
 service workers are found only in Bedford, Auburn, the Magdalen 
 and the Penitentiary, the first three being represented by one case and 
 the fourth by two cases. It is of interest to note that these two classes 
 of occupations which require most training, namely, clerical work and 
 professional service, have the smallest percentage who first went into 
 this work in Auburn, the next larger percentage in the Workhouse, the 
 next in Bedford, the Magdalen, Probation, and the largest percentage 
 in the Penitentiary. Personal service workers constitute only a small 
 percentage of each institutional group and the variations between these 
 are not large. In no group except the Penitentiary were there any 
 women who first went to work as charwomen. The miscellaneous 
 group contains cases from all institutions in not markedly varying 
 percentages. It is particularly worthy of note that within each group 
 the percentage who have never worked is very small. The Magdalen 
 has no women who have never worked, while the other five groups have 
 
OCCUPATIONAL HISTORY AND EFFICIENCY 315 
 
 percentages ranging from 3.0 to 5.1 per cent. In our total group of 
 cases, we may note that the largest percentage, 38.5, went first into 
 housework, and that the next largest percentage was engaged first in 
 factory work. The other occupations have very much smaller per- 
 centages. 
 
 (2) Kind of Work in Latest Joh.—li we compare Table 108 with 
 the foregoing table, we will note that in each of the institutional 
 groups, except Probation, the percentage of domestic service workers 
 is still highest, but that in all the groups except the Magdalen and Peni- 
 tentiary there seems to be a slight decrease in the percentage of domes- 
 tic service workers in the latest job. The factory workers have in- 
 creased, on the other hand, in the percentage found in Bedford, Mag- 
 dalen and the Workhouse, but have decreased rather noticeably in 
 Auburn and the Penitentiary. These two latter institutions, however, 
 have increased in the numbers foUnd in the group of home workers, and 
 this is probably due to the fact that a large percentage of these women 
 are married and have a tendency, where it is necessary for them to help 
 
 TABLE 108 
 
 KIND OF WORK IN LATEST JOB 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups 
 (Includes only those who have had more than one job.) 
 
 
 iNSTITnTIOXAL GbOUPS 
 
 
 Kind of Work 
 
 Bed- 
 ford 
 
 Auburn 
 
 Mag- 
 dalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Proba- 
 tion 
 
 Total 
 
 Domestic 
 
 35.5 
 36,6 
 
 43.8 
 9.6 
 13.7 
 12.3 
 2.7 
 1.4 
 2,7 
 1.4 
 4.1 
 1.4 
 1.4 
 
 38.7 
 40.3 
 1.6 
 9.7 
 3.2 
 4.8 
 1.6 
 
 39.4 
 16.0 
 5.3 
 8.5 
 9,6 
 .4.3 
 3.2 
 
 '3.'2 
 5.3 
 1.1 
 
 37.4 
 17.2 
 5.1 
 20.2 
 6.1 
 3.0 
 1.0 
 
 2.0 
 2.0 
 1.0 
 
 15.7 
 28.9 
 6.0 
 8.4 
 13.3 
 8.4 
 9.6 
 
 '4;8 
 
 ,i'2 
 
 34 9 
 
 Factory Work 
 
 24 2 
 
 Home Work 
 
 5 2 
 
 Restaurant and Hotel Work . . . 
 Work in Stores 
 
 6.5 
 6.5 
 4.3 
 1.1 
 1.1 
 1.1 
 2.2 
 2.2 
 
 11.1 
 7 1 
 
 Vaudeville Performers 
 
 4 4 
 
 Clerical Work 
 
 3 2 
 
 Professional Service 
 
 4 
 
 Personal Service 
 
 2 6 
 
 Charwomen 
 
 2 
 
 Miscellaneous 
 
 1 2 
 
 
 
 Never Worked 
 
 3.2 
 
 5.5 
 
 
 4.3 
 
 5.1 
 
 3.6 
 
 3 8 
 
 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 
 
 Number of cases 
 
 93 
 
 73 
 
 62 
 
 94 
 
 99 
 
 83 
 
 504 
 
 
 
 
 
316 STUDY OF WOMEN DELINQUENTS IN NEW YORK. 
 
 piece out the family income, to do work which will not take them away 
 from home. Restaurant and hotel workers, it will be noted, have in- 
 creased in each institutional group, while workers in stores have de- 
 creased in each group except the Penitentiary. Vaudeville performers 
 have increased in each group except the Workhouse. The clerical work- 
 ers have decreased in percentage in each of the groups except Auburn 
 and Probation, in each of which there is a slight increase. In the same 
 way, the professional service workers entirely drop out of all groups 
 except Bedford and Auburn, which have the same number of cases 
 doing professional service in the latest job as in the first job. Personal 
 service remains nearly unchanged and the number of charwomen in- 
 creases only sHghtly in each group. The miscellaneous workers remain 
 practically the same. In the total group it is of interest to note that 
 there is not a striking change from the first to the latest jobs in the 
 kind of work done. There is a slight decrease in the number of do- 
 mestic service and factory workers and very slight increases in several 
 of the other groups. The significant thing in the comparison seems to 
 be that there is so little change and that there seems to be a marked 
 tendency for these women to keep within somewhat the same kind of 
 work so far as can be judged by the first and latest jobs. 
 
 (3) Kind of Work at First Conviction. — If we observe the kind of 
 work done at time of the first conviction as gfiven in Table 109, we 
 may note first the very large percentages in each institutional group 
 who were idle at the time of first conviction. "Idle" does not include 
 those who were at home doing the work either in their own or their 
 parents' homes. Such cases are included under "Own housework." 
 Those who have never worked might, it is true, be added either to the 
 "Idle" or "Own housework" group in this table, but have been kept 
 separate so that they may show as a constant group in the later tables 
 dealing with kind of work at specified times. Because such a large 
 percentage were idle at this time, the numbers in many of the occupa- 
 tions are smaller than in the two previous tables. Domestic service 
 still keeps first place in all except the Probation group. The factory 
 workers have decreased in numbers perceptibly and are much smaller 
 in proportion to the domestic service workers than in either the first 
 or latest jobs, previously considered. Home work remains the same 
 for the total group as in the first job, but in the institutional groups 
 Auburn alone stands out as having a larger percentage of home workers 
 than was found in a consideration of the first job. The restaurant and 
 hotel workers, workers in stores, and vaudeville performers have de- 
 
OCCUPATIONAL HISTORY AND EFFICIENCY 317 
 
 creased markedly from the first job. Only two women were doing 
 clerical work at the time of first conviction, while there were twenty 
 who had started their first work in a clerical job. The professional 
 service workers, also, have a very small percentage, only two women 
 at the time of first conviction. The number of women in each of the 
 other three groups are also smaller at time of the first conviction than 
 in the first job. 
 
 TABLE 109 
 
 KIND OF WORK AT TIME OF FIRST CONVICTION 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups 
 
 
 Institutional Gkoupb 
 
 
 Kind of Work 
 
 Bed- 
 ford 
 
 Auburn 
 
 Mag- 
 dalen 
 
 Peni- 
 
 tentiaty 
 
 Work- 
 house 
 
 Proba- 
 tion 
 
 Total 
 
 Domestic Service 
 
 18.8 
 12.9 
 
 20.8 
 1.4 
 
 13.9 
 1.4 
 1.4 
 
 lA 
 4.2 
 
 8.8 
 5.9 
 
 'i!5 
 2.9 
 1.5 
 
 10.0 
 2.2 
 2.2 
 1.1 
 1.1 
 
 ill 
 1.1 
 1.1 
 2.2 
 1.1 
 
 4.8 
 1.6 
 3.2 
 3.2 
 
 1.6 
 
 i'.h 
 
 2.3 
 5.7 
 1.1 
 
 ili 
 
 1.1 
 
 11 2 
 
 Factory Work 
 
 f; 4. 
 
 Home Work 
 
 3 1 
 
 Restaurant and Hotel Work . . . 
 Work in Stores 
 
 3.0 
 2.0 
 1.0 
 
 1.7 
 1 3 
 
 Vaudeville Performers 
 
 g 
 
 Clerical Work 
 
 4 
 
 Professional Service 
 
 Personal Service 
 
 
 'a 
 
 1 
 
 Charwomen 
 
 
 4 
 
 Miscellaneous 
 
 
 4 
 
 
 
 
 Own Housework 
 
 5.0 
 
 54.4 
 
 3.0 
 
 25.0 
 
 25.0 
 
 5.6 
 
 10.3 
 69.1 
 
 8.9 
 
 63.7 
 
 4.4 
 
 3.2 
 
 72.6 
 
 8.1 
 
 9.1 
 
 75.0 
 
 3.4 
 
 10 
 
 Idle 
 
 59 9 
 
 Never Worked 
 
 4.0 
 
 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 
 
 Number of cases 
 
 101 
 
 72 
 
 68 
 
 90 
 
 62 
 
 88 
 
 481 
 
 (4) Kind of Work at Present Conviction. — The following table 
 (Table 110) is of more interest if taken in connection with Table 109. 
 Again, the percentage of idle persons is the most striking thing in the 
 table ; though when compared with the percentage of idle women at 
 time of first conviction it is seen to be somewhat smaller. From these 
 two tables it would seem obvious that the factor of unemployment has 
 an important relation to the problem of deHnquency. It is difficult 
 even to estimate how much of this unemployment was unnecessary 
 since there is no way of satisfactorily verifying attempts to obtain 
 
318 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 work. It seems probable, however, from the statements of the women, 
 and from a survey of industrial conditions in New York City during 
 the time that the present convictions occurred, that there would have 
 been no particular difficulty for many of the women who were idle to 
 have obtained work. 
 
 TABLE 110 
 
 KIND OF WORK AT TIME OF PRESENT CONVICTION 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups 
 
 
 iNSTITDTrONAL GkOTJPS 
 
 
 Kind of Work 
 
 Bed- 
 ford 
 
 Auburn 
 
 Mag- 
 dalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Proba- 
 tion 
 
 Total 
 
 Domestic Service 
 
 14.9 
 10.9 
 
 19,5 
 1.3 
 
 11.7 
 2.6 
 
 1.3 
 
 'i]3 
 
 3.9 
 
 10.7 
 6.7 
 
 'i!3 
 1.3 
 
 17.6 
 1.0 
 1.0 
 1.9 
 1.0 
 
 'i!6 
 
 1.0 
 1.9 
 1.9 
 2.9 
 
 11.2 
 4.1 
 4.1 
 4.1 
 1.0 
 2.0 
 
 'i!6 
 'i;6 
 
 4.5 
 3.4 
 1.1 
 
 1.1 
 
 13 1 
 
 Factory Work 
 
 4 6 
 
 
 2 8 
 
 Restaurant and Hotel Work . . . 
 Work in Stores 
 
 2.0 
 
 2.0 
 
 .7 
 
 
 2.0 
 
 g 
 
 Clerical Work 
 
 4 
 
 Professional Service 
 
 1.0 
 
 4 
 
 Personal Service 
 
 1 3 
 
 Charwomen 
 
 
 4 
 
 Miscellaneous 
 
 
 7 
 
 
 
 
 Own housework 
 
 13.9 
 
 53.5 
 
 3.0 
 
 23.4 
 
 29.9 
 
 5.2 
 
 16.0 
 64.0 
 
 13.6 
 
 51.5 
 
 3.9 
 
 10.2 
 
 56.1 
 
 6.1 
 
 14.6 
 
 69.7 
 
 3.4 
 
 14 9 
 
 Idle 
 
 54 3 
 
 Never Worked 
 
 3.5 
 
 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100 
 
 
 
 Number of cases 
 
 101 
 
 77 
 
 75 
 
 103 
 
 98 
 
 89 
 
 543 
 
 
 
 (5) Prevadling Work. — The prevailing work of these women repre- 
 sents the occupations in which they have been engaged for the longest 
 time. We have divided the work histories of the women in our group 
 into three classes, (1) those who have done only one kind of work, 
 (2) those who have done one kind of work almost entirely with per- 
 haps one or two jobs in other occupations, and (3) those who have 
 had variable records, now in one kind of work, now in another, but 
 with one kind of work which might be called prevailing because the 
 subject had worked longer in that occupation than in any other. By 
 combining the first two classes we cover 71.3 per cent of our cases. 
 The third group was also used in considering prevailing work so that 
 
OCCUPATIONAL HISTORY AND EFFICIENCY 319 
 
 the total number of cases could be used. The prevailing work of this 
 third group always represented thp longest number of months of work/ 
 Table 111 shows that in each group, except Probation, the domestic 
 service workers have the highest percentage. The other workers run in 
 very much the same proportions as for the work in the first job. The 
 comparison of these two tables indicates that, for the most part, there 
 is little change in kind of work in the first job and kind of work done 
 later. There is a slight tendency toward a smaller percentage of do- 
 
 'The few following examples from the Bedford group may serve to indi- 
 cate how the work records vary in the kind of work done. 
 
 Class 1. — Those working in only one occupation. 
 
 a) Woman now 29 years, 10 months. Began work at IS. Singing and 
 
 dancing in vaudeville performances from 1901-1912. Was forced to 
 leave because she was using drugs. Has done no other work. 
 
 b) Woman now 26 years, 10 months. Began work at IS. She had 13 ■ 
 
 jobs, all housework, covering a work period of 6 years. 
 
 Qass 2. — ^Almost entirely one kind of work, with a few exceptions. 
 
 a) Girl 16 years, 9 months. Began work at IS. 
 
 1) Housework, 4 mos. 
 
 2) Housework, 4 wks. 
 
 3) Housework, 8 mos. 
 
 4) Helper in hotel dining room, 7 wks. 
 
 5) Housework, 2 wks. 
 
 6) Cigar-making, 3 wks. 
 
 7) Housework, 1 wk. 
 
 b) Colored woman 23 years, 8 months. Began work at 8 years. 
 
 1) Nursegirl, 1 yr. 
 
 2) Housework, V/z yrs. 
 
 3) Housework, 3 yrs. 
 
 4) Housework, 2 yrs. 
 
 5) Waitress, 3 mos. 
 
 6) Housework, 1 yr. 
 
 7) Housework, 1 yr. 
 
 8) Housework, 1 yr. 
 
 9) Waitress, 1 yr. 
 
 Class 3. — Those with variable records. 
 
 a) Woman 23 years, 2 months. Began work at 17 years. Prevailing 
 work — saleslady in a store. 
 
 1) Saleslady in store, 1 yr. 
 
 2) Home work from embroidery factory, 6 mos. 
 
 3) Served milk in milk station, 4 mos. 
 
 4) Housework, 2 mos. 
 
 5) Saleslady in store, 3 mos. 
 
 b) Girl, 18 years, S mos. Began work at IS. Prevailing work— sorting 
 paper in paper mill. 
 
 1) Housework — nursegirl, 2 wks. 
 
 2) Sorting paper in paper mill, 1 yr. 
 
 3) Banding cigars in cigar factory, 3 mos. 
 
 4) Feeder on forming machine in hat factory, 3 mos. 
 
 5) Banding cigars in cigar factory, 1 wk. 
 
320 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 111 
 
 PREVAILING KIND OF WORK 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups 
 
 
 Institutional Ghoups 
 
 
 Kind of Work 
 
 Bed- 
 ford 
 
 Auburn 
 
 Mag- 
 dalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Proba- 
 tion 
 
 Total 
 
 
 47.5 
 30.7 
 
 43.4 
 14.5 
 13.2 
 5.3 
 4.0 
 1.3 
 2.6 
 1.3 
 5.3 
 1.3 
 2.6 
 
 38.0 
 32.5 
 2.8 
 5.6 
 11.3 
 2.8 
 4.2 
 
 '2!8 
 
 41.6 
 
 24.8 
 2.0 
 5.9 
 5.9 
 4.0 
 3.0 
 2.0 
 3.0 
 2.0 
 2.0 
 
 48.2 
 8.4 
 7.2 
 8.4 
 
 10.8 
 4.8 
 1.2 
 
 '2!4 
 
 '2;4 
 
 22.4 
 30.6 
 3.5 
 8.2 
 11.8 
 7.1 
 7.1 
 
 'i'.7 
 
 1.2 
 
 40.4 
 
 Factory Work 
 
 23.8 
 
 Home Work . 
 
 4.5 
 
 Restaurant and Hotel Work . . . 
 
 3.0 
 8.9 
 3.0 
 1.0 
 1.0 
 1.0 
 
 6.0 
 
 8.7 
 
 
 3.9 
 
 Clerical Work 
 
 3.1 
 
 Professional Service 
 
 g 
 
 Personal Service 
 
 3.1 
 
 Charwomen ... 
 
 .6 
 
 
 1.0 
 
 1.6 
 
 
 
 Never Worked 
 
 3.0 
 
 5.3 
 
 
 4.0 
 
 6.0 
 
 3.5 
 
 3.7 
 
 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 
 
 Number of cases 
 
 101 
 
 76 
 
 71 
 
 101 
 
 83 
 
 85 
 
 517 
 
 
 
 mestic service workers in the first job than in the prevailing work, and 
 there is a corresponding increase for factory workers. 
 
 (c) Comparison of Prevcdling Work of Delinquents and General 
 
 Population 
 
 A comparison at this point of the prevailing work of the delinquent 
 women in this study with the kind of work done by females over IS 
 years of age in New York State as compiled in the 1910 census' shows 
 that there are various differences in kind of occupations between the 
 two groups (see Chart XVIII). The delinquent women have 42.0 per 
 cent of their number domestic service workers, which is 18.3 per cent 
 more than we find in the general population. There is also a larger 
 percentage of factory workers, restaurant and hotel workers and vaude- 
 ville performers among the delinquents than among the general popu- 
 lation. The vaudeville performers furnish to the whole group of de- 
 linquents eight times their proportional representation in the general 
 population. It seems probable that with this group of women the 
 
 ' Op. cit., p. 574. 
 
OCCUPATIONAL HISTORY AND EFFICIENCY 321 
 
 nature of the work is to a large extent an important factor in the de- 
 linquency. A traveling vaudeville company where the girl is thrown 
 into contact with many people and many kinds of men, where there 
 is little privacy and innumerable opportunities to earn money "on the 
 side," affords ample chance for sexual irregularities. The occupations 
 in which the delinquents have less than their expected percentages are 
 among those doing home work, work in stores, and more especially 
 among clerical workers and those in professional service. The clerical 
 workers in this study furnish less than one-quarter of their quota in 
 the general population, while the professional service workers furnish 
 
 PER CENT 
 40 60 
 
 DOMESTIC SERVICE 
 
 FACTORY WORK 
 
 HOME WORK 
 
 RESTAURANT & HOTEL WORK 
 
 WORK IN STORES 
 
 VAUDEVILLE PERFORMERS 
 
 CLERICAL WORK 
 
 PROFESSIONAL SERVICE 
 
 PERSONAL SERVICE 
 
 CHARWOMEN 
 
 MISCELLANEOUS 
 
 b 
 
 DELINQUENT WOMEN 
 
 ] FEMALES OVER 15 IN N. Y. STATE 
 
 XVIII 
 
 Chart XVIII 
 
 Percentage Comparison Showing Prevailing Kind of Work of 517 Women 
 Delinquents in New York State, and Kind of Work Done by Females over 
 IS Years of Age in New York State, as Compiled m the 1910 Census. 
 
322 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 less than one-ninth of what we might expect. The last three groups 
 of personal service workers, charwomen and miscellaneous also have 
 less than their expected representation. The largest differences, there- 
 fore, are as follows : with the exception of vaudeville performers who 
 work under especially bad conditions, the delinquents have a larger 
 representation than we should expect among the occupations which are 
 comparatively unskilled, principally domestic service and, to a lesser 
 degree, factory work, while they have a smaller percentage than the 
 general population in occupations which require special training, such 
 as clerical work and professional service.® 
 
 At this point we might observe the comparisons made between the 
 occupations of a group of 3,229 women offenders in six states and the 
 total female population sixteen years of age and over in the United 
 States. In this study made by Miss Mary Conyngton in 1911 on the 
 "Relations between Occupation and Criminality of Women" ^° she 
 found in comparing the delinquents with the total female wage-earners 
 sixteen years of age and over that "by far the largest proportion of 
 offenders comes from the group engaged in domestic and personal serv- 
 ice, and that the only subdivision under this general head furnishing 
 more than its proportionate share of misdemeanants or criminals is 
 that of servants and waiters. . . . These workers, constituting 24.1 
 per cent of the breadwinners, account for 70.3 per cent of the offenders, 
 or very nearly three times their proportionate share. . . . With prac- 
 tically the same number in each group, the servants and waitresses 
 furnish more than four times as many offenders as those engaged in 
 manufacturing and mechanical pursuits." In referring to a group 
 called "trade and transportation" in the Census classification, and in- 
 cluding in Miss Conyngton's study cash girls, bookkeepers, clerks, 
 saleswomen, stenographers, and telephone operators, she finds that this 
 group of workers "furnish to the whole group of offenders studied 
 only one-third of their proportionate representation. And this strik- 
 ing fact becomes even more striking when it is remembered that the 
 investigation was carried on chiefly in industrial centers, where workers 
 
 'We recognize that the two groups of facts are not on exactly the same 
 basis for comparison. That is, while the Census figures apply to women work- 
 ing in given occupations at a certain time, the prevailing work used here may 
 apply to many work records of years past and, at best, it covers a period of 
 time for each statement of prevailing work. 
 
 ""Report on Condition of Woman and Child Wage-earners in the United 
 States," Vol. XV. "Relation between Occupation and Criminality of Women." 
 By Mary Conyngton. Government Printing Office, Washington 1911. p. 30. 
 
OCCUPATIONAL HISTORY AND EFFICIENCY 323 
 
 of this kind are most numerous and where their temptations are 
 greatest." 
 
 It is not easy to say whether this unduly large percentage of domes- 
 tic service workers among the delinquent group is, in any sense, a re- 
 sult of the occupational influences, or whether those who go into do- 
 mestic service represent the great mass of unskilled, untrained women, 
 from which the most of our delinquents are drawn. It is true that the 
 domestic service workers in our group have not gone so far in school 
 as women in other occupations. This is indicated in Tables 131 and 
 132, which show that for the domestic service workers the average 
 grade finished is 4.16+2.48, while for those in occupations other than 
 domestic service the average grade is 5.03±2.56. A later chapter^^ 
 shows that the domestic service workers are also of lower average 
 mentality than the group of women in occupations other than domes- 
 tic service. From our observation, we are inclined to think that the 
 inherent conditions of work surrounding many of the young, inexperi- 
 enced girls who go into housework may be responsible, to a small ex- 
 tent, for their delinquency, but that the occupational influences for 
 the most part are secondary rather than direct causes of delinquency. 
 It is true that in housework there is not the careful supervision which 
 many factories, offices, shops and stores give, that there is opportunity 
 for petty thieving with little danger of being caught, and in many in- 
 stances the opportunity for meeting men, such as tradesmen, in an inti- 
 mate way which may lead to sexual irregularities. Making allowances 
 for these difficulties in domestic service work which, without doubt, 
 affect a large number of the delinquent group, we believe that the more 
 important phase of the question to be stressed is that a large propor- 
 tion of those in our group going into domestic service are the untrained 
 women of somewhat lower mentality, often impressionable and easily 
 influenced.^^ 
 
 " See Chapter XV, p. 503. 
 
 "A case of interest at this point is that of a young Hungarian girl com- 
 mitted to Bedford for petit larceny. She was stupid, unattractive physically, 
 and in addition had very poor eyesight which her brother told her would ulti- 
 mately make her blind. She became infatuated with a negro man who was at- 
 tentive to her, and who, it is thought, finally persuaded her to steal a diamond 
 ring from her employer for whom she had been doing housework about three 
 months. Her excuse was that the employer "had lots,— rings and jewels and 
 things, and I had nothing." She wanted to make herself look beautiful, she 
 wanted to please the negro, the only man who had ever paid any attention to 
 her, and she was too stupid to realize that the theft as she committed it would 
 surely lead to her being caught. 
 
324 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 Dr. Davis in a study of prostitutes at Bedford writes : 
 
 "Almost all the studies of prostitution heretofore made have noted the 
 high percentage of women who were engaged in domestic service previous 
 to entering the life. So far as my observation goes, I do not believe that 
 this indicates any greater danger from domestic service itself as an occupa- 
 tion than from any other in which unskilled girls engage. Domestic service 
 for women under existing economic conditions corresponds to casual labor 
 for men. It is the job where training and experience are unnecessary in 
 order to find work. Such services would not be desired by families where 
 efficiency is demanded and paid for. A very large proportion of our girls 
 were not competent workers but were girls employed in the lowest stratum 
 of families that employ domestic help at all and where standards of service 
 do not exist. This group includes almost all the colored girls and a con- 
 siderable number of the foreign-born white girls." " 
 
 While we are indicating the prevailing work among the institu- 
 tional groups, it may also be advisable to show the prevailing work 
 among nativity and color groups. Table 112 gives us this information 
 and shows that the native colored have over 80.2 per cent of their 
 number engaged in domestic service, while only 26.7 per cent of the 
 native white were so engaged. The total foreign stand between 
 these two groups with a percentage slightly higher than the per- 
 centage of domestic service workers for the total group. The per- 
 centage of factory workers, again, is less in the foreign bom and native 
 colored groups than among the native white, though very much less 
 in the native colored. It seems very likely that the percentages in the 
 Magdalen group in Table 111 have been affected by the fact that there 
 are no colored women in the Magdalen and that there is, therefore, 
 a tendency for the percentage of domestic service workers to be less 
 and for the factory workers to have a slightly larger number than the 
 same occupations in the other institutional groups. The Probation 
 group which has a small percentage of colored is also undoubtedly 
 affected in the same way. It will be noted that the colored have no 
 representative in the professional service class and one only among 
 the clerical workers. The foreign born also have only one professional 
 woman and three clerical workers, while the native white have one 
 professional woman and four clerical workers. 
 
 " Op. cit, p. 175. 
 
OCCUPATIONAL HISTORY AND EFFICIENCY 325 
 
 TABLE 112 
 
 PREVAILING KIND OF WORK 
 
 Per Cent Distribution of Delinquent Women Classified by Nativity 
 
 and Color 
 
 
 Nativity and Coixjb 
 
 
 Kind of Work 
 
 Total 
 Foreign Born 
 
 Total 
 Native Bom 
 
 Native 
 White 
 
 Native 
 Colored 
 
 Total 
 
 Domestic Service 
 
 Factory Work 
 
 Home Work 
 
 45.0 
 
 21.9 
 
 6.6 
 
 8,6 
 6.6 
 1.3 
 2.0 
 
 .7 
 4.6 
 
 .7 
 2.0 
 
 40.8 
 25.7 
 
 3.8 
 
 5.2 
 
 10.1 
 
 5.2 
 
 3.8 
 
 .9 
 2.6 
 
 .6 
 1.5 
 
 26.7 
 32.9 
 
 4.7 
 
 5.9 
 12.9 
 5.9 
 4.7 
 1.2 
 3.1 
 .4 
 1.6 
 
 80.2 
 5.5 
 1.1 
 
 3.3 
 
 2.2 
 3.3 
 1.1 
 
 1.1 
 1.1 
 
 42.1 
 
 24.6 
 
 4 6 
 
 Restaurant and Hotel 
 
 Work 
 
 Work in Stores 
 
 Vaudeville Performers. 
 
 Clerical Work 
 
 Professional Service. . . 
 
 Personal Service 
 
 Charwomen 
 
 Miscellaneous 
 
 6.2 
 9.1 
 4.0 
 3.2 
 
 .8 
 3.2 
 
 .6 
 1.6 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cases 
 
 151 
 
 346 
 
 255 
 
 91 
 
 497 
 
 The trend of change of work at various times in the work histories 
 may be shown by combining the preceding tables. Table 113 shows 
 the predominance of domestic service workers at all times, and the 
 small percentage throughout of workers in the more skilled occupa- 
 tions of clerical work and professional service. An increase in domes- 
 tic service workers is shown between the first job and the prevailing 
 work, while there is a decrease of factory workers. The markedly 
 high percentage of idle women at the time of conviction, both first 
 and present, is of greatest significance and would bear a much closer 
 study of the causes of unemployment. 
 
 It is interesting to observe that a similar tendency was noted in a 
 study of male felons made by Dr. Glueck in Sing Sing Prison.^* In 
 a discussion of the defective group, he notes that 57.1 per cent were 
 unemployed at time of present conviction, 39.6 per cent of the psycho- 
 pathic group were unemployed and 50 per cent of the Russian group 
 were "unemployed at the time of the crime although 56.8 per cent were 
 skilled mechanics." We do not find, however, with our group of de- 
 
 "Op. cit., pp. 112, 123, 131 and 138. 
 
326 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 linquent women as he did with male felons that "there seems to be a 
 certain correlation between the degree of recidivism in a given group 
 and the extent of unemployment at the time of the commission of the 
 crime." From Tables 109 and 110, it is seen that the Probation group 
 which has the smallest average number of convictions has in each case 
 the highest percentage of idle women. Aside from this point, how- 
 ever, the fact of there being such a large percentage out of work at 
 the time of conviction, among male criminals as well as among our 
 delinquent groups, is of great importance. 
 
 Again, in Miss Conyngton's study," she found that between the 
 earliest occupation and the occupation at latest arrest : 
 
 "The most significant changes are those which have taken place not be- 
 tween the great industrial groups, but from them into the non-gainful pur- 
 suits. ... It will be noticed that each of the groups of gainful occupations 
 diminishes, and that the increase is wholly among those who are either 
 married and keeping house, or who have no lawful pursuit. It is rather 
 curious to find these two groups at the time of the latest arrest so exactly 
 the same in number, but the rate of increase has of course been enormously 
 greater for those having no occupations than for those keeping house." 
 
 In our group of delinquent women we found 3.5 per cent who had 
 never had a "lawful occupation" when we were considering the kind 
 of work in first job (Table 113), while at time of the present convic- 
 tion, corresponding to "latest arrest," there were 54.3 per cent of the 
 women idle and 3.5 per cent who had never worked. This is a very 
 much more striking increase in those without employment than Miss 
 Conyngton finds in the percentages of 0.6 per cent who had no lawful 
 employment when the earliest employment was being considered, and 
 17.5 per cent who had no lawful pursuit at the time of the latest arrest, 
 but the trend is in the same direction in both groups. Also, the in- 
 vestigation mentioned was made from institutional records which are 
 largely 'based on the subject's statement. The percentage of idle, we 
 are convinced from our own investigations, is much larger than the 
 women themselves are willing to admit. 
 
 (d) Relationship Between Kind of Work and Grade Finished 
 
 It may be of interest, since so large a percentage of the delinquent 
 women are engaged in housework, to find the relationship between the 
 kind of prevailing work and the grade finished in school in order to 
 
 " Op. cit., pp. 46-47. 
 
OCCUPATIONAL HISTORY AND EFFICIENCY 327 
 
 TABLE 113 
 
 KIND OF WORK DONE BY DELINQUENT WOMEN AT SPECI- 
 FIED TIMES DURING THEIR WORK HISTORIES 
 
 Per Cent Distribution for Total Group 
 
 
 Specified Times 
 
 Kind of Work 
 
 In First 
 Job 
 
 In Latest 
 Job* 
 
 At Time of 
 
 First 
 Conviction 
 
 At Time of 
 
 Present 
 Conviction 
 
 Prevailing 
 Work 
 
 Domestic Service 
 
 Factory Work 
 
 38.5 
 
 26.9 
 
 3.1 
 
 6.5 
 9.9 
 2.8 
 3.7 
 1.1 
 2.6 
 .4 
 2.0 
 
 34.9 
 
 24.2 
 
 5.2 
 
 11.1 
 7.1 
 4.4 
 3.2 
 .4 
 2.6 
 2.0 
 1.2 
 
 11.2 
 5.4 
 3.1 
 
 1.7 
 1.3 
 .8 
 .4 
 .4 
 1.0 
 .4 
 .4 
 
 13.1 
 4.6 
 2.8 
 
 2.0 
 .7 
 .9 
 .4 
 .4 
 
 1.3 
 .4 
 .7 
 
 40.4 
 
 23.8 
 
 4.5 
 
 6.0 
 8.7 
 3.9 
 3 1 
 
 Home Work 
 
 Restaurant and Hotel 
 Work 
 
 Work in Stores 
 
 Vaudeville Performers . . 
 Clerical Work 
 
 Professional Service. . . . 
 
 Personal Service 
 
 Charwomen 
 
 .8 
 
 3.1 
 
 6 
 
 Miscellaneous 
 
 1 6 
 
 
 
 Own Housework 
 
 Idle 
 
 "sis' 
 
 "s'.s 
 
 10.0 
 
 59.9 
 
 4.0 
 
 14.9 
 
 54.3 
 
 3.5 
 
 
 Never Worked 
 
 3.7 
 
 Total . . 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100 
 
 
 
 Number of cases 
 
 543 
 
 504 
 
 481 
 
 543 
 
 517 
 
 * Includes only those wbo have had more than one job. 
 
 determine whether women in certain occupations tend to have had 
 more or less education, and whether those in domestic service, par- 
 ticularly, are characterized as is often stated by reaching a much lower 
 grade than those in other occupations. The correlation ratio, showing 
 the differences in grade finished corresponding to differences in prevail- 
 ing work (Table 114), is .38±.042. This shows a fairly high relation- 
 ship, indicating that the occupational groups differ in respect to school 
 attainment. From the means for the grades we can roughly see what 
 average attainment is characteristic of each occupation. Charwomen, 
 miscellaneous and professional workers are too few in number to be 
 of any value since so few cases might easily have occurred by chance. 
 Bearing in mind the small numbers, we find that the order of grade 
 reached by women in stated occupations, beginning with the lowest 
 grade and running upwards, is as follows : 
 
328 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 1. Charwomen 7. Personal service 
 
 2. Miscellaneous 8. Vaudeville performers 
 
 3. Domestic service 9. Work in stores 
 
 4. Home work 10. Clerical work 
 
 5. Factory work 11. Professional service 
 
 6. Restaurant and hotel work 
 
 From this it would seem, throwing out the first two, that the do- 
 mestic service workers have finished the lowest mean grade, 4.1, the 
 home and factory workers the next higher grade, and the clerical 
 workers the highest mean grade of significance, 6.5. The bulk of our 
 women, therefore, the domestic service and factory workers, fall in 
 the two lower grade groups, indicating that the women who go into 
 these occupations have not had the educational advantages that some 
 of the other occupational groups have, such as workers in stores, 
 clerical workers, etc. We are not justified, however, in assuming that 
 the fact of more or less educational equipment is a determinant of 
 the occupation chosen. The crude relationship, measured by the ratio 
 of .38, may very probably be accounted for by a basic relationship be- 
 tween intelligence and kind of work. We shall see (p. 501) that the 
 ratio of nature of prevailing work on intelligence (Test Aggregate) is 
 ,46zt.043, and that the correlation coefficient of intelligence (Test 
 Aggregate) with grade finished is .7S±.034 (see p. 497). The statistical 
 data give us no grounds for stating which is the primary factor, in- 
 telligence or education, but the logical presumption would be in favor of 
 the former, as a constitutional factor which might well have affected 
 both the degree of education and the choice of a job. 
 
 WAGE 
 
 The study of wage is in many ways the most difficult of all occu- 
 pational data to work vvith, because of the difficulty in getting accurate 
 wage information either from the subject or by verification, and in 
 estimating what the living would be in all jobs such as domestic serv- 
 ice, practical nursing, etc., where living is in addition to wage. Because 
 the time represented by the first job varied in individual cases by many 
 years, it seemed impossible to fix any arbitrary estimate of living which 
 would be fair to all. A woman who did housework ten years ago for 
 a poor family in the country for $3 a week would be receiving much 
 less in monetary value than she would be earning today with the same 
 wage, if she were working for a wealthy family in the city, since the 
 estimate of cost of living would vary so between the two times and 
 
s 
 § 
 
 
 
 sll 
 
 rHIMrHO'l<C<|ini>MC0ei5 
 
 s 
 
 
 
 1^1 
 
 Ttl-*-*iOcOilO«O00>O.HIN 
 
 •* 
 
 
 » 
 
 CCO5C<lTH00i>iOcOlCC0I> 
 
 CO 
 
 
 CDON<MC0i-lr-l ,-1 
 
 iH 
 
 
 
 i-H r-H 
 
 •* 
 
 
 Eh 
 
 
 
 
 
 
 T-< »H tH iH tH (N 
 
 t> 
 
 
 T3 iTCQ 
 
 
 
 
 
 (nI>^b* 
 
 iH »H f-H 
 
 CO 
 
 
 -*^ i!CQ 
 
 
 
 
 
 -^w 
 
 CO i-i i-( 
 
 00 
 
 
 n 
 
 OOIINIOOCOIO <N 
 
 CO 
 
 
 
 
 .-i i-t 
 
 1* 
 
 
 i> 
 
 (N CD CO <N CO >n (M 
 
 CD 
 
 
 
 
 i-l 
 
 CO 
 
 
 a 
 
 
 
 
 
 
 
 
 
 n 
 
 
 
 
 
 S 
 
 CD 
 
 ,-ioeocoi>iMiM'-i>o 
 
 "4 
 
 
 
 1-1 cq 
 
 lO 
 
 
 
 
 
 
 o. 
 
 ra 
 
 
 
 
 
 ° 
 
 « 
 
 CO O i-H CO lO (N CO 
 
 o 
 
 1^ 
 
 
 (N(N 
 
 CO 
 
 O 
 
 o 
 
 
 
 
 
 
 
 •f? 
 
 
 ■^ 
 
 rH CO CO CO lO IN 1-1 T-H rH 
 
 o 
 
 o 
 
 
 
 CO I-H 
 
 
 H 
 
 
 
 
 
 
 CO 
 
 lO lO T-H I-H tH (M iH l-t t-l 
 
 00 
 
 
 
 
 (N.-I 
 
 
 
 N 
 
 OOOIINt-i (N .-( 
 
 CO 
 
 
 
 
 »-l 
 
 
 
 - 
 
 CO-* 1-1 i-< 
 
 iH 
 
 
 
 & 
 
 t-- cocoes 1-1 (NIMCO 
 
 ^ 
 
 
 
 
 
 'J4 '■ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 '. o . 
 
 
 
 
 
 
 
 
 
 
 
 
 ■^ ■ 
 
 
 
 
 
 
 
 
 
 
 
 
 otei 
 mers 
 
 ?■ 
 
 
 
 
 
 
 
 
 S 
 
 ■a :g 
 
 53.53 
 
 
 
 
 
 
 
 ■•tJ cot! 
 
 
 
 
 
 
 Domestic Ser\ 
 Factory Work 
 Home Work. . 
 Restaiu:ant an 
 Work in Store 
 Vaudeville Pe 
 Clerical Work 
 Professional S 
 Personal Serv 
 
 o 
 
 ,1 
 
 tr 
 
 
 
 1 
 
 
 
 i 
 
 
 f[ioM. sutn 
 
 BA£ 
 
 "d 
 
 
 
 
 
 
 (N 
 
 
 Tti 
 
 ii 
 
 O 
 
 
 •« 
 
 
 
 
 t^ 
 
 <a 
 
 
 
 
 
 II 
 
 'm 
 
 ST 
 
 (d 
 
 M 
 
 u 
 
 
 
 
 
 ^ 
 
 iJ 
 
 
 
 
 
 ■M 
 
 
 > 
 
 'S 
 
 a 
 
 -S 
 
 a 
 
 ta 
 
 CD g 
 
 1 
 
 (N.g 
 
 S 
 
 II 3 
 
 
 h fl 
 
 
 o 
 
 ^ 
 
 
 
 CD'S 
 
 1 
 
 lo.S 
 
 
 ■*!« 
 
 s 
 
 11^ 
 
 
 .a 
 s 
 
 go 
 
 ^, •• 
 
 ^ 
 
 
 h 
 
 
 
 T)-S 
 
 t* 
 
 
 
 ii 
 
 1 
 
 «^ 
 
 f 
 
 -§£ 
 
 
 
 
 
 ou 
 
 O 
 
 329 
 
330 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 places. We have, therefore, ' throughout, kept separate the wages of 
 those engaged in domestic service, including, in addition to those doing 
 general housework, nursegirls, waitresses or cooks in private homes, 
 lady's maids and housekeepers. There are also included here two prac- 
 tical nurses who, in addition to weekly wage, received living. All of 
 the wages given for this group, therefore, are only the money received 
 and do not include the allowance for maintenance which must be added 
 in order to make these wages comparable with the non-housework 
 group. We have presented percentage tables on the wage received by 
 domestic service workers and those in occupations other than domestic 
 service, by institutional groups, for the first and latest jobs, and at first 
 and present conviction. Although the division of each group into two 
 parts makes the number of cases very small, we have presented the 
 frequency tables on wage in order to show the composition of the 
 total groups, but have made few comments on the individual tables. 
 
 (a) Wage Earned at Specified Tvmes 
 
 (1) Wage in First Job. If we turn to the weekly wage earned in 
 the first job by women in domestic service, we will note that for the 
 
 TABLE 116 
 
 WEEKLY WAGE IN FIRST JOB 
 
 Per Cent Distribution by Institutional Groups of Women Employed 
 
 at Domestic Service and Allied Occupations Where 
 
 Living Was in Addition to Wage 
 
 
 
 
 Institutional Gboufs 
 
 
 
 
 Weekly Wage 
 
 
 
 
 
 
 
 
 in 
 
 
 
 
 
 
 
 Total 
 
 First Job 
 
 
 
 
 Peni- 
 
 Work- 
 
 
 
 Plus Living 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 tentiaiy 
 
 house 
 
 Probation 
 
 
 Living only. . . . 
 
 17.1 
 
 29.6 
 
 11.1 
 
 6.7 
 
 18.8 
 
 
 15.6 
 
 $1 to $2 
 
 19.5 
 
 7.4 
 
 5.6 
 
 13.3 
 
 9.4 
 
 
 11.3 
 
 2 " 3 
 
 26.8 
 
 18.5 
 
 27.8 
 
 40.0 
 
 12.5 
 
 25.0 
 
 25.0 
 
 3 " 4 
 
 12.2 
 
 25.9 
 
 38.9 
 
 6.7 
 
 21.9 
 
 41.7 
 
 20.6 
 
 4 " 5 
 
 19.5 
 
 11.1 
 
 5.6 
 
 23.3 
 
 21.9 
 
 16.7 
 
 17.5 
 
 5 " 6 
 
 
 7.4 
 
 11.1 
 
 3.3 
 
 6.3 
 
 8.3 
 
 S.O 
 
 6 " 7 
 
 2.4 
 
 • • • < 
 
 
 .... 
 
 
 8.3 
 
 1.3 
 
 7 " 8 
 
 2.4 
 
 
 
 
 3.1 
 
 
 1.3 
 
 8 " 9 
 
 
 • > . . 
 
 .... 
 
 
 3.1 
 
 
 .6 
 
 9 " 10 
 
 
 
 .... 
 
 6.7 
 
 3.1 
 
 
 
 1.9 
 
 Total... 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cases 
 
 41 
 
 27 
 
 18 
 
 30 
 
 32 
 
 12 
 
 160 
 
OCCUPATIONAL HISTORY AND EFFICIENCY 331 
 
 TABLE 116 
 
 WEEKLY WAGE IN FIRST JOB 
 
 Per Cent Distribution by Institutional Groups of Women Employed at 
 Occupations other than Domestic Service 
 
 
 
 
 iNSTITnTIONAL GbOUPS 
 
 
 
 Weekty Wage 
 
 
 
 
 
 
 
 
 in 
 
 
 
 
 
 
 
 Total 
 
 Firat Job 
 
 
 
 
 Peni- 
 
 Work- 
 
 
 
 
 Bedford 
 
 Aubu 
 
 m Magdalen 
 
 tentiary 
 
 house 
 
 Probation 
 
 
 $0 to $1 
 
 
 8. 
 
 
 
 2.0 
 
 5.9 
 
 2.3 
 
 2.5 
 
 1 " 2 
 
 2.0 
 
 
 
 .... 
 
 5.9 
 
 
 1.3 
 
 2 " 3 
 
 
 16. 
 
 ) 5.4 
 
 
 6.9 
 
 
 3.3 
 
 3 " 4 
 
 ii.8 
 
 8. 
 
 ) 21.6 
 
 8.0 
 
 11.8 
 
 11.6 
 
 12.1 
 
 4 " 5 
 
 17.7 
 
 8. 
 
 ) 8.1 
 
 6.0 
 
 11.8 
 
 18.6 
 
 12.1 
 
 5 " 6 
 
 23.5 
 
 24. 
 
 3 21.6 
 
 10.0 
 
 11.8 
 
 11.6 
 
 16.7 
 
 6 " 7 
 
 17.7 
 
 16. 
 
 3 18.9 
 
 16.0 
 
 14.7 
 
 14.0 
 
 16.3 
 
 7 " 8 
 
 5.9 
 
 8. 
 
 D 5.4 
 
 14.0 
 
 2.9 
 
 20.9 
 
 10.0 
 
 8 " 9 
 
 11.8 
 
 
 10.8 
 
 6.0 
 
 8.8 
 
 4.7 
 
 7.5 
 
 9 " 10 
 
 3.9 
 
 4. 
 
 
 8.0 
 
 2.9 
 
 2.3 
 
 3.8 
 
 10 " 11 
 
 2.0 
 
 
 2.7 
 
 4.0 
 
 2.9 
 
 2.3 
 
 2.5 
 
 11 " 12 
 
 2.0 
 
 4.i 
 
 
 4.0 
 
 
 
 1.7 
 
 12 " 13 
 
 
 4.( 
 
 D 2.7 
 
 8.0 
 
 5.9 
 
 2.3 
 
 3.8 
 
 13 " 14 
 
 
 
 2.7 
 
 
 2.9 
 
 2.3 
 
 1.3 
 
 15 " 16 
 
 
 
 
 2.0 
 
 
 4.7 
 
 1.3 
 
 20 " 21 
 
 .... 
 
 
 
 
 2.0 
 
 
 
 .4 
 
 22 " 23 
 
 
 
 
 
 2.0 
 
 
 
 .4 
 
 24 " 25 
 
 
 
 
 
 
 2.9 
 
 
 .4 
 
 25 " 26 
 
 2.0 
 
 
 
 
 2.0 
 
 
 2.3 
 
 1.3 
 
 27 " 28 
 
 • • ■ • 
 
 
 
 
 2.0 
 
 
 .... 
 
 .4 
 
 35 " 36 
 
 .... 
 
 
 
 
 2.0 
 
 2.9 
 
 .... 
 
 .8 
 
 45 " 46 
 
 
 
 
 
 
 2.0 
 
 
 
 
 .4 
 
 Total... 
 
 100.0 
 
 100. ( 
 
 ) 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cases 
 
 61 
 
 25 
 
 37 
 
 50 
 
 34 
 
 43 
 
 240 
 
 total group the range is from living only to $10 a week plus living. 
 The range varies, however, between the groups, with Auburn and 
 the Magdalen having the shortest range, the Penitentiary and the 
 "Workhouse the longest.^' 
 
 The first wages in the occupations other than domestic service, it 
 will be seen from Table 116, have a much longer range than domestic 
 service wages, running up to $45 in the Penitentiary. In this group 
 of wages as well, the Penitentiary and the Workhouse have the long- 
 
 "The Workhouse figures for wage are probably slightly higher in propor- 
 tion to the other groups because of the fact that it was practically impossible 
 to verify the most of this information, making it necessary to accept the subject's 
 statement, which invariably tends to be higher. 
 
•§ 
 
 CO 
 
 .s 
 
 0) 
 
 o 
 t- o 
 «-* ^ 
 
 o 
 
 > 
 
 s 
 ^ 
 
 I 
 
 a 
 o 
 
 i 
 t 
 
 to 
 
 o 
 
 .a 
 
 d £ 
 
 ^ I 
 o ° 
 
 o 
 d 
 
 I 
 t 
 
 3 
 
 m 
 
 ^f 
 
 
 
 05 
 
 
 
 Nconoooo 
 
 o 
 
 
 
 ■^' -^ Tt* Th CO (N 
 
 •*' 
 
 
 
 
 
 . 
 
 
 (NCOM-*>-l 
 
 fc 
 
 9 
 
 CO T-t tH Nt^ 
 
 ^ 
 
 (M(NCOtHCS 
 
 <N 
 
 CO 
 
 fH 
 
 
 i-i 
 
 ^ 
 
 
 ti . 
 
 
 
 
 
 fSoj 
 
 tH 1-1 
 
 M 
 
 
 
 5td 
 
 
 
 
 
 ■* 
 
 
 
 
 
 M • 
 
 
 
 
 
 IxCQ 
 
 
 O 
 
 
 
 •§w 
 
 
 
 « 
 
 
 C4 
 
 
 
 • CO 
 
 
 
 
 
 u • 
 
 
 
 e@ 
 
 
 I^CQ 
 
 1-1 (M 
 
 M 
 
 
 
 •sw 
 
 
 
 
 
 .H 
 
 
 
 
 00 
 
 i-H 00 (N Tt< (M 1-1 
 
 
 
 r» 
 
 rH i-( 05 rH <N 1-1 iH 
 
 cn 
 
 1-1 
 
 
 
 
 
 
 
 
 9» 
 
 Q 
 13 
 
 © 
 
 1-1 ■ IM(N COi-l 
 
 05 
 
 
 U 
 
 
 
 
 
 (H 
 
 
 
 
 
 M 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 EC4 
 
 
 
 
 00 
 
 
 »o 
 
 r-t CQ lO lO 1-1 CO 
 
 t^ 
 
 
 
 
 
 i-H 
 
 (N 
 
 
 
 
 e© 
 
 
 
 
 
 
 
 
 
 ■* 
 
 
 
 (N-^IXN-*^ 
 
 OS 
 
 
 •^ 
 
 
 1-H 
 
 IN 
 
 CO 
 
 iH CO>dNC0CO 
 
 t- 
 
 00 
 
 
 
 f-l 
 
 (N 
 
 
 
 
 
 e© 
 
 
 
 CD 
 
 ■ 
 
 
 
 
 .-( 
 
 lo 
 
 
 
 
 
 - CO 
 
 
 
 
 
 e© 
 
 
 I-l 
 
 (NiH (N 
 
 •O 
 
 
 
 * 
 
 iH COCN(NeOcD 
 
 1> 
 
 lO 
 
 
 o 
 
 
 T— 1 
 
 u 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ■.g 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 :a 
 
 
 
 
 
 
 
 
 
 
 
 
 
 : + 
 
 
 
 
 
 
 
 
 
 
 
 
 or 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 005GOt><£>lCTt<COC^r- 
 "T 1 1 1 1 1 1 1 1 1 
 
 ^ 
 
 
 ■ 1 1 1 1 1 1 1 1 1 
 g001>tO«3T)(|M3(N^C 
 
 
 -»2 
 
 ( 
 
 8n 
 
 An 
 
 + 
 
 )t 
 
 3^ 
 
 iVi 
 
 *B 
 
 m 
 
 I 
 
 
 coco 
 
 II II 
 
 n a 
 o o 
 
 .■as 
 oo 
 
 m QQ 
 
 (A cS 
 
 aa 
 33 
 
 0000 g 
 
 oo u 
 
 ■fl ■N lO 
 
 05C0 "2 
 
 II II « 
 
 CO (=• P" p 
 
 n3' — ' a 
 
 «© ^ g ? 
 
 b « + •§ 
 
 ojw a 
 
 c^ » £ 
 CO R^ 5 
 O u>o3 s 
 
 II O McH -a 
 
 ■ g ^ 
 II S "^ 32 ■" 
 
 i-^ II ii - 
 
 &■ S o '^ 
 
 §1 i 
 
 o 
 
 ■3 J 
 o o 
 
 00 
 
 332 
 
OCCUPATIONAL HISTORY AND EFFICIENCY 333 
 
 est range. The mean wages for the total groups will be considered 
 later. 
 
 It is commonly thought that the child who leaves school early and 
 goes directly to work, gets into unskilled work and therefore earns 
 less than the child who goes to school until he has reached a higher 
 grade and so stands a better chance of getting into more skilled em- 
 ployment. To determine whether there is such a relationship in our 
 group of women between the grade finished and the weekly wage in the 
 first job, the following tables are presented. 
 
 Table 117 shows that for those who first found work in domestic 
 service, the mean grade finished, 4.10, has a relatively large standard 
 deviation of 2.72. The average first wage for those whose first em- 
 ployment was in domestic service is $3.06±1.93. The coefficient of 
 correlation is .14±.087 and the correlation ratios are .23 ±.08 and 
 .29ih.08. The amount of relationship, as we might expect, is small 
 since there is more or less of a standard wage for women in domestic 
 service which tends to change little even with the worker's increase 
 of experience and years. The more striking fact about the relation- 
 ship is the tendency for the women with the least education to earn ' 
 the smaller amounts. Women who earn less than two dollars a week 
 at domestic service show a slightly earlier average grade at leaving 
 school than those who earn more than that amount. 
 
 If we turn to the relationship between grade finished and wage in 
 the first job where this work was other than housework, we find that 
 the coefficient of correlation is .17+.067. The regression of first wage 
 on grade, showing the average change in wage for changes in grade 
 moves in a slightly irregular but fairly consistent progression. The 
 ratio ol .29±.063 indicates that there is a small relationship here. The 
 regression of grade on first wage, however, shows a much more irregu- 
 lar line of the means for grade finished. The ratio of grade on first 
 wage, .19±.067, indicates a small degree of relationship between first 
 wage and grade finished in school. Unlike the situation for domestic 
 workers the relationship between grade finished and wage is more 
 apparent for those who receive the higher wages. There seems to be 
 little difference in the amount of schooling for women receiving less 
 than nine dollars per week, but women receiving more than nine 
 dollars per week show an appreciably higher average grade attainment. 
 
 This low relationship between extent of education and first wage 
 is not surprising. On the whole, the factor of education would seem 
 logically to have less relationship to the first wage than to the later 
 
CO 
 
 ^ 
 
 O 
 
 o 
 
 V 
 
 1— > 
 
 m 
 
 
 
 1f> 
 
 
 u 
 
 
 •& 
 
 a 
 
 4) 
 
 1 
 
 bfl 
 
 o 
 
 1 
 
 >> 
 
 P! 
 
 
 ^ 
 
 qj 
 
 K 
 
 •^ 
 
 s- 
 
 § 3 
 
 
 o 
 
 ^ 
 
 o 
 
 
 ^ 
 
 cd 
 
 c3 
 
 1* 
 
 CO 
 
 ^ 
 
 .s 
 
 ^ 
 
 
 
 
 n 
 
 « 
 
 ,ta 
 
 
 pci 
 
 
 
 ;§ 
 
 o 
 
 psi 
 
 i 
 
 
 
 ■^ 
 
 1 
 
 5 
 
 d 
 
 IS 
 
 (i> 
 
 -w 
 
 
 
 <u 
 
 
 ^ 
 
 a 
 
 1 
 
 
 
 
 
 ,>> 
 
 g 
 
 g 
 
 O 
 
 9> 
 
 
 •a 
 
 
 ^ 
 
 ill 
 
 O rHOOtOUSM 
 
 f2 
 
 
 H 
 
 
 i 
 
 fe 
 ^ 
 
 rH i-H CO 05 t-t 
 
 a 
 
 • 
 
 
 <N 
 
 IN 
 
 (N 
 
 
 w 
 
 T-l 
 
 >hCQ 
 
 1-H rH 1-H 
 
 CO 
 
 GO 
 
 i-H i-HCOt-t ^rH 
 
 §3 
 
 c* 
 
 T-1 ,-1 -^NlOOOi-l 
 
 ?3 
 
 s 
 
 <£> 
 
 fH t-H 
 
 CO 
 
 03 
 
 «3 
 
 C*3 
 
 g? 
 
 (N 
 
 ^ 
 
 r-t i-H OOOOW 
 
 S 
 
 1—1 
 
 CO 
 
 1-1 rH r-1 M 00 ■-1 
 
 S5 
 
 in 
 
 <N 
 
 rH W3COT-H 
 
 CO 
 
 1-1 
 
 o 
 
 - 
 
 rt INCO 
 
 CO 
 
 o 
 
 h 
 
 .-HIM 00 in (N 
 
 rH 
 
 CO 
 
 
 
 V. 
 
 
 
 -4- 
 
 1 
 
 
 
 
 
 
 
 COCOOt^-^i-IOOlONOltOM 
 COCO«IIN(N(Mi-l,-H,-l 
 
 T 1 1 1 1 1 1 1 1 1 1 1 
 
 83Bj(i^aHi 
 
 H 
 
 COS; 5 2 -, 
 
 334 
 
OCCUPATIONAL HISTORY AND EFFICIENCY 335 
 
 earnings, since without special training the majority of people start 
 work with a low wage. 
 
 (2) Wage in Latest Job. The following tables on wage in latest 
 job for the two occupational groups are presented merely to show 
 the range of wages within the institutional groups and for reference. 
 The range does not go above that for wage in first job in any group 
 except Auburn, where one woman was earning from fifteen dollars 
 to sixteen dollars a week plus living in her latest job. This woman, 
 however, was a midwife and might be expected to have a slightly 
 higher wage than regular domestic service workers. The 21.4 per cent 
 in the Probation group who received living only is made up of only 
 three cases, so that the importance of this percentage is lessened by 
 observing the total number of cases. 
 
 TABLE 119 
 
 WEEKLY WAGE IN LATEST JOB 
 
 Per Cent Distribution by Institutional Groups of Women Employed at 
 
 Domestic Service and Allied Occupations Where Living Was 
 
 in Addition to Wage 
 
 
 
 
 iNSTITUTIONAL GbOUPS 
 
 
 
 
 WeeMy Wage 
 
 
 
 
 
 
 
 
 in 
 
 
 
 
 
 
 
 Total 
 
 Last Job 
 
 
 
 
 Peni- 
 
 Work- 
 
 
 
 
 Bedford 
 
 Aiiburn 
 
 Magdalen 
 
 tentiary 
 
 house 
 
 Probation 
 
 
 Living only. . . . 
 
 16.7 
 
 
 8.3 
 
 4.2 
 
 
 21.4 
 
 8.1 
 
 $1 to 2 
 
 6.7 
 
 
 
 4.2 
 
 3.6 
 
 
 3.3 
 
 2 " 3 
 
 
 20.0 
 
 25.0 
 
 4.2 
 
 17.9 
 
 14.3 
 
 11.4 
 
 3 " 4 
 
 20.0 
 
 6.7 
 
 33.3 
 
 25.0 
 
 21.4 
 
 14.3 
 
 20.3 
 
 4 " 5 
 
 33.3 
 
 20.0 
 
 
 20.8 
 
 17.9 
 
 14.3 
 
 20.3 
 
 5 " 6 
 
 6.7 
 
 20.0 
 
 8.3 
 
 20.8 
 
 21.4 
 
 21.4 
 
 16.3 
 
 6 " 7 
 
 
 13.3 
 
 16.7 
 
 8.3 
 
 10.7 
 
 7.1 
 
 8.1 
 
 7 " 8 
 
 16.7 
 
 13.3 
 
 8.3 
 
 8.3 
 
 
 7.1 
 
 8.9 
 
 8 " 9 
 
 
 
 
 
 3.6 
 
 
 .8 
 
 9 " 10 
 
 
 
 
 4.2 
 
 3.6 
 
 
 1.6 
 
 15 " 16: 
 
 
 
 6.7 
 
 
 
 
 .... 
 
 .8 
 
 Total . . . 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cases 
 
 30 
 
 15 
 
 12 
 
 24 
 
 28 
 
 14 
 
 123 
 
 For the occupations other than domestic service, the latest wage 
 in all of the groups except the Penitentiary has a wider range than 
 the first wage, with less massing around the lower wages. We should 
 expect to find this tendency for the beginning and end of a work record 
 to vary in wage-earning ability. 
 
336 .STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 120 
 
 V7EEKLY WAGE IN LATEST JOB 
 
 Per Cent Distribution by Institutional Groups of Women Employed at 
 Occupations other than Domestic Service 
 
 Weekly Wage 
 
 in 
 
 Last Job 
 
 iNBTITUrAONAL GROUPS 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Probation 
 
 Total 
 
 $0 to SI 
 
 1 " 2 
 
 2 " 3 
 
 3 " 4 
 
 4 " 5 
 
 5 " 6 
 
 6 " 7 
 
 7 " 8 
 
 8 " 9 
 
 9 " 10 
 
 10 " 11 
 
 11 " 12 
 
 12 " 13 
 
 13 " 14 
 
 14 " 15 
 
 15 " 16 
 
 16 " 17 
 
 17 " 18 
 
 18 " 19 
 
 19 " 20 
 
 20 " 21 
 23 " 24 
 25 " 26 
 
 27 " 28 
 
 28 " 29 
 
 30 " 31 
 
 31 " 32 
 
 32 " 33 
 65 " 66 
 
 2.0 
 
 7.8 
 
 5.9 
 
 17.7 
 
 19.6 
 
 9.8 
 
 9.8 
 
 11.8 
 
 3.9 
 
 '2!6 
 
 '26 
 
 2.0 
 2.0 
 2.0 
 
 2.0 
 
 3. 
 14. 
 14. 
 14. 
 
 7. 
 
 ii'. 
 
 14. 
 
 .7 
 
 
 8 
 
 2 
 
 8 
 
 16. 
 
 8 
 
 22 
 
 4 
 
 11 
 
 
 11 
 
 8 
 
 2 
 
 8 
 
 5. 
 
 
 11. 
 
 7 
 
 
 
 "2! 
 
 
 2. 
 
 
 2. 
 
 4 
 
 '5! 
 
 7 
 
 '2. 
 
 2.1 
 
 '2.1 
 6.3 
 6.3 
 
 16.7 
 8.3 
 
 10.4 
 
 16.7 
 8.3 
 
 ioii 
 
 2.1 
 
 2.1 
 '2.1 
 
 2.1 
 2.1 
 2.1 
 
 2.3 
 
 2.3 
 
 2.3 
 
 4.7 
 
 4.7 
 
 23.3 
 
 9.3 
 
 4.7 
 
 18.6 
 
 14.0 
 
 2.3 
 
 4.7 
 
 2.3 
 
 2.3 
 
 2.3 
 
 6.4 
 
 12.8 
 
 19.2 
 
 12.8 
 
 10.6 
 
 8.5 
 
 4.3 
 
 4.3 
 
 4.3 
 
 6.4 
 
 4.3 
 2.1 
 2.1 
 
 2.1 
 
 .0 
 
 .8 
 
 .8 
 
 2.8 
 
 5.2 
 
 10.3 
 
 18.3 
 
 11.1 
 
 8.7 
 
 12.7 
 
 8.7 
 
 2.8 
 
 4.4 
 
 1.2 
 
 .4 
 
 2.0 
 
 .8 
 
 .4 
 
 1.2 
 
 .4 
 
 .8 
 
 .4 
 
 2.8 
 
 .4 
 
 .4 
 
 .8 
 
 .4 
 
 .4 
 
 .4 
 
 Total . 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cases 
 
 51 
 
 27 
 
 36 
 
 48 
 
 43 
 
 47 
 
 252 
 
OCCUPATIONAL HISTORY AND EFFICIENCY 337 
 
 (3) Wage at First Conviction. — The wage at first conviction for 
 those engaged in domestic service is based on 44 cases, with only two 
 cases in the Probation group. It is obviously absurd to draw even 
 the most general conclusions from so few cases in the institutional 
 groups. The table shows a shorter range of wage than the other 
 domestic service groups in first and latest jobs. 
 
 TABLE 121 
 
 "WEEKLY WAGE AT TIME OF FIRST CONVICTION 
 
 Per Cent Distribution by Institutional Groups of Women Employed at 
 
 Domestic Service and Allied Occupations Where 
 
 Living Was in Addition to Wage 
 
 
 
 
 
 
 
 
 Weekly Wage at 
 
 
 
 
 
 
 
 
 Time of First 
 
 
 
 
 
 
 
 Total 
 
 Conviction 
 
 
 
 
 Peni- 
 
 Work- 
 
 
 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 tentiary 
 
 liouse 
 
 Probation 
 
 
 Living only .... 
 
 5.0 
 
 
 25.0 
 
 
 25.0 
 
 
 6.8 
 
 U to $2 
 
 15.0 
 
 
 
 20 
 
 
 
 
 
 
 9.1 
 
 2 " 3 
 
 
 33.3 
 
 50.0 
 
 20 
 
 
 
 60 
 
 
 
 50.6 
 
 20.5 
 
 3 " 4 
 
 20.0 
 
 11.1 
 
 25.0 
 
 
 
 
 
 
 13.6 
 
 4 " 6 
 
 25.0 
 
 11.1 
 
 
 
 20 
 
 
 
 
 
 
 15.9 
 
 6 " 6 
 
 20.0 
 
 22.2 
 
 
 
 40 
 
 
 
 
 
 50.0 
 
 20.5 
 
 6 " 7 
 
 
 11.1 
 
 
 
 
 
 25 
 
 
 
 
 4.6 
 
 7 " 8 
 
 16.0 
 
 
 
 
 
 
 
 
 
 6.8 
 
 8 " 9 
 
 
 
 
 
 
 
 
 
 
 .0 
 
 9 " 10 
 
 
 
 
 
 
 
 
 
 
 .0 
 
 15 " 16 
 
 
 11.1 
 
 
 
 
 
 
 
 
 2.3 
 
 Total . . . 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cases 
 
 20 
 
 9 
 
 4 
 
 5 
 
 4 
 
 2 
 
 44 
 
338 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 The wage table showing earnings of those in occupations other 
 than domestic service at the time of first conviction is based on only 
 52 cases, with so few numbers in the institutional groups that again no 
 more comment is justifiable than that there were no women earning 
 under three dollars a week and that the range of wages is shorter in 
 most of the institutional groups than for the previous considerations 
 of non-housework jobs. 
 
 TABLE 122 
 
 WEEKLY WAGE AT TIME OF FIRST CONVICTION 
 
 Per Cent Distribution by Institutional Groups of Women Employed at 
 Occupations other than Domestic Service 
 
 Institutional Ghotjps 
 
 Weekly Wage at 
 
 Time of First 
 
 Conviction 
 
 80 to $1 
 
 1 " 2.... 
 
 2 " 3.... 
 
 3 " 4.... 
 
 4 " 5.... 
 
 5 " 6.... 
 
 6 " 7.... 
 
 7 " 8.... 
 
 8 " 9.... 
 
 9 " 10.... 
 10 " 11.... 
 
 14 " 15.... 
 
 15 " 16.... 
 17 " 18.... 
 20 " 21.... 
 22 " 23.... 
 25 " 26.... 
 30 " 31.... 
 
 Total. 
 
 Number of cases 
 
 Bedford 
 
 25.0 
 
 25.0 
 
 25.0 
 
 6.3 
 
 6.3 
 6.3 
 
 6.3 
 
 100. 
 
 16 
 
 Auburn 
 
 12.5 
 12.5 
 12.5 
 12.5 
 12.5 
 
 25!6 
 
 12.5 
 100.0 
 
 Magdalen 
 
 25 
 25 
 12, 
 
 12. 
 
 12. 
 
 12. 
 
 100.0 
 
 Peni- 
 tentiary 
 
 12.5 
 
 25;6 
 
 12.6 
 25.0 
 12.5 
 
 12.5 
 
 100.0 
 
 Work- 
 house 
 
 33. 
 33! 
 33! 
 
 100.0 
 
 Probation 
 
 11.1 
 
 22.2 
 22.2 
 
 100.0 
 
 Total 
 
 .0 
 
 .0 
 
 .0 
 
 1.9 
 
 13.5 
 
 15.4 
 
 19.2 
 
 13.5 
 
 3.9 
 
 11.5 
 
 5.8 
 
 1.9 
 
 1.9 
 
 1.9 
 
 3.9 
 
 1.9 
 
 1.9 
 
 1.9 
 
 100.0 
 
 52 
 
OCCUPATIONAL HISTORY AND EFFICIENCY 339 
 
 (4) Wage at Latest Conviction. — If we turn to the wage at latest 
 conviction for the domestic service workers, we see in Table 123 that 
 the numbers working in, domestic service at this time are about the 
 same as at first conviction, and that the distribution of wages among 
 the institutional groups are not radically different. 
 
 TABLE 123 
 
 WEEKLY WAGE AT TIME OF PRESENT CONVICTION 
 
 Per Cent Distribution by Institutional Groups of Women Employed at 
 
 Domestic Service and Allied Occupations Where 
 
 Living Was in Addition to Wage 
 
 
 
 
 Institutional Gbodps 
 
 
 
 
 Weekly Wage at 
 
 
 
 
 
 
 
 
 Time of Last 
 
 
 
 
 
 
 
 Total 
 
 Conviction 
 
 
 
 
 Peni- 
 
 Work- 
 
 
 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 tentiary 
 
 house 
 
 Probation 
 
 
 Living only. . . . 
 
 15.4 
 
 
 26.0 
 
 
 
 
 6.7 
 
 $1 to $2 
 
 15.4 
 
 
 
 9.i 
 
 
 
 
 6.7 
 
 2 ' 
 
 3 
 
 
 25.0 
 
 
 
 42.9 
 
 50 
 
 
 
 13.3 
 
 3 ' 
 
 4 
 
 15.4 
 
 
 50 
 
 9.1 
 
 14.3 
 
 
 
 13.3 
 
 4 ' 
 
 5 
 
 30.8 
 
 12.6' 
 
 
 9,1 
 
 . i . • 
 
 
 
 13.3 
 
 5 ' 
 
 6 
 
 
 25.0 
 
 
 36.4 
 
 
 50 
 
 
 
 15.6 
 
 6 ' 
 
 7 
 
 . 
 
 12.5 
 
 25.0 
 
 18.2 
 
 28.6 
 
 
 
 13.3 
 
 7 ' 
 
 8 
 
 23.1 
 
 12.5 
 
 
 18.2 
 
 
 
 
 13.3 
 
 8 ' 
 
 9 
 
 
 
 
 
 14.3 
 
 
 
 2.2 
 
 9 ' 
 
 10 
 
 
 
 
 
 
 
 
 .0 
 
 15 " 16 
 
 
 12.5 
 
 
 
 
 
 
 2.2 
 
 Total. .. 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cases 
 
 13 
 
 8 
 
 4 
 
 11 
 
 7 
 
 2 
 
 45 
 
340 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 Table 124 shows the distribution of wages at time of latest con- 
 viction for the 57 workers in occupations other than domestic service. 
 The wages run higher than at the time of first conviction though the 
 cases at the upper limit are scattering. 
 
 TABLE 124 
 
 WEEKLY WAGE AT TIME OF PRESENT CONVICTION 
 
 Per Cent Distribution by Institutional Groups of Women Employed at 
 Occupations other than Domestic Service 
 
 Weekly Wage at 
 
 InstitutionaI/ Groups 
 
 
 Time of Last 
 Cdnviction 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Probation 
 
 Total 
 
 $0 to $1 
 
 1 " 2 
 
 2 " 3 
 
 3 " 4 
 
 4 " 5 
 
 5 " 6 
 
 6 " 7 
 
 7 " 8 
 
 8 " 9 
 
 9 " 10 
 
 10 " 11 
 
 11 " 12 
 
 12 " 13 
 
 13 " 14 
 
 14 " 15 
 
 15 " 16 
 
 17 " 18 
 
 20 " 21 
 
 22 " 23 
 
 25 " 26 
 
 30 " 31 
 
 32 " 33 
 
 65 " 66 
 
 '7 
 
 '7 
 
 15 
 
 46 
 
 7 
 
 '7 
 '7 
 
 7 
 
 7 
 4 
 2 
 
 7 
 
 7 
 7 
 
 12 
 12 
 12 
 12 
 12 
 
 25 
 12 
 
 5 
 5 
 5 
 5 
 5 
 
 6 
 5 
 
 28 
 28 
 14 
 
 14 
 
 i4 
 
 6 
 6 
 3 
 3 
 
 3 
 
 '9 
 
 is 
 
 18 
 18 
 
 is 
 
 9 
 
 '9 
 
 i 
 
 2 
 2 
 
 2 
 
 2 
 1 
 
 i 
 
 is 
 
 27 
 '9 
 
 9 
 9 
 9 
 
 '9 
 '9 
 
 2 
 3 
 
 i 
 1 
 1 
 1 
 
 i 
 i 
 
 i4 
 
 14 
 14 
 28 
 
 i4 
 i4 
 
 ,3 
 .3 
 .3 
 .6 
 
 [3 
 '3 
 
 .0 
 1.8 
 1.8 
 1.8 
 3.5 
 8.8 
 22.8 
 15.8 
 3.5 
 10.5 
 5.3 
 1.8 
 5.3 
 1.8 
 1.8 
 1.8 
 1.8 
 1.8 
 1.8 
 1.8 
 1.8 
 1.8 
 1.8 
 
 Total . . . 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cases 
 
 13 
 
 8 
 
 7 
 
 11 
 
 11 
 
 7 
 
 57 
 
OCCUPATIONAL HISTORY AND EFFICIENCY 341 
 
 (5) Average Wage for Prevadlmg Work. If, instead of taking the 
 wage of these women at any specified time, we consider the average 
 wage for the prevailing work, we have an index of their general 
 earning capacity. The average wage is obtained by allowing for the 
 time element — that is, making a simple weighted average with the 
 length of time in each of the jobs in the prevaiHng work the weight. 
 The following table shows that the average weekly wage for domestic 
 service workers begins at living only in all groups except the Magdalen 
 and runs to no more than nine dollars in three of the groups and not 
 over seven dollars a week in the other two. Living, of course, is in 
 addition to these wages. 
 
 TABLE 125 
 
 Average Weekly Wage for Prevailing Work of Women Employed at 
 
 Domestic Service and Allied Occupations where 
 
 Living was in Addition to Wage 
 
 Per Cent Distribution by Institutional Groups 
 
 Average 
 
 Institutional Groups 
 
 
 Weekly Wage 
 plus Liying 
 
 Bedford 
 
 Aubitrii 
 
 Magdalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Probation 
 
 Total 
 
 Living only. . . . 
 $1 to $2 
 
 2 " 3 
 
 3 " 4 
 
 4 " 5 
 
 5 " 6 
 
 6 " 7 
 
 7 " 8 
 
 8 " 9 
 
 8.7 
 15.2 
 34.8 
 
 8.7 
 19.6 
 
 4.4 
 
 8.7 
 
 11.1 
 
 3.7 
 18.5 
 29.6 
 22.2 
 14.8 
 
 33^3 
 33.3 
 
 27.8 
 5;6 
 
 3.1 
 
 6.3 
 
 25.0 
 
 18.8 
 
 25.0 
 
 9.4 
 
 'i'.i 
 
 9.4 
 
 5.7 
 
 2.9 
 14.3 
 20.0 
 34.3 
 14.3 
 
 5.7 
 
 2^9 
 
 9.5 
 
 i4!3 
 9.5 
 
 38.1 
 4.8 
 
 19.1 
 
 'i'.h 
 
 6.7 
 
 6.2 
 24.2 
 18.4 
 24.2 
 11.2 
 
 6.2 
 .6 
 
 2.8 
 
 Total . . . 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cases 
 
 46 
 
 27 
 
 18 
 
 32 
 
 35 
 
 21 
 
 179 
 
 The following table (Table 126), on the other hand, has a much 
 longer range, from $1 to $46, with the Penitentiary and the Workhouse 
 having the longest range. The Probation group has the fewest cases 
 in the lowest wage groups, and has none who have earned less than $4 
 a week for their average wage where their prevailing work was in an 
 occupation other than domestic service. 
 
342 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 126 
 
 Average Weekly Wage for Prevailing Work of Women Employed at 
 Occupations other than Domestic Service 
 
 Per Cent Distribution by Institutional Groups 
 
 Average 
 Weekly Wage 
 
 Institutional Groups 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Probation 
 
 Total 
 
 $1 to 82 
 
 2 " 3 
 
 3 " 4 
 
 4 " 5 
 
 5 " 6 
 
 6 " 7 
 
 7 " 8 
 
 8 " 9 
 
 9 " 10 
 
 10 " 11 
 
 11 " 12 
 
 12 " 13 
 
 13 " 14 
 
 14 " 15 
 
 15 " 16 
 
 16 " 17 
 
 17 " 18 
 
 19 " 20 
 
 20 " 21 
 
 22 " 23 
 
 24 " 25 
 
 25 " 26 
 
 27 " 28 
 
 28 " 29 
 
 30 " 31 
 
 32 " 33 
 
 35 " 36 
 
 45 " 46 
 
 Total... 
 
 Number of cases 
 
 2.1 
 
 i'.i 
 
 18.8 
 
 18.8 
 
 20.8 
 
 14.6 
 
 6.3 
 
 2.1 
 
 4.2 
 
 2.1 
 
 2.1 
 
 2.1 
 
 2.1 
 
 'y.i 
 
 9. 
 
 4. 
 27. 
 18. 
 
 9. 
 
 9. 
 
 4. 
 
 
 2 
 
 .1 
 
 5 
 
 .6 
 
 11. 
 
 .3 
 
 19. 
 
 .2 . 
 
 13. 
 
 .1 
 
 16. 
 
 .1 
 
 8. 
 
 .6 
 
 8. 
 
 
 2. 
 
 6 
 
 2. 
 
 
 2. 
 
 
 '2. 
 
 i 
 
 '2. 
 
 6 
 
 
 4:A 
 
 4.4 
 
 6.5 
 
 10.9 
 
 15.2 
 
 8.7 
 
 13.0 
 
 6.5 
 
 4.4 
 
 4.4 
 
 2^2 
 
 2.2 
 
 2.2 
 
 4!4 
 
 2^2 
 2.2 
 2.2 
 
 2.2 
 2.2 
 
 7.1 
 10.7 
 10.7 
 14.3 
 
 7.1 
 17.9 
 
 3.6 
 
 7.1 
 
 '3.6 
 
 3.6 
 
 3.6 
 3.6 
 
 3.6 
 3.6 
 
 4.6 
 
 11.4 
 
 11.4 
 
 25.0 
 
 13.6 
 
 9.1 
 
 4.6 
 
 2.3 
 
 2.3 
 
 2.3 
 
 2.3 
 
 4.6 
 2.3 
 
 4.6 
 
 14 
 14 
 
 15.6 
 
 10.3 
 
 7.1 
 
 4,5 
 
 1.8 
 
 3.1 
 
 .9 
 
 .5 
 
 1.3 
 
 .5 
 
 .5 
 
 1.3 
 
 .9 
 
 .9 
 
 .5 
 
 3.6 
 
 .5 
 
 .5 
 
 .5 
 
 .5 
 
 .9 
 
 .5 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 48 
 
 22 
 
 36 
 
 46 
 
 28 
 
 44 
 
 224 
 
 Charts XlXa and XlXb show graphically how the average wage 
 for prevailing work varies between the domestic service workers and 
 those in other occupations. The charts are based on the figures for the 
 total group, and indicate that while the range of wage for the domestic 
 service workers is short, that for the workers in other occupations is 
 very long, broken and irregular. 
 
 The preceding tables on wage, which have been included largely 
 for reference to determine how the total group is made up, will be 
 
u 
 > 
 
 m 
 
 CO 
 
 o 
 
 Q 
 
 a 
 
 ? 
 
 
 u 
 
 CO 
 
 H 
 
 CO 
 
 S 
 o 
 
 Q 
 
 I 
 
 Z 
 
 o 
 z 
 
 I 
 
 c 
 
 ■ i 
 
 o 
 
 ij 
 
 H 
 (S 
 
 < 
 OS 
 
 > 
 
 « I— 
 
 J « 1- 
 
 ^1 S 
 
 3 
 "3. 
 
 343 
 
344 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 127 
 
 WEEKLY WAGE EARNED BY DELINQUENT WOMEN AT SPECI- 
 FIED TIMES DURING THEIR WORK fflSTORIES 
 
 Per Cent Distribution of Women Employed at Domestic Service and 
 Allied Occupations Where Living Was in Addition to Wage 
 
 
 Specified Times 
 
 Weekly Wage 
 
 In 
 
 First Job 
 
 In 
 
 Latest Job 
 
 At Time of 
 
 First 
 Conviction 
 
 At Time of 
 
 Present 
 Conviction 
 
 Prevailing 
 Work 
 
 
 15.6 
 
 11.3 
 
 25.0 
 
 20.6 
 
 17.5 
 
 5.0 
 
 1.3 
 
 1.3 
 
 .6 
 
 1.9 
 
 8.1 
 
 3.3 
 
 11.4 
 
 20.3 
 
 20.3 
 
 16.3 
 
 8.1 
 
 8.9 
 
 .8 
 
 1.6 
 
 .8 
 
 6.8 
 
 9.1 
 
 20.5 
 
 13.6 
 
 15.9 
 
 20.5 
 
 4.6 
 
 6.8 
 
 ■2!3 
 
 6.7 
 6.7 
 13.3 
 13.3 
 13.3 
 15.6 
 13.3 
 13.3 
 2.2 
 
 ■212 
 
 6.7 
 
 $1 to $2 .... 
 
 6.2 
 
 2 " 3 
 
 24.2 
 
 3 " 4 
 
 18.4 
 
 4 " 5 
 
 24.2 
 
 5 " 6 
 
 11.2 
 
 6 " 7 
 
 6.2 
 
 7 " 8 
 
 .6 
 
 8 " 9 
 
 2.8 
 
 9 " 10 
 
 
 15 " 16 
 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cases 
 
 160 
 
 123 
 
 44 
 
 45 
 
 179 
 
 
 2.56 
 ±.147 
 
 3.94 
 + .203 
 
 3.66 
 
 ±.384 
 
 4.29 
 ±.400 
 
 3.25 
 
 (Tm 
 
 + .131 
 
 
 
 (7 
 
 1.85 
 + .104 
 
 2.25 
 + .143 
 
 2.55 
 
 ±.272 
 
 2.68 
 ±.283 
 
 1.75 
 
 (Ja 
 
 + .109 
 
 
 
 of much more interest if these totals are combined, showing what 
 wages the women engaged in domestic service and those in other occu- 
 pations were earning at specified times during their work histories, and 
 for their prevailing work. Table 127 presents the wages for women 
 in domestic service and gives in addition to the actual percentages of 
 women earning given wages the mean wage and standard deviation for 
 each group. The mean wage in the first job, $2.56±1.85 is, it will 
 be seen, considerably less than the mean wage in the latest job, 
 $3.94±:2.25. This trend is in the direction we should expect, 
 that one should earn larger wages with increase of experience. 
 The comparison of the mean wage earned at first conviction and in 
 the first job indicates that these women were earning more at the 
 time of first conviction, though there was a wider dispersion of wages. 
 
OCCUPATIONAL HISTORY AND EFFICIENCY 345 
 
 TABLE 128 
 
 WEEKLY WAGE EARNED BY DELINQUENT WOMEN AT SPECI- 
 FIED TIMES DURING THEIR WORK HISTORIES 
 
 Per Cent Distribution of Women Employed at Occupations other than 
 
 Domestic Service 
 
 Weekly Wage 
 
 Specified Times 
 
 In 
 First Job 
 
 In 
 
 Latest Job 
 
 At Time of 
 
 First 
 Conviction 
 
 At Time of 
 
 Present 
 Conviction 
 
 Prevailing 
 Work 
 
 $0 to $1. 
 
 1 
 
 2 
 3 
 
 4 
 5 
 6 
 7 
 8 
 9 
 10 
 11 
 12 
 13 
 14 
 15 
 16 
 17 
 18 
 19 
 20 
 22 
 23 
 24 
 25 
 27 
 28 
 30 
 31 
 32 
 35 
 45 
 65 
 
 2. 
 
 3. 
 
 4. 
 
 5. 
 
 6. 
 
 7. 
 
 8. 
 
 9. 
 10. 
 11. 
 12. 
 13. 
 14. 
 15. 
 16. 
 17. 
 18. 
 19. 
 20. 
 21. 
 23. 
 24. 
 25. 
 26. 
 28. 
 29. 
 31. 
 32, 
 33. 
 36. 
 46. 
 66. 
 
 Total. 
 
 Number of cases . 
 
 Mean. 
 0-™... 
 
 (T. . 
 CT'o. 
 
 2.5 
 
 1.3 
 
 3.3 
 
 12.1 
 
 12.1 
 
 16.7 
 
 16.3 
 
 10.0 
 
 7.5 
 
 3.8 
 
 2.5 
 
 1.7 
 
 3.8 
 
 1.3 
 
 i'.s 
 
 A 
 A 
 
 '".i 
 
 1.3 
 
 .4 
 
 100.0 
 
 240 
 
 6.78 
 + .356 
 
 5.51 
 + .251 
 
 .8 
 
 .8 
 
 2.8 
 
 5.2 
 
 10.3 
 
 18.3 
 
 11.1 
 
 8.7 
 
 12.7 
 
 8.7 
 
 2.8 
 
 4.4 
 
 1.2 
 
 .4 
 
 2.0 
 
 .8 
 
 .4 
 
 1.2 
 
 .4 
 
 .8 
 
 2.8 
 .4 
 .4 
 .8 
 .4 
 .4 
 
 100.0 
 
 252 
 
 9.17 
 + .409 
 
 6.495 
 ±.0837 
 
 1.9 
 13.5 
 15.4 
 19.2 
 13.5 
 
 3.9 
 11.5 
 
 5.8 
 
 1.9 
 1.9 
 
 1.9 
 
 3,9 
 1.9 
 
 1.9 
 i'.9 
 
 100.0 
 
 52 
 
 8,52 
 + .789 
 
 5.69 
 
 + .558 
 
 1,8 
 1.8 
 1.8 
 3.5 
 8.8 
 22.8 
 15,8 
 3,5 
 10.5 
 5,3 
 1.8 
 5.3 
 1.8 
 1,8 
 1.8 
 
 'i!8 
 
 1.8 
 1.8 
 
 1.8 
 
 1.8 
 
 100,0 
 
 57 
 
 10.2 
 ±1.26 
 
 9.54 
 ±.893 
 
 .0 
 
 .5 
 
 .5 
 
 4.0 
 
 9.4 
 
 14.7 
 
 14.7 
 
 15.6 
 
 10.3 
 
 7.1 
 
 4.5 
 
 1.8 
 
 3.1 
 
 .9 
 
 .5 
 
 1.3 
 
 .5 
 
 .5 
 
 .0 
 
 .3 
 
 .9 
 
 .9 
 
 .0 
 
 .5 
 
 .6 
 
 .5 
 
 .5 
 
 .5 
 
 .0 
 
 ,5 
 
 .9 
 
 .5 
 
 .0 
 
 1 
 
 100.0 
 
 224 
 
 8.90 
 ±.441 
 
 6.60 
 ±312 
 
346 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 It must be remembered, however, that a very large percentage of 
 women who were included in the wage table for the first job were not 
 working at the time of the first conviction and were, therefore, not 
 included in the table under discussion. A comparison of the mean 
 wa^es at first and latest convictions also shows that the women were 
 earning more at latest conviction than at the first, and more than at 
 any of the other times which we have specified. The average wage 
 for prevailing work which applies to 179 women is $3.25±1.7S. That 
 is, the average wage for the prevailing work, while higher than the 
 average wage in the first job, is lower than the average wage in the 
 latest job, or the wage at the time of first or present conviction. 
 Throughout, the average wage for women employed at domestic service 
 is low, with a maximum of a mean weekly wage of $4.29±2.68. There 
 is also little variation between the lowest wage of $2.56 and the high- 
 est wage of $4.29. 
 
 Table 128 presents the wages earned by women in occupations other 
 than domestic service, and the average wages for each of the specified 
 times. The trend is in the same direction as that in Table 127, 
 with the lowest mean wage of $6.78 in the first job, in contrast to the 
 higher wage of $9.17 in the latest job. The standard deviation, how- 
 ever, is relatively large in both of these cases. The wage at first con- 
 viction, $8,52, is based on 52 cases only and though the mean is smaller 
 than the average wage of $10.20, earned at present conviction, the 
 standard deviation of the latter is so large, proportionately, as to lessen 
 markedly the importance of the larger mean. The mean of the average 
 wage for prevailing work is $8.90rfc6.60. The average wage for pre- 
 vailing work, therefore, would seem to be larger than either the wage 
 in the first job or the wage at first conviction. On the whole, the wages 
 are low, and the frequency tables show a wide scattering with only a 
 few cases at the upper wage limit. As shown in Table 113, the pre- 
 vailing work of 40.4 per cent of the delinquent women was housework, 
 while 56.1 per cent of the women were engaged in other work. This 
 40 per cent of women were earning an average of $3.25 a week plus 
 living at domestic service, while over 50 per cent of the women were 
 earning an average of $8.90 a week. 
 
 Since the numbers of women in institutional groups are too few 
 for comparison of the means when divided into occupations, a corre- 
 lation ratio has been worked between the institutional groups and the 
 average weekly wage for prevailing work. The ratio of .25 ±.070 as 
 given in Table 129 shows that between institutional groups and the 
 
Means 
 
 (Average 
 
 Weekly Wage 
 
 Work) 
 
 CO N 00 CD t* U3 
 ^■*00OOci5 
 
 CO CO CO -*-*^ 
 
 
 
 1 
 
 SS?52g5§S 
 
 1—1 
 
 1 
 
 
 
 1 
 
 g 
 
 1 
 
 1 
 
 o 
 •5 
 
 6 -8 
 
 eOrHrH 
 
 lO 
 
 8 -Z 
 
 1-H 
 
 1-t 
 
 Z -9 
 
 -* rt N-* 
 
 )— 1 
 1—1 
 
 9 -S 
 
 IM ■* «3 CO "3 ■-I 
 
 s 
 
 B -* 
 
 05C0 OOINOO 
 
 % 
 
 t -E 
 
 TtHOOCD«>l>(N 
 
 CO 
 CO 
 
 s -z 
 
 CO W3 ':D 00 IC CO 
 
 ^ 
 
 Z -X 
 
 I>i-H IM.-I 
 
 1— 1 
 I— I 
 
 m-os 
 
 ■*C0 .-ICi|C<l 
 
 l-H 
 
 
 ^ 
 
 1 
 
 1 
 
 
 1 
 1 
 
 1 
 
 
 Q! 
 
 "ci 
 
 O 
 
 H 
 
 
 I 
 
 SO 
 
 dn 
 
 3TJ 
 
 OJ 
 
 3 
 
 I 
 
 
 
 o 
 o 
 
 a 
 
 "3.2 
 
 a 
 o 
 
 9 > 
 
 0) ^ 
 
 OS'S 
 
 .o 2 
 
 
 > S 
 
 347 
 
348 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 average weekly wage for women whose prevailing work was domestic 
 service, there is a small but probably significant relationship. Refer- 
 ence to the average wages earned by the women in the several institu- 
 tional groups shows that the order of arrangement of the institutional 
 groups in accordance with increasing average weekly wage for pre- 
 vailing work where that work is housework is as follows: (1) Bed- 
 ford, (2) Auburn, (3) Magdalen, (4) Penitentiary, (5) Workhouse, 
 (6) Probation. 
 
 In a correlation ratio between the institutional groups and average 
 weekly wage for prevailing work of women employed at occupations 
 other than domestic service, we find that ■)j=.25ih.063. This ratio also 
 would indicate that, the relationship between the two factors is small 
 but probably significant. The order of arrangement of the institutional 
 groups in accordance with the increasing average weekly, wages varies 
 from the order of arrangement in Table 129, and is as follows: (1) 
 Bedford, (2) Magdalen, (3) Auburn, (4) Probation, (5) Workhouse, 
 (6) Penitentiary. 
 
 From the two tables discussed above (Tables 129 and 130), it 
 seems evident that there is a tendency for the Bedford women to re- 
 ceive a lower average wage than the women of the other groups. 
 There are variations between the wages at various times in all of the 
 other groups except the Workhouse, which comes near the highest 
 wage limit in both domestic service and occupations other than do- 
 mestic service. 
 
 (&) Relationship Between Average Wage and Social Factors 
 
 (1) Relationship Between Average Wage and Grade Finished. — In 
 Tables 117 and 118 we discussed the relationship between the grade 
 finished and the first wage received, showing that the grade finished 
 has a very slight relationship to the wage one is first able to earn in 
 either domestic service or in occupations other than domestic service. 
 If we consider the relationship between the grade finished and the 
 average wage for prevailing work of those employed in domestic serv- 
 ice and in occupations other than domestic service, we find evidence 
 of somewhat higher relationships. Table 131 shows a coefficient of 
 correlation of .29 ±.077. This indicates a tendency for those who 
 reach the higher grades to earn a higher average wage for prevailing 
 work, where the prevailing work is domestic service. 
 
e 
 
 CO 
 
 g 
 
 I 
 I 
 
 a 
 
 I 
 .o 
 
 
 I 
 
 73 
 
 I'll 
 
 I 
 o 
 
 Z^-9^ 
 
 S^-£* 
 
 e^-T* 
 
 I»-6£ 
 
 62-LS 
 
 ZE-9E 
 
 98-£E 
 
 ££-!£ 
 
 I£-6e 
 
 6Z-IZ 
 
 t^OJt-'HOOl 
 
 iZ-SZ 
 
 SZ-EZ 
 
 £Z-TS 
 
 IZ-6T 
 
 ai-zi 
 
 ZT-ST 
 
 9I-£I 
 
 £1-11 
 
 TI-6 
 
 6 -I 
 
 I -s 
 
 S -£ 
 
 Et-I« 
 
 i-H iH M 
 
 NfHi-l-^^N 
 
 CC'H^OIMCD 
 
 O^OS'-'t-t^ 
 
 OSONOOt^O 
 
 OCCtD^»ON 
 
 ^'2 * H o o 
 
 adnoiQ 
 
 349 
 
•o 
 
 a> 
 
 >, 
 
 o 
 
 
 o. 
 
 a 
 
 (H 
 
 S 
 
 
 
 ^1 
 
 T3 o 
 
 o +» 
 
 ^g 
 
 SP'-O 
 
 C S3 
 
 g<! 
 
 
 IH «] 
 
 ^^ 
 
 If 
 
 32 
 
 4> V 
 
 s ^^ 1 
 
 S &§ 5 
 
 ri * 9 -zs 
 1 |-^ 1 
 
 H^g-H 
 
 
 go 
 
 •at* 
 
 S <» 
 
 
 •si 
 
 •«« 
 
 a> a> 
 
 .3 .a 
 
 '3 s 
 
 faCO 
 
 a> o 
 
 •o-^d 
 
 cj to 
 
 IH 0) 
 
 o a 
 
 fl5 
 
 Sp 
 
 
 Its 
 
 <u 
 
 ,a 
 
 d 
 
 o 
 
 •a 
 
 
 
 d) 
 
 s 
 
 o 
 
 u 
 
 
 S|| 
 
 
 "5 
 
 
 u3eooooOTi< 
 
 to 
 
 
 
 
 
 1(3 U5 •* ■* CO M eo 
 
 tH 
 
 
 
 ss| 
 
 
 
 
 
 
 
 
 CO 
 
 
 ■a 
 
 
 cq ■* >o "3 r-i .-1 
 
 a 
 
 «o 
 
 tH'-hOS 
 
 
 H 
 
 i-H CO IM CO i-H rH 
 
 ■^ 
 
 CO 
 
 
 
 1-i 
 
 »» 
 
 
 ^ is-a'S 
 
 f-( 
 
 I— t 
 
 
 
 
 
 
 
 
 iglxM^ 
 
 
 
 00 
 
 Second 
 Year 
 High 
 
 School 
 
 
 o 
 
 
 
 
 
 " ^ 
 
 
 
 
 
 ^ 
 
 
 -p uji o 
 
 
 
 
 
 SS»S 
 
 .-1 (NN "-1 
 
 CD 
 
 
 
 S!StS| 
 
 
 
 
 00 
 
 rt >-l(Nr-ieOrH 
 
 OJ 
 
 
 
 
 
 
 
 
 
 (N 
 
 g 
 
 t^ 
 
 i-H i-liHCONN --I 
 
 I-H 
 I-H 
 
 ■*' 
 
 M 
 
 
 
 
 »» 
 
 1 
 
 
 
 
 
 
 
 
 >o 
 
 » 
 
 
 ■-I (M USiMr-l I-H 
 
 <N 
 
 
 
 (D 
 
 
 I-H 
 
 tH 
 
 
 
 
 
 e« 
 
 I§ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 
 
 U5 
 
 CO C» (M 1> .-H .-1 
 
 (N 
 
 CO 
 
 
 
 
 00 
 
 
 
 C<1I>!DININ 
 
 05 
 
 
 
 ^ 
 
 
 T-H 
 
 si 
 
 
 
 
 >o 
 
 
 CO 
 
 N .-1 00 CO >0 CMM 
 
 •* 
 
 
 
 
 N 
 
 CO 
 
 
 
 
 
 «@ 
 
 
 
 
 cq 
 
 
 
 (N w N lO t~ rH (N 
 
 o 
 
 
 
 
 
 IM 
 
 gi 
 
 »-l 
 
 rH(N"-H i-l'-l 
 
 to 
 
 CO 
 
 CO 
 
 
 
 
 
 e& 
 
 * 
 
 rH rHINrH-^lNTH 
 
 N 
 
 N 
 
 
 o 
 
 
 iH 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 <u 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 UD^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 =3J^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^g 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ^fe: 
 
 
 
 
 
 
 
 
 
 
 
 
 a 
 
 S3 M 
 
 
 
 O3001>tO>OTi(c0C<li- 
 
 ^ 
 
 ^ s 
 
 
 
 S= = = - = = : = 
 
 ^ 
 
 
 
 
 oot-«>nTtieoc<i<HC 
 
 
 ^1 
 
 
 
 «© 
 
 
 
 
 
 
 1.^ 
 
 MOM anipBAMJ 101 
 
 
 
 eS 
 
 m 
 
 L A 
 
 pp 
 
 aji 
 
 V9 
 
 3e 
 
 I9A 
 
 V 
 
 
 
 1 
 
 P] 
 
 g 
 
 
 
 s 
 
 g 
 
 
 
 ■^ 
 
 "ofl 
 
 
 
 n 
 
 m 
 
 ^ 
 
 ■* 
 
 
 r>. 
 
 o 
 
 o 
 
 ■H 
 
 -H 
 
 m 
 
 >o 
 
 (N 
 
 CO 
 
 o 
 
 .4 
 
 
 & 1 
 
 i 
 
 ^ 
 ^ 
 
 t 
 
 o 
 
 1 
 
 B 
 ji 
 
 ^ o 
 
 S«3*00 
 
 b 2P Kg-"* S 
 .Si>5'"'iB_; 5 
 ag I ' S" 
 
 ,« So 
 
 ^ £41 o a |>>B - 
 
 350 
 
OCCUPATIONAL HISTORY AND EFFICIENCY 351 
 
 In Table 132 a similar relationship is shown between the grade 
 finished and the average wage for prevailing work of women in occu- 
 pations other than domestic service. The coefficient of correlation 
 (.25±:.066) gives evidence of a small, significant relationship in this 
 case also. It may be noted that there is a distinct rise in weekly earn- 
 ings for the women who have had more schooling than the average 
 amount for the group. On the whole, we may state that there is a 
 tendency among all of the working women in our group for those with 
 more education to earn a higher wage in their prevailing work than 
 those with less education. 
 
 (2) Relationship Between Average Wage and Number of Convic- 
 tions. — In trying to determine whether there is any relationship between 
 the extent of delinquency measured by number of convictions and 
 economic efficiency as measured by average wage for prevailing work, 
 the partial correlation coefficient of number of previous convictions 
 and average weekly wage for prevailing work, for constant age, has 
 been used. Before considering this coefficient, it may be of interest to 
 present Tables 133 and 134, showing the relationship between age and 
 average wage for the two occupational groups. 
 
 Table 133 shows this relationship for women employed at domestic 
 service, giving a correlation coefficient of zero. The regression of 
 wage for prevailing work on age is, however, non-linear; the correla- 
 tion ratio for this is .27 ± .069. The tendency here is toward a slight in- 
 crease in wage from the earliest ages and a decrease in wage in the high- 
 est ages. The tendency between ages 26 and 44 is not clear except in 
 so far as the wage is, in general, higher than for either the very young 
 or the very old. The regression of age on wage for prevailing work 
 with a correlation ratio of .11 ±.07 shows no trend of significance. 
 
 If we turn to the correlation of age with average weekly wage of 
 women in occupations other than domestic service, we find more rela- 
 tion between age and wage for prevailing work than in the previous 
 comparison. The regression of age on wage for prevailing work 
 shows that the women receiving less than $7 a week are below the 
 average in age. The correlation ratio is .37±.058. There is a clearly 
 marked tendency for these women as they grow older to increase 
 in wage-earning capacity. From these two tables it would seem that 
 domestic service is like a blind alley for the women who do that work 
 mainly for a living, while the workers in other occupations tend to 
 increase their wages as they grow older. 
 

 •i 
 
 o 
 
 h 03^0 
 OmED u 
 
 "aj 
 
 
 
 
 'm 
 
 b- -H © t* "3 "S "3 ■* ^ 
 
 WitOiOOiH 
 
 rt »-iMiO^M 
 
 i-H ^ ^ CC CC CO -^ 
 
 «Tj<O.Hl-t 
 
 
 3110^ 3UI^'BA9I J JOJ 
 93B^ Jtl3199^ 93BI9AV 
 
 D D 
 g o 
 
 •c-c 
 
 gg 
 
 ■^ CO 
 
 coco 
 00 
 
 S>!<3 
 
 II 4<'t! .5 
 
 II ^ 
 oS 
 
 Is 
 
 a* 
 
 *"'B a 
 
 oj P "S 
 
 
 'O'w«o 5J flj 2 
 
 (4 > II .£ fe •'^ 
 ..oa " 
 
 S.2 
 
 II 
 
 as 
 s§ 
 00 
 
 
 
 i 
 
 1^ 
 
 .■a 
 
 boo 
 
u 
 
 V 
 
 IH 
 
 01 
 CO 
 u 
 
 •■g 
 
 9) 
 
 o 
 
 I 
 
 pj 
 a> 
 
 (U 
 
 .£3 
 
 a 
 
 I 
 
 u 
 
 g5 
 
 ci « 
 
 i CI 
 
 ti 
 
 1 1" 
 
 a, a 
 
 «o 
 
 "Si 
 
 2 o i-i ^ 
 
 63 0% 9S 
 
 93 o» ss 
 
 £S o? 03 
 
 OS o» i* 
 
 Z^ oj ^^ 
 
 WOIJI^ 
 
 «■ o» 86 
 
 8E o» S£ 
 
 38 o* ZE 
 
 ZS o» 6Z 
 
 6B 0^ gz 
 
 9Z o? EZ 
 
 £3 o» OB 
 
 OZ o? il 
 
 II o% n 
 
 M ■* (N CO 
 
 CO CO M iH 
 
 (N 1-1 « 
 
 M Tjt lO O eo 
 
 CO ■'^ CO W Tj< 
 
 rH CO t* OS CO iH « 
 
 fH i-t CO CO CD CD U3 
 
 ^ 00 M O N M 
 
 oosoo^■cD"5^we^^H 
 
 IH 
 
 0)MC*CDIDtJ<CONiHO 
 
 3[io^ BuiKOAaJd: JO} 
 
 353 
 
 
 II g 
 
 a a 
 o o 
 
 •E-a 
 
 •c-s 
 oo 
 
 a 
 
 03 OS 
 
 aa 
 ■^■^ 
 
 OO 
 
 iH*» 
 rH(N 
 
 II II 
 
 &a 
 
 ?l 
 
 O oj 
 
 
 •■ ^ 
 
 03 V 
 
 
 lag; 
 '§ ■■ 
 
 03 -tJ 
 
 '3,2 
 
 is 
 
 O O 
 
 OO 
 
09 
 
 I 
 
 u 
 O 
 
 o 
 t» 
 
 o 
 
 t 
 
 d 
 a> 
 
 a 
 
 
 
 •a 
 
 SOT 
 ^ o 
 
 bo 
 S 
 
 « 
 
 bO 
 <« 
 
 d 
 
 § 
 
 1) 
 
 s 
 
 o 
 U 
 
 S 
 
 •s ^ 
 
 
 II 
 
 32.3 
 
 29.2 
 
 31.5 
 24.9 
 22.8 
 
 s 
 s 
 
 
 
 .2 
 
 
 ^ 
 
 S 
 
 s 
 
 OrtOOOO(NOrtrt(N00rtlNWlrt-*m^»MgO(N 
 
 I 
 
 4 
 
 89 0* S9 
 
 
 
 
 en 
 
 S9 o» 29 
 
 iH 
 
 iH 
 
 B9 o» 6ff 
 
 - 
 
 " 
 
 6B o» 95 
 
 , - 
 
 »-l 
 
 9ff0ȣs 
 
 '•* 
 
 
 
 8S 0* OS 
 
 tHiH 1-1 IN 
 
 10 
 
 OS o» Z* 
 
 ^« 
 
 CD 
 
 
 i^ o» f * 
 
 i-Hi-l i-t 
 
 CO 
 
 W 0* If 
 
 -- ^ 
 
 CO 
 
 I* o» 8E 
 
 IN rHrH 
 
 •* 
 
 88 o» S8 . 
 
 ^ W TlICO 
 
 OS 
 
 
 B8 °% ZS 
 
 _co^«- 
 
 CO 
 
 l-< 
 
 Z8 0* 6Z 
 
 -if WNfflCOtH 
 
 
 01 
 
 
 6Z o» 92 
 
 «rt eo«OT(i 
 
 ^ 
 
 92 0* EZ 
 
 y-< rtf-H ^,-1 i-H tH « 10 i-H CO 1-1 
 
 3 
 
 10 
 
 EZ <A OZ 
 
 rH rt eq rtrt CDTHOTtl 
 
 S; 
 
 OZ o» il 
 
 ^ rtCOOWlN^ 
 
 S 
 
 
 ZI o» H 
 
 t- 
 
 t- 
 
 
 
 = ::: = = = = = ::: -::::.: = ::: 
 
 H 
 
 a. 
 C 
 c 
 a 
 
 ? 
 
 Is 
 
 ■a^aHs, amnoAaij joj sSBj^i jCi3[a3^ aSBisAy 
 
 i" 
 
 354 
 
OCCUPATIONAL HISTORY AND EFFICIENCY 355 
 
 Of special importance for estimating the relationship between 
 economic eii&ciency and criminal record is the correlation between 
 average weekly wage for prevailing work and number of previous con- 
 victions. Table 135 gives the correlation between number of previous 
 convictions and average weekly wage for women employed in domestic 
 service and allied occupations. The coefficient of correlation here is 
 .046±.075. Obviously this is insufficient to indicate the presence of 
 any relationship. Because of the probability of a non-linear relation- 
 ship the correlation ratios were computed. Both ratios are sufficiently 
 small in relation to their standard deviations so that they are of little 
 importance as further evidences of relationship. We have gone a step 
 further and attempted to discover whether any significant relationship 
 would appear if allowance were made for the age factor. Computing 
 the partial correlation we find: r (average weekly wage with number 
 of previous convictions, age constant) ^.047 it .075." The upshot of 
 our analysis of these data is, therefore, the indication that there is no 
 relation between average wage and number of convictions if we con- 
 sider those women whose prevailing work has been of the nature of 
 domestic service. 
 
 Table 136 figures the relationship between weekly wage and num- 
 ber of previous convictions for women in occupations other than 
 housework. The correlation coefficient of — .067±.067 is insufficient 
 to indicate a relationship of any significance. Moreover, as in the case 
 of the women employed at domestic service, the correlation ratios need 
 not be taken into consideration because of the small values of these 
 ratios and the lack of regular trend in the lines of means. Computing 
 the partial correlation coefficient, with age constant, we get the follow- 
 ing result: r (average weekly wage with number of previous convic- 
 tions, age constant) = — .169±.065.i* From this we might conclude 
 that, if allowance be made for age, there is evidence of a very slight 
 negative correlation between average wage, at occupations other than 
 domestic work, and number of convictions. This does not show 
 whether this slight relationship is due to the fact that women with 
 
 " The data used in computing this coefficient are as follows : 
 r (wages with convictions) = .046 
 r (wage with age) = .000 
 
 r (convictions with age) =.17S 
 
 "The data used in computing this coefficient are as follows: 
 r (wage with convictions) = — -067 
 r (wage with age) = .264 
 
 r (convictions with age) = .331 
 
«. £ 3 e. 
 
 d ^ 
 
 l| 
 
 9>H 
 
 Means 
 (Number 
 of Previous 
 Convic- 
 tions) 
 
 to 
 
 
 1-H 
 
 03 
 
 f-H 
 
 t-H 
 
 
 
 1 
 
 'o "o' 
 
 r-l 
 
 's ^' 
 
 i^ 
 
 
 iH 
 
 CO 
 
 i-I 
 
 f— 1 
 
 1 
 
 1 
 
 00 
 
 ^ 
 
 i-H 
 
 
 t* 
 
 
 o 
 
 CO 
 r-i 
 
 
 o 
 
 rH 
 
 
 o 
 
 3 
 
 tH 
 
 r-t 
 
 CO 
 
 1-1 
 
 
 o 
 
 i-H 
 
 i-H 
 
 ,H 
 
 - 
 
 T-H 
 
 tH 
 
 o 
 
 1-H 
 
 T-1 
 
 OS 
 
 1— t 
 
 1-H 
 
 CO 
 
 (M 
 
 N 
 
 t* 
 
 IN 
 
 (M 
 
 «5 
 
 ^ IM T-l 
 
 -^ 
 
 IS) 
 
 OS 
 
 si 
 
 "3 
 
 1-t iH ^ 
 
 T— 1 
 
 ■* 
 
 Tft 
 
 ^ rH i-H i-t i-H 
 
 lO 
 
 W 
 
 r-( (N lO w 
 
 1—1 
 
 o 
 
 1— I 
 
 i 
 
 (N 
 
 CO CO 00 00 
 
 (N 
 
 
 CO 
 CO 
 
 - 
 
 ,-1 ■* Cd to O 50 
 
 (N 
 
 
 ■* 
 
 i 
 
 o 
 
 ■* 1> O 00 I-H O lO 
 tH T-H r-l C<l 
 
 to 
 
 g 
 
 05 
 
 
 o05odi>50"5TtIeo(N 
 
 g001>tOiO-*CO<Nrt 
 
 T-H 
 
 o 
 
 1 
 
 1^1 
 
 3iTO;a aan!BA9a J JOJ aaBjSi A-pt99j)i aaojOAv 
 
 
 
 o 
 
 o 00 
 
 iH *^ 
 
 I 
 
 
 o> 
 to 
 
 ■■^ 
 INti 
 
 ■* § 
 
 -£? 
 
 g 
 
 Til 
 
 3 
 
 
 a 
 
 ^ 
 
 n 
 
 Q> 
 
 
 
 1 
 
 ,i4 
 
 d 
 
 ^ 
 
 too fliogoo 
 
 ' pa 
 
 fl o° 
 o«H 
 
 3 g 
 Pi's -3' 
 
 «*H ^ * 
 
 O ^"g 
 
 ^ o c 
 ^ abs. 
 
 .2 
 
 IIS"! 
 
 O d P M 
 
 ■s s 
 
 a > 
 I?; <! 
 
 356 
 
I 
 
 ■a 
 
 > 
 
 Pi 
 u 
 •S, « 
 
 « 
 
 « m 
 «s.2 
 
 0'.{3 
 •a cS 
 u A 
 
 13 
 
 5 " 
 
 50 
 
 +j 
 
 n «i 
 g-d 
 
 (U ff 
 A" "3. 
 
 ^1 
 oM 
 
 i-i 
 
 ■i 
 
 a 
 d 
 
 I 
 I 
 
 t! 
 
 o 
 
 <n fc e O 
 
 %^6 
 
 S 
 o 
 
 •a 
 
 O 
 
 H 
 
 00 « T|f Tjt 
 
 .-H CO tH CO CO T-t 
 
 CO 1-ii-ieO Nt^COOJOlN 
 
 OSt^incO.-Iost*»OCOTHOH«-iOCOiHOJt-'OCO'-<Oit-"3eO 
 ^Ttt-*'*-*COCOCQCOCO(M(N(MCMINrtTH,Hrt.-H 
 
 r*wcoi-<aii>»ocoi-Host-u3eoi-(Ost^»ncO'-*Oifr*"5M'- 
 
 "*Tj(-<*<'^C0COC0C0C0IN(NiNMW'-ii-ii-ii-Hr-i 
 
 357 
 
 
 gs 
 
 -<!'*' 
 
 •2-§ 
 
 §§ 
 35 
 
 00 
 
 ■«■« 
 
 <ooo 
 
 C B- 
 
 ;o 
 
 ■tfs 
 
 
 
 
 ^■43 
 
 II 
 
 
 b 
 
 3H 
 
 X 
 
 
 
 
 II 
 
 £ 3 
 
 
 
 "11 
 
 T.S. 
 
 
 i''s 
 
 
 
 
 
 r^S 
 
 ■^^ 
 
 
 
 S^ 
 
 ^^ 
 
 Sfe 
 
 go 
 
 ■^1 
 
 ll 
 
 °l 
 
 
 S'g 
 
 g^ 
 
 
 w 
 
 §gg 
 
 as 
 
 .2 ^ 
 
 H"- 
 
 
 V 
 
 
 s:^ 
 
 s 50 1 
 
 
 'aS'i 
 
 fc 
 
 ^^r 
 
 ^ Q> 
 
 ^ « 
 
 •1 
 
 ^7-^ 
 
 Si 
 
 ■3,3 
 
 O O 
 00 
 
358 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 better earning capacity get into trouble slightly less often because 
 of their better financial status, or to the fact that a woman who gets 
 into trouble frequently thereby interrupts her working career and 
 lowers her rate of earning. 
 
 (c) Brief Comparison of Average Wage with Wage of Certain Groups 
 in General Population 
 
 The number of our women in any one occupation is too small for 
 significant comparison with the wage studies in given occupations 
 which have been made within recent years. If we compare the totals 
 of our domestic service and occupations other than domestic service, 
 the basis of comparison is changed, because we can not get wage 
 statistics in New York State for each of the occupations included in 
 this total. In spite of the discirepancies in the basis of comparison, we 
 shall present a few of the wage figures for certain occupations in New 
 York State and neighboring states, to see if the trend of wages among 
 our delinquents differs from them to any marked degree. Unfortu- 
 nately, there are no reliable figures on wages for women in domestic 
 service so that a comparison of the wages of this group will not be 
 possible. Domestic service is still an occupation where the individual 
 barters with the individual for the sale and purchase of labor. In cer- 
 tain sections of urban communities there is what approaches a stand- 
 ard wage, but these standards may vary radically within the different 
 parts of the same city. 
 
 In making any comparisons, the average wage for prevailing work 
 of the delinquent women will be used, since this wage is probably the 
 least affected by the time element which must be considered in dealing 
 with individuals of varying ages. The wage at time of present con- 
 viction, of course, applies to work done within the years 1915-1916, 
 but Since so many of the women (over 50 per cent) were not 
 working, it seems more advisable to use a wage which will include 
 a larger number of women. We are citing various isolated studies for 
 specified occupations in order to estimate roughly 'how the earning 
 capacity of our women stands in comparison with that of any groups 
 of the general population. (See table, p. 359.) 
 
 This crude method of estimating comparative wages indicates that 
 our group of women has a smaller percentage earning under $8 a 
 week in their prevailing work than the box-makers, those in candy 
 factories and in small neighborhood or department stores. Those in 
 the large department stores in New York and those in the dress and 
 
OCCUPATIONAL HISTORY AND EFFICIENCY 359 
 
 Percentage Earning Weekly Wages: 
 
 T, ,. , „, Under $4 Under $6 Under $8 Under $10 
 
 Delinquent Women 
 
 Average wage for prevailing 
 
 work (not domestic service) . . S.O 29.1 59.4 76.8 
 
 Box Industry, N. Y. State, 1914"° 8.8 38^1 642 854 
 
 Confectionery Industry, N. Y. 
 
 Stores^"'*^' ^^""^ ^^'^ ^'^■^ ^°-^ ^-^ 
 
 Large department store 53.0 
 
 Small neighborhood store 68.0 '.'.'.'. 
 
 5 and 10 cent store 99.0 .'..'. 
 
 Dress and Waist Industry, N. Y. 
 City, 1913™ 
 
 Week-workers 0.2 5.3 21.1 39.3 
 
 Clothing Factories, New Jersey, 
 
 1913-1914 "' 4.2 23.6 55.9 91.7 
 
 Clothing Factories, Massachusetts, 
 1913 ""• 
 
 Men's Clothing 1.9 13.5 42.8 74.0 
 
 Women's Clothing 1.1 13.6 39.6 66.8 
 
 waist industry in New York have a smaller percentage earning under 
 $8 a week than has our group of delinquent women. The two indus- 
 tries cited in neighboring states. New Jersey and Massachusetts, show 
 that there is a smaller percentage than in our delinquent group, earning 
 the small wage of less than $8 or less than $10 a week in the clothing 
 industry. From this we may see that the wages in the different in- 
 dustries noted vary considerably, but that aside from the workers in 
 the dress and waist industry, our group of women shows no appreci- 
 ably higher percentage earning under $8 a week, and in several instances 
 the percentage of delinquent women earning over $8 a week is larger 
 than the percentage of workers earning this amount in the occupations 
 cited. In the same report from which the above figures were taken ^^ 
 the statement is made: "There is ample ground for the conclusion 
 that from two-thirds to three-fourths of women workers in factories, 
 stores, laundries, and in industrial wage-earning occupations generally, 
 work at wages under $8 a week. Since practically all findings of 
 
 "Most of the above figures were taken from "Conditions of Labor in Amer- 
 ican Industries" by W. Jett Lauck and Edgar Sydenstricker. Funk and Wag- 
 nails Co., N. Y. C, 1917, pp. 53-57. The reports cited are: 
 
 a. Fourth Report of the New York Factory Investigating Commission, 
 
 Feb. 15, 1915. Vol. 2. Albany, J. B. Lyon Co. 
 
 b. U. S. Bureau of Labor Statistics : Wages, etc., in the dress and waist 
 
 industry in New York City, 1913 (Women). 
 
 c. New Jersey, Bureau of Statistics : Annual report on the statistics of 
 
 manufactures 'for the year ending Oct. 31, 1913 (16 years and 
 over). 
 
 d. Massachusetts, Bureau of Statistics : Annual report on the statistics 
 
 of manufactures for the year 1913 (18 years and over). 
 
360 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 minimum wage commissions and boards in the United States and the 
 estimates of investigators agree that the independent woman worker 
 can not live decently and without detriment to her health on less than 
 $8 a week, this fact has been given a great deal of emphasis during 
 recent years."^" 
 
 We might also compare with the wages of our group the wages of 
 workers in silk mills and in department stores in 1908, These figures 
 are given in a study of these industries in the report on the Condition 
 of Woman and Child Wage-earners in the United States.^^ The first 
 study gives in detail the percentage of women over fifteen earning 
 specified wages in the silk mills in New Jersey and Pennsylvania. The 
 second study gives the percentages of women employees earning speci- 
 fied wages in department and retail stores in New York City, Chicago 
 and Philadelphia. The total workers in stores include cash girls, 
 messengers, inspectors, bundle wrappers, packers, saleswomen, office 
 employees and buyers. 
 
 Percentage Earning Weekly Wages: 
 Under $8 Under $10 
 Delinquent Women (not in domestic service) 59.4 76.8 
 
 Workers in silk mills in New Jersey 5S.4 69.8 
 
 Workers in silk mills in Pennsylvania 87.S 96.4 
 
 Employees in department and retail stores in New 
 
 York City, Chicago and Philadelphia 57.5 76.1 
 
 From these investigations made several years ago it seems evident 
 that, except for the workers in silk mills in Pennsylvania who were 
 earning very low wages, the delinquent group is like the workers in 
 silk mills in New Jersey and the employees in stores in the percent- 
 age of women earning under $8 a week, though there are slightly 
 more of the delinquents than of the other two groups earning the lower 
 wages. The delinquents have a somewhat larger percentage than the 
 workers in silk mills earning under $10 a week, and only a very 
 slightly larger percentage than the workers in department stores. The 
 report under consideration gives the average weekly wage of the total 
 group of workers in department stores as $7.93. In the delinquent 
 group the average weekly wage for prevailing work is $8.90 with the 
 large standard deviation of $6.60. The longer range of wages of the 
 delinquent group would indicate that its standard deviation is larger 
 than that for the group of women in stores. The average wage for 
 
 " Op. cit., p. 43. 
 
 " U. S. Department of Labor, Bureau of Labor Statistics : "Surhmary of the 
 Report on Condition of Woman and Child Wage-earners in the United States." 
 pp. 196 and 216, 
 
OCCUPATIONAL HISTORY AND EFFICIENCY 361 
 
 the saleswomen alone is $8.84 a week, but these wages also have a 
 short range, so that the standard deviation is probably small. From 
 this scattering material we have gathered on wage, it would seem that 
 our delinquent group has earned an average wage as high as and in a 
 few cases higher than the weekly wage earned by women in specified 
 occupations in New York State and certain neighboring states. The fol- 
 lowing discussion will show the estimate of regularity of work of the 
 delinquent women, and will indicate that a very high percentage have 
 worked less than one half of the time, so that their actual average 
 income has been much less than a statement of wage would indicate. 
 
 Before proceeding to discuss the regularity of work, it may be of 
 interest to note Miss Conyngton's findings regarding low wages as a 
 cause of delinquency. She writes : ^^ "Not one worker assigned 
 poverty or lov/ wages as a direct and immediate cause of immorality. 
 It was agreed that indirectly their influence is great, but in the whole 
 inquiry only five cases were found in which the workers reporting 
 them believed that the women bad been driven into wrong-doing by 
 want." 
 
 ESTIMATE OF REGULARITY OF WORK 
 
 We have attempted to "size up" the work records of the women 
 whom we have studied intensively, both by an estimate of the regu- 
 larity of employment and by a total estimate of the work record. Un- 
 fortunately, it is not possible to present similar figures for either of 
 these for the general population. We have tried to get figures from 
 employment agencies which would have a record of individual appli- 
 cants for a period of years so that we might determine how much of 
 the time the rank and file who applied for work were employed. These 
 figures could not be obtained, however, and so we present our data 
 as merely descriptive of our total group. In considering the regularity 
 of work we have aimed to show how much of the time in a woman's 
 work history she was actually working when she had no other legiti- 
 mate means of support. We have grouped the women who have done 
 any work into five classes showing the amount of time working. In 
 order to determine this, we have observed the range of the expected 
 work period, — that is, from the date of the first job to present con- 
 viction, or to the time to which she would be expected to work, and 
 have based the regularity on the length of time working during this 
 
 "Op. cit.. p. 82. 
 
362 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 period. Neither the time when married and keeping house, nor the 
 time in institutions has been counted as idle. The five classes are as 
 follows : 
 
 1. Those who have worked practically none, but may have held one or 
 two jobs for short times. 
 
 a. A typical case in this class is the colored girl in Bedford, 23 
 years old, who began work at 16 years as nursegirl for a 
 family on Long Island. She worked for 2 months in 1909 
 and then ran away from her employer. Since that time she 
 has prostituted practically all of the time and has done no 
 legitimate work, 
 b. Another girl in the Workhouse, 20 years of age, began to 
 work at 16 and held 3 short jobs in quick succession,— 2 weeks 
 in a factory for manufacturing pennants, 1 week as attendant 
 in an orphan asylum, and a few days coloring films. Her en- 
 tire work period covered less than a month while the expected 
 work period was 4 years. 
 
 2. Those who have worked about one-quarter of the time. 
 
 a. Here we might cite the girl of 17 years who started work at 
 16 years and during the year of her expected work period 
 held .S jobs as follows: 
 
 1. Seamstress in waist factory, 3 weeks. 
 
 2. Removing foil from scrap candy, 1 month. 
 
 3. Spinner in yarn factory, 7 days. 
 
 4. Housework, 2 weeks. 
 
 5. Seamstress in factory, 1 week. 
 That is, she worked about 11 weeks or approximately one- 
 fourth of the time she was supposed to be working. 
 
 3. Those who have worked about one-half of the time. 
 
 4. Those who have worked about three-quarters of the time. 
 
 5. Those who have worked practically all of the time. The number 
 of women in this class is very few and nearly all are those 
 who have had excellent work records. The individual records will 
 not be enumerated here, since in nearly every case they are a repe- 
 tition of one job after another with very short or no intervals be- 
 tween the jobs. 
 
 If we turn to Table 137, we may see how the institutional groups 
 vary in the estimate of regularity of work. It is obvious, first of all, 
 that the percentages are very much alike in the various institutional 
 groups. Probation, it is true, has a smaller percentage who did very 
 little work, and it also has a larger percentage in the two upper classes. 
 
OCCUPATIONAL HISTORY AND EFFICIENCY 363 
 
 The Workhouse has the smallest percentage of women in the two upper 
 classes, and this serves to justify the impression one gets by reading 
 over the work records of the Workhouse women. Their own state- 
 ments of number of jobs held and length of time in them is so indefi- 
 nite, particularly with the older women, that many of the jobs were 
 verified only with great difficulty and were then found to have been 
 held for only a few days. For the women in .the Workhouse, which 
 has the largest number of recidivists, the work history is a story of 
 many repetitions,— coming out of the Workhouse, finding a position for 
 a few days, committing an offense against the law again, being recom- 
 mitted to the Workhouse, and so on. For the total group of women 
 there are in the first three classes, 60.9 per cent who worked approxi- 
 mately less than half the time, and who could not have earned enough 
 in the time working to tide them over the rest of the time. Only about 
 3.5 per cent were surely able to support themselves, while 35.6 per cent 
 by working about three-quarters of the time may have had sufficient 
 money by careful saving to tide them over the time of idleness.^^ 
 
 TABLE 137 
 
 REGULARITY OF EMPLOYMENT 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups 
 
 
 iNSTITtJTIONAL GROUPS 
 
 
 Regularity 
 
 Bed- 
 ford 
 
 Auburn 
 
 Mag- 
 dalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Proba- 
 tion 
 
 Total 
 
 Worked almost none 
 
 Worked about one - fourth of 
 time . . 
 
 11.2 
 
 23.5 
 26.5 
 
 36.7 
 2.0 
 
 10.1 
 
 17.4 
 29.0 
 
 42.0 
 1.5 
 
 6.9 
 
 27.8 
 31.9 
 
 30.6 
 2.8 
 
 10.0 
 
 22.2 
 27.8 
 
 32.2 
 
 7.8 
 
 8.8 
 
 35.0 
 25.0 
 
 30.0 
 1.3 
 
 3.9 
 
 16.7 
 31.2 
 
 42.9 
 5.2 
 
 8.6 
 23.9 
 
 Worked about one-half of time . 
 
 Worked about three-quarters 
 of time . 
 
 28.4 
 35.6 
 
 Worked nearly all of time 
 
 3.5 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 
 
 Number of cases 
 
 98 
 
 69 
 
 72 
 
 90 
 
 80 
 
 77 
 
 486 
 
 ^ It should be remembered that the time in idleness does not include the time 
 keeping house for one's family or parents, or the time in institutions, so that 
 idleness without money means no legitimate means of support. 
 
364 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 It may be of interest here to add Table 138 showing the distribu- 
 tion of estimate of regularity of work among the nativity groups. 
 From this table it seems that the foreign born have worked much 
 more of the time than either the native white or native colored. That 
 is, 44.8 per cent of the foreign born were employed less than half of 
 the time while 66.5 per cent of the native white and 70.5 per cent of 
 the native colored fall in this group. The native colored have the poor- 
 est showing in the regularity of work. 
 
 TABLE 138 
 
 REGULARITY OF EMPLOYMENT 
 
 Per Cent Distribution of Delinquent Women Classified 
 by Nativity and Color 
 
 
 Nativity and Color 
 
 
 Regularity 
 
 Total 
 
 Foreign 
 
 Born 
 
 Total 
 Native 
 Born 
 
 Native 
 White 
 
 Native 
 Colored 
 
 Total 
 
 
 4.2 
 15.4 
 25.2 
 
 48.3 
 7.0 
 
 10.5 
 27.2 
 29.8 
 
 30.4 
 2.1 
 
 10.6 
 25.2 
 
 30.7 
 
 31.5 
 2.0 
 
 10.2 
 33.0 
 27.3 
 
 27.3 
 2.3 
 
 8.7 
 
 Worked about one-fourth of time . 
 Worked about one-half of time. . . 
 Worked about three-quarters of 
 
 23.7 
 28.6 
 
 35.5 
 
 Worked nearly all of time 
 
 3.5 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 
 
 Number of cases 
 
 143 
 
 342 
 
 254 
 
 88 
 
 485 
 
 
 
 TOTAL ESTIMATE OF WORK RECORD 
 
 The total estimate of the efficiency of the work history, consider- 
 ing all phases of the work record, is based on the following factors: 
 
 1. Prevailing wage. 
 
 2. Regularity of vyork. 
 
 3. Ability as reported by employers, etc. 
 
 4. Increase or decrease in wage, regularity of work and ability as 
 shown by a survey of the entire work record. 
 
OCCUPATIONAL HISTORY AND EFFICIENCY 365 
 
 The estimate we have used is divided into five classes which have 
 been made up on the following basis. The points suggested under 
 each class are not arbitrary and there are innumerable combinations of 
 good and poor qualities in certain individual work histories which 
 might arise to place the estimate in one or another class. The follow- 
 ing scheme was used only as a means of showing the trend of the 
 most of the records.^* 
 
 1. Very poor. May include cases with 
 
 a. Very low prevailing wage (not enough to live on indepen- 
 dently). 
 
 b. Very irregular work record (class' 1 or 2). 
 
 c. Ability very poor as reported by employers. 
 
 d. No increase to higher wages, etc., and cases with actual de- 
 Crease. 
 
 2. Poor. 
 
 a. Low prevailing wage. 
 
 b. Irregular work record (class 2 or 3). 
 
 c. Ability poor. 
 
 d. Very little or no increase or development in work record. 
 
 3. Mediocre. Includes women who are probably self-supporting, but 
 barely so. 
 
 a. Wage large enough to live on but with no possibility of saving. 
 
 b. Fairly regular work record (about class 4). Includes cases 
 of seasonal unemployment where women are out of work 
 through no fault of *-heir own. 
 
 c. Mediocre ability. Able to keep job requiring no skill. 
 
 d. Possibly slight increase in wage. 
 
 4. Good. Women who are thoroughly self-supporting and good workers. 
 
 a. Good prevailing wage. 
 
 b. Regular work (class 5 or occasionally 4). 
 
 c. Well spoken of by employers. 
 
 d. Record shows fairly consistent development in wage, regularity 
 and ability. 
 
 5. Very good. Exceptionally good work record. 
 
 a. Consistently high wage. 
 
 b. Regular work (class S or occasionally 4). 
 
 c. Very well recommended by employer. 
 
 d. Consistent development. 
 
 "These estimates were made separately on each institutional group by two 
 persons and checked by a third person so that the method of treating each 
 case was uniform. 
 
366 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 We shall cite a few examples of work records which have been esti- 
 mated to belong in one or another of these five classes and which are 
 typical of many cases in each of the institutional groups. 
 
 Class I. A girl 16 years, 2 months old at time of conviction for 
 vagrancy, (prostitution), is very low grade mentally and comes from 
 most abominable home surroundings. She began to work when 14, 
 has an expected work period of 2 years and worked 7 months. Her 
 jobs are as follows : 
 
 a. General housework for Mrs. M., M , New York, 2 months, 
 
 1915, earning $6 a week. (Her own statement which could 
 not be verified and which is probably untrue.) 
 
 b. Helped with housework, Mrs. M , S , New York, 1 
 
 week, 1915, earning $1 a week. Was discharged because she 
 was so unsatisfactory. Could not learn how to do anything, 
 was very careless and dirty. 
 
 c. General housework, Mrs. S , M , New York, 2 months, 
 
 1916, earning $1 a week. Says she left because there were 
 too many boarders. Employer does not remember her. 
 
 d. Winder in silk mill, M , New York, 3 days, Dec. 20-23, 
 
 1916, at rate of $4.50 a week. Was very unsatisfactory. 
 
 e. Piece worker on fireworks, 19 days. May 29— June 16, 1917, 
 at $6 a week. Was very unsatisfactory, flirted with men and 
 could not do the work. Was arrested while working here. 
 This girl earned not more than $75 during two years. She worked 
 about one-fourth of the time, her employers report her as unable 
 to do the work, and her record shows no consistent increase in 
 wage, or ability to do work. 
 
 Class 2. A girl, 16 years, 9 months of age, was com.mitted to Bed- 
 ford for vagrancy (prostitution). She is also of low grade men- 
 tality. She started work at 15 and has an expected work period 
 of 21 months, during which she worked aboqt 16 to 17 months. 
 Her work was as follows: 
 
 a. General housework, 4 months, 1915, earning $3 a week. Said 
 she left because work was too far from home. (Could not 
 be verified.) 
 
 b. General housework, 4 weeks, 1916, at $2 a week. Said she 
 left because she could not eat Jewish food. (Could not be 
 .verified.) 
 
 c. General housework, 7 months, 1916, earning from $1.50 to 
 $2 a week. Was discharged for dishonesty and employer said 
 she grew so careless in her work she could not keep her. 
 
OCCUPATIONAL HISTORY AND EFFICIENCY 367 
 
 d. Helper in hotel dining room, 7 weeks in spring of 1916 and 
 4 weeks in the summer, earning $2.50 to $3 a week. Her 
 work was reported as unsatisfactory. She was absent-minded 
 and inefficient. 
 
 e. General housework, 2 weeks, 1916, earning $3 a week. Was 
 discharged as "unsatisfactory and very dirty." 
 
 f. Cigar-making in factory, in July, 1916 at $3 a week. Was 
 discharged as unsatisfactory. 
 
 g. Cleaning house, 4 days, November, 1916, at $.35 for half a 
 day's work. Was very dirty and unreliable and thoroughly 
 unsatisfactory. 
 
 This girl earned a low prevailing wage of about $2.33 weekly plus 
 maintenance, she worked about three-quarters of the time, but her 
 ability was nil and she was discharged from four places. Her work 
 shows no consistent development. Because of her fairly regular 
 work record she was placed in class 2, instead of class 1 where her 
 ability alone would place her. 
 
 Class J. A girl, 18 years, 5 months of age, was convicted of va- 
 grancy (prostitution). She is very bright, quick and learns easily. 
 She began work when 15, and in an expected work period of 38 
 months worked about half of that time. 
 
 a. Nursegirl, 2 weeks in September, 1913, at $2 a week. Was 
 discharged because she was unreliable and unsatisfactory. 
 
 b. Sorting, paper in paper mill 1 year, from May, 1914, to May, 
 1915, earning $5 a week. Employer says she was "equal to 
 average." 
 
 c. Banding cigars in factory, 3 months at $6 a week. Left be- 
 cause work made her ill. Employer says she was "O.K. but 
 not steady." 
 
 d. Feeder on forming machine in hat factory for 3 months, earn- 
 ing $10 a week. Was very good at first but finally had to be 
 discharged on account of her character. 
 
 e. Banding cigars for 1 week previous to arrest at $6 a week. 
 This girl averaged $5 a week in her prevailing work, she worked 
 about half of the time, had good ability if she would use it, and 
 showed possibilities of increase in earning capacity and length of 
 time in jobs. Her whole work record, though unsatisfactory be- 
 cause of the two discharges, might be called mediocre. 
 
 Class 4. A woman of 29 years, 10 months, was convicted of pos- 
 sessing opium. She started work at IS and worked for 11 years, 
 singing and dancing in vaudeville performances, Though there wa5 
 
368 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 only a slight increase in her wage up to $25 a week and traveling 
 expenses, she made a name for herself in many of the~ small places 
 near New York and was very successful as a dancer. She was 
 placed in class four, because she had a fairly high average wage, be- 
 cause she had ability, and because she worked steadily. up until the 
 last 4 years before her conviction when she acquired a drug habit. 
 
 Class 5. A woman, 25 years, 7 months of age, was convicted of vio- 
 lating the Tenement House Law. Sh'e began work at 14 and held 13 
 jobs, covering 9 years, during her work period of 11 years. The work 
 record is rather too long and too much the same to give all of the 
 details but the main trends are as follows: ten of the thirteen jobs 
 were in general housework, the other three being waitress, usher in 
 theater for two weeks, and laundry work in a laundry for six weeks. 
 The wage begins at $2 a week in the first jobs and runs progressively 
 through $4 a week, $5 a week, $6 a week and $7 a week and living, 
 which she had been earning in her two latest jobs. Several of the 
 records we were not able to verify because the employers had moved, 
 but those from whom we heard, speak of her as ''very satisfactory," 
 "excellent," "the best worker we ever had," and "A, No. 1." Though 
 this girl was out of work practically two years out of the 11, she was 
 placed in the highest class because of her steadily increasing wage, 
 and her unusually good ability which was spoken of by every em- 
 ployer. 
 
 (a) Distribution for Institutional Groups 
 
 If we apply this estimate to the institutional groups, we find irt 
 Table 139 that there is considerable variation between the groups 
 though there is the same trend running through each. As with the esti- 
 mate of regularity of work, the Probation group has the smallest per- 
 centage, only 6,5 per cent, in the very poor class, while the Workhouse 
 runs up to 17.5 per cent for the same class. It is also of interest to 
 note that the Workhouse and Bedford, those whose records were 
 known least and those which were known most thoroughly, have no 
 representative in the very good class. Only L7 per cent of the total 
 group have exceptionally good work records and only 12.3 per cent a 
 record that might be called good. Roughly, by the method of our esti- 
 mate, the women in classes 1 and 2 could not possibly be self-support- 
 ing if they lived on what they earned from their legitimate employ- 
 ment. Those in class 3 might, with great care and with no occasion 
 for large additional expenditures, get along on their earnings, while 
 those in the two upper classes ought to be able to live on their wages. 
 
OCCUPATIONAL HISTORY AND EFFICIENCY 369 
 
 By this standard, we find that for the total group of women, 49.7 per 
 cent, almost half of the women, have poor or very poor work records 
 and could not possibly be self-supporting on the amount they have 
 earned, 36.4 per cent of the total group have mediocre records and 
 could with great care have lived on their earnings, while 14.0 per cent 
 have good work records and could probably be called entirely self-sup- 
 porting. It is a striking commentary on the economic competence of 
 this group of delinquent women that as a group ,they have worked so 
 irregularly and that half of them at least have done so little and such 
 poor work that they could not have lived on their earnings. 
 
 TABLE 139 
 
 ESTIMATE OF WORK RECORD 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups 
 
 Estimate 
 
 Inbtitdtional Ghoups 
 
 
 of 
 Work Record 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Probation 
 
 Total 
 
 Very poor 
 
 Poor 
 
 14.3 
 
 50.0 
 
 31.6 
 
 4,1 
 
 13.2 
 33.8 
 36.8 
 11.8 
 4.4 
 
 11.8 
 
 39.7 
 
 36.8 
 
 8.8 
 
 2.9 
 
 12.2 
 32.2 
 27.8 
 26.7 
 1.1 
 
 17.5 
 36.3 
 36.0 
 11.3 
 
 6.6 
 27.3 
 53.2 
 10.4 
 
 2.6 
 
 12.7 
 37.0 
 36.4 
 12.3 
 1.7 
 
 Fair 
 
 Good 
 
 Very good 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cases 
 
 98 
 
 68 
 
 68 
 
 90 
 
 80 
 
 77 
 
 481 
 
 (b) Distribution for Nativity Groups 
 
 If we turn to Table 140, showing the distribution of women in 
 nativity groups by the estimate of the work record, we note first 
 of all the very small percentage of the foreign born who fall in the 
 poorest class, the larger percentage of the native white in this class 
 and the very much higher percentage, almost five times as many women, 
 among the native colored. The native colored have no representa- 
 tives in the exceptionally good work records, while the native white 
 have only .8 of a per cent and the foreign born 4.3 per cent. If 
 we use the larger groupings by combining classes 1 and 2, and 4 and 
 5, we find that the foreign born have 32.2 per cent in the class which 
 
370 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 could not possibly be self-supporting from the amount earned in legit- 
 imate work, while the native white have 44.8 per cent in this class and 
 the colored 62.5 per cent. In the mediocre class, it will be seen, the 
 foreign born have 43.6 per cent, the native white 34.5 per cent and the 
 native colored 30.7 per cent. In the upper class of those probably- 
 self-supporting, we find 25.0 per cent among the foreign born, 10.3 
 per cent among the native white and 6.8 per cent among the native 
 colored. From this it would seem that the foreign group are more 
 efficient ^nd regular workers, if we view their work records as a whole, 
 than either the native white or the native colored, and that the native 
 colored have the most unsatisfactory work records. In order to ex- 
 plain this, we might turn back to Table 138 which shows that the 
 foreign born have worked much more of the time than have the native 
 bom. Table 112 also shows us that the prevailing work of the foreign 
 born differs quite markedly from the prevailing work of either the 
 native white or native colored, with slight differences in certain of the 
 occupations but very striking differences in occupations such as do- 
 mestic service or factory work.^^ 
 
 (c) Relationship to Factors Within Work Record 
 In this connection we shall try to determine whether there is any 
 relationship between the nature of prevailing work and the estimate 
 of the work record. The relationship of these two factors is tabulated 
 in Table 141. From this it is seen that there is no striking relation- 
 ship. Perhaps the most noteworthy thing about the table is the fact 
 that both domestic service and factory work are represented by indi- 
 viduals of all degrees of efficiency as indicated by the total work his- 
 tory. The failure of the clerical workers to show any individuals 
 with very good or very poor records is striking, as is the fact that of 
 the three professional women of the group none show a record better 
 than fair. 
 
 ""In answer to the objection which may be raised that the foreign born 
 seem to have a better work record because we have been able to verify less of 
 their work record and they tend to exaggerate their ability in their own state- 
 ment of work, we give the following figures showing whether the data on which 
 the estimates are based are largely verified or largely unverified. This shows 
 that about the same percentage of records are verified among the foreign born 
 and the native colored, while the native white have a larger percentage verified. 
 
 Largely Verified Largely Unverified 
 
 Foreign Born 55.0 4S.0 
 
 Total Native Born 64.7 3S.3 •' 
 
 Native White (ff_g 32.1 " 
 
 Native Colored SS'7 443 
 
 Total Group gfg 3g3 ^" 
 
OCCUPATIONAL HISTORY AND EFFICIENCY 371 
 
 TABLE 140 
 ESTIMATE OF WORK RECORD 
 
 Per Cent Distribution of Delinquent Women Classified by 
 Nativity and Color 
 
 
 Natititt and Colok 
 
 
 Estimate of Work Eeoord 
 
 Total 
 Foreign Born 
 
 Total 
 Native Born 
 
 Native 
 White 
 
 Native 
 Colored 
 
 Total 
 
 Very poor 
 
 4.3 
 27.9 
 43.6 
 20.7 
 
 4.3 
 
 16.2 
 
 40.6 
 
 33.5 
 
 8.8 
 
 .6 
 
 14.3 
 
 40.5 
 
 34.5 
 
 9.5 
 
 .8 
 
 21.6 
 
 40.9 
 
 30.7 
 
 6.8 
 
 12.7 
 37.1 
 36.5 
 12,3 
 1 7 
 
 Fair 
 
 Good 
 
 Verv sood . 
 
 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cases 
 
 140 
 
 340 
 
 252 
 
 88 
 
 480 
 
 TABLE 141 
 Nature of Prevailing Work in Relation to Estimate of Work Record 
 
 Domestic service 
 
 Factory work 
 
 Home work 
 
 Restaurant and hotel work . 
 
 Work in stores 
 
 Vaudeville performers 
 
 Clerical work 
 
 Professional service 
 
 Personal service 
 
 Charwomen 
 
 Miscellaneous 
 
 Estimate of Wobk Record 
 
 Very 
 Poor 
 
 Poor 
 
 Total . 
 
 26 
 12 
 1 
 5 
 6 
 1 
 
 1 
 
 1 
 
 2 
 
 2 
 
 57 
 
 81 
 
 43 
 
 4 
 
 10 
 
 13 
 
 3 
 
 8 
 
 Fair 
 
 167 
 
 63 
 
 49 
 
 11 
 
 11 
 
 16 
 
 7 
 
 4 
 
 2 
 
 3 
 
 1 
 
 2 
 
 169 
 
 Good 
 
 21 
 
 14 
 
 3 
 
 4 
 
 3 
 
 Very 
 Good 
 
 Total 
 
 57 
 
 193 
 
 120 
 
 19 
 
 30 
 
 40 
 
 19 
 
 13 
 
 3 
 
 13 
 
 3 
 
 5 
 
 458 
 
372 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 It is of interest here to see if there is a relationship between the kind 
 of work one does when she first starts working and the estimate of 
 the work record which follows. Table 142 shows this relationship. As 
 in the preceding table there is little indication of any striking relation. 
 One starting in almost any occupation has a fair chance of making" 
 either a good or a poor work record, although the chances are over- 
 weighted toward the latter probability except in case of two of the 
 very small groups. 
 
 TABLE 142 
 
 Estimate of Work Record in Relation to Kind of Work in First Job 
 
 
 
 Estimate of Wokk Eecokd 
 
 
 
 Very 
 Poor 
 
 Poor 
 
 Fair 
 
 Good 
 
 Very 
 Good 
 
 Total 
 
 
 
 23 
 19 
 2 
 5 
 4 
 2 
 
 1 
 1 
 1 
 2 
 
 76 
 45 
 
 5 
 15 
 19 
 
 1 
 10 
 
 4 
 
 3 
 
 70 
 
 57 
 
 6 
 
 7 
 
 14 
 
 5 
 
 4 . 
 2 
 3 
 1 
 3 
 
 21 
 13 
 1 
 2 
 5 
 5 
 5 
 2 
 2 
 
 1 
 
 2 
 2 
 
 3 
 1 
 
 191 
 
 
 Factory work 
 
 136 
 
 
 
 14 
 
 •g 
 
 •-3 
 
 Restaurant and hotel work .... 
 
 Work in stores 
 
 Vaudeville performers 
 
 29 
 45 
 13 
 
 
 Clerical work 
 
 19 
 
 1 
 
 Professional service 
 
 Personal service 
 
 Charwomen 
 
 Miscellaneous 
 
 5 
 
 11 
 
 2 
 
 9 
 
 Total 
 
 60 
 
 177 
 
 172 
 
 57 
 
 8 
 
 474 
 
 (d) Relationship ta Other Social Factors 
 
 A reason often advanced by child labor propagandists for much 
 of the economic inefficiency is that those who start work earliest have 
 no opportunity for training, that they start work in an unskilled oc- 
 cupation, get into a rut, and are never able to progress. As they grow 
 older they become slower and are gradually pushed out to more and 
 more casual work, while younger persons fill their places. In order to 
 
OCCUPATIONAL HISTORY AND EFFICIENCY 373 
 
 see whether in our group of delinquent women there is any tendency 
 for those who start work earliest to have any less satisfactory work 
 records, the following correlation table is presented (Table 143). By 
 calculating the correlation ratio we find that 77 =.14 with a standard 
 deviation of .047. From this it would seem that there is a significant 
 but a small relationship between the age at starting work and the esti- 
 mate of the work record. The means of the age at starting work 
 show that for the three lowest classes of the estimate of the work 
 record, the average ages at starting work are about the same, though 
 there is a very slight increase in age from class 1 to class 3. The age at 
 starting work is appreciably higher, 20 years, in both classes 4 and S, 
 showing that those who have been most competent industrially have 
 started work at a later age than those who have had very poor work 
 records. As stated before, the relationship is low and would seem to 
 indicate that the factor of starting work early, though it has a slight 
 effect on the later efficiency of the work record, is not one of the im- 
 portant factors operating in the determination of a very good or very 
 poor record. 
 
 Another factor which is often considered as influencing the ef- 
 ficiency of a work record is the amount of schooling, and as a measitre 
 of this we shall use grade finished. Table 144 shows the relationship 
 between the grade finished and the estimate of the work record. The 
 correlation ratio of .13±.049 indicates that there is a genuine but a 
 small relationship. The means show only a small increase in the 
 changes in grade for the changes in efficiency of record, but the trend 
 of the average grade increases in regular order from the lowest to 
 the highest class of the estimate. The factor of amount of schooling, 
 therefore, like the factor of age at starting work, probably has had 
 only slight influence in determining the character of the work record in 
 the case of the women under consideration. 
 
 There is still another factor which may have a significant relation- 
 ship to the estimate of the work record, and that is the estimate of 
 conditions in the home during childhood and adolescence. A girl 
 brought up in the lowest kind of home surroundings may be made rela- 
 tively much more incompetent for good work later on by these condi- 
 tions than she would be fitted for good work by a long period of 
 schooling, for example. Table 145 presents the coefficient of contin- 
 gency of the estimate of the work record with the estimate of home 
 conditions, and shows by the coefficient of C=.24 that there is a small 
 relationship between the two factors, with a tendency for those with 
 
I 
 
 .o 
 
 d 
 o 
 
 I 
 
 "53 
 
 s 
 
 U 
 
 a 
 
 ^^•5 
 
 oouseo-H 
 
 g2 
 
 Mea 
 (Age 
 Start 
 Wor 
 
 OOI>l>b- 
 
 t~ 
 
 C<1(N rH i-H i-( 
 
 1-t 
 
 <n 
 
 
 
 "d 
 
 00 CD-* OS «: 
 
 CO 
 
 H 
 
 lOttiiou; 
 
 
 rH 1-H 
 
 
 
 Z9-8B 
 
 T-1 
 
 »-< 
 
 SSH^S 
 
 
 o 
 
 fS-OS 
 
 
 o 
 
 0B-9» 
 
 IN i-H 
 
 CO 
 
 9»-Zf 
 
 Nrt 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 tS 
 
 e*-8£ 
 
 ■-Hr-IM 
 
 lO 
 
 
 
 
 
 
 
 
 6= 
 
 8£-fE 
 
 T— 1 Y-4 »-H 1-H 
 
 ■* 
 
 
 
 
 
 
 
 
 
 
 
 
 
 § 
 
 f'S-OS 
 
 i-H CO CO CO i-< 
 
 r-1 
 
 
 
 
 
 m 
 
 
 
 
 S 
 
 0S-9Z 
 
 lOOOO 
 
 02 
 1-H 
 
 a 
 
 
 
 
 
 
 
 
 <! 
 
 
 
 
 
 9z-ez 
 
 ■*cooo 
 
 s 
 
 ze-8i 
 
 l>t-CO(N 
 
 Ol 
 
 
 
 (MIMrt 
 
 !0 
 
 8I-tI 
 
 »OO5i-t00C<l 
 
 lO 
 
 
 
 (N 00 00 CO 
 
 CO 
 (N 
 
 J'l-OI 
 
 .-H-^fN ■*01 
 
 o 
 
 
 
 (N(N 
 
 CO 
 
 01-9 
 
 tOIN 
 
 00 
 
 
 
 
 
 u 
 
 
 
 § 
 
 
 
 c 
 r 
 
 '^ 
 
 
 b£ 
 
 
 
 e 
 
 : 1 
 
 
 ^li i I 
 
 
 >OSft> 
 
 
 pjooaa 3|joji\ 
 
 
 
 
 }0 8» 
 
 BUI 
 
 HB 
 
 a 
 
 
 o 
 
 41 
 
 •a 
 o 
 
 1-H i-H -a 
 
 b d 
 o 
 
 •9 
 1? 
 
 I- 
 
 II 
 
 oa-3 
 -.^.^ 
 
 bOO 
 
 374 
 
H 
 
 h) 
 
 ^ 
 
 u 
 
 (U 
 
 Pi 
 
 ■a 
 
 o 
 
 o 
 
 4) 
 
 CO 
 
 "3 
 
 I 
 
 en 
 
 .a 
 
 10 
 
 ■a 
 
 K 
 
 •d 
 cd 
 u 
 
 a 
 
 & 
 
 -•J 
 
 V 
 
 § 
 
 o 
 U 
 
 I 
 
 ■a 
 
 
 Ml 
 
 
 o 
 
 US 
 
 
 
 (M 
 
 
 
 1 
 
 t- 
 
 ^ 
 
 1-t i-H 
 
 s? 
 
 i 
 
 i 
 
 1 
 § 
 
 ■a 
 a 
 
 
 iH 
 
 (N 
 
 C<) IM 
 
 
 i> 
 
 
 
 o 
 
 
 (M T-H 
 
 CO 
 
 ml 
 
 (M 
 
 
 CO IN 
 
 I— 1 
 
 00 
 
 00 
 
 
 a- 
 
 t> CO 
 
 CO 
 
 ^ 
 
 I> 
 
 
 CC 
 
 T— 1 fH 
 
 to 
 
 £; 
 
 <o 
 
 r-i 
 
 >* 
 
 ^ s 
 
 oc 
 
 S 
 
 to 
 
 rH 
 
 -* 
 
 2 ?§ 
 
 -* 
 
 g 
 
 -* 
 
 1-H 
 
 oc 
 
 2 ^ 
 
 IM 
 
 1— 1 
 
 o 
 
 CO 
 
 CO 
 
 
 IC 
 
 CO CD 
 ■-^ IM 
 
 "3 
 
 ? 
 
 (M 
 
 
 Tf 
 
 o CO 
 
 U2 
 
 IM 
 
 CO 
 
 r-t 
 
 iH 
 
 tH 
 
 (N IM 
 
 ■* 
 
 o 
 
 & 
 
 
 •c 
 
 IM i-H 
 
 ^ 
 
 ^ 
 
 
 
 i 
 
 ) 
 
 ^ 1 
 
 c 
 
 1 
 
 1 
 
 1 
 > 
 
 a 
 > 
 
 : 1 
 
 
 
 3(10 
 
 pio 
 
 
 usa 
 
 
 
 OJ 
 
 ■* 
 
 ^ 1 
 
 - I 
 
 II I 
 
 " -a 
 
 o 
 u 
 a; 
 
 -a 
 I 
 
 0) 
 
 to s 
 
 a 
 t 
 •a 
 
 1 
 
 "5 Q 
 - a; 
 
 b S 
 
 s ^ 
 
 CO ei « 
 ll| I 
 
 d h 
 
 o3r5 
 
 • • -*^ 
 
 .a fl 
 a. 2 
 
 E 
 
 03 
 
 go ^ 
 
 6o s 
 
 375 
 
376 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 good home conditions to have a better work record than those with poor 
 
 home conditions. 
 
 TABLE 146 
 
 Contingency of Estimate of Work Record with Estimate of 
 Home Conditions 
 
 
 
 Estimate of Wobk Recobd 
 
 
 
 Very Poor 
 
 Poor 
 
 Fair 
 
 Good 
 
 Very Good 
 
 
 •s:l 
 is 
 
 Very poor 
 
 Poor 
 
 Fair 
 
 Good 
 
 7 
 
 24 
 
 27 
 2 
 
 15 
 
 67 
 
 77 
 7 
 1 
 
 5 
 
 64 
 
 90 
 
 9 
 
 1 
 
 5 
 13 
 30 
 
 9 
 
 1 
 6 
 
 1 
 
 32 
 159 
 230 
 
 28 
 
 w 
 
 Very good. . . . . 
 \ 
 
 2 
 
 Total 
 
 60 
 
 167 
 
 159 
 
 57 
 
 8 
 
 451 
 
 Coefficient of contingency of estimate of work record with estimate of home 
 conditions, C == . 24 
 
 If we summarize the factors we have considered which might have 
 a' possible causal influence in determining the kind of work record, we 
 find that there are no very high relationships demonstrable but that 
 each factor has a slight and probably significant relationship in the de- 
 termination of the efficiency of the work record. These low relation- 
 ships will be noted again briefly. 
 
 1. There is a small relationship between the estimate of the work 
 record and the nature of prevailing work, indicating that there is a 
 tendency for those who do certain kinds of work to have a more 
 satisfactory work history than those doing other kinds of work. 
 
 2. There is a still smaller relationship between the estimate of the 
 work record and the kind of work in the first job indicating that there 
 is a slight tendency for those who first go into certain kinds of work 
 to have more efficient work records than those who start in with other 
 work. 
 
 3. There is only a small relationship between the age at starting 
 work and the estimate of the work record with a tendency for those 
 who start work earliest to do the least satisfactory work after that. 
 
 4. The element of grade finished in school has a small relationship 
 
OCCUPATIONAL HISTORY AND EFFICIENCY 377 
 
 to the estimate of the work record, and indicates that there is a slight 
 tendency for those who have finished more grades in school to do 
 better work than those who have had less schooling. 
 
 5. The relationship between estimate of work record and estimate 
 of home conditions is small and indicates that the factor of good or 
 poor home conditions, though they may effect to some extent the 
 efficiency of the work record, do not do so to any marked degree. 
 
 SUMMARY 
 
 From the data which have been presented in this chapter and which 
 have aimed to show the development of the occupational histories of 587 
 delinquent women, from the age at starting work to the estimate which 
 summarizes the most important factors of the entire work record, we 
 can not draw as definite conclusions as we should like because of the 
 difficulty of obtaining comparative data for the general female popula- 
 tion of New York State. We may note, however, the main factors we 
 have observed in the work records of this group of delinquent women. 
 
 First of all, we find that our group of women has 14.8 per cent 
 starting work before they were fourteen, while the general female 
 population has only 0.1 of a per cent working between ten and fourteen 
 years of age. This indicates a trend for these delinquent women to 
 go to work earlier than the general female population in New York 
 State. The age at starting work, however, allowing for the factor of 
 present age, seems to have no significant relationship to the age at 
 first conviction. 
 
 Data offered on the kind of work done show that the domestic serv- 
 ice workers have much more than their expected percentage, if com- 
 pared with the women over fifteen in New York State. The vaudeville 
 performers, also, have a larger percentage among the delinquents, while 
 the more skilled workers, such as those in clerical work and profes- 
 sional service, have a much smaller representation among the delin- 
 quent group than among the general population. The same trend has 
 been found in other studies, particularly that of Miss Mary Conyng- 
 ton, in the series of "Woman and Child Wage-earners in the United 
 States." " A correlation between grade finished and nature of work 
 shows that there is a significant relationship in the tendency of the 
 occupational groups to differ in respect to school attainment. The 
 domestic service workers have reached a lower average grade than the 
 women in any other occupations except charwomen and the miscel- 
 
 -" Op. cit., p. 30, 
 
378 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 laneous group, both of which have too small numbers to be significant. 
 
 One of the most striking facts in observing the nature of work is 
 to find how large a part of the women were idle at the time they com- 
 mitted the oflfenses which led to both their first and present convictions. 
 This tendency has also been noted by Miss Conyngton ^^ and by Dr. 
 Glueck in a study of male felons in Sing Sing.^* 
 
 The study of wage in our groups is difficult because the women are 
 of varying ages and their work records cover different periods of 
 time. Recognizing this unavoidable difficulty, we have, however, com- 
 puted the average wage for the prevailing work of those in occupa- 
 tions other than housework, and find by comparing this with certain 
 scattering wage studies in specified employments that the women in 
 our group probably earned no less than the women in the occupations 
 noted. From the figures given in the reports used no exact compari- 
 son is possible, but in general we may note that the average wage of 
 the delinquents is probably not so low as to be directly accountable 
 for their getting into difficulties with the law. 
 
 In a correlation between grade finished and average weekly wage 
 for prevailing work there seems to be a significant relationship, with 
 a tendency for both the domestic service workers and those in occu- 
 pations other than domestic service tp receive a higher average wage if 
 they have finished the higher grades. No significant difference was 
 demonstrable in a correlation between grade finished and first wage. 
 
 Referring to the correlation between average weekly wage and 
 number of convictions, we find no evidence of a significant relation- 
 ship in the case of women in domestic service, and evidence of only 
 a. slight degree of association, in the direction of negative correlation, 
 for women in other occupations. The striking relationship which 
 might be anticipated between proved earning capacity and degree of 
 recidivism is notably lacking. 
 
 In summarizing the regularity of work we have used an estimate 
 which shows that 60.9 per cent of the women have worked half of 
 the time or less, while only 3.5 per cent have worked nearly all of the 
 time. Upon applying this estimate to the color and nativity groups we 
 find that the foreign born have worked with much more regularity, 
 while the native colored have the poorest showing in this respect. 
 
 A summary of all of the factors in the work record is made in 
 the estimate of the work record, which indicates that on the whole the 
 
 =' Op. cit., pp. 46-47. 
 "" Op. cit., p. 138. 
 
OCCUPATIONAL HISTORY AND EFFICIENCY 379 
 
 work records are poor, only 14.0 per cent of the total group falling in 
 the class above mediocre, while 49.7 per cent fall in the two groups of 
 very poor and poor work records. Here, also, the foreign born show 
 better work records than either the native white or native colored. 
 
 In an attempt to find what factors in the personal histories of these 
 women have a significant relationship to the efficiency of the work 
 records, we have computed the correlation between estimate of work 
 record a^d age at starting work, grade finished in school and estimate 
 of early home conditions. In each of these correlations a very small 
 but probably significant relationship is demonstrable. The relationship 
 is so small, however, that it would be difficult to say any real causal 
 influence existed. 
 
 On the whole, the records show, for the most part, poor ability, 
 great irregularity, and a low wage, though probably no lower wage than 
 the bulk of unskilled workers in the same occupations outside are 
 earning. It would seem that the delinquents in this study are drawn 
 almost entirely from this large class of the unskilled workers, but that 
 this fact is not, as might be argued by the Socialist School, the funda- 
 mental cause of the delinquency. We have noted elsewhere the low 
 grade attained in school as compared with the majority of children 
 who have been in public school the same length of time, the low men- 
 tality as measured by intellectual tests, the poor home conditions and 
 lack of training. It would, therefore, be impossible for us to state 
 that the economic relations of these women are in any sense exclusively 
 the causative factors of delinquency. In many cases there is, un- 
 doubtedly, a close association between the limited, poorly paid, indus- 
 trial opportunities and the continuance of illegal acts. In many cases, 
 the economic possibilities for the father or mother *in the family have 
 been so poor that the women in this study did not have proper educa- 
 tional advantages as children and were forced into work with no 
 future and a small wage. For the most part, however, we believe that 
 the elements of occupational influences and low wages are not the 
 direct and immediate causes of the delinquency, though indirectly 
 their influence is great. 
 
CHAPTER XII 
 HISTORY OF SEX IRREGULARITIES 
 
 THE study of delinquent women in several institutions, some of 
 which take the petty offender and some the criminal who has com- 
 mitted the more serious offenses, shows that the history of these wom- 
 en with reference to sexual irregularity divides 'itself into three general 
 classes. First there is the professional prostitute who earns her entire 
 living by this method. Secondly, there is the woman who prostitutes 
 herself either regularly or irregularly, but who earns money in addition 
 to this by stealing or by other illegal methods. And, thirdly, there is 
 the woman who may have been sexually promiscuous, but not for 
 money; or the woman who may have lived with one or more consorts 
 for any length of time. In this discussion, we shall attempt to show 
 the status of degree of sexual irregularity among the total group of 
 women, and by institutional groups, giving whatever data we have 
 been able to gather about the first contacts with sex life. Since the 
 prostitutes constitute the most important part of the sex problem, they 
 will be separately considered and the main factors in their social his- 
 tories compared with similar factors in the lives of prostitutes studied 
 elsewhere. 
 
 FIRST SEX EXPERIENCE 
 
 (a) Age at First Sex Offense 
 
 The age at first sex oifense refers to the time that the first illicit 
 sexual intercourse occurred, whether the woman consented at that 
 time or was raped. The distribution by four year age groups given 
 in Table 146 is for the age at first sex offense of the women in in- 
 stitutional groups. The range of years runs from 6 to 51 years and 
 the age group from 14 to 18 years has the largest number of cases. 
 In the lowest age group Bedford has four cases, one who was raped at 
 six, and one at eight years, and two who consented to have sexual 
 intercourse when they were nine years old. The one Auburn case 
 which falls in this group was nine years of age when she was raped 
 by her cousin in whose home she was visiting. Only one case of the 
 
 38a 
 
HISTORY OF SEX IRREGULARITIES 
 
 381 
 
 Probation women falls in this lowest age group, and she was raped 
 at nine years. The higher age groups have representatives, of course, 
 only in the institutions which have women of the older age. 
 
 The comparison of the means and standard deviations for the age 
 at first sex offense shows that Bedford has the youngest mean age at 
 first sex offense with the smallest standard deviation and that, in order 
 of increasing average ages, follow the Magdalen, Probation, the Peni- 
 tentiary, Workhouse and Auburn. The range of average ages is from 
 16.2 years in Bedford to 21.4 years in Auburn. The average age at 
 first sex offense for the total group is 18.72±.312 years with a stand- 
 ard deviation of 5.82±.221 years. 
 
 TABLE 146 
 AGE AT FIRST SEX OFFENSE 
 
 Per Cent Distribution of Delinquent Women 
 
 in Institutional Groups 
 
 
 iNSTITnTIONAL GbOUPS 
 
 
 Age at First 
 
 
 
 
 
 
 
 
 Sex 0£fense 
 
 
 
 
 Peni- 
 
 Work- 
 
 
 
 
 Bedford 
 
 Auburn 
 
 ^fagdalen 
 
 tentiary 
 
 house 
 
 Probation 
 
 
 6 to 10 yet 
 
 irs 4.4 
 
 2.4 
 
 
 
 
 1.9 
 
 1.7 
 
 10 " 14 ' 
 
 13.0 
 
 2.4 
 
 5.1 
 
 4.6 
 
 7.8 
 
 
 6.3 
 
 14 " 18 ' 
 
 54.4 
 
 38.0 
 
 55.9 
 
 42.0 
 
 39.2 
 
 35.2 
 
 45.7 
 
 18 " 22 ' 
 
 22.8 
 
 19.1 
 
 27.1 
 
 24.0 
 
 25.5 
 
 37.0 
 
 25.9 
 
 22 " 26 ' 
 
 3.3 
 
 9.5 
 
 8.5 
 
 16.0 
 
 7.8 
 
 11.1 
 
 8.6 
 
 26 " 30 ' 
 
 1.1 
 
 9.5 
 
 1.7 
 
 6.0 
 
 3.9 
 
 13.0 
 
 5.2 
 
 30 " 34 ' 
 
 1.1 
 
 11.9 
 
 1.7 
 
 4.0 
 
 3.9 
 
 1.9 
 
 3.5 
 
 34 " 38 ' 
 
 
 2.4 
 
 
 
 5.9 
 
 
 1.2 
 
 38 " 42 ' 
 
 
 2.4 
 
 
 4.0 
 
 5.9 
 
 
 1.7 
 
 50 " 54 ' 
 
 
 2.4 
 
 
 
 
 
 .3 
 
 Total . 
 
 .. 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of ca 
 
 ses 92 
 
 42 
 
 59 
 
 50 
 
 51 
 
 54 
 
 348 
 
 Mean 
 
 .. 16.19 
 
 21.4 
 
 17.34 
 
 19.76 
 
 20.7 
 
 19.67 
 
 18.72 
 
 am 
 
 . . + . 380 
 
 ±1.27 
 
 ±.416 
 
 ±.864 
 
 ±1.07 
 
 + .581 
 
 + .312 
 
 
 
 
 ff 
 
 3 65 
 
 8.25 
 ±.900 
 
 3.20 
 ±.294 
 
 6.11 
 ±.611 
 
 7.63 
 ±.756 
 
 4.27 
 ±.411 
 
 5.82 
 
 (r„ 
 
 . . + .269 
 
 ±.221 
 
 
 
 
 (b) Rape or Consent 
 
 Table 147 shows the percentages of women who were forced and 
 those who consented to have first sexual intercourse. In this table 
 
382 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 the percentages of women who were raped vary markedly between 
 institutional groups. The Workhouse has the smallest percentage who 
 were forced to have sexual intercourse and the percentages increase 
 progressively through Auburn, the Penitentiary, Probation, Magdalen 
 and Bedford. In the total group, 22.6 per cent of the women were 
 raped while 77.4 per cent willingly consented to have sexual inter- 
 course. A large part of the 22.6 per cent is made up of the women 
 falling in the youngest age groups shown in Table 146. 
 
 TABLE 147 
 
 RAPE OR CONSENT AT FIRST SEX OFFENSE 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups 
 
 
 Institutional Groupb 
 
 
 Rape or Consent 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Probation 
 
 Total 
 
 Rape 
 
 Consent 
 
 35.9 
 64.1 
 
 9.3 
 90.7 
 
 34.5 
 65.5 
 
 16.3 
 
 83.7 
 
 8.6 
 91.5 
 
 18.9 
 81.1 
 
 22.6 
 
 77.4 
 
 Total. ... 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of oases 
 
 92 
 
 43 
 
 58 
 
 49 
 
 59 
 
 53 
 
 354 
 
 Classified by nativity and color, the distribution of age at first sex 
 offense shows that the range of years for the foreign born is from 12 
 to 51 years, for the native white from 8 to 39 years, and for the 
 native colored from 6 to 39 years. There is a concentration of cases 
 in the lower age groups of the native colored, with only 7.3 per cent 
 who committed their first sex offense when they were thirty or over. 
 The native white have their concentration of cases in the age groups 
 slightly above those noted for the native colored, but have 4.0 per cent 
 who were first sexually immoral when they were 30 or over. The 
 foreign born, on the other hand, show few cases in the earljer age 
 groups, and have 13.3 per cent who were 30 or over at time of their 
 first sex offense. It would seem from this that there is a tendency 
 for the native colored to get into sexual difficulties at the younger 
 ages, and for the foreign born to commit their first sex offenses much 
 later in life. 
 
HISTORY OF SEX IRREGULARITIES 
 
 383 
 
 TABLE 148 
 
 AGE AT FIRST SEX OFFENSE 
 
 Per Cent Distribution of Delinquent Women Classified by 
 Nativity and Color 
 
 
 Nativkty and Color 
 
 
 Age at First Sex Offense 
 
 Total 
 Foreign Born 
 
 Total 
 Native Born 
 
 Native 
 White 
 
 Native 
 Colored 
 
 Total 
 
 6 to 10 years 
 
 10 " 14 " 
 
 14 " 18 " 
 
 18 " 22 " 
 
 22 " 26 " 
 
 26 " 30 " 
 
 30 " 34 " 
 
 34 " 38 " 
 
 38 " 42 " 
 
 50 " 54 " 
 
 '2.7 
 
 33.3 
 
 33.3 
 
 12.0 
 
 5.3 
 
 6.7 
 
 1.3 
 
 4.0 
 
 1.3 
 
 2.2 
 7.4 
 49.3 
 23.5 
 7.7 
 5.2 
 2.6 
 1.1 
 1.1 
 
 2.0 
 
 5.9 
 
 47.8 
 
 26.1 
 
 8.4 
 
 5.9 
 
 2.0 
 
 1.5 
 
 .5 
 
 2.9 
 
 11.6 
 
 53.6 
 
 15.9 
 
 5.8 
 
 2.9 
 
 4.4 
 
 '2.'9 
 
 1.7 
 
 6.3 
 
 45.8 
 
 25.7 
 
 8.7 
 
 5.2 
 
 3.5 
 
 1.2 
 
 1.7 
 
 .3 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 
 
 Number of cases 
 
 75 
 
 272 
 
 203 
 
 69 
 
 347 
 
 (c) Sex Instruction 
 
 Since the question of giving proper sex instruction to children is 
 being stressed at present throvtghout the country, it is of interest to see 
 how many of the women in this study were instructed by persons who 
 were competent to give children the information they ought to have 
 about sex matters. Sex instruction, as we have used it, refers to the 
 minimum of even the simplest and most general statements concerning 
 reproduction and sex matters. In many cases the instruction given 
 by the mother, for instance, was nothing more than a warning to "keep 
 away from the fellows now" with a few admonitions about the con- 
 sequences. We have included as sex instruction, however, even these 
 crudest and most unintelligent attempts to tell the girl what things 
 she must be careful about. The number who have received intelligent 
 instruction on sex matters from those who are in a proper position to 
 tell them is much smaller, therefore, than the number we have included. 
 
 However, it is interesting to note in Table 149 the large percentage 
 of these women who received no adequate sex knowledge except from 
 girl friends, by reading or by observation. In the total group, 52.7 
 per cent of the women received a certain amount of sex knowledge 
 
384 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 149 
 
 SOURCE OF SEX INSTRUCTION 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups 
 
 
 Institutional GROtrPS 
 
 
 Source of Sex Instruction 
 
 Bed- 
 ford 
 
 Auburn 
 
 Mag- 
 dalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Proba- 
 tion 
 
 Total 
 
 
 45.5 
 
 58.2 
 
 57,6 
 
 45.6 
 
 60,5 
 
 51.7 
 
 52.7 
 
 
 
 
 29,3 
 1,0 
 5.1 
 
 10.1 
 
 37.3 
 
 '4c'.5 
 16,4 
 
 45,5 
 
 'i'5 
 10,6 
 
 31.3 
 
 1.2 
 13,1 
 
 41,9 
 
 isle 
 
 27.6 
 10,3 
 
 35.3 
 
 Father 
 
 2 
 
 JJustjand 
 
 3 9 
 
 Other relatives, 
 
 13 3 
 
 
 
 Friends 
 
 46.7 
 
 37,4 
 
 28,7 
 
 9,6 
 
 3,5 
 
 32,7 
 
 26.2 
 
 
 
 
 5,1 
 
 16.2 
 
 15.2 
 
 5,1 
 
 6,1 
 
 i9A 
 9,0 
 6,0 
 3,0 
 
 24^2 
 1,5 
 1.5 
 1,6 
 
 '9.6 
 
 3!5 
 
 5.2 
 
 24.1 
 
 1.7 
 
 1.7 
 
 1.8 
 
 "Friends" 
 
 15 3 
 
 Girl friends in school 
 
 Girl friends at work 
 
 5.0 
 2 4 
 
 Girl friends in institutions . . . 
 
 1.7 
 
 Teacher or matron in institution 
 
 4.0 
 1,0 
 2,0 
 1,0 
 
 '3!6 
 1,5 
 
 1,5 
 
 ioie 
 
 1.5 
 
 'i!2 
 
 42.2 
 
 1,2 
 
 1,2 
 U.9 
 
 1.7 
 
 i6!3 
 3.6 
 
 1.5 
 
 .2 
 
 Employer 
 
 .7 
 
 Reading or observation 
 
 Self or "none" 
 
 17.7 
 1.1 
 
 
 
 
 Total 
 
 100,0 
 
 100,0 
 
 100.0 
 
 100,0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 
 
 Number of cases 
 
 99 
 
 67 
 
 66 
 
 83 
 
 86 
 
 68 , 
 
 459 
 
 
 
 from members of their families. The number receiving no such train- 
 ing until they were told by their husbands should be thrown out of 
 any consideration of sex instruction that was of value to the child. 
 Instruction by friends includes that of friends met on the street, 
 friends in school, at work, in institutions. A teacher or matron 
 in an institution or a physician are legitimate sources of information, 
 and the three cases noted who were instructed by an employer were 
 young nurse-girls, who were very intelligently told of sex matters at 
 the time menstruation was established. As a whole, we may divide 
 the total group into (1) those who were given sex instructions, even 
 very inadequate instructions, by persons who were in a position to 
 give this information intelligently, and (2) those who were improperly 
 instructed by girls of their own age, or by reading and observation; 
 
HISTORY OF SEX IRREGULARITIES 385 
 
 which was unsuitable. Redistributed in this way for the total group 
 the percentages fall as follows: 
 
 (1) Sex instructions given by proper source, including mother, father 
 and other relatives; teacher or matron in institution; physician 
 and employer=S1.2 per cent. 
 
 (2) Sex instructions given by improper source or too late, and those 
 who had no instructions except the things they observed=:48.9 
 per cent. 
 
 By this division it appears that almost half of the cases "picked up" 
 whatever knowledge they have of sex matters, through sources which 
 could not present the matter in the right way. This is, we realize, a 
 very liberal estimate for those who have had sex instructions, since 
 many we have included in the first group as instructed by "members 
 of family" were told very little that would be of benefit to them. 
 
 EXTENT OF SEXUAL IRREGULARITY 
 
 (a) Distribution by Institutional Groups 
 
 If we proceed to a discussion of the extent of sexual irregularity 
 which followed these first sex offenses, we may note in Table 150 that 
 the first class, the prostitutes, have the bulk of the cases in each group 
 except Auburn and the Penitentiary, both of which have the largest 
 percentage of felons. By "prostitute" we mean one who has practised 
 indiscriminate sexual intercourse for money. We have included here 
 not only the professional prostitutes who made their entire living in 
 this manner, but also the irregular prostitutes who supplemented other 
 earnings by prostitution. The second class which we have called "sex- 
 ually irregular" includes women who may have been sexually promis- 
 cuous at times, but not for money ; or women who may have lived with 
 one or more consorts for any length of time. "Occasional sex offend- 
 ers" includes those who may have had illicit sexual intercourse a few 
 times, but who are only occasional and random offenders. "Never a 
 sex offender" includes those who have never had illicit sexual inter- 
 course. 
 
 Table 150 followed by Chart XX shows that the percentages of 
 women in these various classes vary considerably. The Workhouse 
 has the highest number, 93.0 per cent of its total, who were prostitutes. 
 The Probation group which follows has, as would be expected, a high 
 percentage of prostitutes, since the court from which we selected the 
 Probation cases was designed particularly for women of the prostitute 
 
386 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 class. Bedford has the third highest percentage of prostitutes and 
 the Magdalen the fourth. As previously noted, the Penitentiary and 
 Auburn have an appreciably smaller prostitute class than any of the 
 other groups, but the percentages of women in these two institutions 
 who fall in the sexually irregular class is larger than in any of the 
 other institutional groups. The occasional sex offenders have a small 
 percentage in each group except the Workhouse, which has none. 
 Those who have never had illicit sexual intercourse constitute 35.5 
 per cent of the Penitentiary group, 23.4 per cent of the women in 
 Auburn, 9.2 per cent in the Magdalen, 7.0 per cent in the Workhouse, 
 5.6 per cent in the Probation group and 2.0 per cent in Bedford. In 
 the total group we find that 66.2 per cent of the women are prostitutes, 
 17.8 per cent sexually irregular, 2.0 per cent occasional sex offenders, 
 and 14.0 per cent who have never been sex offenders. 
 
 If we divide our institutional groups into a more general classi- 
 fication of (1) those who have been prostitutes and sexually irregu- 
 lar, and (2) those who have been random sex offenders only or never 
 sex offenders, we find that while Auburn and the Penitentiary still 
 keep the lowest places in the first group, their percentages are, never- 
 theless, very high. That. is, in the first class defined above. Auburn 
 has 72.8 per cent and the Penitentiary 63.5 per cent, in contrast to 
 
 TABLE 160 
 
 EXTENT OF SEXUAL IRREGTJLARITY 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups 
 
 
 Institutional Groups 
 
 
 Extent of Sexual Irregularity 
 
 Bed- 
 ford 
 
 Auburn 
 
 Mag- 
 dalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Proba- 
 tion 
 
 Total 
 
 Prostitutes 
 
 77.3 
 
 17.8 
 3.0 
 2.0 
 
 32,5 
 
 40.3 
 
 3.9 
 
 23.4 
 
 68.6 
 
 21.1 
 1.3 
 9.2 
 
 35.5 
 
 28.0 
 
 .9 
 
 35.6 
 
 93.0 
 7.0 
 
 87.8 
 
 3.3 
 3.3 
 
 5.6 
 
 66 2 
 
 Sexually Irregular (not for 
 money) 
 
 17 8 
 
 Occasional Sex Offenders 
 
 Never a Sex Offender . . 
 
 2.0 
 14.0 
 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 inn n 
 
 
 
 Number of cases 
 
 101 
 
 77 
 
 76 
 
 107 
 
 100 
 
 90 
 
 551 
 
 

 ! I - 
 
 8 - 
 
 if- 
 
 tb 
 
 1- 
 
 s - 
 
 i -'" 
 
 if 
 
 1 5 S 
 
 ■ ■ ■ 
 
 
 
 li. 
 
 DO • 
 
 1 
 
 PENITENTIARY 
 
 PBB CUNT 
 
 1 
 
 s 
 
 
 
 5 
 
 
 is, 
 
 I.I 
 
 i , 
 
 h 1 
 
 I.I 
 
 n ; 
 
 'f! i 
 
 S 
 
 o 
 t-« 
 bo 
 
 1 i 
 
 
 
 3 
 
 
 387 
 
388 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 95.1 per cent in Bedford, 93.0 per cent in the Workhouse, 91.1 per 
 cent in the Probation group and 89.6 per cent in the Magdalen. The 
 fact that in our total group only 16.0 per cent of the cases have never 
 been sex offenders, or only occasionally so, is sufficient evidence that 
 the sex problem is one of the most prevalent ones in a study of de- 
 linquent women, and that even the more serious offenders are a large 
 part of the problem, though to a lesser extent than the petty offenders. 
 A similar comparison of extent of sexual irregularity for color and 
 nativity groups shovvrs that the foreign born have the smallest per- 
 centage of prostitutes, the native white the next larger percentage 
 and the native colored the largest percentage. The percentage of sex- 
 ually irregular women, on the other hand, is largest in the foreign 
 born group, next smaller among the native whites and smallest among 
 the native colored. The foreign born have a very much larger per- 
 centage of women who have never been sex offenders than either of 
 the other groups, and the native colored have the smallest percentage. 
 If we divide the classes, of sexual irregularity into two groups as we 
 did with the institutional groups, we find that among those who have 
 been either prostitutes or sexually irregular, though not for money, 
 the foreign bom have 66.6 per cent, the native white 90.5 per cent 
 
 TABLE 161 
 
 EXTENT OF SEXUAL IRREGULARITY 
 
 Per Cent Distribution of Delinquent Women Classified by 
 Nativity and Color 
 
 
 Nativity and Color 
 
 
 Extent of Sexual Irregularity 
 
 Total 
 Foreign Born 
 
 Total 
 Native Born 
 
 Native 
 White 
 
 Native 
 Colored 
 
 Total 
 
 Prostitutes 
 
 46.1 
 
 19.5 
 
 3.0 
 31.4 
 
 75.1 
 
 17.0 
 
 1.6 
 6.3 
 
 71.8 
 18.7 
 
 1.8 
 
 7.8 
 
 84.7 
 12.2 
 
 1.0 
 
 2.0 
 
 66.2 
 17.8 
 
 2.0 
 
 14.0 
 
 Sexually Irregular (not 
 for money) 
 
 Occasional Sex Offend- 
 ers 
 
 Never a Sex Offender . . 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100,0 
 
 100.0 
 
 
 Number of cases 
 
 1*9 
 
 382 
 
 284 
 
 98 
 
 551 
 
HISTORY OF SEX IRREGULARITIES 
 
 389 
 
 and the colored 96.9 per cent. From any combination of classes of 
 sexual irregularity, it would seem that the foreign bom have a much 
 smaller percentage of women who are professional prostitutes, and a 
 much larger percentage of those who are not sex offenders. An indica- 
 tion of this was found in Chapter VI, in considering the nature of 
 present offense for the women classified by nativity and color. There, 
 also, it was evident that the foreign born had a much smaller per- 
 centage of women convicted of offenses against chastity than had 
 the native white or native colored. 
 
 (b) Length of Time in Prostitution 
 
 In order to show the range of time during which the women we 
 are studying have been prostitutes, or sexually irregular, we present 
 Tables 152 and 153. Table 152, showing the length of time that cer- 
 tain of the women prostituted themselves, covers a range of from a few 
 days to 26 years. The first group — under two years — includes all who 
 have.just started the life of a prostitute, as well as those who have been 
 
 TABLE 162 
 
 TOTAL LENGTH OF TIME IN PROSTITUTION 
 
 Per Cent Distribution of Professional and Irregular Prostitutes 
 in Institutional Groups 
 
 
 Institutional Gboups 
 
 
 Length of Time 
 
 Bed- 
 ford 
 
 Auburn 
 
 Mag- 
 dalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Proba 
 tion 
 
 Total 
 
 
 35.2 
 
 29.6 
 
 21.1 
 
 6.6 
 
 7.1 
 
 36!4 
 17.4 
 
 17.4 
 4.4 
 8.7 
 8.7 
 8.7 
 
 ■4;4 
 
 54.6 
 20.5 
 
 6.8 
 11.4 
 
 2.3 
 
 2.3 
 
 15.4 
 23.1 
 
 7.7 
 7.7 
 3.9 
 15.4 
 7.7 
 7.7 
 3.9 
 
 16.7 
 
 20.0 
 
 16.7 
 
 13.3 
 
 10.0 
 
 13.3 
 
 1.7 
 
 1.7 
 
 1.7 
 
 3.3 
 
 1.7 
 
 75. £ 
 
 14. S 
 
 6.1 
 
 4.] 
 
 36.6 
 
 2 to 4 vears 
 
 22.7 
 
 4 " 6 " 
 
 13.6 
 
 6 " 8 " 
 
 9.2 
 
 8 " 10 " 
 
 5.1 
 
 10 " 12 " 
 
 5.1 
 
 12 " 14 " ; 
 
 1.4 
 
 2.2 
 
 14 " 16 " 
 
 2.2 
 
 16 " 18 " 
 
 
 1.1 
 
 18 " 20 " 
 
 
 .7 
 
 20 " 22 " 
 
 
 1.1 
 
 22 " 24 " 
 
 
 
 24 " 26 " 
 
 
 .4 
 
 
 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100. ( 
 
 D 100.0 
 
 
 
 Number of cases 
 
 71 
 
 23 
 
 44 
 
 26 
 
 60 
 
 49 
 
 273 
 
 
 
390 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 in it any time up to two years. The number of years in prostitution has 
 the longest range in the Penitentiary, Workhouse and Auburn, and the 
 shortest in Bedford, the Magdalen and Probation where the average 
 age is youngest and where the women have not had a chance to be in 
 this life for a long period of time. The very much shorter range in 
 the Probation group is explained by the fact that the women in this 
 group are supposed, for the most part, to be first offenders, in the 
 sense of this being their first conviction, and are chosen from the 
 most promising women who go through the court. This process of 
 selection naturally eliminates the older women who have been prosti- 
 tutes for many years and who would be entirely hopeless on probation. 
 In the total group, the concentration of number of years in prostitu- 
 tion is among the smaller groups of years, — up to six years. 
 
 (c) Length of Time Sexually Irregular 
 
 The total length of time these women have been sexually irregular, 
 including in addition to the time in prostitution the years sexually 
 promiscuous, though not for money, shows that the range of years by 
 
 TABLE 163 
 
 TOTAL LENGTH OF TIME SEXUALLY IRREGULAR 
 
 Per Cent Distribution by Institutional Groups of Professional and 
 Irregular Prostitutes and Those Sexually Irregular 
 
 
 
 
 Length of Time 
 
 Bed- 
 ford 
 
 Auburn 
 
 Mag- 
 dalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Proba- 
 tion 
 
 Total 
 
 Less than 2 years 
 
 39.1 
 27.6 
 20.7 
 4.6 
 5.8 
 1.2 
 1.2 
 
 6.5 
 
 45.7 
 
 13.0 
 
 10.9 
 
 2.2 
 
 4.4 
 
 8.7 
 
 4.4 
 
 2.2 
 
 '2!2 
 
 51.7 
 
 26.7 
 
 6.7 
 
 8.3 
 
 3.3 
 
 "i.i 
 
 1.7 
 
 26.1 
 34.8 
 6.5 
 4.4 
 2.2 
 8.7 
 4.4 
 4.4 
 4.4 
 
 '212 
 
 '2^2 
 
 16.7 
 
 20.0 
 
 16.7 
 
 13.3 
 
 10.0 
 
 13.3 
 
 1.7 
 
 1.7 
 
 1.7 
 
 3.3 
 
 1.7 
 
 76.5 
 
 13.7 
 
 5.9 
 
 3.9 
 
 36.9 
 
 2 to 4 years 
 
 27 4 
 
 4 " 6 " 
 
 12 6 
 
 6 " 8 ■' 
 
 7.4 
 
 8 " 10 " 
 
 4 3 
 
 10 " 12 " , 
 
 4.3 
 
 12 " 14 " 
 
 2.3 
 
 14 " 16 " 
 
 1.7 
 
 16 " 18 " 
 
 
 1.4 
 
 18 " 20 " 
 
 
 .6 
 
 20 " 22 " 
 
 
 .9 
 
 22 " 24 " 
 
 24 " 26 " 
 
 
 "is 
 
 
 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 
 
 Number of cases 
 
 87 
 
 46 
 
 60 
 
 46 
 
 60 
 
 51 
 
 350 
 
 
 
HISTORY OF SEX IRREGULARITIES 391 
 
 two-year groups is the same as in Table 152, but that there is slightly 
 more of a concentration in the lower groups. The three institutions 
 made up of women of older average age have a longer range of years 
 in which they have been either sexually promiscuous (not for money) 
 or prostitutes. 
 
 SPECIAL STUDY OF PROSTITUTES 
 
 (a) Age at Entering Prostitution 
 
 The part of the sex problem in which we are particularly inter- 
 ested is that of the prostitute, the woman, who, for the most part, 
 gives up other means of earning money for this easier and more lucra- 
 tive way. The most of the data we present for this will be for the 
 total group of prostitutes, since, if divided by institutional groups, the 
 numbers are too small for significant use. Bearing in mind that 66.2 
 per cent of our total group are prostitutes, we shall proceed to the 
 consideration of age of these women at the time they began to pros- 
 titute. Table 154 gives by three-year age groups the distribution for in- 
 
 TABLE 164 
 
 AGE AT ENTERING PROSTITUTION 
 
 Per Cent Distributioii of Prostitutes in Institutional Groups 
 
 
 
 
 Institutional Gbodps 
 
 
 
 
 Age at 
 
 
 
 
 
 
 
 
 Entering 
 
 
 
 
 
 
 
 Total 
 
 Prostitution 
 
 
 
 
 Peni- 
 
 Work- 
 
 
 
 • 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 tentiary 
 
 house 
 
 Probation 
 
 
 10 to 13 years 
 
 
 5.3 
 
 
 4.4 
 
 
 
 .8 
 
 13 " 16 " 
 
 25.6 
 
 21.1 
 
 2.3 
 
 13.0 
 
 5.5 
 
 
 11.7 
 
 16 " 19 " 
 
 30,6 
 
 31.6 
 
 40.9 
 
 13.0 
 
 16.4 
 
 27.8 
 
 27.3 
 
 19 " 22 " 
 
 27.8 
 
 21.1 
 
 31.8 
 
 21.7 
 
 25.5 
 
 22.2 
 
 26.1 
 
 22 '■ 25 " 
 
 6.9 
 
 5.3 
 
 18.2 
 
 30.4 
 
 12.7 
 
 19.4 
 
 14.1 
 
 25 " 28 " 
 
 5.6 
 
 5.3 
 
 2.3 
 
 
 12.7 
 
 22.2 
 
 8.4 
 
 28 " 31 " 
 
 2.8 
 
 5.3 
 
 4.6 
 
 17.4 
 
 5.6 
 
 5.6 
 
 5.6 
 
 31 " 34 " 
 
 1.4 
 
 
 
 
 9.1 
 
 
 2.4 
 
 34 " 37 " 
 
 
 5.3 
 
 
 
 7.3 
 
 
 2.0 
 
 37 " 40 " 
 
 
 
 
 
 1.8 
 
 2.8 
 
 .8 
 
 40 " 43 " 
 
 
 
 
 
 1.8 
 
 
 .4 
 
 49 " 52 " 
 
 
 
 
 
 1.8 
 
 
 .4 
 
 Total. .. 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Number of cases 
 
 72 
 
 19 
 
 44 
 
 23 
 
 55 
 
 36 
 
 249 
 
 Mean age at entering prostitution. 
 
 (Total group) 20 . 68 + . 350 
 
 5.53±.249 
 
392 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 stitutional groups of the age at entering prostitution. The youngest 
 age, ten years, applies to a woman in Auburn who has had a long 
 record not only of prostitution but of other illegal acts. The one 
 Penitentiary case in this lowest age group was only twelve years when 
 she began to prostitute. The one woman in the oldest age group was 
 49 when she began to prostitute. Her husband had just died and she 
 knew of no easier way to earn her living. The numbers in the insti- 
 tutional groups, as seen in Table 154, are too small for significant use 
 of the mean and standard deviation, so these are presented only for 
 the total. The average age at entering prostitution for the total group 
 is 20.68+5.53. (See also Chart XXI.) 
 
 Per 
 Cent 
 
 AGE 
 
 Chart XXI 
 
 Age at Entering Prostitution 
 
 Per cent distribution for total group (249 cases). 
 
 This average age at entering prostitution is somewhat higher than 
 the age found by Dr. Davis ^ both in her study of Bedford prostitutes 
 and of street cases. The comparison of the means for age afentering 
 prostitution are as follows : 
 
 269 Bedford prostitutes 187 years. 
 
 1106 street cases 19.44 years! 
 
 249 prostitutes in this study ...[, .20.68 years. 
 
 The higher age at entering prostitution for our group is probably 
 due to the fact that we have women of many ages from institutions 
 .^Op. cit., pp. 216 and 245. 
 
HISTORY OF SEX IRREGULARITIES 393 
 
 which have varying percentages of prostitutes. Likewise our group 
 contains many of the more serious offenders who did not become pros- 
 titutes until later in life when that became incidental to the other illegal 
 acts which they were committing. 
 
 (1) Age at Entering Prostitution for Nativity Groups. If we ob- 
 serve the age at entering prostitution in the nativity and color groups, 
 as shown in Table 155, we see that the range of years for the foreign 
 born is longer than for the native white and native colored, and that 
 there is, therefore, less concentration about the younger age groups. 
 The means and standard deviations show that the average age at en- 
 tering prostitution is lowest among the native colored, 19.87zt5.13. 
 The native white have the next larger average age at entering prosti- 
 tution, 20.31 ±4.63, while the foreign born have the oldest average age, 
 23.0+7.87 years. 
 
 TABLE 166 
 
 AGE AT ENTERING PROSTITUTION 
 
 Per Cent Distribution of Prostitutes for Total Group 
 Classified by Nativity and Color 
 
 
 Nativity and Color 
 
 
 Age at Entering Prostitution 
 
 Total 
 Foreign Born 
 
 Total 
 Native Born 
 
 Native 
 White 
 
 Native 
 Colored 
 
 Total 
 
 10 to 13 years 
 
 13 " 16 " 
 
 16 " 19 " 
 
 19 " 22 " 
 
 22 " 25 " 
 
 25 " 28 " 
 
 28 " 31 " 
 
 31 " 34 " 
 
 34 " 37 " 
 
 37 " 40 " 
 
 40 " 43 " 
 
 49 " 52 " 
 
 ii!6 
 
 18.6 
 25.6 
 18.6 
 2.3 
 4 7 
 2.3 
 7.0 
 4.7 
 2 3 
 2 3 
 
 1.0 
 
 11.7 
 
 29.3 
 
 25.9 
 
 13.8 
 
 9.8 
 
 6.9 
 
 2.0 
 
 1.0 
 
 .7 
 
 10.0 
 
 30.0 
 
 26.0 
 
 12.7 
 
 12.0 
 
 6.0 
 
 2.0 
 
 .7 
 
 1.8 
 
 16.4 
 
 27.3 
 
 25.5 
 
 14.6 
 
 3.6 
 
 5.5 
 
 3.6 
 
 1.8 
 
 .8 
 
 11.7 
 
 27.4 
 
 25.8 
 
 14.1 
 
 8.5 
 
 5.7 
 
 2.4 
 
 2.0 
 
 .8 
 
 .4 
 
 .4 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100. o' 
 
 100.0 
 
 Number of cases . . . . 
 
 43 
 
 205 
 
 150 
 
 55 
 
 248 
 
 Mean 
 
 ff m 
 
 23.0 
 ±1 20 
 
 20.20 
 ±333 
 
 20.31 
 
 ±.378 
 
 19.87 
 ±692 
 
 20.68 
 + 352 
 
 
 
 (J- 
 
 7.87 
 + .849 
 
 4.77 
 ±•236 
 
 4.63 
 ±.267 
 
 5.13 
 ±.489 
 
 5 54 
 
 0-ff 
 
 ±249 
 
394 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 Chart 22 shows graphically the ages at which the foreign born and 
 native bom entered prostitution. The dots indicating the means and 
 the arrows indicating the standard deviations illustrate how much 
 older, on the average, the foreign born were when they began to pros- 
 titute. 
 
 Per 
 Cent 
 
 Native born 
 
 •• — • ^ Foreign boro 
 
 Chart XXII 
 
 Age at Entering Prostitution 
 
 Percentage comparison between native and foreign born. 
 
 The differences we have just noted are crude, however, and are 
 not of great significance unless we can demonstrate whether or not 
 they might have occurred by chance. Accordingly in Table 156 the 
 means and standard deviations of the age at entering prostitution have 
 been computed for the native white and native colored groups. The 
 difference of the means is .44 and the difference of the standard 
 deviations, — .51. From Table 156 we see that the former differ- 
 ence is only .56 times the standard deviation of the difference and 
 therefore not demonstrably valid. 
 
 Since we have shown that there is no valid difference between the 
 central tendencies of the native white and native colored with refer- 
 ence to their ages at entering prostitution, it will not be necessary for 
 us to compare either of these groups separately with the foreign born 
 if we use the total native born for comparison. Table 157 gives us 
 this comparison and shows that there is a large crude difference both 
 
HISTORY OF SEX IRREGULARITIES 
 
 395 
 
 TABLE 166 
 
 NATIVE WHITE AND NATIVE COLORED 
 
 Comparison of the Means and Standard Deviations of Age at Entering 
 Prostitution for the Native White and the Native Colored 
 
 
 Native 
 White ■ 
 
 Native 
 Colored 
 
 Difference 
 
 d 
 0-d 
 
 Chances that real 
 
 difference does 
 
 not exist are 1 
 
 in: 
 
 Mean 
 
 2a. 31 
 ±.378 
 
 19.87 
 ±.692 
 
 .44 
 
 .56 
 
 3 
 
 (Tm 
 
 
 
 (7 .... ' 
 
 4.63 
 ±.267 
 
 5.13 
 
 ±.489 
 
 -.50 
 
 .91 
 
 
 ffa 
 
 6 
 
 
 
 Cases 
 
 150 
 
 55 
 
 
 
 
 
 
 of the means and standard deviations. The ratios that have been 
 computed show these differences to be genuine, with a tendency for 
 the foreign born to go into prostitution at a later age, and to have a 
 wider scattering than the native born. The differences which we 
 iind to be valid between the foreign and native in respect to this part 
 of their sex life is of particular interest since it was shown in an 
 earlier discussion that the foreign born have a much smaller per- 
 centage of prostitutes than the other groups have, and also because 
 they show a smaller percentage of women whose present conviction was 
 for an offense against chastity. 
 
 TABLE 157 
 
 TOTAL NATIVE BORN AND TOTAL FOREIGN BORN 
 
 Comparison of the Means and Standard Deviations of Age at Entering 
 Prostitution for the Total Native Bom and Total Foreign Bom 
 
 
 Total 
 Native Born 
 
 Total 
 Foreign Born 
 
 Difference 
 
 d 
 
 Chances that real 
 
 difference does 
 
 not exist are 1 
 
 in: 
 
 Mean 
 
 20.20 
 ±.333 
 
 23.0 
 ±1.20 
 
 -2.80 
 
 2.25 
 
 82 
 
 ffm 
 
 
 
 
 (7 
 
 4.77 
 ±.236 
 
 7.87 
 ±.849 
 
 -3.10 
 
 3.52 
 
 5000 
 
 0-<r 
 
 
 Cases 
 
 205 
 
 43 
 
 
 
 
 
 
01 
 
 g 
 
 •43 
 u 
 
 1 
 
 O 
 
 01 
 
 O 
 
 a> 
 
 I 
 
 00 °5 Qi 
 
 "^ -2 S 
 ago 
 
 >g 
 
 0) 
 
 a) 
 
 o 
 
 t3 
 
 o 
 O 
 
 a o 
 
 
 9* o* Sf 
 
 Sf o% 0* 
 
 0* o» ZS 
 
 ZE o» »£ 
 
 ^o»iE 
 
 le ot 82 
 
 82 0? SZ 
 
 9Z o* ee 
 
 BZ o» 61 
 
 61 o» 91 
 
 91 <<» £1 
 
 SI 0% 01 
 
 N i-lFHrH 
 
 tH M.-«C0»O« 
 
 >-( i-lrHC0i-t«0>50O 
 
 ff404nc4ra>ocoeou3a> 
 
 i-ioi N'SieOF-it* 
 
 S§! 
 
 HrtlN-^Ot* 
 
 20t*«0|2^'^^'^'^®oot^'^'0'*eo«i-to 
 
 suoipiAuoa snoiABij JO jgqranN 
 
 396 
 
 s 
 
 il 
 
 ■^1 
 
 "9 
 
 C4ci 
 
 ■p 
 
 09 (Q 
 
 si 
 
 nn 
 
 «4 
 
 II II 
 
 9S 
 
 as 
 ll 
 
 
 •S-B 
 
 as 
 
 .sal 
 
 •s 
 
 -*a g 
 
 :*■§ 
 
 ■3*9 
 
 •■y 
 
 EB£ 
 
 
 
 
 
 •C3 
 
 00 
 
 
 
 
 
 . 
 
 
HISTORY OF SEX IRREGULARITIES 397 
 
 (2) Age at Entering Prostitution, and Number of Convictions. It 
 may be of value to know whether there is any relationship between 
 the age at entering prostitution and other social factors. It is of 
 especial importance to know whether there is any tendency for those 
 who enter prostitution when very young to have more convictions 
 than those who became prostitutes later in life. To see whether there 
 is any association between these two, Table 158 is presented, and gives 
 the coefficient of correlation between age at entering prostitution and 
 number of previous convictions. The coefficient of correlation .001 ± 
 .063 indicates that there is no genuine relationship here. Referring to 
 the correlation ratios we note that both of these are high enough so 
 that they would indicate a very slight degree of relationship were it 
 not that the lines of the means are so irregular that it is impossible to 
 locate any significant trend of relation. It seems probable, therefore, 
 that any relationship between the age at entering prostitution and 
 number of convictions is so small that it is of little significance. The 
 very slight trend which may be observed is for those who started to 
 prostitute at a very early age to have the largest number of convictions. 
 
 (3) Age at Entering Prostitution, and Grade Finished. Of con- 
 siderable importance, also, is the association between the age at en- 
 tering prostitution and the grade finished in school. Table 159 shows 
 a correlation coefficient of .20 ±.066 indicating a significant relation- 
 ship, with a slight tendency for those who began to prostitute them- 
 selves at the youngest ages to have completed the lower grades only 
 and vice versa. The line of the means of age at entering prostitution 
 on grade finished is so irregular that it is difficult to find any definite 
 progression of change in age at entering prostitution for change in 
 grade finished. The regression line of the other ratio, however, grade 
 finished on age at entering prostitution, shows a small but consistent 
 trend of changes in grade finished for changes in age at entering pros- 
 titution. 
 
 (4) Age at Entering Prostitution, and Estimate of Home Condi- 
 tions. — We might very well expect that there would be a relationsnip 
 between the age at entering prostitution and the estimate of the early 
 home conditions. This we ifind to be the case in Table 160. The cor- 
 relation ratio of .29±.067 indicates that there is a significant rela- 
 tionship. Reference to the mean ages at entering prostitution, also, 
 shows that the average age at entering prostitution for those who 
 come from the poorest homes is 16.5 years. These average ages at 
 entering prostitution become progressively older as we go to the next 
 
o 
 
 I 
 
 en 
 
 CO 
 
 •a 
 
 u 
 
 to C5 
 
 I 
 I 
 
 .a 
 o 
 
 "5) 
 
 15 
 
 o 
 
 * Si-a.-a 
 
 g: 
 
 CO^OJOlW "3 •"< 
 
 
 CONCOiHi-i 
 
 NiH^ N 
 
 ■^ rHrtCCrHM i-l 
 
 ^NCOMiHtH N 
 
 MrHMrHNr 
 
 C^ rHiHW^lN U3r-I« 
 
 1-1 lO n iH N lO ■* rH 
 
 H N C<l ■* (O 1-1 rt 
 
 »OiHCO'* Tt<Cqi-<rH 
 
 I. 
 
 -I 
 
 Tji N lO lO lO 1-1 r-t 
 
 ^Cqi-lrHTt< 
 
 « ^rH 
 
 Wffltiim ; : : : • : I j ; 
 
 « 5,0) § 
 
 »^>*>i :::::;::: 
 
 ■^CONl-H 
 
 398 
 
 I g 
 
 Q O 
 D la 
 
 n n 
 
 CO ;o 
 q q 
 
 HI M 
 
 D U S 
 
 P B- -g 
 
 •a a .a 
 
 S-3 -T) 
 
 Si 
 
 ■s 
 
 S 'C ■^ 
 
 o S o 
 
 1 a a -e s 
 
 ! ° I S I 
 
 ; 41 1 1 5 
 
 «.» ° a 
 
 ' B -a — 
 
 "I 
 <! CD 
 
 ^1| 
 ■3 <10 
 
 11 
 
 I 
 
 I 
 
 a 
 
•a 
 
 
 S% 
 
 
 
 i 
 
 III 
 
 CDOOU3 
 
 «3 
 
 1" 
 
 ssss 
 
 s 
 
 
 
 
 
 dfl, 
 
 
 
 
 -i 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 H 
 
 
 
 
 b. 
 
 pH 
 
 r-t 
 
 
 
 
 
 CO 
 
 
 o 
 
 
 
 
 
 >ra 
 
 
 o 
 
 
 CO 
 
 
 
 
 »H 
 
 
 
 
 
 
 
 ^ 
 
 ^ 
 
 
 CO 
 
 
 
 N 
 
 
 o 
 
 
 
 
 
 .H 
 
 
 o 
 
 
 
 
 
 g 
 
 i-Hwcq 
 
 Tjt 
 
 a> 
 
 iH 
 
 
 e 
 
 c^ 
 
 
 
 ?s 
 
 CO« 
 
 w 
 
 6: 
 
 CO 
 
 
 1 
 
 
 
 
 
 
 S 
 
 w- 
 
 M 
 
 in 
 
 iHTjfCO 
 
 00 
 
 ■* 
 
 
 
 
 
 
 
 g 
 
 CO 
 
 U3C0 
 
 CO 
 
 
 w 
 
 
 
 
 
 
 
 
 (N 
 
 Oi-^ 
 
 CO 
 
 w 
 
 (N 
 
 
 
 
 
 
 
 
 
 
 1 
 
 
 
 
 
 
 00 
 
 
 
 
 
 (MI>b-{N 
 
 00 
 
 
 
 
 
 00 
 
 -*OS<N 
 
 lo 
 
 
 
 
 
 
 
 ^ 
 
 
 i-H 
 
 
 
 
 
 CS) 
 
 
 
 
 
 
 
 <N 
 
 
 
 
 
 -* 
 
 ^lOIN 
 
 00 
 
 
 
 
 
 
 (NCO'-i 
 
 CO 
 
 
 
 
 
 N 
 
 ^ 
 
 w 
 
 
 
 
 
 ^ 
 
 
 o 
 
 
 i-H 
 
 
 
 O 
 
 ^ 
 
 ^ 
 
 
 
 
 
 
 
 tJ 
 
 
 
 h 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 bO 
 
 
 
 
 
 
 
 
 ei.6 s £ 
 
 ■» -3 
 
 
 
 
 
 
 
 >Of^PLHK* 
 
 E- 
 
 
 enoT!jipuo 
 
 
 
 
 
 
 auioH jc 
 
 1 
 
 
 
 a? 
 
 EiU 
 
 ^i' 
 
 »i 
 
 I 
 
 1 
 
 "a 
 
 t.3 
 
 boo 
 
 <!0 
 
 399 
 
400 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 higher classes of home conditions, indicating a definite trend for the 
 average age at entering prostitution to increase in years with changes 
 for the better in early home conditions. 
 
 (b) Other Social Factors in Lives of Prostitutes 
 
 Although the age at first sex offense is not so important as the 
 age at entering prostitution because it may be a random or uninten- 
 tional affair, while the other has intent and is, in most cases, a delib- 
 erate choice, we shall consider it here because of its importance in rela- 
 tion to the age at entering prostitution. The age at first sex offense 
 has a range of from eight to forty years for the prostitute group with, 
 however, only one case in each of these extreme groups. The average 
 age at first sex offense for the prostitutes is 18.46±.30 with a standard 
 deviation of 5.36. This average age is younger than the average age 
 at entering prostitution, 20.68 years, and might indicate that the first 
 offense is in many ways only casual, but that often it undoubtedly 
 serves to break down the barriers of restraint sufficiently to make the 
 more serious offenses come easily later. 
 
 If we proceed to other social factors in the lives of the prostitute 
 group, we may note first the average age of this class at the time of 
 the present conviction. This age we find to be 27.13±.443 years, with 
 a standard deviation of 8.46±.313. In comparison with the total 
 group which, however, is heavily weighted with the prostitutes '^ we 
 find that the average age of the prostitutes is younger, that of the 
 total group being 28.79 years with a standard deviation of 10.13 
 years. Dr. Davis' study, previously noted,^ shows the following com- 
 parison for average age at time of present conviction: 
 
 647 Bedford prostitutes 20.09 years. 
 
 598 street cases 22.66 years. 
 
 365 prostitutes in this study 27.13 years. 
 
 This, again, shows we are dealing with an older group than that 
 studied by Dr. Davis and that the age factor undoubtedly enters in to 
 make significant differences in the most of the comparisons. 
 
 The birthplace of the prostitute group shows the following: 
 
 Total foreign born 21.4 per cent 
 
 Total native born 78.6 per cent 
 
 Native vifhite 66.8 per cent 
 
 Native colored 22.8 per cent 
 
 "A comparison of our total group with the prostitutes in the total group is 
 obviously of little significance since the prostitutes make up 66.2 per cent of 
 the total. 
 
 ' Op. cit., pp. 216 and 232. 
 
HISTORY OF SEX IRREGULARITIES 401 
 
 In comparison witli our total group, the percentage of foreign 
 born has decreased to a large extent. This results in the foreign born 
 prostitutes having, in a more marked degree than the total group, a 
 smaller representation than we should expect from their percentages 
 m the general population. Dr. Davis found among the Bedford 
 prostitutes 24.11 per cent foreign born, 62.75 per cent native white and 
 13.14 per cent native colored.* These figures are not at great vari- 
 ance with ours, except as they show- a higher percentage of native 
 white and a somewhat smaller percentage of native colored. 
 
 The data on schooling for the prostitute group show that the aver- 
 age number of years they have been in school is 6.82 with a standard 
 deviation of 2.71 years. This mean is slightly larger than that for 
 the total group in which the average number of years is 6.50 with a 
 standard deviation of 3.05. For the prostitute group there are 4.5 per 
 cent who never attended school and the range of years in school is 
 from no schooling to thirteen years. If we observe the grade finished 
 in school, we find that the average grade finished by the prostitutes is 
 4.74 with a standard deviation of 2.44. The frequency distribution 
 shows a range of cases from those who never finished the first grade 
 to two women who finished high school. One of these women who 
 went through high school was sentenced to the Magdalen on a prosti- 
 tution charge; the other is in the Probation group, for the same 
 offense. Both are bright, fond of pretty things, and went into prosti- 
 tution that they might be able to get what they wanted with less effort 
 than they would have needed to expend by working eight hours a day 
 and earning an honest living. 
 
 The work record of the prostitute group is a large subject in 
 itself, and affords an interesting field for detailed case studies. Since 
 this is not practicable at this point, we shall give the same type of 
 treatment to this group that was used in discussing the work histories 
 of the total group. The age at starting work has a range of from 8 
 to 58 years for the prostitute group and the average age at starting 
 work is 17.40±.35l years, with a standard deviation of 6.23. The 
 percentage starting work under fourteen years is 16.6 per cent, almost 
 exactly the same percentage as for the total group. 
 
 Table 161 gives the kind of work done by prostitutes at various 
 
 times during their work careers. The kind of work in the first job 
 
 indicates that the largest percentage of the prostitutes first went into 
 
 domestic service and the next largest percentage into factory work. 
 
 *Loc. cit., p. 198. 
 
402 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 161 
 
 KIND OF WORK DONE BY PROSTITUTES AT SPECIFIED TIMES 
 DURING THEIR WORK fflSTORIES 
 
 Per Cent Distribution for Total Group 
 
 
 Specified Times 
 
 Kind of Work 
 
 In 
 First Job 
 
 In 
 Latest Job* 
 
 At Time of 
 
 First 
 Conviction 
 
 At Time of 
 
 Present 
 Conviction 
 
 Frevailine 
 Work 
 
 Domestic service 
 
 40.8 
 
 23.7 
 
 2.8 
 
 6.8 
 
 10.7 
 
 3.4 
 
 4.5 
 
 .6 
 2.3 
 
 .3 
 1.7 
 
 '2^5 
 
 32.1 
 
 27.0 
 
 4.5 
 
 12.6 
 7.5 
 5.4 
 3.3 
 .3 
 2.1 
 1.5 
 1.2 
 
 .... 
 
 9.0 
 5.5 
 1.0 
 
 1.9 
 
 1.3 
 
 1.0 
 
 .3 
 
 ".Z 
 
 ".Z 
 
 6.8 
 
 69.0 
 
 3.5 
 
 9.2 
 5.3 
 1.4 
 
 2.2 
 .6 
 .6 
 .3 
 
 '".?, 
 
 "!6 
 
 12.3 
 
 64.1 
 
 2.5 
 
 40.8 
 23.1 
 
 
 3.9 
 
 Restaurant and hotel 
 
 6.9 
 
 Work in stores 
 
 Vaudeville performers.. 
 Olprioal work 
 
 9.6 
 4.8 
 3.6 
 
 Professional service 
 
 Personal service 
 
 Charwomen . ...... 
 
 .3 
 
 2.7 
 .6 
 
 
 1.5 
 
 Own housework 
 
 Idle 
 
 
 Never worked . .... 
 
 2.7 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 
 
 Number of cases 
 
 355 
 
 334 
 
 310 
 
 357 
 
 335 
 
 ^Includes only those who have had more than one job. 
 
 In comparing this with the latest work done we find a decrease in 
 domestic service workers and an increase in those who did factory 
 work. The kind of work done at the time of the first conviction is 
 of particular interest in showing the number of women who were idle. 
 The per cent of prostitutes (69.0) who were idle at this time is larger 
 than the per cent of the total group (59.9) who were idle at the time of 
 first conviction. The high percentage of domestic service workers, 
 with factory workers having second place, still obtains at time of the 
 first conviction. If we turn to the kind of work done at the time of 
 the present conviction, we may still observe the very large percentage 
 of idle women, though not quite so many as at the time of the first 
 conviction. The percentage of women who were doing their own 
 housework shows a fairly large increase between the time of first and 
 present conviction and would indicate that a certain number of the 
 women had married in this interim. On the whole, there seems to be 
 
HISTORY OF SEX IRREGULARITIES 
 
 403 
 
 little change in the kind of work done during the time between these 
 two convictions. The prevailing work shows, again, the high percent- 
 age of domestic service and factory workers in the prostitute group, 
 and for the other kinds of work a striking similarity to the percentages 
 in the same occupations at the time of the first job. 
 
 In general, the discussion of the kind of work done shows prin- 
 cipally that there were more women in domestic service than in any 
 other occupation, and that the factory workers have the next highest 
 percentage. The number of idle women at the time of the first and 
 present convictions is of particular importance, especially in connec- 
 tion with the estimate of regularity of work which will be considered 
 later. In the prostitute group, as with the total group, we find a very 
 small percentage of skilled workers, such as women in professional 
 service and clerical work. 
 
 The wages which were earned by these women at the specified 
 times are given in Tables 162 and 163. Table 162 shows the range 
 of wages for women employed at domestic service in the first job, 
 the latest job, work at time of first and present conviction, and the 
 
 TABLE 162 
 
 WEEKLY WAGE EARNED BY PROSTITUTES AT SPECIFIED 
 TIMES DURING THEIR WORK fflSTORIES 
 
 Per Cent Distribution of Women Employed at Domestic Service and 
 Allied Occupations Where Living Was in Addition to Wage 
 
 
 Specified Times 
 
 Weekly Wage 
 
 In 
 
 First Job 
 
 In 
 
 Latest Job 
 
 At Time of 
 
 First 
 Conviction 
 
 At Time of 
 
 Present 
 Conviction 
 
 Prevailing 
 Work 
 
 Livine onlv 
 
 15.8 
 
 8.8 
 
 24.6 
 
 21 .'9 
 
 18.4 
 
 4.5 
 
 1.8 
 
 1.8 
 
 .9 
 
 1.8 
 
 10.0 
 
 3.3 
 
 10.0 
 
 20.0 
 
 22.2 
 
 17.8 
 
 7.8 
 
 6.7 
 
 1.1 
 
 1.1 
 
 10.7 
 10.7 
 14.3 
 14.3 
 17.9 
 21.4 
 3.6 
 7.1 
 
 7.7 
 7.7 
 11.5 
 11.5 
 19.2 
 11.5 
 15.4 
 11.5 
 3.8 
 
 7.3 
 
 $1 to 82 
 
 6.5 
 
 2 " 3 
 
 21.0 
 
 3 " 4 
 
 21.8 
 
 4 " 5 
 
 25.0 
 
 5 " 6 
 
 8.0 
 
 6 " 7 
 
 7.3 
 
 7 " 8 
 
 .8 
 
 8 " 9 
 
 2.4 
 
 9 " 10 
 
 
 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 
 
 Number of cases 
 
 114 
 
 90 
 
 28 
 
 26 
 
 124 
 
404 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 163 
 
 WEEKLY WAGE EARNED BY PROSTITUTES AT SPECIFIED 
 TIMES DURING THEIR WORK HISTORIES 
 
 Per Cent Distribution of Women Employed at Occupations 
 other than Domestic Service 
 
 
 Spboipied Times 
 
 Weekly Wage 
 
 In 
 First Job 
 
 In 
 
 Latest Job 
 
 At Time of 
 
 First 
 Conviction 
 
 At Time of 
 
 Present 
 Conviction 
 
 Ptevailing 
 Woik 
 
 80 to $1 
 
 1.9 
 1.3 
 2.5 
 
 14.6 
 
 14.6 
 
 12,1 
 
 15.3 
 
 8.9 
 
 8.9 
 
 2.6 
 
 3.2 
 
 1.3 
 
 5.1 
 
 1.9 
 
 'i!9 
 
 '".& 
 
 1.3 
 
 .6 
 
 'i'.s 
 
 ".6 
 
 .6 
 
 2.3 
 
 6.8 
 
 8.6 
 
 18.8 
 
 11.4 
 
 8.6 
 
 12.0 
 
 9.7 
 
 3.4 
 
 4.6 
 
 1.1 
 
 .6 
 
 .6 
 1.1 
 .6 
 .6 
 .6 
 
 '2!9 
 
 "!6 
 
 "!6 
 
 16.7 
 16.7 
 10.0 
 13.3 
 3.3 
 13.3 
 10.0 
 
 Z.S 
 3.3 
 
 s.i 
 'sis 
 
 '3.3 
 
 6.4 
 25.8 
 16.1 
 3,2 
 9.7 
 9.7 
 3.2 
 ,3.2 
 
 '3!2 
 3.2 
 
 '3;2 
 
 3.2 
 3.2 
 
 
 1 " 2 
 
 
 2 " 3 
 
 .7 
 
 3 " 4 
 
 5.3 
 
 4 " 5 
 
 12.5 
 
 5 " 6 
 
 11.9 
 
 6 " 7 
 
 13.9 
 
 7 " 8 
 
 13.2 
 
 8 " 9 
 
 10.6 
 
 9 " 10 
 
 6 6 
 
 10 " 11 
 
 5.3 
 
 11 " 12 
 
 2.0 
 
 12 " 13 
 
 4 6 
 
 13 " 14 
 
 1 3 
 
 14 " 15 
 
 .7 
 
 15 " 16 
 
 2.0 
 
 16 " 17 
 
 
 17 " i8 
 
 7 
 
 18 " 19 
 
 
 19 " 20 
 
 1 3 
 
 20 " 21 
 
 7 
 
 23 " 24 
 
 ■' 
 
 24 " 25 
 
 7 
 
 25 " 26 
 
 4 
 
 27 " 28 
 
 7 
 
 31 " 32 
 
 
 35 " 36 
 
 1 3 
 
 65 " 66 
 
 
 
 
 Total .... 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 
 Number of cases 
 
 157 
 
 175 
 
 SO 
 
 • 
 
 31 
 
 151 
 
 average wage for prevailing work. The numbers were too few to 
 justify computing the means and standard deviations for certain of the 
 wage groups. The wages show a wider range, however, for the first 
 and latest jobs than for the job at first and present conviction or for 
 the prevailing work. The one woman who raised the range to $16 for 
 the total group (Chapter XI, Table 127) is not a prostitute and there- 
 fore does not fall in this group. 
 
HISTORY OF SEX IRREGULARITIES 405 
 
 The wages earned by prostitutes who were employed at occupa- 
 tions other than domestic service at specified times show a range simi- 
 lar to that for the total group (Chapter XI, Table 128). The one 
 prostitute who was earning $65 at the time she was sentenced to Bed- 
 ford was a dancing instructor in New York City. She also "picked up" 
 private pupils on the roof gardens and made a great deal of money by 
 teaching them how to dance. The absence of any cases earning the 
 lowest wages at the time of first and present conviction would indi- 
 cate that the women who have the lowest wage-earning capacity drop 
 out of work at those times. 
 
 As we should expect, from the percentages of women who were 
 idle at the time of first and present convictions, the work records show 
 great irregularity. The percentages of prostitutes in the five classes of 
 the estimate are as follows : 
 
 Worked almost none 11.4 per cent 
 
 " about one-quarter of time 27.6 " " 
 
 about one-half of time 27.6 " 
 
 " about three-quarters of time 31.6 " " 
 
 nearly all of time 1.8 " " 
 
 By this method of estimation, it appears that 66.6 per cent of the 
 women worked half of the time or less. For the total group, the per- 
 centage who worked half of the time or less was 60.9, even with the 
 prostitutes included in this total. It would seem that the prostitutes 
 do not combine with the kind of life they are living any serious at- 
 tempt to work. This is, of course, obvious after talking with many 
 of them and after working over their histories for any length of time. 
 Sometimes a prostitute has a nominal trade which she pretends to fol- 
 low, in order to protect herself in case of arrest. There are rare 
 cases, also, of girls who work during the day at legitimate work and 
 go out for purposes of prostitution occasionally at night in order to 
 supplement their earnings. There is in the Bedford group a girl of 
 eighteen who worked during the day in a department store for $7 a 
 week. She found she could not buy the clothes she wanted for that 
 money and so she began to go out with a "friend" and then with an- 
 other for two or three nights a week. By this method, she was able to 
 have what she wanted, and she was able to make her mother think she 
 was earning the extra money in the store. 
 
 The estimate of the work record shows a predominance of poor 
 work. The prostitutes distribute themselves in these five classes as 
 follows : 
 
406 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 Very poor 15.8 per cent 
 
 Poor 40.3 " " 
 
 Mediocre 35.3 " " 
 
 Good 8.1 ;; ;; 
 
 Very good 6 
 
 This estimate shows that 56.1 per cent of the women were entirely 
 unable to live on their earnings because of the irregularity, inefficiency 
 and low wages involved in the work record. There were only two cases 
 in the estimate of very good, — one in the Magdalen and one in the 
 Probation group. The latter one had just gone into prostitution and 
 was arrested the second or third time she solicited a man on the 
 street. As a whole, the work records are poor and show great insta- 
 bility.^ 
 
 Iii considering the criminal record of the prostitute class, we shall 
 first observe the age at first conviction. The range is from eight years 
 to 61 years, with eighteen cases or 4.2 per cent of the total who were 
 convicted as juvenile delinquents under sixteen years of age. The 
 mean age at first conviction for the total group is 24.69±.427, and 
 there is a standard deviation of 8.12 years. 
 
 The offenses which caused the first conviction fall in the prostitute 
 
 group as follows: 
 
 Offenses against the Person 8 per cent 
 
 Offenses against Chastity 61.6 " " 
 
 Offenses against Family and Children .8 " " 
 
 Offenses against Regulations for Public Health, etc 6.2 " " 
 
 Offenses against Administration of Government 
 
 Offenses against Property Rights 5.4 " " 
 
 General Criminality 25.0 " " 
 
 The offenders against chastity have the largest percentage, with rela- 
 tively small percentages in each other group except the general crimi- 
 nality group containing the semi- juvenile offenders. It is interesting 
 to compare the nature of the first offenses of this special group with 
 the first offenses of the total group of which, we must remember, the 
 prostitutes form a large pairt. In the total group the offenders against 
 chastity at time of the first conviction were only 44.3 per cent. The 
 offenders against property rights, on the other hand, had 18.8 per cent 
 of the cases, while among the prostitutes this class has only 5.4 per 
 cent. One is reminded here of the remark so often made by many of 
 the prostitutes who take men 'to a furnished room or a hotel but never 
 to creep houses where they are to be robbed : "Well," they will say in 
 trying to justify their own deeds, "I may be a prostitute, but I'm not 
 a thief. What I get money for, I earn, and a thief, — she just steals 
 and don't give nothing in return." 
 
HISTORY OF SEX IRREGULARITIES 407 
 
 The number of convictions for this group shows that the range is 
 from those who have never had any previous conviction (39.8 per 
 cent of the cases) to one woman in the Penitentiary who has had 
 eighteen previous convictions. The average number of previous con- 
 victions is 1.83dz.l35 with a standard deviation of 2.62. The women 
 in the total group who have the larger number of convictions, 20 and 
 31, have their history of deliquency complicated by excessive alcohol- 
 ism. As noted in Chapter VII the Intoxication Group in the Work- 
 house, though its mean number of convictions is about four times as 
 large as that for the total group, has only four instances of convictions 
 for offenses against chastity out of the 212 convictions for the total 
 group. 
 
 If we turn to the early home conditions of this group, we find that 
 in the total estimate the homes of the prostitutes seem to be poorer 
 than for the total group. The distribution is as follows : 
 
 Very Poor Poor Mediocre Good Very Good 
 ■Economic Status in Early Home . . S.S 39.8 42.1 12.2 .4 
 
 Moral Standards in Early Home ... 11.4 29.1 44.1 1S.4 
 
 Parental Supervision in Early Home 16.9 50.8 29.9 2.4 
 
 Estimate of Early Home 8.3 44.1 42.9 4.7 
 
 As with the total group the lack of parental supervision seems to be 
 one of the most striking things. The economic standards, again, ap- 
 proximate the total estimate more than any other factor entering into 
 the estimate, which would seem to indicate that the economic factor is 
 very important in the consideration of what makes up a good or a 
 poor home. The methods of making up the estimates and illustrative 
 cases for each class have been given in Chapter IX. 
 
 A somewhat isolated item, but one which is of importance because 
 of the common belief that excessive drug and alcohol using are com- 
 mon among prostitutes, is the consideration of habits. The following 
 distribution, if compared with Tables 32, 34 and 35, Chapter VII, 
 shows that there is among the prostitute group a larger percentage 
 of women who use alcohol, drugs and tobacco than there is in the 
 total group. The distribution for the prostitute group is as follows : 
 
 Alcohol Drugs Tobacco 
 
 Non-users 49.6 77.2 67.1 
 
 Moderate 28.3 ... 7.5 
 
 Excessive 22.1 22.8 25.4 
 
 There is, of course, much overlapping present here, many of the same 
 women being addicted to more than one habit. 
 
408 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 VENEREAL DISEASE 
 
 The subject of venereal disease among the women in our institu- 
 tional groups has been included in this chapter dealing with the sex 
 history, because, in most cases, there is a close connection between 
 irregularities in sex life and the diseases usually resulting from this 
 life. There are, however, several exceptions in our group of women 
 who, so far as we know, have not been sexually promiscuous, but who 
 have contracted syphilis from their husbands. Unfortunately, we 
 have been unable to obtain adequate data for comparison on all of 
 the institutional groups regarding the amount of syphilis and gonor- 
 rhea. The Probation women were given no blood tests and we have 
 data on only the few cases in this group where the clinical manifesta- 
 tions of venereal disease were so evident that the girl had to be sent 
 to a hospital for treatment during her probation period. These cases, 
 however, are too few in number to use. The 22 Penitentiary women 
 who had blood tests given have been used for comparison with the 
 other groups, though this is, perhaps, unfair since only the women 
 were tested who were suspected of being syphilitic or women who 
 were in need of other treatment and to whom the blood test was given 
 as an incidental thing. Each of the cases we have used in making up 
 Table 164 has had specimens of blood taken and a series of com- 
 plement fixation tests performed to determine by the Wassermann test 
 whether the reaction was positive, that is, whether there was evidence 
 of syphilis, and by the complement fixation test for gonorrhea whether 
 gonorrhea was present as shown by a positive reaction. We have 
 considered as a positive reaction all reactions of 4-f-, 3-j- or 2+. A 
 doubtful reaction includes l-f- or±, and negative includes only those 
 cases in which there was no reaction. 
 
 Table 164 shows the percentages of women in institutional groups 
 who had positive, negative or doubtful reactions to the complement 
 fixation tests for syphilis and gonorrhea. The Magdalen has the 
 smallest percentage of women who had a positive reaction to the 
 Wassermann test for syphilis. This is probably determined- by the 
 fact of the Magdalen being a private institution which has the right 
 to refuse any cases so that an elimination of the most diseased girls 
 takes place. Auburn has the next larger percentage of women with 
 a positive reaction, and this we might expect because of the fact that 
 Auburn has a small percentage of prostitutes. Bedford and the 
 Workhouse follow with the next largest percentages of women with 
 
HISTORY OF SEX IRREGULARITIES 409 
 
 TABLE 164 
 
 VENEREAL DISEASE 
 
 Per Cent Distribution of Delinquent Women in Institutional Groups by 
 
 Results of Wassermann and Complement Fixation Tests 
 
 for Syphilis and Gonorrhea* 
 
 
 Wassermann Test 
 
 Complement Fixation Test 
 
 Institutional Groups 
 
 Pos- 
 itive 
 
 Nega- 
 tive 
 
 Doubt- 
 ful 
 
 Number 
 of Cases 
 
 Pos- 
 itive 
 
 Nega- 
 tive 
 
 Doubt- 
 ful 
 
 Number 
 of Cases 
 
 Bedford 
 
 36.0 
 31.0 
 24.3 
 40.9 
 39.3 
 
 48.0 
 62.1 
 64.3 
 59.1 
 47.2 
 
 16.0 
 
 6.9 
 
 11.3 
 
 13.5 
 
 100 
 58 
 70 
 22 
 89 
 
 2.0 
 '3.6 
 
 67.7 
 68.6 
 82.4 
 
 68!7 
 
 30.3 
 31.4 
 17.7 
 
 27]7 
 
 99 
 
 Auburn 
 
 35 
 
 Magdalen 
 
 68 
 
 Penitentiary 
 
 
 Workhouse . . . 
 
 83 
 
 
 
 Total 
 
 33.9 
 
 54.3 
 
 11.8 
 
 339 
 
 1.8 
 
 71.6 
 
 26.7 
 
 285 
 
 
 
 ♦Positive includes reactions to the Wassermann or Complement Fixation Testa of i-\-, S+, or 2+. 
 Doubtful includes reactions of 1 -|- and +. Negative includes only '=^. 
 
 positive reactions and the Penitentiary has the largest percentage. 
 This high percentage in the Penitentiary, as has been stated before, 
 is probably due to the fact that only the cases which were suspected 
 of being syphilitic were given the blood tests. In the total group 
 of delinquent women studied, we find that 33.9 per cent have a posi- 
 tive reaction to the Wassermann test for syphilis, — ^that 54.3 per cent 
 show no reaction and that 11.8 per cent give a doubtful reaction. 
 
 If we turn to the percentage of women in institutional groups 
 who show positive, doubtful or negative reactions to the complement 
 fixation tests for gonorrhea, we see that the results are quite differ- 
 ent. To explain the obviously misleading percentages, we might 
 state that the lack of positive reactions in the Magdalen group is due 
 to the fact that a girl who is suffering from gonorrhea is not allowed 
 to be kept in the institution and is transferred to the City Hospital for 
 cure. We were able to obtain no records of this test for the Peni- 
 tentiary group and found records on only 35 of the Auburn women. 
 This part of the data is, therefore, unsatisfactory and probably quite 
 unreliable, except for the Bedford and Workhouse groups where re- 
 turns of this test were made on approximately as many cases as for 
 the Wassermann test. By examining the vaginal smears it is also true 
 that many cases of gonorrhea are found which might not show in the 
 blood, particularly if the girl had been recently infected. As the 
 
410 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 figures stand, 1.8 per cent of the total group examined gave a positive 
 reaction to the complement fixation test for gonorrhea, 71.6 per cent 
 gave a negative reaction, and 26.4 per cent a doubtful reaction. 
 
 The percentage of women among the prostitute group infected 
 with venereal disease is, as we might expect, higher than for the total 
 group. The figures for the prostitutes are as follows : 
 
 Wassermann and 
 Complement Fixation Tests for: 
 Syphilis Gonorrhea 
 
 Positive 42.5 per cent 1.9 per cent 
 
 Negative 41.6 " " 68.7 " " 
 
 Doubtful 15.9 " " 29.4 " " 
 
 These figures show a large percentage, 42.5 per cent, who gave a posi- 
 tive reaction to the Wassermann test for syphilis. In comparison with 
 466 prostitutes studied by Dr. Davis,^ all of whom were given Wasser- 
 mann tests, it appears that in the prostitute group we are studying 
 there is a larger percentage who have positive reactions for syphilis 
 alone. The figures for the Bedford women in Dr. Davis' study are 
 as follows for the Wassermann reaction: 
 
 Syphilis Gonorrhea 
 
 Positive 37.7 per cent 29.0 per cent 
 
 Negative 58.6 " " 50.0 " " 
 
 Doubtful 3.6 " " 21.0 " " 
 
 . In this connection it is important to note that the discrepancies oc- 
 curring between our group and other studies, in the matter of venereal 
 disease, are due largely to the changes in laboratory technique for the 
 various studies. We have observed in one institution, such as Bed- 
 ford for instance, that the reported percentages of infected cases 
 varied markedly as the blood examinations were made by one or an- 
 other laboratory, or subjected to modifications of the examination. 
 This undoubtedly explains the much higher percentage of cases of 
 gonorrhea found in Dr. Davis' study than among our cases. 
 
 In any event, the percentage of women in institutions, infected 
 with syphilis, is large. It is appalling to contemplate the chance for 
 infection by many of these women who are promiscuous in their sex- 
 ual relations and who have no moral scruples about spreading the 
 disease they have contracted. An example of the harm which may 
 be done by one person infected with syphilis is found in the history 
 of a woman known as the "Leopard" because of the spots on her 
 skin. She was a prostitute in Chinatown, New York City, for years 
 
 ^ hoc. cit., pp. 189 and 190. 
 
HISTORY OF SEX IRREGULARITIES 411 
 
 and contracted syphilis. She allowed herself to get into a frightful 
 state, until the Chinamen who had been supporting her took up a col- 
 lection of money and offered to send her to a hospital for treatment. 
 She refused to go and continued to have sexual relations with young 
 boys who did not know about the disease and who were attracted by her 
 attentions to them. She undoubtedly infected a great many persons, 
 many of them boys whom she deliberately chose, knowing her condi- 
 tion, in order that she might show the Chinamen she was still able to 
 "get" some one. 
 
 SUMMARY 
 
 In summarizing the data that have to do with the history of sex 
 irregularities, we have seen that the prostitutes constitute 66.2 per 
 cent of our total group and that those who have been sexually irregu- 
 lar, though not for money, make up 17.8 per cent more. Only 16.0 
 per cent of the women were not serious sex offenders. These per- 
 centages vary between institutional groups, the two institutions consist- 
 ing of the largest percentage of felons having the smallest percentage 
 of prostitutes but the largest percentage of those sexually irregular. 
 
 In a comparison between the nativity and color groups it appears 
 that the foreign born have the smallest percentage of prostitutes and 
 the largest percentage of non-sex-offenders, while the native colored 
 have the most prostitutes and the fewest,- — only two cases, — who were 
 never sex offenders. 
 
 In considering the age at entering prostitution, it also appears 
 that the foreign born do not begin prostitution until a later age, and 
 that there is a valid difference between the foreign born and the total 
 native born with respect to the age at entering prostitution. There 
 seems to be no similar significant difference between the native white 
 and native colored. 
 
 A rough attempt, because such information is hard to obtain, to 
 find how many of these women had sex instructions from any one 
 who was competent to give them such training, shows that 48.9 per 
 cent, at a very liberal estimate, were given no sex instructions except 
 what they "picked up" from girl friends, from reading or their own 
 observation. 
 
 In attempting to show the relationship between age at entering 
 prostitution and other social factors we find that : 
 
 (1) There is a very small and possibly not significant relationship 
 between the age at entering prostitution and the number of previous 
 
412 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 convictions, with, however, a tendency for those who enter prostitution 
 when youngest to have the largest number of convictions. 
 
 (2) There is a small but significant relationship between the age 
 at entering prostitution and the grade finished in school, with a 
 tendency for those who entered prostitution at the earliest ages to have 
 finished the lower grades in school, and vice versa. 
 
 (3) There is a significant relationship demonstrated in the cor- 
 relation ratio between age at entering prostitution and estimate of 
 home conditions. The mean ages at entering prostitution show a 
 progressive increase from the lowest class of the home conditions 
 through the highest class. 
 
 In the data which have been presented on the prostitute class, we 
 have shown that, for the most part, these women came from poor 
 homes, that their work was irregular and poor, that they are in fairly 
 large numbers addicted to drugs, alcohol and tobacco, and that they 
 have a large percentage of women infected with venereal disease. 
 Detailed comparisons with the total group, regarding schooling and 
 criminal record, particularly, are difficult to make since the total group 
 is so heavily weighted with the prostitutes who constitute 66.2 per cent 
 of its number. 
 
 The venereal disease in the group studied, based on returns from 
 the Wassermann and the complement fixation tests for syphilis and 
 gonorrhea, shows that there are varying percentages within the groups, 
 giving positive reactions, and that the total group has 33.9 per cent 
 with a positive reaction to the Wassermann test for syphilis, 54.3 per 
 cent with no reaction, and 11.8 per cent in which the reaction was 
 doubtful. The percentages, stating the various reactions to the com- 
 plement fixation test for gonorrhea show 1.8 per cent with a positive 
 reaction, 71.6 per cent with a negative reaction, and 26.7 per cent with 
 a doubtful reaction. 
 
CHAPTER XIII 
 
 MENTAL CAPACITY: COMPARISON WITH THE GENERAL 
 
 POPULATION 
 
 THE question of the relation between criminality and mental 
 capacity, or, as it is generally conceived, between criminality 
 and mental inefficiency, is one of primary importance. Goring's judg- 
 ment concerning the importance of this relationship is well indicated 
 in the following statement, summarizing the results of his investiga- 
 tions : 
 
 "Our final conclusion is that English criminals are selected by a phys- 
 ical condition and a mental constitution which are independent of each 
 other — that the one significant physical association with criminality is a 
 generally defective physique; and that the one vital mental constitutional 
 factor in the etiology of crime is defective intelligence." ^ 
 
 Other equally positive statements might be cited from many sources, 
 both in the literature of criminology and in that of feeble-mindedness.^ 
 In most of these discussions, however, the tendency has been to make 
 general statements based entirely on descriptive data concerning the 
 group studied, or, and this is more unfortunate, to make comparisons 
 with the general population based on a rather vague hypothesis of the 
 mental ability of the "average" non-delinquent individual. Where 
 the information sought is concerned with the inter-relations inherent 
 in the group, — such, for example, as the relationship between an in- 
 dividual's mental capacity and his criminal record, it is of course 
 necessary to restrict comparisons to the group studied, but where the 
 object is to determine whether the delinquent group as such is to be 
 distinguished from the general population with respect to mental 
 
 *0/i. cit., p. 263. 
 
 ' See, for example, the following : 
 
 Ellis, Havelock. "The Criminal." 1895. pp. 133-139. 
 
 Aschaffenburg, G. "Crime and Its Repression." Trans, by Albrecht, A. 
 1913. pp. 178-180. 
 
 Tredgold, A. F. "Mental Deficiency." 2nd Ed. 1914. pp. 319-324. 
 
 Goddard, H. H. "Feeble-mindedness. Its Causes and Consequences." 1914. 
 pp. 8-lS. ■ 
 
 413 
 
414 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 capacity, it is essential that we have reliable information concerning 
 reasonably representative samples of the whole population. The dearth 
 of such information affords the ready explanation for its almost com- 
 plete omission from the literature. Nevertheless it must be evident 
 that without the possibility of some such comparison between offenders 
 against the law and the rest of the population we can not progress far 
 in our evaluation of the significance of mental inferiority as a con- 
 tributory factor in criminality. 
 
 The most serious effort at comparison of the two groups from this 
 point of view is that of Goring, who finds his common ground in the 
 figures giving percentages of "weak-minded" individuals among crimi- 
 nals and in the general population. On the basis of the most careful 
 estimates available he offers, as a minimum figure for the percentage 
 of prisoners who are weak-minded, 10 per cent, and, as a maximum, 
 20 per cent. For comparison with these he cites the enumeration 
 of mental defectives made in 1908 by the British Royal Commission 
 for the Care and Control of the Feeble-Minded, in which they esti- 
 mated that .46 per cent of the general population were mentally de- 
 fective. By these figures the preponderance of mental defect among 
 the prisoners is strikingly suggested. 
 
 In Chapter IV we have discussed Goring's use of the method of 
 estimates and the limitations on the objectivity of this method. This 
 point applies with special force to an investigation covering as many 
 different types of groups as are included under the general population. 
 A further factor enters in to lessen the certainty of the comparison. 
 In spite of the fact that the Royal Commission performed its work 
 with the utmost care it necessarily lacked the sort of opportunity for 
 the direct observation of each individual which the prison situation 
 makes possible. This would tend to produce an under-estimation of 
 the number of weak-minded in the general population. It remains for 
 later investigations to discover whether the degree of difference, which 
 Goring maintains exists between the two groups, represents any seri- 
 ous degree of over-statement. 
 
 The psychometric methods, on which we have based our study, do 
 not entirely escape the limitations discussed above. While free from 
 the subjective variations incident to the method of estimates, they meet 
 the same obstacle of inadequate information concerning the popula- 
 tion at large. The extrerne difficulty of securing such representative 
 data is readily apparent. Through the mechanism of the common 
 schools it is reasonably possible to assemble data concerning relatively 
 
MENTAL CAPACITY 415 
 
 unselected groups of children. But in adult life individuals tend to 
 separate along very definite lines of cleavage, so that when we at- 
 tempt to secure unselected adult groups we find ourselves obtaining 
 groups of business men, or of college students, or of factory opera- 
 tives, or of some other specialized segment of the population. The 
 measurement of intelligence is obviously not a form of record taking 
 which can ever be included in a general census enumeration, which 
 would afford the only complete data. 
 
 At last, however, mass data of importance, if not of conclusive 
 finality, concerning the adult male population are available as a result 
 of the examination of army recruits in this country during 1917 and 
 1918. This constitutes our most satisfactory standard of reference, 
 in spite of certain limitations affecting the interpretation of these data, 
 to be discussed later. Next in importance to these are the data ob- 
 tained by Dr. WooUey on girls of sixteen to eighteen years of age who 
 came under the supervision of the Cincinnati Bureau of Vocational 
 Guidance. The make-up of this group will be discussed in connection 
 with its use. Other minor groups are also noted for comparison, but 
 are far more restricted in their significance than the two just men- 
 tioned. 
 
 In the following section we offer such data regarding non-delin- 
 quent adults as have been presented by other investigators in such 
 form that comparison with our material is possible. With the data 
 which we have obtained on Test Aggregate as a measure of mental 
 capacity (see p. 58) we have, as we have said, no possibility of 
 comparison with other groups. Its use is, therefore, restricted to that 
 of making comparisons within the delinquent group. Similarly, we 
 are unable to utilize the most comprehensive material available from 
 the army testing, because it is based on the special group tests— alpha 
 and beta— designed for the use of the United States army and re- 
 stricted to such use during the period of the war. The three measures 
 which furnish common ground for comparison with other groups are 
 the Stanford-Binet Scale, the Yerkes-Bridges Point Scale, and the 
 tests of the Woolley series.^ 
 
 In our comparison with outside groups, both criminal and non- 
 criminal, we have restricted ourselves entirely to consideration of 
 
 'We shall not discuss the results of this series, since an extensive compari- 
 son of results obtained from a group of Bedford women with data obtamed 
 by Dr. Woolley from working girls of fourteen and fifteen years of age has 
 already been published from this laboratory. See Weidensall, Jean: The 
 Mentality of the Criminal Woman." 1916. 
 
416 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 other adult groups, that is, to those made up of individuals of 16 years 
 or over. We have, therefore, omitted all reference to the excellent 
 studies made by Healy, Williams, Haines, Grace Fernald, Kelly, and 
 others. 
 
 COMPARISON BASED ON STANFORD-BINET 
 
 Because of the special importance of the army data as affording 
 standards of reference we shall consider first the relationship between 
 the results obtained with delinquent women and the data available 
 from the army testing.* Common ground for this comparison is 
 found in the Stanford-Binet, which was used with both groups. The 
 army data were obtained from nine camps in different sections of the 
 country. The groups were taken at random as they came for exam- 
 ination.^ When the results were in^ records of individuals born in 
 non-English-speaking countries were rejected, in order that it might be 
 certain that the scores were not affected by marked language handicap 
 in any case. Only white recruits were included in these groups. A 
 total of 653 Stanford-Binet records of white individuals born in Eng- 
 lish-speaking countries was thus obtained. 
 
 Before proceeding to our comparison the limitations affecting this 
 material should be noted. The group under consideration is at least 
 two steps removed from being a random sample of the total white 
 population born in English-speaking countries, of which we wish to 
 consider it typical. In the first place there is the uncertainty as to 
 whether the group of 653 is itself representative of the enlisted per- 
 sonnel of the army. Some information on this point was available, 
 since the other tests which were used so extensively with the army 
 were also used with this group. On the basis of this evidence it was 
 felt that data from this group were reasonably representative of the 
 army as a whole, at least in so far as this was expressed in terms of 
 the central tendency. 
 
 In the second place there is uncertainty as to how far the enlisted 
 personnel of the army represents the general population. The fact 
 that the personnel of the army under consideration was assembled 
 under the operation of the Selective Service Act tends to make it 
 
 *For an account of the results of the psychological examinations given in 
 the army see "Psychological Examining in the tfnited States Army," Memoir 
 of the American Academy of Science. 
 
 'This is important as distinguishing these data from the much larger body 
 of Stanford-Binet records amassed in the routine of the army examining but 
 obtained from men who were conspicuously low in the group tests. The latter 
 material is obviously valueless for our present purposes. 
 
MENTAL CAPACITY 417 
 
 more representative than would have been the personnel of tne 'army 
 recruited during times of peace. Even so there were selective 
 processes at work, the effect of which it is impossible adequately to 
 evaluate. Only males were included, and these fell mainly within 
 a certain range of ages. Moreover, selection took place which pre- 
 sumably -did not occur evenly over the intelligence scale. From the 
 upper levels men had been removed by selection as officers, by ad- 
 mission to officers' training camps or to special forms of service 
 requiring superior ability, and in numerous other ways. From the 
 lower levels men had been rejected through the discarding of the 
 obviously defective and incompetent by the local draft boards and 
 by the medical examining boards at the various cantonments. These 
 and other limiting factors prevent the army data from serving as 
 an exact picture of the general population. Even if we might assume 
 that elimination had been equal from the two extremes of the scale — 
 an assumption which we have no justification for making — ^the general 
 form of the distribution curve would still be influenced to some degree 
 by these selective processes. 
 
 Even in view of the limitations noted above, however, we recur 
 to our earlier assertion that the army data afford our best indica- 
 tions at present as to the mental caliber of the general population. 
 It is very probable, in fact, that they will continue to be our best 
 source of information concerning adults for some time to come. 
 
 With reference to the delinquent group, the most important 
 limitation to bear in mind is the fact that we are dealing with the 
 delinquent women who have been caught, and not with all who are 
 delinquent. Presumably, those who have succeeded in escaping con- 
 viction arc somewhat superior in intelligence. How much superior 
 they are, we have no means of knowing. None of the conclusions 
 reached in this chapter should be given wider application than to the 
 group of convicted women delinquents of New York State. 
 
 Table 165 gives the percentage distribution and constants for the 
 army group examined by the Stanford-Binet. The mean mental age 
 is here shown to be 13.4 years, a surprisingly low figure in view of 
 Terman's initial assumption that the average adult has a mental age 
 of 16 years." It requires, however, no argument to show that the 
 group of adults available for his use at the time when this assumption 
 was made, constituted as it was of 32 high school students over 
 
 "Terman, L. M., etc. "The Stanford Revision and Extension of the Binet- 
 Simon Scale for Measuring Intelligence." pp. 49ff. 
 
418 'STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 166 
 
 MENTAL CAPACITY AS MEASURED BY STANFORD-BINET 
 
 Per Cent Distribution of Delinquent Women and of Anny 
 Group, vdth Constants 
 
 
 White, 
 
 English-speaking Cases Only 
 
 • 
 
 Mental Age 
 
 Delinquent Women 
 
 Army Group 
 
 19 to 
 
 20 
 
 .2 
 
 .4 
 
 1.0 
 
 2.2 
 
 6.0 
 
 6.7 
 
 10.5 
 
 13.9 
 
 17.0 
 
 19.9 
 
 13.4 
 
 6.7 
 
 1.3 
 
 .7 
 
 .8 
 
 18.0 to 
 
 19.0 
 
 5.2 
 
 17.0 to 
 
 18.0 
 
 7.2 
 
 16 to 
 
 17 
 
 8.3 
 
 15 to 
 
 16 
 
 9.6 
 
 14.0 to 
 
 15.0 
 
 U.8 
 
 13.0 to 
 
 14.0 
 
 10.6 
 
 12.0 to 
 11 to 
 
 13.0 
 
 12 
 
 12.4 
 10,6 
 
 10.0 to 
 9.0 to 
 
 11.0 
 
 10.0 
 
 10. 1 
 9.5 
 
 8.0 to 
 7.0 to 
 
 9.0 
 
 8 
 
 3.4 
 .2 
 
 6.0 to 
 
 7.0 
 
 .3 
 
 5.0 to 
 
 6.0 
 
 .2 
 
 
 
 
 Total 
 
 100.0 
 
 100.0 
 
 Number of cases 
 
 447 
 
 653 
 
 Mean 
 
 O-m 
 
 11.8 
 ±.106 
 
 13.4 
 ±.111 
 
 
 
 0- 
 
 2.24 
 ±.075 
 
 2.85 
 
 (To- • 
 
 ±.079 
 
 
 
 sixteen years of age and 30 business men of moderate success and very 
 limited educational advantages, is very much fvirther removed from 
 being a random sample of the population than is the army group under 
 consideration. This point is readily admitted by Terman, who has 
 re-formulated his views, in the light of the more recent data, in the 
 following statement written under date of August 20, 1918, which he 
 authorizes us to^ quote : 
 
 "It has become evident, from the results of psychological work in 
 the army, that 'Average Adult' intelligence is considerably lower thaa 
 the score of 16 on the Stanford revision. 
 
 "Because of the numerous factors of selection operative, I hesitate to 
 venture an estimate as to what the real average adult score is. I do hot 
 think it is below 14, but I doubt whether it is much above this figure. 
 
MENTAL CAPACITY 419 
 
 "Even if the true 'Average Adult' score were known, this, of course, 
 would not tell us exactly at what point mental growth ceases, because there 
 is still the question of determining the true average score for unselected 
 children just below the age of maturity. 
 
 "I believe that the peculiar and unsatisfactory conditions under which 
 the army tests have been made tend to make the average score of the 
 recruit lower than it would be under normal testing conditions." 
 
 The above statement shows that Terman judged the army data 
 which we are using, and which constituted a part of the data before 
 
 P«r 
 
 Cent "^^^""^ Army Group 
 
 I ~"| B-^— -■■• Womea DelloqueoU 
 
 11 12 13 
 
 MENTAL AGE 
 
 Chart XXIII 
 
 Mental Capacity as Measured by Stanford-Binet 
 
 Percentage comparison .between delinquent women and army recruits. White, 
 English-speaking cases only. 
 
 him at this time, to err in the direction of under-estimation of the 
 mentality of adults in general, though not to the same degree that 
 the previous figures erred in the opposite direction. 
 
 For comparison with army data we offer, in Table 165, the data 
 on the white, English-speaking women from all six delinquent groups.^ 
 It is evident from consideration both of the distribution and of the 
 means that the delinquent group is inferior mentally to the army 
 group. 
 
 Comparison of the distributions is facilitated by Chart XXIII. 
 While the range of the delinquent group is practically coextensive with 
 
 ' The various groups contribute to this total in the following proportions : 
 Bedford Auburn Magdalen Penitentiary Workhouse Probation 
 217 29 52 56 38 55 
 
 It is evident that elimination of both the colored and the non-English-speak- 
 ing reduces the numbers to very small figures except in the case of Bedford. 
 The large numbers in the latter group are due to the fact that figures on the 
 larger Bedford group (see p. 17) were available. 
 
420 STUDY OF WOMEN DELINOUENTS IN NEW YORK 
 
 that of the army group, it is apparent that the delinquent group is 
 more heavily weighted at the lower end of the scale and less heavily 
 at the upper end. If we take, for example, two arbitrary dividing 
 lines and consider the percentages of each group falling (1) below 10 
 years mental age and (2) above 15 years, we obtain the following 
 
 figures : 
 
 Below 10 years Above IS years 
 mental age mental age 
 
 Delinquent Women 22.1% 3.8% 
 
 Army group 13-6% 21.5% 
 
 ,The difference is quite striking in favor of the army group, 
 especially for the comparison of percentages of individuals who 
 rank high on the intelligence scale. If we consider the proportions 
 falling in the lower portion of the scale, in comparison with Goring's 
 figures (see p. 414), it is at once apparent that our figures, while- 
 quite convincing as to the fact of difference, indicate a far less 
 extreme degree of divergence between the delinquent and the non- 
 delinquent group than do his. Even if allowance be made for the pos- 
 sible slight inferiority of the present army group in comparison with 
 the general population, no such extreme variation as Goring reports 
 would appear. 
 
 Superiority of the army group to the delinquent women ap- 
 pears, also when we compare the means of the two groups. (See 
 Table 166.) The difference in mental age of 1.6 years may be 
 accepted as valid, beyond any reasonable question, since it amounts 
 to more than ten times the standard deviation of the difference. 
 
 TABLE 166 
 
 MENTAL CAPACITY AS MEASURED BY STANFORD-BINET 
 
 Comparison of Means and Standard Deviations of Delinquent Women 
 
 and of Army Group 
 White, English-speaking Cases Only 
 
 
 Delinquent 
 Women 
 
 Army Group 
 
 Difference 
 
 d 
 0-d 
 
 Chances that real 
 
 difierenee doea 
 
 not exist are 1 
 
 in: 
 
 Mean Mental Age. . 
 tTm 
 
 11.8 
 ±.106 
 
 13.4 
 ±.111 
 
 -1.6 
 
 10.4 
 
 m 
 
 
 
 <T 
 
 2.24 
 ±.075 
 
 2.85 
 ±.079 
 
 - .61 
 
 5.6 
 
 00 
 
 fflT .• 
 
 
 
 
 Cases 
 
 447 
 
 653 
 
 
 
 
 
 
MENTAL CAPACITY 
 
 421 
 
 TABLE 167 
 
 MENTAL CAPACITY AS MEASURED BY STANFORD-BINET 
 
 Per Cent Distribution of Delinquent Women by Institutional 
 Groups, with Constants 
 
 English-speaking Cases Only 
 
 
 iNSTITnTIONAL GbOXIPS 
 
 
 Mental Age* 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 Peni- 
 tentiarj' 
 
 Work- 
 house 
 
 Probation 
 
 Total 
 
 19 
 
 .4 
 
 ".8 
 
 1.2 
 
 4.6 
 
 2.7 
 
 9.6 
 
 11.5 
 
 16.2 
 
 21.2 
 
 17.3 
 
 9.6 
 
 3.5 
 
 1.5 
 
 'i'.7 
 
 4.7 
 14.0 
 18.6 
 18.6 
 23.3 
 9.4 
 7.0 
 
 '3.8 
 
 3.8 
 15.4 
 15.4 
 17.3 
 13.4 
 15.4 
 9.6 
 5.7 
 
 'i!3 
 
 1.3 
 
 3.9 
 
 7.8 
 
 11.7 
 
 7.8 
 
 11.7 
 
 15.6 
 
 19.5 
 
 9.1 
 
 6.6 
 
 3.9 
 
 'i'.i 
 
 i'.i 
 
 2.7 
 
 9.6 
 
 12.3 
 
 15.1 
 20.5 
 19,2 
 11,0 
 2,7 
 1,4 
 
 3.4 
 6.7 
 6.7 
 11.7 
 
 13.3 
 16.6 
 18.3 
 18,3 
 1,7 
 1.7 
 
 .2 
 
 .4 
 
 .8 
 
 1 8 
 
 18 
 
 17 
 
 16 
 
 15 
 
 5 1 
 
 14 
 
 5 6 
 
 13 
 
 10 4 
 
 12 
 
 12 9 
 
 11 
 
 16 
 
 10 
 
 20 2 
 
 9 
 
 15 2 
 
 8 
 
 8,0 
 
 7 
 
 2 6 
 
 6 
 
 .9 
 
 5 
 
 
 
 
 Total.... 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100,0 
 
 100.0 
 
 Cases 
 
 260 
 
 43 
 
 52 
 
 77 
 
 73 
 
 60 
 
 565 
 
 
 
 Mean 
 
 11.2 
 + .136 
 
 11.7 
 ±.270 
 
 12.3 
 
 ±.287 
 
 12.1 
 + .283 
 
 11.1 
 ±.253 
 
 11.9 
 
 ±,285 
 
 11.5 
 
 (Tui 
 
 + .094 
 
 
 
 
 2.19 
 
 + .086 
 
 1.77 
 + .191 
 
 2.07 
 ±.203 
 
 2.48 
 ±.200 
 
 2.16 
 ±.179 
 
 2.21 
 ±.201 
 
 2.24 
 
 (To- 
 
 + .067 
 
 
 
 *A mental age of 5 here signifies from 5.0 to 6.0, etc. The mid-value of the intervals are, therefore, 
 S.5, 6.5, etc. 
 
 We note also, by reference to Table 166, that the variability of 
 the army group is wider than that of the delinquent group and that 
 we are justified in accepting this difference as valid. This is especially 
 important in view of the fact that all the other comparisons which 
 we are able to present between delinquent and non-delinquent groups 
 show the opposite tendency with regard to variability (see Tables 
 170 to 174), that is, the delinquent group .shows consistently the 
 wider dispersion. The reduced variability of the other non-delinquent 
 groups would arise naturally from the fact that they are picked, as 
 
422 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 we have noted previously, from a more or less restricted portion of 
 the general population. 
 
 In Table 167 the percentage distribution and constants for the 
 various institutional groups are oiifered. These figures apply to the 
 actual groups as examined, except for the necessary elimination of 
 cases handicapped by language difficulty. In general, throughout 
 this study we have presented the institutional groups as units, without 
 elimination of the colored, because the latter constitute an integral 
 part of the institutional population. It seems desirable, therefore, to 
 offer the data on mentality on the same basis. It should be remem- 
 bered, however, in considering the figures of Table 167 that all groups 
 except the Magdalen include colored women. The final column 
 gives data on the total English-speaking group, which differs from 
 our preceding total group by the presence of the colored. Reference 
 to the figures show that the difference between the two groups is 
 small. The explanation lies in the fact that the colored portion of 
 the group, though appreciably lower mentally than the white (see 
 pp. 480-4), constitutes a relatively small proportion of the total. 
 
 Among the institutional groups two are of particular interest; the 
 Bedford and the Magdalen. The former deserves special attention 
 because the figures offered may be accepted as peculiarly representa- 
 tive of the institution (see pp. 39-40). Not only do they cover larger 
 actual numbers, but they represent a longer period of study, and were 
 obtained under the most satisfactory conditions of any of our groups. 
 Since the mean for the Bedford group (11.2 years) is one of the 
 lowest two, among the institutional groups, it is obvious that the 
 Bedford women would appear even more convincingly inferior to the 
 army group than did the delinquent group previously considered. 
 Table 168 shows that such is the case. (The elimination of the 
 colored women from the Bedford group would not affect the com- 
 parison to any important degree, owing to the small numbers of 
 colored cases. The mean mental age for the Bedford white is 11.4 
 years, differing from the army mean by two years.) The distribution 
 of the Bedford group is shown graphically in Chart XXVII for pur- 
 poses of another comparison. 
 
 The special interest in the Magdalen Home is based on the fact 
 that it ranks highest in mentality among the institutional groups on 
 the present basis of measurement. Comparison with the array group 
 is, therefore, important in order to discover whether any of the 
 groups of delinquent women reach the mental level of the group 
 
MENTAL CAPACITY 
 
 423 
 
 TABLE 168 
 
 MENTAL CAPACITY AS MEASURED BY STANFORD-BINET 
 
 Comparison of Means and Standard Deviations of Army Group 
 and Bedford Women 
 
 English-speaking Cases Only 
 
 
 Bedford 
 Women 
 
 Army Group 
 
 Difference 
 
 d 
 
 Chances that real 
 
 difference does 
 
 not exist are_l 
 
 in: 
 
 Mean Mental Age. . 
 (J-m 
 
 11.2 
 ±.136 
 
 13.4 
 ±.111 
 
 -2.2 
 
 12.6 
 
 00 
 
 (J 
 
 2.19 
 ±.096 
 
 2.85 
 ±.079 
 
 - .66 
 
 5.3 
 
 00 
 
 a„ 
 
 
 
 
 
 260 
 
 653 
 
 
 
 
 
 
 which we are accepting tentatively as representative of adults in 
 general. Table 169 presents the data for this comparison. The 
 difference is still in favor of the "general" group, and, although the 
 amount of difference is reduced to 1.1 years, this is sufficient to justify 
 the assertion of a genuine distinction between the two groups. Since 
 there are no negroes in the Magdalen group no qualification of the 
 results from this point of view is needed. Again the army group 
 is more variable by a significant degree. 
 
 TABLE 169 
 
 MENTAL CAPACITY AS MEASURED BY STANFORD-BINET 
 
 Comparison of Means and Standard Deviations of Army Group 
 and Magdalen Home Group 
 
 White, English-speaking Cases Only 
 
 
 Magdalen 
 Women 
 
 Army Group 
 
 Difference 
 
 d 
 
 Clianoea that real 
 
 difference does 
 
 not exist are 1 
 
 in: 
 
 Mean Mental Age . . 
 ffm 
 
 12.3 
 
 ±.287 
 
 13.4 
 ±.111 
 
 -1.1 
 
 3.6 
 
 5500 
 
 
 2.07 
 ±.203 
 
 2.85 
 .079 
 
 - .78 
 
 3.6 
 
 5500 
 
 cr<r 
 
 
 Cases 
 
 52 
 
 653 
 
 
 
 
 
 
424 .STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 The evidence at hand, therefore, shows all the delinquent groups 
 to be appreciably inferior mentally to the army group and to have a 
 narrower range of variability. If we grant the probable truth of the 
 contention that the army figures fail to do full justice to the general 
 population with respect to mentality, it is apparent that we have 
 understated by at least a slight degree the amount of this difference. 
 Of its existence there can, we think, be no reasonable question. 
 
 We have previously cited the figures on the mental ages of adults 
 offered by Terman (see pp. 417-8). Without lingering over these data 
 we may consider briefly the relationship of our delinquent women to 
 his 62 adults. In view of the evident superiority of his group to 
 adults in general, it is not surprising to find a considerable difference 
 between the means of the two groups in favor of Terman's adults, 
 as well as a decided difference in the degree of variability in favor 
 of the delinquent group (see reference to latter point, pp. 421-2). The 
 constants for Terman's 62 adults are as follows: 
 
 Mean Mental Age = 16.1 ± .146 
 a = 1.15 ± .103 
 
 In view of the recognition by Terman that these figures are not 
 adequately representative of the general population further elabora- 
 tion of the comparison is unnecessary. 
 
 COMPARISON BASED ON YERKES-BRIDGES POINT SCALE ■ 
 
 Turning to the figures based on the data from the Yerkes-Bridges 
 Point Scale we discover four groups of non-delinquent adults avail- 
 able for comparison as follows: (1) 25 male students in the Boston 
 Y. M. C. A., (2) 25 mill operatives, (3) 109 Cincinnati working 
 girls, (4) a combination of the above working girls with 78 girls 
 who had continued in school. The first two groups mentioned can 
 lay no claims to being representative and will, therefore, be con- 
 sidered only in passing. The fifst of these, the Y. M. C. A. student 
 group, obviously consists of superior individuals as Yerkes himself 
 points out.* He states that "each of these individuals had had a 
 partial or complete high school education, and several of them are 
 college graduates." The mean for this group (94.66 ± .926) naturally 
 falls entirely outside the range of means for the delinquent groups, 
 among which the Magdalen shows the highest mean value (76.58 ± 
 
 'Yerkes, R. M., Bridges, J. W. and Hardwick, R. S. "A Point Scale for 
 Measuring Mental Ability," 191S, pp. 91-92. 
 
MENTAL CAPACITY 425 
 
 1.65). The standard deviation on the other hand is conspicuously 
 low, 4.63 ±.685, as compared, for example, with that for Magdalen 
 of 12.03 ±1.166. This narrow variability emphasizes again the re- 
 stricted range of selection of the student group in question. 
 
 The mill operative group cited by Yerkes » shows no such obvious 
 feason for being considered a superior group. We know only 
 that they were mill operatives (male), that their ages fall between 
 17 and 27 and that no one of them had had better than grammar 
 school education and few had completed the work in grammar 
 school. The group is too small to have significance for our purposes, 
 in view especially of its lack of definition. The constants are, 
 however, presented herewith. They show the same direction of 
 difference from the delinquent- groups as do the figures for the student 
 group just discussed. 
 
 YERKES' MILL OPERATIVES 
 
 Mean Score in Points = 88.42 ± L64 
 o-= 8.19 ±1.83 
 Number of Cases ' =25 
 
 The data from the Cincinnati groups of girls constitute our only 
 useful source of comparison based on the Yerkes-Bridges Point 
 Scale. We have obtained, through the courtesy of Dr. Woolley, the 
 records of the testing by this scale of 187 white girls, sixteen, seven- 
 teen and eighteen years of age, made at the Cincinnati Bureau of 
 Vocational Guidance. This is composed of two groups. One group 
 of 109 was taken from the working girls, i.e., those that had left 
 school at sixteen to go to work, and one of 78 from the school 
 group, i.e., those that had continued in school beyond the time 
 when they were legally permitted to leave. This fact in itself, Dr. 
 Woolley believes, indicates that the latter is a superior group, and 
 since the members of the working girl group were picked out as being 
 somewhat inferior, she considers that the combined group "should 
 make a fairly good average for the whole community." Whether or 
 not such a statement is entirely justified, we are interested to see how 
 our data compare, first, with the combination of her groups, and 
 second, with the group of working girls recognized as inferior. All 
 cases in the Cincinnati groups had completed at least the 6th grade. 
 This of itself would tend to eliminate the very inferior. 
 
 Table 170 presents the data on these two groups, and also, for 
 comparison, the figures for the white, English-speaking portion of 
 
 ' Op. cit., pp. 90-91. 
 
426 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 170 
 
 MENTAL CAPACITY AS MEASURED BY YERKES-BRIDGES 
 POINT SCALE 
 
 Per Cent Distribution of Delinquent Women and of Two Groups of 
 
 Cincinnati Girls — ^Working Girls and Combination of Working 
 
 and School Girls — with Constants 
 
 White, English-speaking Cases Only 
 
 
 Delinquent Women 
 
 Cincinnati Gbotjps 
 
 Score in Points • 
 
 Woridng Girls 
 
 Working and 
 School Girls 
 
 96-100 
 
 .8 
 
 6.1 
 
 13.4 
 
 18.4 
 
 15.5 
 
 16.7 
 
 11.1 
 
 7.7 
 
 5.2 
 
 3.1 
 
 1.1 
 
 .6 
 
 .2 
 
 2.8 
 
 11.9 
 
 21.1 
 
 24.8 
 
 16.5 
 
 9.2 
 
 8.3 
 
 3.7 
 
 'i!8 
 
 13.9 
 
 90-96 
 
 27.8 
 
 84-90 
 
 18.7 
 
 78-84 
 
 16.6 
 
 72-78 
 
 9.6 
 
 66-72 
 
 5.4 
 
 60-66 
 
 4.8 
 
 54^60 
 
 2.1 
 
 48-54 
 
 
 42-48 
 
 1.1 
 
 36-^2 
 
 
 30-36 
 
 
 24-30 
 
 
 
 
 Total 
 
 100.0 
 
 100.0 
 
 100.0 
 
 
 
 Cases 
 
 478 
 
 109 
 
 187 
 
 
 
 Mean 
 
 71.80 
 ±.608 
 
 78.83 
 ±1.07 
 
 84 62 
 
 ffm 
 
 + 830 
 
 
 
 a 
 
 13.29 
 ±.416 
 
 11.18 
 
 ±.832 
 
 11 36 
 
 a a 
 
 + 707 
 
 
 
 the total delinquent group. The distributions of these groups are 
 also shown graphically in Chart XXIV. Inspection of the chart shows 
 a marked difference between the delinquent group and each of the 
 Cincinnati groups. We are at once impressed by the greater flatness 
 of the delinquent curve, as well as by the location of the massing. 
 We note that the dot indicating location of the mean falls lower in 
 case of the delinquents, and that the span of the arrows, showing the 
 range of the standard deviation above and below the mean, is wider. 
 It is noteworthy also that the range of the delinquent group runs 
 lower than that of either Cincinnati group. So marked a difference 
 
Per 
 Cent 
 
 DELINQUENT WOMEN 
 
 Per 
 Cent 
 
 Per 
 Cent 
 
 WORKING GIRLS (Cincinnati) 
 
 
 
 
 
 
 
 
 
 ' — 1 
 
 -, — 1,11 — > — r— ^ — P--^-; 
 
 PM^ 
 
 P-J 
 
 24 30 36 
 
 48 54 60 66 72 78 84 90 96 102 
 
 POINTS '^"^ ^ 
 
 L. 
 
 WORKING AND SCHOOL GIRLS (Cincinnati) 
 
 36 
 
 48 54 60 ee 
 
 POINTS 
 
 Chart XXIV 
 Mental Capacity as Measured by Yerkes-Bridges Point Scale 
 
 Per cent distribution of delinquent women and of two groups of Cincinnati 
 girls: (1) working girls, and (2) working and school girls. White, English- 
 speaking cases only, 
 
 427 
 
428 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 in this respect undoubtedly reflects the eliminations from the lower 
 portion of the Cincinnati group involved in the requirement' of 6th 
 grade completion. Turning to statistical comparison to determine 
 how far we are justified in accepting the evidence of inspection, we 
 learn from Table 171 that the inferiority of the delinquent group to 
 both of the Cincinnati groups is established beyond any reasonable 
 question. The difference in variability of the two groups, in favor 
 of the greater spread of the delinquent group, appears also to be 
 almost certainly valid. 
 
 TABLE 171 
 
 MENTAL CAPACITY AS MEASURED BY YERKES-BRIDGES 
 
 POINT SCALE 
 
 Comparison of Means and Standard Deviations of Delinquent Women • 
 
 with those of (i) Cincinnati Working Girls and (2) Working 
 
 and School Girls (Cincinnati) 
 
 White, English-speaking Cases Only 
 
 
 Delinquent 
 Women 
 
 Working 
 Girls 
 
 Difference 
 
 d 
 
 Chances that real 
 
 difference does 
 
 not exist are 1 
 
 in: 
 
 Mean 
 
 71.80 
 ±.608 
 
 78.83 
 ±1.071 
 
 -7.03 
 
 5.7 
 
 00 
 
 0-m 
 
 
 
 
 a 
 
 13.29 
 ±.416 
 
 11.18 
 
 ±.832 
 
 2.11 
 
 2.3 
 
 86 
 
 da 
 
 
 
 
 Cases 
 
 478 
 
 109 
 
 
 
 
 
 
 
 Delinquent 
 Women 
 
 Working and 
 School Girla 
 
 
 
 
 Mean 
 
 (Tm 
 
 71.80 
 ±.608 
 
 84.62 
 ±.830 
 
 -12.82 
 
 12.4 
 
 QC 
 
 
 
 0. .. 
 
 13.29 
 ±.416 
 
 11.36 
 ±.707 
 
 1.93 
 
 2.4 
 
 107 
 
 ffo-. . 
 
 
 
 Cases 
 
 478 
 
 187 
 
 
 
 
 For further comparison with the Cincinnati groups we offer the 
 data for the several units of the total delinquent group (Table 172). 
 As in the case of the Stanford-Binet comparison the Bedford and 
 the Magdalen groups are of special interest to us, and for the reasons 
 
MENTAL CAPACITY 
 
 429 
 
 previously given. The distributions of these two groups are shov^^n 
 graphically in Chart XXV. 
 
 Since Bedford ranks mentally as one of the lowest of the de- 
 linquent groups it is to be expected that it should be even more 
 clearly distinguished from the Cincinnati groups than is the total 
 white delinquent group. Reference to Table 173 shows that even 
 when compared with the poorer of the Cincinnati groups the Bed- 
 ford group is inferior mentally by a significant difiference. 
 
 TABLE 172 
 
 MENTAL CAPACITY AS MEASURED BY YERKES-BRIDGES 
 
 POINT SCALE 
 
 Per Cent Distribution of Institutional Groups, with Constants 
 
 English-speaking Cases Only 
 
 
 Institdtional Groops 
 
 
 Score in Points 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Probation 
 
 Total 
 
 96-100 
 
 90-96 
 
 84-90 
 
 .4 
 
 3.2 
 
 11.6 
 
 13.3 
 
 15.1 
 
 19.6 
 
 11.9 
 
 9.1 
 
 7.4 
 
 5.3 
 
 1,4 
 
 1.4 
 
 .4 
 
 '2.Z 
 
 20.5 
 
 25.0 
 
 13.6 
 
 9.1 
 
 13.6 
 
 9.1 
 
 2.3 
 
 2.3 
 
 2.3 
 
 1.9 
 
 5.7 
 
 26.4 
 
 24.5 
 
 13.2 
 
 9.4 
 
 7.5 
 
 5.7 
 
 3.8 
 
 1.9 
 
 14.0 
 14.0 
 15.0 
 
 12.7 
 7.6 
 
 10.1 
 6.3 
 2.5 
 1.3 
 
 2.7 
 
 ■gis 
 
 18.7 
 13.3 
 10.7 
 
 9.3 
 13.3 
 
 8.0 
 12.0 
 
 1.3 
 
 i'.Z 
 
 k'.h 
 
 8.5 
 
 22.5 
 
 18.3 
 
 12.7 
 
 15.5 
 
 8.5 
 
 2.8 
 
 1.4 
 
 1.4 
 
 .7 
 
 5.3 
 
 13.2 
 
 78-84 
 
 17.0 
 
 72-78 
 
 15.0 
 
 66-72 
 
 15.2 
 
 60-66 
 
 11.2 
 
 54-60 
 
 9.4 
 
 48-54 
 
 6.1 
 
 42-48 
 
 4.8 
 
 36-42 
 
 1.3 
 
 30-36 
 
 .7 
 
 24-30 
 
 .3 
 
 
 
 Total .... 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 100.0 
 
 Cases 
 
 285 
 
 44 
 
 53 
 
 79 
 
 75 
 
 71 
 
 607 
 
 
 
 Mean 
 
 68.53 
 ±.811 
 
 72.73 
 ±1.877 
 
 76.58 
 ±1.650 
 
 73.64 
 + 1.629 
 
 66.61 
 ±1.706 
 
 72.61 
 ±1.441 
 
 70.44 
 
 Cm 
 
 ±.563 
 
 (T 
 
 13.69 
 + .535 
 
 13.45 
 ±1.280 
 
 12.03 
 ±1.166 
 
 14.48 
 ±.908 
 
 14.78 
 ±.972 
 
 12.14 
 ±.928 
 
 18.88 
 
 (J a 
 
 ±360 
 
 
 
 A very different situation exists, however, in case of the Magdalen 
 group. The appearance of the distribution curve (Chart XXV) is 
 markedly different from that of the other delinquent groups, with 
 
430 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 the sudden piling up of cases toward the upper portion of the scale. 
 The range of the Magdalen cases extends no lower than that of 
 the Cincinnati groups. The mean falls lower in the Magdalen group 
 than the mean for either of the other two but the difference is not 
 
 Per 
 
 Cent 
 
 BEDFORD 
 
 Per 
 
 Cent 
 
 Chart XXV 
 
 Mental Capacity as Measured by Yerkes-Bridges Point Scale 
 
 Per cent distribution of Bedford and Magdalen institutional groups— English- 
 speaking cases only. 
 
 great enough so that we can determine by inspection whether or 
 not it might have occurred by chance. From Table 174 we learn that 
 we need not question the validity of the difference between the 
 means of the Magdalen and the combined group of school and work- 
 ing girls. In comparison with the Cincinnati working girl grouf 
 
MENTAL CAPACITY 
 
 431 
 
 TABLE 173 
 
 MENTAL CAPACITY AS MEASURED BY YERKES-BRIDGES 
 
 POINT SCALE 
 
 Comparison of the Means and Standard Deviations of Bedford Women 
 
 with those of (i) Cincinnati Working Girls and (2) Working 
 
 and School Girls (Cincinnati) 
 
 English-speaking Cases Only 
 
 
 Bedford 
 
 Working 
 
 Girls 
 
 (Cincinnati) 
 
 Difference 
 
 d 
 
 Chances that real 
 
 difference does 
 
 not exist are 1 
 
 in: 
 
 Mean 
 
 68.53 
 ±.811 
 
 78.83 
 ±1.071 
 
 -10.30 
 
 7,67 
 
 
 (Tni 
 
 
 
 
 (T .. 
 
 13.69 
 ±.535 
 
 11.18 
 ±.832 
 
 2.51 
 
 2.54 
 
 182 
 
 <Ta 
 
 
 
 Cases 
 
 285 
 
 109 
 
 
 
 
 
 
 
 Bedford 
 
 Working and 
 Scliool Girls 
 
 
 
 
 Mean 
 
 68.53 
 ±.811 
 
 84.62 
 ±.830 
 
 -16.090 
 
 13.86 
 
 00 
 
 ffm 
 
 
 
 
 (7 
 
 13.69 
 ±.535 
 
 11.36 
 ±.707 
 
 2.34 
 
 2.64 
 
 244 
 
 (Tff 
 
 
 
 
 Cases 
 
 285 
 
 187 
 
 
 
 
 
 
 however, the slight difference between the means can not be stressed 
 as significant, since, as great a difference as this might occur, by 
 chance, as often as 1 in 8 times. No demonstrable difference in 
 mentality appears, therefore, between the best of the delinquent 
 ' groups and the poorest of the non-delinquent. This is not at variance 
 with our other findings, since the extensive overlapping of range of 
 delinquent and non-delinquent groups has been apparent in all our 
 comparisons. We should anticipate that, if we increased the num- 
 ber and variety of our samples of non-delinquents, we should find 
 groups appreciably lower than certain of the delinquent groups. 
 
 Table 174 shows also that there is no demonstrable difference in 
 variability between the Magdalen and either of the Cincinnati groups. 
 
432 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 COMPARISON BY MEANS OF WOOLLEY SERIES 
 
 As we stated earlier in this chapter, we shall not attempt to make 
 any comparison of the results obtained by us, through application 
 of the Woolley series of tests to delinquent groups, with Dr. Wool- 
 ley's results from working girls, since an extended study along these 
 lines has already appeared from this laboratory. As indicative of 
 the general results of this comparison, we quote the following 
 paragraph : 
 
 "Thus when the mentality of the criminal women is measured by the 
 number of tests in which their ability is at or above that of the median 
 working girl of fifteen, again about a third of them (35 per cent) fall 
 within the range of the better fifty per cent of the working girls of 
 
 TABLE 174 
 
 MENTAL CAPACITY AS MEASURED BY YERKES-BRIDGES 
 
 POINT SCALE 
 
 Comparison of Means and Standard Deviations of Magdalen Home 
 
 Women with Those of (i) Cincinnati Working Girls and (2) 
 
 Working and School Girls (Cincinnati) 
 
 White, English-speaMng Cases Only 
 
 
 Magdalen 
 
 Working 
 Girls 
 
 Difference 
 
 d 
 (7a 
 
 Chances that real 
 
 difference does 
 
 not exist are 1 
 
 in; 
 
 Mean 
 
 76.58 
 ±1.660 
 
 78.83 
 ±1.071 
 
 -2.25 
 
 1.13 
 
 8 
 
 (Tm 
 
 
 
 
 a , 
 
 12.03 
 ±1.166 
 
 11.18 
 
 ±.832 
 
 .85 
 
 .59 
 
 
 (7<r 
 
 4 
 
 
 
 
 53 
 
 109 
 
 
 
 
 
 
 
 Magdalen 
 
 Working and 
 School Girls 
 
 
 
 
 Mean 
 
 76.58 
 ±1.650 
 
 84.62 
 ±.830 
 
 -8.04 
 
 4.4 
 
 147,000 
 
 (Tm 
 
 
 
 a 
 
 12.03 
 ±1.166 
 
 11.36 
 ±.707 
 
 .67 
 
 .49 
 
 3 
 
 <7„ 
 
 
 
 
 Cases 
 
 53 
 
 187 
 
 
 
 
MENTAL CAPACITY 433 
 
 fifteen, about two-thirds below the median, 54 per cent poorer than all 
 but the poorest 25 per cent of the standard, 30.7 per cent as poor as 
 the poorest 5.7 per cent and 17 per cent poorer than any of the working 
 girls."" 
 
 It is evident that the data from this source are likewise con- 
 firmatory of the trend toward a relatively low average mentality in 
 the delinquent group. 
 
 SUMMARY 
 
 Summarizing, then, the results of such comparisons as we have 
 been able to make of the mental capacity of delinquent women with 
 that of non-delinquents, it appears : 
 
 (1) That the average mental capacity of the delinquent women 
 whom we have examined is lower than that of any groups of non- 
 delinquent adults with regard to whom we have data.^^ 
 
 (2) That, however, the above statement does not imply a selec- 
 tion of individuals entirely from the lower end of the scale of in- 
 telligence for the delinquent group. There is, in fact, an extensive 
 amount of overlapping of the delinquent with the non-delinquent 
 groups. The range of the delinquent group was found to be prac- 
 tically coextensive with that of the army group, our most representa- 
 tive sampling of the general population. Further, the difference be- 
 tween the means of the delinquent and the non-delinquent groups, 
 while affording adequate indication of a distinction between the two 
 groups, is not extreme in amount. In other words, this is definitely 
 not a case of an "all or none" relationship. 
 
 (3) The data are somev/hat ambiguous regarding the relative 
 variability of the two groups. From consideration of our most repre- 
 sentative material, the army data, we conclude that the variability 
 of the delinquent group is less than that of the adult population in 
 general. Reference to the other non-delinquent groups, however, 
 shows that the variability is in all these cases lower than that of the 
 delinquent group. The evidence at hand, therefore, indicates that a 
 group of non-delinquent adults, picked at random from any spe- 
 cialized segment of the population, is likely to show a narrower 
 dispersion than a group of adult delinquents. Our data are, however, 
 
 " Op. cit., p. 274. 
 
 " It should be remembered that the statements of this summary refer to the 
 white, English-speaking portions of the delinquent and the non-delinquent 
 groups. 
 
434 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 far too slight to justify us in maintaining that this rule will always 
 hold. Fundamentally, of course, it would depend upon the breadth 
 of the category which determined the basis of selection of the non- 
 delinquent group. 
 
 It is evident from the foregoing statement that our findings are 
 in accord with Goring's as regards the fact of a difference and its 
 direction. They indicate, however, a slighter degree of difference than 
 he implies. They fail absolutely to justify the view expressed recently 
 by certain propagandists that delinquency and defective intelligence 
 are practically synonymous, and that, accordingly, solving the prob- 
 lem of mental deficiency will solve the problem of delinquency. 
 
CHAPTER XIV 
 
 MENTAL CAPACITY: INTER-COMPARISON OF DELIN- 
 QUENT GROUPS 
 
 IN the preceding chapter we have endeavored to discover the re- 
 lationship between our group of dehnquent women and the general 
 population, with respect to intelligence. In the present chapter we 
 shall be concerned with a comparisbn of different delinquent groups 
 with one another, a problem of importance from several points of 
 view. If differences, significant in amount, are found to exist in the 
 case of delinquent groups now available for consideration, a knowl- 
 edge of the amounts of such differences may suggest the nature 
 of the limitations to be imposed upon the universality of conclusions 
 based on a study of any particular group or combination of groups. 
 Further, it is important to know whether differences of any consid- 
 erable size are found between the different types of groups under 
 consideration, and also to know whether the institutions which we 
 have chosen to represent particular types of institutions bear a close 
 resemblance to others of these types. It is also desirable, wherever 
 possible, to compare groups of women delinquents with corresponding 
 groups of men delinquents, with a view to discovering to what extent 
 any particular findings must be qualified as valid for offenders of 
 one sex but not for those of the other sex. 
 
 The various comparisons suggested above will be carried out, so 
 far as the available data permit, in the present chapter. Scantiness 
 of data for comparison, however, is such as greatly to limit the 
 thoroughness of our treatment. 
 
 COMPARISON OF GROUPS CONSIDERED IN PRESENT INVESTIGATION 
 
 The first step in the comparison of different delinquent groups 
 with respect to general intelligence is the inter-comparison of the 
 six groups which constitute the units from which our total is made 
 up. As the simplest method of indicating whether intelligence had 
 been a dominating factor in determining the disposition of cases, we 
 
 435 
 
s 
 
 S6 <n E6 
 
 86 o» 16 
 
 16 o* 68 
 
 !0»Z8 
 
 Z8 o» S8 
 
 S8 o» E8 
 
 88 o? 18 
 
 U °% 91 
 
 91 o% SI 
 
 SI o» U 
 
 69 o% 19 
 
 19 0% 69 
 
 S9 o% 89 
 
 89 o* T9 
 
 T9 o» 6B 
 
 6S o» ZS 
 
 ZS o» BS 
 
 fiS o» SS 
 
 ES 0% 19 
 
 TB o? 6^ 
 
 m o» i» 
 
 ^^i-4t«eoio 
 
 W «rt^ 
 
 paTj^TjiN-^fN 
 
 ^lOTt^«lN^» 
 
 CO MMt»« 
 
 XSO%Gl ■*«3tti(M«3M 
 
 6Z o^ ZZ o> ■* "5 »o i« CO 
 
 tl Tj< iH 00 ■* »o 
 
 U3(N ■*««■* 
 
 N-^COW-^CO 
 
 ^^iHiOfOCD 
 
 COCO CO CO US CO 
 
 O ^NCO 
 
 CD r-t iH ^ U5 f-t 
 
 « P oa g o o 
 
 
 436 
 
 
 II § 
 
 b ci 
 O 
 
 t> to 
 
 »o £ 
 
 III 
 il 
 
 ^^ 
 '"■§ 
 
 ca f^ 
 Uc3 
 
 £3 
 
 0} o 
 
 HO 
 
MENTAL CAPACITY 
 
 437 
 
 computed a correlation ratio with Test Aggregate as the quantitative 
 variable and the six groups of delinquent women as a series of 
 qualitative arrays. The relationship between intelligence and type of 
 sentence is shown in Table 175, from which a correlation ratio of .17 
 ±.043 was obtained. This would indicate that the degree of relation- 
 ship is extremely slight, but significant. With a view to further 
 analysis of these differences we have made more detailed inter-com- 
 parisons among the groups. 
 
 (a) Coniiparison of Total English-Speaking Portions of Groups 
 
 The seven figures of Chart XXVI show the percentage distribution 
 of scores on Test Aggregate for each of the six groups and for 
 the combined total. (These distributions appear in numerical form 
 in Table 175.) An inspection of these shows the most notice- 
 able items to be the wide distribution of the Penitentiary and, to 
 a lesser extent, of the Bedford group; the relatively low, flat curves 
 of these two groups and of the Workhouse as compared with the 
 marked concentration about the mode shown by the Magdalen ; the 
 lack of representation at the poor end for the Magdalen group ; the 
 small percentage found in the high end of the Workhouse group ; and 
 differences in the location of the mean, indicated by the position of 
 the dot on each curve. (Table 176 gives the numerical values 
 of the means and standard deviations for these groups.) We see 
 
 Per 
 Cent 
 
 TOTAL 
 
 Chart XXVI 
 
 Mental Capacity as Measured by Test Aggregate 
 
 Per cent distribution by institutional groups. 
 
Per 
 
 Cent 
 
 Per 
 
 Cent 
 
 Per 
 
 Ceiit 
 
 BEDFORD 
 
 AUBURN 
 
 93 97 
 
 XKVI C 
 
 MAGDALEN 
 
 49 
 
 53 57 61 65 69 73 77 81 85 
 
 SCORE 
 
 Chart XXVI (Continued) 
 
 Mental Capacity as Measured by Test Aggregate 
 
 Per cent distribution by institutional groups. 
 
 438 
 
 97 
 KHVI 
 
Cent 
 
 Pw 
 
 Cent 
 
 PENITENTIARY 
 
 WORKHOUSE 
 
 93 97 
 
 xKvt e 
 
 Per 
 Cent 
 
 49 5? s; 61 
 
 «5 69 73 77 
 
 SCORE 
 
 PROBATION 
 
 , , n 
 
 45 49 53 57 61 65 69 73 77 81 85 89 93 97 
 
 SCORE ■"'^' * 
 
 /Chart XXVI (Continued) 
 Mental Capacity as Measured by Test Aggregate 
 Per cent distribution by institutional groups. 
 439 
 
440 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 176 
 
 MENTAL CAPACITY MEASURED BY TEST AGGREGATE 
 
 Comparison of Constants by Institutional Groups 
 
 English-speaking Cases Only 
 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 Peni- 
 tentiary 
 
 Work- 
 bouse 
 
 Probation 
 
 Total 
 
 Mean 
 
 75.05 
 
 ±.594 
 
 75.38 
 + 1.39 
 
 79.10 
 + 1.06 
 
 76.72 
 
 ±1.24 
 
 73.31 
 ±1.09 
 
 77.49 
 ±1.19 
 
 75 65 
 
 (Tm 
 
 + 406 
 
 
 
 (T 
 
 9.12 
 + .376 
 
 9.01 
 + 1.30 
 
 7.52 
 + .720 
 
 10.67 
 
 ±.738 
 
 9.08 
 ±.650 
 
 8.30 
 ±.763 
 
 9 27 
 
 da 
 
 + 266 
 
 
 
 
 236 
 
 42 
 
 50 
 
 74 
 
 69 
 
 49 
 
 520 
 
 
 
 that the averages for mental capacity run in order of excellence, 
 Magdalen, Probation, Penitentiary, Auburn, Bedford and the Work- 
 house, with the widest steps coming between the Magdalen and 
 Probation and between- Bedford and the Workhouse. The Magdalen 
 and the Workhouse seem then to be indicated as clearly superior 
 and clearly inferior groups respectively. As we have suggested, the 
 Magdalen group would seem to owe its excellence not so much to 
 a piling up of the most superior cases, i.e., those in the highest two 
 intervals, for example, but rather to the lopping off of the inferior 
 end and a high concentration of cases which are well above the 
 average. The Workhouse in its turn would seem to have derived 
 its inferiority from its small percentage of superior women and its 
 uniform distribution above and below the average rather than from 
 any unusual representation in the lowest group. The Penitentiary 
 and Bedford groups include a wide range of cases from those very 
 low to those very superior. They differ from each other in that 
 while Bedford is sparsely represented in the most superior region, 
 the Penitentiary has a greater percentage of cases here than is found 
 even in the Magdalen group. Auburn and the Probation group are 
 alike in that they both show some evidences of bimodality, but the 
 numbers are too few to attempt to draw conclusions. 
 
 In order to find out whether these differences are valid or whether 
 they are only such variations as might occur from sampling, we 
 have made inter-comparisons of the means and standard deviations 
 
MENTAL CAPACITY 
 
 441 
 
 of these prison groups. We print below a table (Table 177) showing 
 the inter-relations expressed in terms of the ratio of the difference of 
 the means to the standard deviation of this difference. (As earlier 
 stated [see p. 35] we are considering as indicative of a valid difference 
 a ratio of two or more times the standard deviation.) 
 
 From this table we see that a difference sufficient to be re- 
 garded as almost certainly valid occurs in only five of the com- 
 parisons, namely: between the Magdalen and Bedford, the Mag- 
 dalen and the Workhouse, the Magdalen and Auburn, the Work- 
 house and Probation, and the Workhouse and the Penitentiary. From 
 this we would infer that intellectual capacity, either primarily or 
 secondarily, may have been a factor in determining the disposition 
 of a case where there was a choice between these institutions. On 
 the whole we note that the institutional groups are mutually over- 
 lapping and that no one group is demonstrably superior or inferior 
 to all the others. As a possible reason to account for the relative 
 superiority of the Magdalen and inferiority of the Workhouse groups, 
 we may say that the Magdalen is the only one of the institutions 
 studied which may itself exercise any choice in the matter of the 
 inmates it receives. The Magdalen, being a private institution, has 
 the privilege of refusing to receive, or of returning as undesirable, 
 cases which seem unpromising. It seems possible, therefore, that 
 this might function as a selective factor in that it would tend to 
 
 TABLE 177 
 
 MENTAL CAPACITY MEASURED BY TEST AGGREGATE 
 
 Inter-comparison of Means of Institutional Groups in Terms of the 
 
 Ratio of the Difference between the Means to the Standard 
 
 Deviation of the Difference 
 
 
 Auburn 
 
 Magdalen 
 
 Penitentiary 
 
 Workliouse 
 
 Probation 
 
 Bedford 
 
 -.22 
 
 -3.33 
 -2.07 
 
 -1.21 
 
 - .18 
 
 1.46 
 
 1.40 
 1.17 
 3.f3 
 
 2.07 
 
 -1.84 
 
 Auburn 
 
 - .12 
 
 Magdalen 
 
 .10 
 
 Penitentiary 
 
 - .45 
 
 Workhouse 
 
 -2.60 
 
 
 
 Explanation of table: The number in each space gives the value for -^ for the 
 
 two institutional groups designated at the head of the column and to the left of the 
 row, respectively. A minus sign indicates that the mean of the institutional group 
 at the left is smaller than the mean of the group heading the column. 
 
442 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 raise the standard mentally as well as otherwise of the girls com- 
 mitted tliere. 
 
 There is another factor also which may conceivably act as a selec- 
 tive agent to account both for the superiority of the Magdalen and 
 the inferiority of the Workhouse. This is the matter of color. There 
 are no colored women committed to the Magdalen, while the Work- 
 house contains a larger percentage than any of the other groups, 36 
 per cent, as against 20 per cent at the Penitentiary, 20 per cent at 
 Bedford, 17 per cent at Auburn, and 9 per cent on Probation. As 
 we shall later show (Table 201), there is a large and unquestionably 
 significant difference between the means of the colored and the white 
 women. 
 
 There is the further fact, which might well carry weight with 
 judges familiar with conditions in the various institutions, that the 
 Workhouse offers least, in the way of opportunities for training, of 
 any of the institutions, with the possible exception of the Penitentiary. 
 
 Considering now the amount of variability in each group as 
 shown by their respective standard deviations, we find that while 
 the evident order runs Penitentiary, Bedford, Workhouse, Auburn, 
 Probation, Magdalen, Table 178 shows that there exists a demon- 
 strably valid difference only between the Penitentiary and Probation 
 and between the Penitentiary and the Magdalen. 
 
 TABLE 178 
 
 MENTAL CAPACITY MEASURED BY TEST AGGREGATE 
 
 Inter-comparison of Standard Deviations of Institutional Groups in 
 
 Terms of the Ratio of the Difference between the Means to 
 
 the Standard Deviation of the Difference 
 
 
 Auburn 
 
 Magdalen 
 
 Penitentiary 
 
 Workhouse 
 
 Probation 
 
 Bedford 
 
 .08 
 
 1.97 
 1.00 
 
 -1.85 
 -1.11 
 -3.05 
 
 .02 
 
 - .05 
 
 -1.62 
 
 1.62 
 
 97 
 
 
 .75 
 
 
 — 74 
 
 Penitentiary 
 
 2 23 
 
 
 78 
 
 
 
 Explanation of table: The number in each space gives the value for 7=- for the 
 
 two institutional groups designated at the head of the column and to the left of the 
 row, respectively. A minus sign indicates that the standard deviation of the in- 
 stitutional group at the left is smaller than that of the group heading the column. 
 
MENTAL CAPACITY 
 
 443 
 
 It seems possible that here again the factor of the ehmination of 
 the colored from the Magdalen and their practical elimination from 
 the Probation group may explain the facts. 
 
 (b) Comparison of Institutional Groups: White Women Only 
 
 In order to be certain whether or not the question of color 
 was a significant factor in determining the degree of difference in 
 mental capacity of these groups, we have calculated new frequency 
 tables, using only the white women of each group. This has un- 
 fortunately made our numbers very small in some groups. 
 
 Table 179 shows the means and standard deviations of each group. 
 The most noticeable change is, as we should expect, the raising of 
 the means of the Workhouse and Penitentiary. 
 
 TABLE 179 
 
 MENTAL CAPACITY MEASURED BY TEST AGGREGATE 
 
 Comparison of Constants by Institutional Groups 
 
 English-speaking White Women Only 
 
 
 Bedford 
 
 Auburn 
 
 Magdalen 
 
 Peni- 
 tentiary 
 
 Work- 
 house 
 
 Probation 
 
 Total 
 
 Mean 
 
 76.00 
 + .647 
 
 75.64 
 + 1.682 
 
 79.10 
 ±1.06 
 
 79.07 
 ±1.323 
 
 76.58 
 
 ±1.497 
 
 77.98 
 ±1.214 
 
 77 03 
 
 (Tm 
 
 + 442 
 
 
 
 
 8.96 
 + .457 
 
 8.90 
 ±1.190 
 
 7.52 
 ±.720 
 
 9.72 
 ±.935 
 
 8.98 
 ±1.058 
 
 8.23 
 
 ±.859 
 
 8.91 
 
 O-ff 
 
 ±.305 
 
 Cases. 
 
 192 
 
 28 
 
 50 
 
 54 
 
 36 
 
 46 
 
 406 
 
 
 
 Table 180 shows the significance of the difference between the 
 means of the various white groups in the same way that Table 177 
 showed 'the validity of the difference between the totals, ie., com- 
 bined colored and white, of each group. 
 
 From this it appears that, when we eliminate the colored women 
 from the comparison, we are no longer able to demonstrate the ex- 
 istence of a difference between the Magdalen and Auburn, the Mag- 
 dalen and the Workhouse, or the Workhouse and the Probation 
 group. Reference to Table 179, in relation to Table 176, makes it 
 clear, however, that the differences referred to have not been eradi- 
 cated by elimination of the colored but merely reduced in amount. 
 
444 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 180 
 
 MENTAL CAPACITY MEASURED BY TEST AGGREGATE 
 
 Inter-comparison of Means of White Women of Institutional Groups in 
 
 Terms of the Ratio of the Difference between the Means to 
 
 the Standard Deviation of the Difference 
 
 
 Auburn 
 
 Magdalen 
 
 Penitentiary 
 
 Workhouse 
 
 Probation 
 
 Bedford 
 
 .20 
 
 -2.50 
 -1.74 
 
 -2.09 
 
 -1.60 
 
 .02 
 
 -.36 
 
 -.42 
 1,37 
 1.25 
 
 — 1.44 
 
 Auburn 
 
 — 1 13 
 
 Magdalen . 
 
 70 
 
 Penitentiary 
 
 .61 
 
 Workhouse 
 
 — 72 
 
 
 
 Explanation of table: The number in each space gives the value for -— - for the 
 
 two institutional groups designated at the head of the column and to the left of the 
 row, respectively. A minus sign indicates that the value of the constant for the in- 
 stitutional group at the left is smaller than that of the group heading the column. 
 
 The validity of the distinction has been affected by the reduction in 
 numbers involved in dropping the colored as well as by the reduction 
 in amount of difference. 
 
 There still remains, however, a difference which is almost cer- 
 tainly valid between the mental capacity of the Magdalen women 
 and that of the white women sent to Bedford. Apparently it is 
 not the presence of the 16 per cent of colored women in the Bedford 
 group which makes it inferior to the Magdalen group. 
 
 There appears also a valid difference, which the presence of 
 the colored women obscured in the previous comparison, between the 
 Penitentiary and the Bedford groups. Elimination of the figures for 
 the colored women from the total for the Penitentiary group raises 
 the mean from 76.72 to 79.07, while the mean for the Bedford 
 group is raised from 75.05 only to 76.00 by the corresponding change. 
 In possible explanation of the superiority, with regard to intelligence, 
 of the white women of the Penitentiary over the white women at 
 Bedford the following points may be noted. 
 
 Table 4 in Chapter VI shows that of the women at Bedford there 
 are 62.7 per cent convicted for offenses against chastity and 16.7 
 per cent for offenses against property, while at the Penitentiary there 
 are only 6.4 per cent convicted for offenses against chastity and 
 56.4 per cent for offenses against property. In the following chapter 
 it is shown that there is a difference in the mental capacity of the 
 
MENTAL CAPACITY 
 
 445 
 
 chastity and property offenders which is in favor of the latter. It 
 would seem, then, that the difference in mentality between Bedford 
 and the Penitentiary may simply reflect this difference in proportions 
 of the two types of offenders mentioned above. 
 
 On the matter of the variability of the group, when we are con- 
 sidering only the white women of each group, we see from Table 
 181 that we lose the distinctions which were seen to be present in the 
 other table (Table 178), between the Penitentiary and the Magdalen, 
 and the Penitentiary and the Probation group. When we consider, 
 then, just the white women oT these groups, we can see no difference 
 in the degree of variability displayed by each. Reduction in number 
 of cases affects this comparison as well as that of the means. 
 
 TABLE 181 
 
 MENTAL CAPACITY MEASURED BY TEST AGGREGATE 
 
 Inter-comparison of Standard Deviations of White Women in Institu- 
 tional Groups in Terms of the Ratio of the Difference 
 between the Means to the Standard Deviation 
 of the Difference 
 
 
 Auburn 
 
 Magdalen 
 
 Penitentiary 
 
 Workhouse 
 
 Probation 
 
 Bedford 
 
 .05 
 
 1.69 
 .99 
 
 - .73 
 
 - .54 
 -1.87 
 
 - .02 
 
 - .05 
 -1.14 
 
 .52 
 
 75 
 
 Auburn 
 
 45 
 
 Magdalen 
 
 — 64 
 
 Penitentiary 
 
 1 17 
 
 Workhouse 
 
 55 
 
 
 
 Explanation of table: The number in each space gives the value for -=- for the 
 
 I'd 
 
 two institutional groups designated at the head of the column and to the left of the 
 row, respectively. A minus sign indicates that the value of the constant for the in- 
 stitutional group at the left is smaller than that of the group heading the column. 
 
 SUMMARY 
 
 We have found only a small relationship between mental capacity 
 and the institution to which the women were committed. Inter- 
 comparison of the groups shows that the means run in order of ex- 
 cellence: Magdalen, Probation, Penitentiary, Auburn, Bedford and 
 Workhouse. An effort to prove the validity of these differences 
 shows that they are significant in only four instances, i.e., between 
 Magdalen and Auburn, Magdalen and Bedford, Magdalen and Work- 
 house, and Workhouse and Probation. Eliminating the colored 
 
446 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 women, whose mental capacity is markedly inferior to the white, we 
 found that a demonstrably valid distinction no longer appeared be- 
 tween the mental capacity of the Magdalen group and that of the white 
 women at Auburn and at the Workhouse; and between the white 
 women at the Workhouse and those on Probation. The difference re- 
 mains as almost certainly valid, however, between the Magdalen group 
 and the white women at Bedford, and a difference previously obscured 
 appeared between the white women at Bedford and those at the 
 Penitentiary in favor of the latter. We suggest for this latter dis- 
 tinction the fact that the Bedford group is made up largely of chastity 
 offenders, and the Penitentiary of property offenders, between whom 
 a difference in mental capacity is found to exist. The difference 
 between the mental capacity of the Magdalen and the white women 
 at Bedford, both of which institutions contain a large percentage of 
 chastity offenders (Bedford 62.7 per cent and Magdalen 77.6 per 
 cent), we attribute possibly to the selection privileges possessed by 
 the Magdalen Home. The difference in the amount of variability 
 displayed in the different groups was seen to be significant only be- 
 tween the Penitentiary and the Magdalen, and the Penitentiary and 
 Probation groups. When the colored women are eliminated from 
 the groups there are no distinctions which approach the limits of cer- 
 tain validity. 
 
 It must not be forgotten that, while the presence of the colored 
 women explains many of these distinctions between groups, they are 
 themselves intrinsic parts of the groups in which they are present 
 and must not be thought of as extraneous when the comparison is 
 between institutional groups per se. 
 
 The main outcome of our comparison of institutional groups is 
 the indication that the similarity of the various groups, as regards 
 both mental level and variabiHty, is more striking than their differ- 
 ences. This similarity of the groups has an important bearing on the 
 general significance of the figures on the mentality of the total group. 
 We have noted, in Chapter II, that our total group does not contain 
 representatives of the women of the different types of institutions 
 in the proportions in which they actually occur in the total delinquent 
 population. Certain groups within our total accordingly exercise an 
 undue weight on the general result. This can not, however, appre- 
 ciably distort the results as to mentality, when the inter-institutional 
 differences are no greater than we find them. 
 
MENTAL CAPACITY 
 
 COMPARISON OF OUR DELINQUENT GROUPS WITH OTHERS 
 
 447 
 
 Turning now to a comparison of our delinquent women with other 
 groups of delinquent adults, we shall consider such groups as we 
 have been able to discover of either men or women, which are com- 
 parable in type with the delinquent groups which we have studied 
 and which have been examined by the same test methods. 
 
 (a) Bedford and Albion Reformatories 
 
 In one of the publications of the Bureau of Analysis and In- 
 vestigations" there is contained a brief report by J, L. Herrick 
 on the testing of 194 inmates at the Western House of Refuge at 
 Albion, New York. These women range in age from sixteen to forty 
 years and are directly comparable with our Bedford women, as 
 the Western House of Refuge handles the same types of cases from 
 the western part of New York State that Bedford does for the east- 
 ern. The Stanford-Binet was given among other tests and Dr. 
 Herrick prints a table of frequencies by mental ages in years. (For 
 corrsponding data on Bedford group, see Table 167.) Chart XXVII 
 presents graphically the distributions of 343 Bedford cases and 194 
 cases from the Albion Reformatory. 
 
 It will be seen that the groups suggest each other in general out- 
 line, the spread of the distribution being very similar, with a standard 
 
 "Herrick, J. L. "Mental Examinations." P. 7, Eugenics and Social Wel- 
 fare Bulletin, No. XI. Publication of the Bureau of Analysis and Investiga- 
 tion. 1917. 
 
 TABLE 182 
 
 Comparison of the Means and Standard Deviations of the Mental 
 
 Capacity Measured by Stanford-Binet Mental Age between 
 
 Bedford Women and Women from the Western 
 
 House of Refuge at Albion 
 
 
 Bedford 
 Women 
 
 Albion 
 Women 
 
 Difference 
 
 d 
 tTd 
 
 Chances that real 
 
 difference does 
 
 not exist are 1 
 
 in: 
 
 Mean 
 
 11.11 
 + .136 
 
 11.60 
 + .155 
 
 -.49 
 
 2.39 
 
 119 
 
 (Tm 
 
 
 
 
 <T 
 
 2.2 
 + .10 
 
 2.16 
 + .110 
 
 .04 
 
 .260 
 
 3 
 
 <^a... 
 
 
 
 
 Cases 
 
 261 
 
 194 
 
 
 
 
 
 
448 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 deviation of 2.20±.096 for one as against 2.16±.110 for the other. 
 There is a difference of almost half a year in the central tendency, 
 11.60+.155 as against 11.11±.136, and this difference is in favor of 
 the Albion women. Table 182 shows the degree of validity that may 
 
 Per 
 Cent 
 
 BEDFORD 
 
 Per 
 
 Cent 
 
 ALBION 
 
 12 13 14 
 
 MENTAL AGE 
 
 Chart XXVII 
 
 Mental Capacity as Measured by Stanford-Binet 
 
 Percentage comparison between Bedford Reformatory women and Albion 
 Reformatory women — English-speaking cases only. 
 
 be attached to these differences. It is evident that the slight amount 
 of difference between the two standard deviations can be accounted 
 for in terms of chance, but that the half year's difference between 
 the means is almost certainly of valid significance. We can say, there- 
 
MENTAL CAPACITY 449 
 
 fore, that while these groups of women at Bedford and Albion cover 
 about the same range of variability in their mental capacity as shown 
 by this scale, the Albion women are on the average slightly superior. 
 It is of interest to note, however, that Bedford has a slight repre- 
 sentation higher on the scale than the Albion group extends. 
 
 (b) Bedford Reformatory and Chicago House of Correction 
 
 The work by Dr. Kohs" at the Chicago House of Correction 
 may also be compared with the Bedford group. The study was made 
 of 335 boys from 17 to 21 years of age, the testing being done with 
 the Goddard modification of the Binet-Simon scale. We had earlier 
 
 TABLE 183 
 
 MENTAL CAPACITY AS MEASURED BY BINET-SIMON 
 (GODDARD REVISION) 
 
 Per Cent Distribution of Bedford Women and of Men at the Chicago 
 House of Correction (Chicago), with Constants 
 
 
 Mental Age 
 
 Number 
 of cases 
 
 Mean 
 
 (X 
 
 
 6— 
 
 7— 
 
 8— 
 
 9— 
 
 lo- 
 
 11— 
 
 12— 
 
 
 Bedford Women 
 
 
 1.0 
 
 5.0 
 
 28.0 
 
 ss. 
 
 28.0 
 
 
 100 
 
 10.32 
 ±.091 
 
 .91 
 ±.065 
 
 House of Correction 
 Men (Chicago) 
 
 .6 
 
 .6 
 
 5.4 
 
 20.0 
 
 31.6 
 
 35.2 
 
 6.6 
 
 335 
 
 10.58 
 ±.069 
 
 1.08 
 ±.042 
 
 in our work at the Laboratory used this scale on a group of 100 
 Bedford cases whose results we can, therefore, compare with those 
 of Dr. Kohs' group. Table 183 presents the data on these two 
 groups for comparison. A point of difference in method enters in, 
 in that Dr. Kohs had added to the Goddard scale tests XV^ and 
 XV* from the original Binet scale to compensate for any accidental 
 variations, while our figures represent the Goddard scale unmodified. 
 This gives the Chicago group a slight advantage over ours. From 
 a comparison of the constants it is seen that there is a small but, 
 as Table 184 shows, an almost certainly significant diflference be-- 
 
 " Kohs, Samuel C. "The Practicability of the Binet Scale and the Question 
 of the Borderline Case.'' Bulletin No. 3, Psychopathic Department Series No. 
 2. Publications of the Research Department, Chicago House of Correction, 
 1915. 
 
450 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 184 
 
 Comparison of the Means and Standard Deviations of Mental Capacity 
 
 Measured by Binet Mental Ages (Goddard) between Bedford Women 
 
 and Men at Chicago House of Correction 
 
 
 Bedford 
 Women 
 
 Chicago 
 House of Cor- 
 rection Men 
 
 Difference 
 
 d 
 0-d 
 
 Chances that real 
 
 difference does 
 
 not exist are 1 
 
 in: 
 
 Mean 
 
 10.32 
 ±.091 
 
 10.58 
 ±.059 
 
 -.26 
 
 2.43 
 
 133 
 
 (T^ 
 
 
 
 
 ff 
 
 .91 
 ±.065 
 
 1.08 
 ±.042 
 
 -.17 
 
 2.15 
 
 63 
 
 cr<r 
 
 
 
 
 
 100 
 
 335 
 
 
 
 
 
 
 tween the means and standards deviations of these two groups. Con- 
 sidering the slight advantage which accrues to the Chicago group 
 as a result of the difference in procedure of testing, we should not feel 
 justified, however, in stressing so slight a difference. We should, 
 therefore, feel safe in saying that, measured by the Goddard modifi- 
 cation of the Binet-Simon scale, our Bedford group appears no better 
 and probably not appreciably worse than the men tested by Dr. 
 Kohs at the Chicago House of Correction. 
 
 (c) Auburn Women and Auburn Men 
 
 For purposes of comparison with the group of Auburn women, 
 we are relying mainly on data obtained by Dr. Frank L. Heacox on 
 a group of men at Auburn Prison. We are indebted to him for 
 the advance use of data on 473 men tested by the Yerkes-Bridges Point 
 Scale and 100 men tested by the Stanf ord-Binet. Chart XXVIII shows 
 the distribution, in terms of points on the Yerkes-Bridges Point 
 Scale, of our 44 English-speaking Auburn women in comparison 
 with that of 473 consecutive admissions (with minor exceptions), 
 of English-speaking men in the men's prison at Auburn. The marked 
 irregularity of the curve for the Auburn women is explicable on 
 the ground of the small number of cases. Aside from this difference 
 between the curves there is no important difference in general out- 
 line. The crude mode falls in the same interval, and the direction of 
 skewness is the same, although the range of the men's scores is slightly 
 greater in both directions. The means are practically identical, 
 
Cent 
 
 AUBURN WOMEN 
 
 Per 
 
 Cult 
 
 M 30 36 « 48 S4 60 66 72 
 
 POINTS 
 
 AUBURN MEN 
 
 ^ 
 
 84 90 96 102 
 
 Chart XXVIII 
 Mental Capacity as Measured by Yerkes-Bridges Point Scale 
 Percentage comparison between Auburn women and Auburn men — English- 
 speaking cases only. 
 
 TABLE 185 
 
 Comparison of the Means and Standard Deviations of Mental Capacity 
 
 Measured by Yerkes-Bridges Point Scale between Women 
 
 at Auburn Prison and Men at Auburn Prison 
 
 
 Auburn 
 Women 
 
 Auburn 
 Men 
 
 Difference 
 
 d 
 0-d 
 
 Chances that real 
 
 difference does 
 
 not exiat are 1 
 
 in : 
 
 Mean (Points) 
 
 (Tm 
 
 72.73 
 
 ±1.88 
 
 72.66 
 ±.622 
 
 .07 
 
 .03 
 
 2 
 
 
 
 a ■ . . . 
 
 12.45 
 
 ±1.28 
 
 13.53 
 + .440 
 
 -1.08 
 
 .80 
 
 4 
 
 Co- 
 
 
 
 
 Cases 
 
 44 
 
 473 
 
 
 
 
 
 
 451 
 
452 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 72.73 + 1.87 as against 72.66zh.622. Table 185 shows that this in- 
 finitesimal difference is insignificant, the chances being 1 in 2 that 
 as much difference could occure by pure chance. The divergence 
 
 Per 
 Cent 
 
 Per 
 Cent 
 
 
 
 
 
 
 
 AUBURN WOMEN 
 
 H 
 
 
 
 
 
 
 K 52 58 64 70 76 
 
 94 100 106 112 118 124 
 
 AUBURN MEN 
 
 Chart XXIX 
 
 Mental Capacity as Measured by Stanford-Binet 
 
 Percentage comparison between Auburn women and Auburn men — English- 
 speaking cases only. 
 
 between the standard deviations, i.e., between 13.53±.440 and 
 12.45 ±1.280, might be expected to occur through sampling 1 time in 4. 
 Comparison of the same group of Auburn women with a group 
 of 100 Auburn men, both measured by the Stanford-Binet intelligence 
 quotient, is presented graphically in Chart XXIX. Intelligence quotient 
 
MENTAL CAPACITY 
 
 453 
 
 rather than mental age has been used, since we have given the former 
 measure the preference, except where it has been necessary to use 
 mental age for comparison with other data. Referring to Chart XXIX 
 we note that the greater range of variability in the group of Auburn 
 men is more evident here than in the comparison on the Yerkes- 
 Bridges Scale. 
 
 Especially is this noticeable at the superior end, which extends 
 for three intervals beyond anything which we obtained in our group. 
 It is only fair to recall here that three of the apparently brightest and 
 best educated women in our Auburn group refused to be examined 
 and that this perhaps helps to explain our lack of representation at 
 this end. Whatever the explanation, it is evident that our group, 
 as it stands, forms a more compact unit than that of the men. Table 
 186 shows that while there is no difference demonstrable between the 
 means, there is an almost certain difference between the respective 
 degrees of variability shown in the two groups 
 
 TABLE 186 
 
 Comparison of the Means and Standard Deviations of Mental Capacity 
 Measvired by Stanford Binet I. Q. between Women at Auburn 
 Prison and Men at Auburn Prison 
 
 
 Auburn 
 Women 
 
 Auburn 
 Men 
 
 Difference 
 
 d 
 0-d 
 
 Chances that real 
 
 difference does 
 
 not exist are 1 
 
 in: 
 
 Mean 
 
 72.73 
 + 1.709 
 
 72.70 
 + 1.506 
 
 .03 
 
 .02 
 
 2 
 
 (Tm 
 
 
 
 
 (T 
 
 11.21 
 ±.736 
 
 15.06 
 + 1.076 
 
 -3.85 
 
 2.95 
 
 625 
 
 0-ff 
 
 
 
 
 Cases 
 
 43 
 
 100 
 
 
 
 
 
 
 We had hoped to make some comparison of our data on this 
 group with the results obtained by Dr. Glueck in his work at Sing 
 Sing,^^ but, although he states that he has employed both the Yerkes- 
 Bridges and the Stanford-Binet scales, he gives no figures showing 
 distributions. He diagnoses as defective 98 of his native-born cases 
 and says that while he bases this diagnosis on their previous history 
 and present reactions as well as on the laboratory findings, none of 
 them possessed a degree of intelligence beyond that of the average 
 
 "Op. cit. 
 
454 STUDY OF WOMEN DELINyUKJNTiJ IN NEW YORK 
 
 twelve-year-old American child. We regret also that we can not make 
 a comparison between our results on this group and the testing of 
 330 inmates of the Massachusetts State Prison which was made by 
 Rossy ^* in 1917. He has used the Yerkes-Bridges Point Scale as a 
 
 Per 
 Cent 
 
 PENITENTIARY WOMEN (New York) 
 
 Per 
 Cent 
 
 MEN (Ohio) 
 
 P '~ 
 
 < • 
 
 ' 
 
 
 11 
 
 78 
 
 90 
 
 Mir 2 
 
 24 30 36 42 48 54 eo 66 
 
 POINTS 
 
 Chart XXX 
 Mental Capacity as Measured by Yerkes-Bridges Point Scale 
 Percentage comparison between New York Penitentiary women and Ohio 
 State Penitentiary men— EngUsh-speaking cases only. 
 
 measure and prints his frequencies so that comparison would have 
 
 been simple were it not for the fact that he has used the mental age 
 
 "Rossy, C. S. "Report of the First Three Hundred Cases Examined at the 
 Massachusetts State Prison." Bulletin No. 17 of the Mass. State Board of 
 Insanity. Jan., 1916. 
 
MENTAL CAPACITY 
 
 455 
 
 TABLE 187 
 
 Comparison of the Means and Standard Deviations of Mental Capacity 
 
 Measured by Yerkes-Bridges Point Scale between New York 
 
 Penitentiary Women and Ohio State Penitentiary Men 
 
 
 New York 
 
 Penitentiary 
 
 Women 
 
 Oliio 
 
 Penitentiary 
 
 Men 
 
 Difference 
 
 a 
 
 0^ 
 
 Chances that real 
 
 difference does 
 
 not exist are 1 
 
 in: 
 
 Mean (Points) 
 
 (Tm 
 
 73.64 
 ±1.63 
 
 77.10 
 ±1.29 
 
 -3.46 
 
 1.67 
 
 21 
 
 
 
 cr 
 
 (7(r 
 
 14.48 
 ±.908 
 
 12.01 
 ±.761 
 
 2.47 
 
 2.08 
 
 63 
 
 
 
 Cases 
 
 79 
 
 87 
 
 
 
 
 
 
 as a measure without stating from which one of Yerkes' curves he has 
 derived these ages. 
 
 (d) New York Penitentiary Wofnen cmd Ohio Penitentiary Men 
 
 For purposes of comparison with our Penitentiary group, we 
 have rehed mainly on the study by Dr. Haines, who tested 100 con- 
 secutive admissions to the Ohio State Penitentiary by the Yerkes- 
 Bridges Point Scale.^^ Of these only 87 are included in the data 
 given, as 13 were too foreign to make such examination reliable. 
 Chart XXX shows the curve for this group in comparison with that for 
 our group of Penitentiary women. The most striking difference 
 between the two curves is in respect to the degree of concentration 
 of cases about any given point. Our Penitentiary group shows an 
 approximate evenness of distribution over quite a wide range, whereas 
 Haines' group shows a sharp concentration about a mode in ihe 
 upper portion of the scale. The range of our group is wider than 
 that of the other, by virtue of the fact that it extends lower in the 
 intelligence scale. Although there appears to. be a slight difference 
 between the means in favor of Haines' group, reference to Table 
 187 shows that this difference is probably to be accounted for by 
 sampling. As regards the amount of variability, however, there is an 
 almost certain degree of difference in the direction of the greater 
 variability of our group. 
 
 "Haines, Thomas H. "Feeble-mindedness Among Adult Delinquents." 
 Jour, of Criminal Law and Criminology. 1917. Vol. 7, No. 5. 
 
456 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 188 
 
 Comparison of the Means and Standard Deviations of Mental Capacity 
 Measured by Yerkes-Bridges Point Scale between New York Work- 
 house Women and Columbus (Ohio) Workhouse Men 
 
 
 New York 
 
 Workhouse 
 
 Women 
 
 Columbus 
 
 Workhouse 
 
 Men 
 
 Difference 
 
 d 
 
 Chances that real 
 
 difference does 
 
 not exist are 1 
 
 in: 
 
 Mean (Points) 
 
 (Tm 
 
 66.61 
 + 1.71 
 
 71.88 
 + 1.46 
 
 -5.27 
 
 2.35 
 
 106 
 
 
 
 0" . . . . . 
 
 14.78 
 + .972 
 
 14.57 
 ±.901 
 
 .21 
 
 .155 
 
 2 
 
 (Ta 
 
 
 
 
 Cases 
 
 75 
 
 100 
 
 
 
 
 
 
 The work done at the Joliet Penitentiary by Louise E. Ordahl and 
 George OrdahP^ would seem to furnish data on a similar group, 
 but their method of combining two scales, the Stanford and Kuhl- 
 man revisions, make their results impossible of comparison. 
 
 (e) Workhouse Women {New York) and Workhouse Men {Ohio) 
 
 The figures which are available for comparison with our Work- 
 house group are the results of a study made by R. A. Gilliland on 
 100 men at the Workhouse at Columbus, Ohio.^^ Their ages varied 
 from 18 to 53; 28 were colored and 72 were white. The cases 
 were chosen by the crimes committed in proportion to their repre- 
 sentation in the entire group, the only exception being that more 
 than the correct percentage of vagrancy cases were chosen. Chart 
 XXXI shows the distribution of the scores on the Yerkes-Bridges Point 
 Scale, by which they were tested, and, for comparison, the distribu- 
 tion of scores made on this scale by our group of Workhouse women. 
 
 Inspection of the curves shows a certain similarity as regards 
 the spread of the curves, but a greater concentration of cases in the 
 upper end of the Ohio group. From Table 188 we see that there 
 exists an almost certainly valid difference between the central tenden- 
 
 " Ordahl, Louise E. and Ordahl, George. "A Study of 49 Female Con- 
 victs." Journal of Delinquency. 1917. Vol. II, No. 6, pp. 331-351 
 /^'I*li''r'^"'^' -^^ ^- "^^^ Mental Ability of 100 Inmates of the Columbus 
 (O.) Workhouse." Journal of Criminal Law and Criminology. Vol 7 No. 6, 
 March, 1917. ' 
 
MENTAL CAPACITY 
 
 457 
 
 Cent 
 
 25 
 
 Per 
 Cent 
 
 ^ 
 
 WORKHOUSE WOMEN (New York) 
 
 (WORKHOUSE MEN AND^ WOMEN (Ohio) 
 
 Chart XXXI 
 Mental Capacity as Measured by Yerkes-Bridges Point Scale 
 
 Percentage comparison between New York Workhouse women and Ohio 
 Workhouse group composed of men and women— English-speaking cases only. 
 
 cies of the two curves, with the New York group inferior to the 
 other, but that, as regards the degree of variability shown, no differ- 
 ence can be demonstrated which is not explicable by sampling. 
 
 Summary 
 
 Summarizing these results, we can see, then, that if we assume as 
 adequate measures of intellectual capacity the various scales that 
 have been used, we can make the following statements: 
 
458 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 The Bedford women, as compared with those in a similar institu- 
 tion at Albion, differ by a small but almost certainly valid amount 
 as to their averages, but no difference between the amount of varia- 
 bility around the average can be demonstrated for the two groups. 
 Both groups approach symmetry in their distribution. 
 
 In comparison of the Bedford women with the group of young 
 men studied by Kohs at the Chicago House of Correction, both the 
 mean and the degree of scatter about the mean are found to be slightly 
 greater for the latter group. This difference appears to be significant 
 though perhaps to be accounted for as a result of the somewhat 
 greater leniency in examination in the case of Kohs group. It seems, 
 therefore, that in comparison with these two other groups of offenders 
 of the reformatory type, our women are not markedly inferior. 
 
 When we compare the group of women committed to Auburn 
 Prison with Heacox's study of male inmates of the same place, we 
 see that no difference can be demonstrated in the average of their 
 mental capacity by either of the two measures employed, but that by one 
 of these scales a difference in variability is indicated. It would seem, 
 therefore, that the average mental capacity of the male and female 
 convicts in this institution is not demonstrably different but that the 
 men convicts would appear to exhibit a slightly greater variability 
 than the women, at least as measured by the Stanford Revision. 
 
 Regarding the two Penitentiary groups, it appears that Haines' 
 group is less spread out than the New York Penitentiary women and 
 somewhat better on the average, though the vaUdity of the latter dis- 
 tinction is open to question. 
 
 The New York Workhouse women seem to be clearly inferior to 
 the men studied by Gilliland in a similar institution. Their average 
 on the Yerkes-Bridges Point Scale is more than five points lower and 
 this difference has been shown to be almost certainly valid. (We may 
 say, moreover, that our mean would undoubtedly have been lower if 
 the women committed to the Workhouse for intoxication had been 
 included in our group. Our sample of 19 such cases shows a mean of 
 64.52.) The variation in the amount of scatter between these groups 
 is insignificant and easily accounted for by sampling. 
 
 Owing to the small amount of material available on delinquent 
 groups other than ours, the scanty information regarding the selective 
 processes operative in determining the make-up of these groups, and 
 the uncertainty as to whether the technique of administration of the 
 tests was in every case comparable with ours, we offer no general 
 
MENTAL CAPACITY 459 
 
 interpretation of the above comparisons. One negative inference sug- 
 gested by the material at hand might, however, be noted. In view 
 of the slightness and statistical uncertainty of the majority of the dif- 
 ferences observed between our groups and others from similar types 
 of institutions, it would appear that ours are not eccentric to any 
 marked degree. 
 
CHAPTER XV 
 
 MENTAL CAPACITY: RELATED TO VARIOUS FACTORS 
 WHICH AFFECT THE DELINQUENT GROUP 
 
 IN the following discussion we shall limit ourselves to comparisons 
 made possible by our own data on the women of the six delinquent 
 groups studied. We shall consider especially the relationship of men- 
 tal capacity, as measured by the showing- in certain tests, to the main 
 social factors in the records of these women. 
 
 MENTAL CAPACITY IN RELATION TO CRIMINAL CAREER 
 
 (a) Mental Capacity and Nature of Offense 
 
 The contention is frequently heard that certain types of offenders 
 are mentally superior to those guilty of other kinds of delinquencies. 
 It is possible to obtain some information with regard to the accuracy 
 of this contention, at least as it applies to women delinquents, through 
 the mechanism of correlation. In Table 189 we have used as a quan- 
 titative variable the mental capacity measured by Test Aggregate and 
 as a qualitative classification that employed by the New York City 
 Police Department (see Chapter VI, p. 85). The correlation ratio 
 which this table yields is .23 ±.049, which would seem to indicate 
 that there probably exists a small but significant degree of association 
 between the intellectual capacity of a woman and the kind of crime, 
 classified in this way, that she commits. The means run, in order of 
 excellence, — Offenses against Property Rights, against Regulations for 
 Public Health, against Chastity, against the Family, General Criminal- 
 ity, Offenses against the Person, and Offenses against Administration 
 of Government. The numbers in some of the groups are, of course, 
 too few to make the means significant. 
 
 If we consider, however, the two classes of offenses for which we 
 have a reasonable number of cases, those against property and those 
 against chastity, we obtain the curves figured in Chart XXXII. The 
 central tendency of the property offenders is noticeably higher than that 
 of the chastity offenders, 78.48±.993 as over against 75.52±.619, but 
 
 460 
 
u 
 
 4) 
 
 o 
 I 
 
 
 
 ■^ 0) 
 
 ■SP 
 
 So 
 
 ^^ 
 
 o 
 
 I 
 
 I 
 
 & 
 
 o 
 U 
 
 & 
 
 f 
 
 
 s-f 
 
 
 
 
 
 
 
 
 
 i 
 
 
 ■* 
 
 10 
 
 « 
 
 m 
 
 OS 
 
 lO 
 
 CO 
 
 !S 
 
 
 at 
 
 10 
 
 iH 
 
 lO 
 
 t> 
 
 <x> 
 
 (—1 
 
 Ui 
 
 
 s^l 
 
 
 
 I> 
 
 
 
 
 
 t^ 
 
 
 
 
 
 
 
 
 
 
 
 ^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 O) 
 
 S 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 B 
 
 
 .-. 
 
 
 
 
 
 
 CO 
 
 i 
 
 Z6^6 
 
 
 « 
 
 
 
 
 CO 
 
 
 W3 
 
 S6-S6 
 
 
 f-l 
 
 
 (N 
 
 
 -■ 
 
 
 •* 
 
 £6-16 
 
 
 CO 
 
 
 
 
 N 
 
 
 lO 
 
 T6-68 
 
 
 00 
 
 iH 
 
 CO 
 
 
 M 
 
 
 s 
 
 6S-IS 
 
 
 CD 
 
 
 ^ 
 
 
 »o 
 
 - 
 
 CD 
 
 IS-QS 
 
 ^ 
 
 « 
 
 
 
 
 «5 
 
 
 S 
 
 S8-S8 
 
 -H 
 
 ■^ 
 
 
 CO 
 
 
 t- 
 
 
 IM 
 
 E8-IS 
 
 
 Cfl 
 
 
 - 
 
 
 IC 
 
 
 C<1 
 
 XS-61 
 
 -^ 
 
 8 
 
 <N 
 
 CO 
 
 
 ■* 
 
 " 
 
 CO 
 
 61-11 
 
 
 IX 
 
 
 lO 
 
 
 OS 
 
 
 (N 
 
 
 11-91 
 
 N 
 
 s 
 
 -H 
 
 CO 
 
 
 o 
 
 -- 
 
 o 
 
 CO 
 
 
 91-U 
 
 
 I-t 
 
 
 N 
 
 
 N 
 
 
 ?5 
 
 Si,-ll 
 
 
 t-H 
 
 (H 
 
 Hi 
 
 
 l> 
 
 CO 
 
 CO 
 CO 
 
 Ii-69 
 
 
 « 
 
 
 
 .H 
 
 CO 
 
 w 
 
 ^ 
 
 69-Z9 
 
 M 
 
 « 
 
 
 (N 
 
 -- 
 
 ■* 
 
 - 
 
 ^ 
 
 Z9-e9 
 
 
 IM 
 
 <N 
 
 CO 
 
 
 
 CO 
 
 o 
 
 B9-S9 
 
 
 l> 
 
 
 M 
 
 
 (N 
 
 
 CO 
 
 E9-I9 
 
 
 eo 
 
 
 
 
 -* 
 
 
 o 
 
 T9-6S 
 
 
 U3 
 
 
 
 
 CO 
 
 
 o 
 
 6S-iS 
 
 
 OT 
 
 
 
 
 -« 
 
 iH 
 
 lO 
 
 zs-ss 
 
 ■H 
 
 IT 
 
 
 
 
 
 
 l> 
 
 SB-SS 
 
 
 o 
 
 ES-IS 
 
 1-1 
 
 iH 
 
 IS-6?' 
 
 - 
 
 - 
 
 6f-a 
 
 - 
 
 iH 
 
 
 , 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 cj 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 a 
 
 
 > 
 
 > .2 
 
 ?? 
 
 f3 
 
 
 
 
 
 
 
 t> 
 
 ' ^ 
 
 "g 
 
 •s 
 
 ^ 
 
 
 
 
 
 
 li 
 
 
 3 
 
 '3 
 
 tH 
 
 
 
 
 
 
 P- 
 
 a 
 
 D 
 
 1 
 
 D 
 
 P 
 
 'S 
 
 a. 
 
 ■fe 
 
 C 
 
 'S 
 
 Is 
 
 ■a 
 
 ixi 
 
 i 
 
 .s 
 o 
 
 
 
 
 
 
 ? 
 
 a 
 
 mX 
 
 o^S 
 
 s 
 
 -^ 
 
 
 
 
 p 
 
 e: 
 
 f 
 
 S*^ 
 
 gc 
 
 a 
 
 S3 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 It: 
 
 (t 
 
 * 
 
 in 
 
 to 
 
 to 
 
 
 
 
 
 c 
 
 C 
 
 c 
 
 O 
 
 O 
 
 u 
 
 O 
 
 
 
 5_ ^+1 
 
 .. o 
 
 B-2 
 
 461 
 
462 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 Per 
 Cent 
 
 ^^^■^-■" Property 
 ^j— — — Chastity 
 
 Chart XXXII 
 
 Mental Capacity as Measured by Test Aggregate 
 
 Percentage comparison between offenders against property and offenders 
 against chastity. English-speaking cases only. 
 
 the amount of variability displayed is not greatly different for the two 
 groups, 9.15±.584 for the property offenders and 8.95±.442 for the 
 chastity offenders. It might seem to be of interest to note that the 
 scores of the offenders against chastity follow more closely the lines of 
 the normal curve while the curve of property offenders tends to be 
 
 TABLK 190 
 
 OFFENDERS AGAINST PROPERTY AND OFFENDERS 
 AGAINST CHASTITY 
 
 Comparison of the Means and Standard Deviations of Mental Capacity, 
 as Measured by Test Aggregate, of Offenders against Property 
 and Offenders against Chastity 
 Total Group. English-speaking Cases Only 
 
 
 OffendeTS 
 against 
 Property 
 
 Offenders 
 against 
 Chastity 
 
 Difference 
 
 d 
 (Td 
 
 Chances that real 
 
 difference does 
 
 not exist are 1 
 
 in: 
 
 Mean 
 
 78.48 
 ±.993 
 
 75.52 
 ±.619 
 
 2.96 
 
 2.53 
 
 175 
 
 Cm 
 
 
 
 
 (J 
 
 9.15 
 ±.584 
 
 8.95 
 ±.422 
 
 .20 
 
 .28 
 
 3 
 
 Oa 
 
 
 
 
 Cases 
 
 85 
 
 209 
 
 
 
 
 
 
MENTAL CAPACITY 463 
 
 bimodal, but the cases of the latter are probably too few to make this 
 significant. From Table 190, it appears that this observed superiority 
 of the offenders against property may be regarded as almost certainly 
 significant. For the slight difference which exists in the standard de- 
 viations, however, no valid significance can be demonstrated. 
 
 We have next taken the entire group and instead of using this six- 
 fold division have classified their offenses simply as misdemeanors and 
 felonies. The two groups thus obtained are distributed, as to mental 
 capacity, as shown in Chart XXXIII. Although there is a noticeable 
 
 Per 
 
 Ceat 
 
 i 1-^ i__. 
 
 Misdemeanaot* 
 
 Chart XXXIII 
 
 Mental Capacity as Measured by Test Aggregate 
 
 Percentage comparison between felons and misdemeanants. English-speak- 
 ing cases only. 
 
 difference between the crude modes of these two groups, in favor of the 
 felons, the actual difference in the means is slight. We see by Table 
 191 that this difference of .46 in favor of the felons, taken in con- 
 nection with its standard deviation is too small to be demonstrated as 
 valid and that the slight difference in the degree of variability can also 
 be accounted for by sampling. There would seem to be, then, no 
 demonstrable difference between the means and the standard deviations 
 of those women guilty of felonies and those convicted of misdemean- 
 ors. 
 
 An effort to compare our results with Goring's with reference to 
 nature of offense was quite abortive, due to the extreme difference 
 in classification. 
 
464 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 (b) Mental Capacity and Extent of Criminal Record 
 
 As described in Chapter VI, we have used as a measure of an in- 
 dividual's criminal record; first, the number of previous convictions 
 and, second, the number of months incarcerated. The relationship be- 
 tween mental capacity and the former measure is shown in Table 192. 
 The correlation coefficient is negligible in size (—.05 ±.052) and would 
 indicate absence of any correlation between these two factors. The 
 correlation ratios (.ISih.OSO and .14±.050) suggest the existence of 
 at most a small degree of relationship between the mental capacity of 
 a woman and the number of times she has been convicted. Examining 
 the scatter table and noticing the lack of a regular trend in the line 
 of means, however, we become dubious of the existence of even the 
 small relationship suggested, although there is a faint indication of a 
 tendency toward lower mentality with increase in number of convic- 
 tions. 
 
 It is evident at once that the ages of the women in question would 
 have to be taken account of here. A partial correlation with age con- 
 stant, however, gave a coefficient practically identical with the former 
 coefficient, — .053 as against - .054. We might, therefore, assume that 
 the matter of age is not a contributing factor in this relationship. 
 
 If we next consider the relation of intellectual capacity as shown 
 by Test Aggregate to the period of time spent in penal institutions. 
 
 TABLE 191 
 
 FELONS AND MISDEMEANANTS ON PRESENT CONVICTION 
 
 Comparison of the Means and Standard Deviations of Mental Capacity, 
 
 as Measured by Test Aggregate, of those Classified as Felons 
 
 and Misdemeanants on Present Conviction 
 
 English-speaking Cases Only 
 
 
 Felons 
 
 Misde- 
 meanants 
 
 Difference 
 
 d 
 ffd 
 
 Chances that real 
 
 difference does 
 
 not exist are 1 
 
 in: 
 
 Mean 
 
 76.1 
 ±1.20 
 
 75.57 
 ±.531 
 
 .53 
 
 .40 
 
 3 
 
 fl-m 
 
 
 
 
 a 
 
 9.57 
 ±.958 
 
 9.36 
 ±323 
 
 .21 
 
 .20 
 
 , 2 
 
 <Ta 
 
 
 
 
 Cases 
 
 64 
 
 311 
 
 
 
 
 
 
to 
 
 o 
 O 
 m 
 
 S 
 o 
 
 0) 
 
 •a 
 
 
 ^ 
 
 V bJ) 
 
 l>. 
 
 -s 
 
 ^ 
 
 « 
 
 
 Q< 
 
 TJ 
 
 "f 
 
 § 
 
 i^ 
 
 9 
 
 a 
 
 in 
 
 bil 
 
 ei 
 
 a 
 
 a> 
 
 M 
 
 ^ 
 
 u 
 ns 
 
 o 
 
 g 
 
 I 
 
 § 
 
 i 
 
 
 ill 
 
 W CO 00"3X^COU3 
 
 
 
 ■i 
 
 o 
 Eh 
 
 
 
 §? 
 
 rtOINOOOrtOrH(NMC0CCOt^M.-lTj1C0t*t^ 
 iHWCO^OSCO 
 
 1 
 
 o 
 o 
 
 <! 
 
 1 
 
 Z6 o» 56 
 
 r-t(N(N 
 
 U3 
 
 X 
 
 S6 o» £6 
 
 (N ^rH 
 
 '^ 
 
 £6 o» 16 
 
 (Mirq th 
 
 in 
 
 16 0? 68 
 
 rH rTlCOCOl^lO 
 
 s 
 
 iH 
 
 68 0% IS 
 
 r-( iHiHi-i iHlOO 
 
 to 
 
 (M 
 
 IS <n S8 
 
 ^ 1-1 tHiMMt^o 
 
 s 
 
 „ 
 
 S8 0} £8 
 
 TtlMcDCO 
 
 
 - 
 
 £8 0% 18 
 
 rHUSNNOOOS 
 
 S 
 
 GO 
 
 T8 0% ei 
 
 ^ « rnrHTjJ^lOCO 
 
 M > 
 
 rH 
 
 6/ 0* 11 
 
 iHNiH i-(.HM(NCOO 
 
 
 t-- 
 
 11 o? SZ 
 
 rt cq coosin 
 
 O 
 CO ^ 
 
 i-H 
 
 SI 0% £i 
 
 .-H .-( rH r-i .-( lO t» U5 
 
 CI 
 
 O) 
 
 £i ot U 
 
 .H i-(»-) TH^eoO'-t 
 
 
 ^ 
 
 U o» 69 
 
 ^ tH,-I INM COI>T-( 
 
 Sl 
 
 ^ 
 
 69 0? 19 
 
 iHiH (MNi-tt»0 
 
 s;. 
 
 (N 
 
 19 0* S9 
 
 ,-H ^rH (NM^IN 
 
 s 
 
 •H 
 
 fi9 0% £9 
 
 ^ tH IC(Nt11 
 
 CO 
 
 CO 
 
 £9 o» 19 
 
 i-H (NCCtHCC 
 
 si 
 
 - 
 
 19 0% 6S 
 
 tH T-tr-t CCTf 
 
 sj 
 
 ^ 
 
 6S 0% ig 
 
 f-(«cq 
 
 lO 
 
 OJ 
 
 2.9 0% SS 
 
 (M .HTjf 
 
 '- 
 
 -g o» gg 
 
 
 o 
 
 ggo Ts 
 
 rH 
 
 - 
 
 IS o» 61' 
 
 ^ 
 
 iH 
 
 6* o» Z» 
 
 - 
 
 -^ 
 
 
 
 
 
 :| 
 
 
 
 •'? 
 
 d oi 00 1> O »o' -«fl CO <N rH O OS « l> «D 10 Tjl CO « iH O 
 
 
 
 snoifjoiATio;^ snoiASJ^j jo jaqmn^ 
 
 "5 00 
 
 II II 
 
 2 '^J 
 
 WW 
 oo 
 °.° 
 ■H-H 
 
 00 Til 
 
 §8 t^p 
 
 
 CO 
 
 Si 
 
 O c3 
 
 ■.a ho 
 
 |£ 
 .2 -c « 
 
 '"■e is 
 gS °.-§ 
 
 S gS-9 
 ■ V • 3 ° 
 
 *£^^ ag 
 
 aa d 5 
 £''SS°| 
 wS (S p. 
 
 * a II s^ 
 11^ as 
 
 ■SHZ 
 
 •32 
 IBS 
 
 o o 
 
 oo 
 
 46s 
 
466 STUDY OF WOMEN DELINQUENTS IN NEW YORK' 
 
 we see by Table 193 that here again the coefficient is small and, in 
 consideration of its standard deviation, probably unreliable, .09±.051. 
 Here, as before, we feared that the factor of age might be complicat- 
 ing the relationship, but a partial correlation making age constant 
 brought about no change in the coefficient. The correlation ratios 
 (.16±.050 for Test Aggregate on time served, and .10±.051 for time 
 served on Test Aggregate) can indicate at most only a slight associ- 
 ation. Reference to the means (Table 193) shows that whatever cor- 
 relation exists is positive, that is, that the tendency is toward an 
 association of longer periods of imprisonment with higher levels of 
 mentality. 
 
 Regarding this matter of the relation of intelligence to the criminal 
 record, as shown in these two ways, we quote from Goring. Jt will be 
 remembered that his intelligence scale is a five-fold judgmental one 
 and that his measure of criminal record differs from ours in that he 
 has used as a measure the number of convictions per year of freedom 
 and the number of months imprisoned per year of freedom. (For 
 further explanation see page 104 of this work.) 
 
 "With increasing frequency of conviction, the proportion of convicts 
 who are mentally defective progressively increases from 19 per cent to 
 .36 per cent, and their mean intelligence decreases from -|-.39'to — .27; on 
 the other hand, with increasing periods of imprisonment, the average in- 
 telligence of convicts increases from — .08 to -(-.32 — ^the proportion of the 
 mentally defective correspondingly diminishing from 33.6 per cent to 23.6 
 per cent. Evidently, the characteristic of the penal records of relatively 
 weak-minded offenders is frequency of conviction to short periods of im- 
 prisonment, for trivial offenses ; and the distinguishing features of the penal 
 records of the more intelligent recidivists are fewer convictions, but 
 longer sentences, for serious offences." . . . 
 
 "Correlation of intelligence with frequency of conviction: 
 
 Coefficient r . . . — .16±.03 
 
 Ratio (intelligence upon convictions) .20 
 
 "Correlation of intelligence with time of imprisonment : 
 
 Coefficient r . . . .10+03 
 
 Ratio n (intelligence upon imprisonment) .12 
 
 "It will be seen that the correlation coefficient of intelligence with con- 
 victions is — .16, which measures the extent to which defective intelligence 
 is associated with frequency of conviction; and that the coefficient of im- 
 prisonment with intelligence is -)-.10, which measures the strength of bond 
 uniting length of imprisonment, or sentence, with good intelligence."^ 
 'Op. cit., p. 271. 
 
CO 
 
 I 
 
 o 
 
 
 
 S ^►S 
 
 S3 " 
 
 ■a 
 o 
 
 n 
 
 0) 
 09 
 St 
 
 O 
 
 n 
 
 "Bb 
 
 u 
 
 5 « 
 a> 
 
 ^^ 
 
 3 g a 
 
 1 
 
 d 
 
 I 
 
 § 
 
 « 
 
 o 
 
 
 lif 
 
 
 
 
 i 
 
 
 CO 
 
 
 in M ^ 1-1 O (N <N l>U lO CO « M M 
 
 
 
 Z6 <n S6 
 
 .-iTji 
 
 to 
 
 _«5 
 
 S6 ot 86 
 
 .H CO 
 
 -# 
 
 £6 0? 16 
 
 j W r-l i-<iN 
 
 ua 
 
 16 oi 68 
 
 ^w cqTfiM 
 
 s 
 
 6S o% 18 
 
 iH iHiHOq r-tO 
 
 CD 
 
 IS 0% 98 
 
 i~i r-\ r-llO(M 
 
 S 
 
 'oo 
 
 B8 0* E8 
 
 
 
 E8 0% 18 
 
 .-c^ t-HiOOi 
 
 
 18 0* 6Z 
 
 " ««*- 
 
 CO 
 
 61 0% LL 
 
 .H i-t Ni-ICCiHr-lO 
 
 s 
 
 U o\ Si 
 
 ^ ^ ^^(N^« 
 
 s~ 
 
 
 91 o? EZ 
 
 COM (M-* 
 
 s 
 
 EZ o» TZ 
 
 cq CO 
 
 TZ 01 69 
 
 rHrHrHeO«3l> 00 
 
 69 o* Z9 
 
 i-HrHIN .H(NI>. gj 
 
 Z9 o» S9 
 
 iH rHC0U3 
 
 o 
 
 >- • 
 
 S9 01 89 
 
 iH iHMcn 
 
 CO 
 
 1-H 
 
 E9 °% 19 
 
 WCOCO 
 
 o 
 
 19 o» 6S 
 
 ■^<© 
 
 o 
 
 6S ot Z9 
 
 (NCO 
 
 w 
 
 ZS Of 95 
 
 rHTHW 
 
 l> 
 
 o 
 
 S9 oj ES 
 
 
 o 
 
 ES o» 19 
 
 iH 
 
 w 
 
 IS 01 m 
 
 -H 
 
 rH 
 
 6f oi Z* 
 
 - 
 
 ■H 
 
 
 
 
 1 
 
 1 
 
 paAjag sqiuojii jo raqam^ 
 
 s 
 11 i 
 
 b- s 
 
 CD Qi 
 iO.2 
 
 III 
 
 O Q 
 
 II 
 
 ss 
 
 (D 01 
 
 OB ri 
 
 si 
 
 •Hfl 
 
 oo 
 
 II II 
 
 •■gg' 
 la 
 
 n 03 
 
 .a-g 
 
 no 
 
 03 OT 
 
 OS S 
 
 il 
 
 ■s.g 
 
 O m 3 
 
 o o 
 
 oo 
 
 467 
 
468 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 It will be seen that such relationships as we discovered in our 
 groups tended in the same directions as did Goring's — that is, to a 
 negative association between intelligence and frequency of conviction 
 and a positive correlation between intelligence and total length of 
 imprisonment. In view of the slightness of the relationships indicated 
 by our figures, however, we should not feel justified in making so 
 definite a statement on this point as does Goring. We are inclined 
 to feel, in fact, that his verbal formulation leaves the reader with a 
 sense of a more impressive relationship than is justified by the size of 
 his correlation coefficients. 
 
 In this connection it should also be noted that our situation is some- 
 what different from his, in that a long sentence which is the result of 
 a serious offense in the cases studied by Goring may either have the 
 same significance for our group or may result from relatively trivial 
 causes. This is brought about by the fact, first, that we have included, 
 as Goring evidently has not, the time spent ih juvenile institutions 
 where the sentence is generally for the period of the subject's minor- 
 ity; second, that at least three of the institutions studied receive cases 
 on indeterminate sentence, which means that, of two women brought 
 in for the same offense, one might get 20 days in the Workhouse, and 
 the other three years at Bedford, dependent on whether or not she was 
 considered a "reformatory type." In view of the attendant ambi- 
 guity in this measurement, we have not regarded the number of months 
 served as a very satisfactory measure of a woman's criminal record. 
 
 If we next consider the mental capacity of the women who were • 
 first offenders and those who may be classed as recidivists, we obtain 
 the curve shown in Chart XXXIV. It is evident that in general form 
 the two present a similar distribution. The means are: almost identical, 
 7S.64dt.765 as against 75.68±.627. Table 194 shows that from these 
 figures we can not demonstrate the existence of a difference, the 
 chances being even that such difference could occur by chance. On 
 the matter of variability, it can be seen that the standard deviation of 
 the recidivists is slightly greater than that of the first offenders but this 
 difference is not large, and according to Table 194 its validity is not 
 established. We are, then, unable to demonstrate that there exists 
 any difference, other than might be accounted for by sampling, in 
 either average mental capacity or variability between those women 
 delinquents who are now convicted for the first time and those who 
 have received two or more convictions. It would seem, therefore, 
 that the two classes represent relatively similar groups. 
 
MENTAL CAPACITY 
 
 469 
 
 Per 
 
 Cenl 
 
 First Offenders 
 > Recidivists 
 
 Chart XXXIV 
 
 Mental Capacity as Measured by Test Aggregate 
 
 Percentage comparison between first offenders and recidivists. English- 
 speaking cases only. 
 
 There is nothing in our data to support the assertion of Anderson's^ 
 that "there is a high correlation between the frequency of offense and 
 the mental condition of these individuals ; 39.3 per cent of first offend- 
 ers, 47.2 per cent of second offenders and 84.2 per cent of recidivists 
 were suffering from severe mental handicaps." Our definition of 
 
 TABLE 194 
 FIRST OFFENDERS AND RECIDIVISTS 
 
 Comparison of the Means and Standard Deviations of Mental Capacity, 
 
 as Meastired by Test Aggregate, of First Offenders and Recidivists 
 
 English-speaking Cases Only 
 
 
 First 
 Offenders 
 
 Recidivists 
 
 Difference 
 
 d 
 0-d 
 
 Chances that real 
 
 difference does 
 
 not exist are 1 
 
 in: 
 
 Mean 
 
 75.64 
 ±.765 
 
 76.68 
 ±.627 
 
 -.04 
 
 .042 
 
 2 
 
 (Tm 
 
 
 
 
 (7 
 
 8.96 
 ±.493 
 
 9.66 
 ±.402 
 
 -.70 
 
 1.10 
 
 7 
 
 (Ta 
 
 
 
 
 Cases 
 
 137 
 
 237 
 
 
 
 
 
 
 'Anderson, V. V. "The Immoral Woman." Journal of Criminal Law and 
 Criminology, Vol. 8, No. 6. March, 1918. p. 910. 
 
470 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 recidivist differs from his, since he defines a recidivist as one who has 
 had at least two previous convictions. Moreover, he apparently 
 includes in this percentage of those having serious mental handicaps a 
 small number of psychopathic cases in addition to those mentally in- 
 ferior. Even making all such allowances, our results would, how- 
 ever, appear to be quite at variance with his. 
 
 (c) Mental Capacity and Factors Connected with Beginnings 
 of Criminal Career 
 
 The relationship between the mental capacity of these women and 
 the ages at which they were first convicted is of interest since the 
 common view is to the effect that individuals with inferior mentality 
 are more likely to be led into delinquency at an early age, and, further, 
 are less skilful in evading detection than are their more intelligent 
 associates. Reference to Table 195, which presents these factors in 
 relationship to each other, reveals at once the absence of any readily 
 discernible trend of association between the two variables for the 
 women delinquents whom we have studied. The scatter over all 
 parts of the table is marked, which is reflected in the negligible corre- 
 lation coefficient of —.07 ± .052. Even a closer analysis of the 
 table fails to discover any significant trends of relationship. The cor- 
 relation ratio of test aggregate on age at first conviction (.08 it .052) 
 shows no appreciable increase over the correlation coefficient, mak- 
 ing it apparent that our data afford no grounds for inferring the 
 level of mentality of a woman offender from a knowledge of her age 
 when first convicted. Can we nevertheless discover any consistent 
 variations in age at first conviction with varying levels of mentality? 
 At first consideration of the value of the ratio (.21 ± .050) we might 
 hope to find a relationship of some significance even though small, but 
 reference to the irregularity of the line of means indicated by the 
 values on the lowest row of the table shows that it is impossible even 
 to state the direction of the association — i.e., whether the more intelli- 
 gent women tend to be convicted at an earlier or at a later age than do 
 the less intelligent. 
 
 It is doubtless unnecessary to point out that failure to discover evi- 
 dence of an association between these factors from our data does not 
 prove absence of any such relationship. It may well be obscured by 
 the presence of other factors, such, for example, as difference in social 
 and economic backgrounds, local differences in closeness of police su- 
 pervision, and other equally important conditions. Unfortunately our 
 
< I 
 
 10 01 
 
 V « 
 
 r_t {A 
 
 ■d 
 
 .a 
 
 V 
 
 *3 
 
 3 
 
 04 
 
 
 
 « 
 
 •s 
 
 S 
 
 
 ^ 
 
 s 
 
 ftl 
 
 Z6-S6 
 S6-S6 
 
 E6-I6 
 16-68 
 
 68-i8 
 Z8-S8 
 
 98-E8 
 E8-I8 
 
 T8-6i 
 61-U 
 
 Si-EZ 
 
 Ei-Ii 
 Xi-69 
 
 69-i9 
 i9-S9 
 
 S9-E9 
 
 E9-t9 
 
 19-63 
 
 6S-L9 
 
 ZS-SS 
 
 9S-Ee 
 
 EC-IS 
 
 tS-6* 
 
 6f-lf 
 
 fH 00 »H (H 
 
 <N rt CO fH CO ^ N CO 
 
 (N<M(NCO(N 
 
 COCOCOi-t^CO 
 
 «.H.HN«'*NTj<Tl<rt 
 
 CO CO tH .-I CO lO U3 -^ rH 
 
 ^ W ^ rH iH M CO CO CO Cq ^ U5 i-H tH 
 
 iHrH iH 1-4 (N ^ iH CO -^ Cq ^ f-< C^ iH »H 
 
 HiHiH .-( N CO CO ^ -* CO I- 
 
 N NCq.H 
 
 CO iHCOMiHiHiH t^lO'^rHCq 
 
 CO r-1 i-H f-H W -^ (M to ^ N 
 
 rHiH M ^ Tf W IH CO Tf I- 
 
 ■ f u 
 
 « J = 
 
 r-i ,-H W,-lrM(M 
 
 CO NNN 
 
 -1.^ iH cq,~i N 
 
 f-* rH rtM 
 
 COr-r 
 
 OSt^U3COiHOit*lCMi-lOSt*>'3COr-IOSt^K5COi-HO>t*»OCOi-lOS 
 W -.,«_u_iJ_u_u_u«^«^«nflr,p0C^C<^C^C^^(N^.-Hr^.-lW 
 
 I I I I I I I L I I i I 
 
 OS rs. 113 t^ I— 1 Ui 1.^ "J i-v >— • "JJ I.— "J i'J r- 
 
 
472 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 numbers are not large enough to make possible any control of these 
 external factors by selecting for consideration only the more homo- 
 geneous portions of our group. The most that we can state, therefore, 
 is the fact that the relationship within our group is, at least, not a suffi- 
 ciently dominant influence to stand out as appreciable in amount in 
 face of the other factors operative. 
 
 The situation is evidently different in the group of men convicts 
 studied by Goring.^ He states that he finds a coefficient of .34 for the 
 correlation between mental grade and age at first conviction, which he 
 speaks of as "a relatively high degree of association between defective 
 intelligence and conviction at an early age." Later he says, "We con- 
 clude that undoubtedly the principal factor conducing to the early first 
 conviction of convicts is defective intelligence." 
 
 The relationship between the mental capacity of an individual and 
 the nature of the first offense is shown in Table 196. In comparing 
 this table with Table 189, which shows a similar correlation of mental 
 capacity with the nature of the present offense, it should be remem- 
 bered that many individuals fall inevitably into identical groups in the 
 two tables, since the first offense is actually the present offense in over 
 a third of the group. We note, however, that there is a consistent cut- 
 ting down of the numbers occurring in each of the first six divisions 
 and an increase in the numbers given in the group of general criminal- 
 ity. This, as was explained on page 94, is easily comprehensible when 
 it is understood that under this heading are included all the semi- 
 juvenile offenses such as Disobedient Child, Associating with Vicious 
 and Disorderly Persons, etc., — the type of offense upon which many 
 of the women who are convicted young are first sentenced. In spite 
 of the variation in this arrangement, we find here no great change in 
 the amount of association which can be shown to exist between the 
 variables. The correlation ratio between mental capacity and the na- 
 ture of the first offense is .20 ± .050. 
 
 The only group conspicuous as having mentality above the average 
 is that made up of individuals whose first offense was against property. 
 The three groups which fall appreciably below the general average- 
 offenders against the person, against the family, and against the ad- 
 ministration of government — are too small to carry any weight. 
 
 If we compare the means of each subdivision in the two tables 
 (Table 189 and Table 196) we see that the only change of any size in 
 'Op. cit, p. 282. 
 
ui 
 
 
 
 
 
 
 
 
 
 
 
 1 
 
 
 HI 
 
 ■*. 
 
 ?4 
 
 o 
 
 OS 
 
 
 C4 
 
 01 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 t* 
 
 
 < 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 00 
 
 10 
 
 CO 
 
 
 
 OJ 
 
 w 
 
 w 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 Eh 
 
 
 
 
 
 
 
 
 
 
 i6-S6 
 
 
 (N 
 
 
 
 
 CO 
 
 
 lO 
 
 96-S6 
 
 
 
 
 <M 
 
 
 iH 
 
 - 
 
 -* 
 
 86-16 
 
 
 (M 
 
 
 
 
 - 
 
 (N 
 
 lO 
 
 16-68 
 
 
 00 
 
 -H 
 
 (N 
 
 
 r- 
 
 <N 
 
 s 
 
 68-i8 
 
 
 ■* 
 
 
 « 
 
 
 Tji 
 
 Til 
 
 »o 
 
 i8-S8 
 
 rt 
 
 o 
 
 
 tH 
 
 
 CO 
 
 tn 
 
 s 
 
 
 
 
 
 
 
 
 
 
 
 S8-S8 
 
 iH 
 
 1-t 
 
 
 1-1 
 
 
 CO 
 
 CO 
 
 lO 
 
 
 
 
 
 
 
 
 
 
 
 E8-I8 
 
 
 t> 
 
 
 M 
 
 
 IN 
 
 CD 
 
 1^ 
 
 I8-6i 
 
 
 CO 
 
 ^ 
 
 M 
 
 
 M 
 
 ^ 
 
 o 
 
 i 
 1 
 
 
 
 
 
 
 
 
 
 
 %1.-U 
 
 
 c 
 
 
 
 
 CD 
 
 o 
 
 s 
 
 11-91 
 
 
 c 
 
 - 
 
 M 
 
 
 CO 
 
 t^ 
 
 g 
 
 91-21 
 
 rH 
 
 c 
 
 
 
 
 -- 
 
 o> 
 
 
 21-U 
 
 i-l 
 
 IT 
 
 -- 
 
 CO 
 
 
 CO 
 
 OJ 
 
 CO 
 
 
 
 
 
 
 
 
 
 
 g 
 
 H 
 
 U-69 
 
 
 ;i 
 
 
 « 
 
 
 IC 
 
 o 
 
 g 
 
 
 
 
 
 
 
 
 
 
 H 
 
 69-19 
 
 
 e- 
 
 
 cq 
 
 
 -* 
 
 ■* 
 
 s 
 
 i9-S9 
 
 
 IP 
 
 tH 
 
 - 
 
 
 - 
 
 cq 
 
 s 
 
 S9-E9 
 
 iH 
 
 ■^ 
 
 
 
 
 N 
 
 CD 
 
 CO 
 
 S9-I9 
 
 iH 
 
 ■4 
 
 
 
 
 1-1 
 
 CO 
 
 o 
 
 I9-6S 
 
 
 T}1 .-1 
 
 
 
 -■ 
 
 CO 
 
 2 
 
 69-2.9 
 
 
 ■* 
 
 
 
 
 - 
 
 lO 
 
 19-99 
 
 IM 
 
 -* 
 
 
 
 
 
 !> 
 
 Bff-SB 
 
 
 O 
 
 SS-TS 
 
 -- 
 
 - 
 
 IS-6^ 
 
 - 
 
 -- 
 
 6*-i* 
 
 -> 
 
 ^ 
 
 
 
 
 
 
 s 
 
 "o 
 
 
 
 
 
 
 
 
 
 
 
 CI 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 a 
 
 
 ^ 
 
 1 
 
 t 
 
 ti 
 
 
 
 
 
 
 1 
 
 6 
 
 1 
 1 
 
 ^ 
 
 1 
 
 
 i 
 
 
 
 
 
 •i 
 
 I 
 
 1 
 
 si 
 
 §5 
 
 ■s 
 
 •i 
 
 ■a 
 1 
 
 ct 
 
 -5 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 U 
 
 ^ 
 
 
 
 
 
 
 in>£ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 d 
 
 r" 
 
 ti 
 
 flf^ 
 
 gC 
 
 S 
 
 ^ 
 
 
 
 
 
 
 
 <U 
 
 
 
 
 
 
 
 fn 
 
 fr 
 
 (Q 
 
 fa 
 
 
 lO 
 
 V 
 
 
 
 
 O 
 
 C 
 
 o 
 
 O 
 
 O 
 
 o 
 
 o 
 
 
 
 II 
 
 o 
 o 
 
 II II 
 
 HO 
 
 473 
 
474 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 the mentality of given groups is in this last division of general crim- 
 inality. 
 
 Mean Mental Capacity (Test Aggregate) : 
 
 At First Offense At Present Offense 
 
 Offenses against the Person 68.4 69.4 
 
 Offenses against Chastity 75.2 7S.S 
 
 Offenses against Family 73.0 71.6 
 
 Offenses against Public Health 75.9 75.3 
 
 Offenses against Administration of Gov- 
 ernment 67.0 67.9 
 
 Offenses against Property Rights 79.2 78.5 
 
 General Criminality 74.9 70.6 
 
 When we consider the group under the more general classification 
 of misdemeanants and felons on first conviction, we find that the dif- 
 ference is slight in amount and not demonstrably certain (see Table 
 197). Neither is there a demonstrable difference in the variability of 
 the two groups. It will be noted that there is an appreciably smaller 
 proportion of individuals guilty of felonies on first convictions than on 
 later conviction. 
 
 TABLE 197 
 FELONS AND MISDEMEANANTS ON FIRST CONVICTION 
 
 Comparison of the Means and Standard Deviations of Mental Capacity, 
 
 as Measured by Test Aggregate, of Those Classified as Felons 
 
 and Misdemeanants on First Conviction 
 
 English-speaking Cases Only 
 
 
 Felona 
 
 Misde- 
 meanants 
 
 Difference 
 
 d 
 (Td 
 
 Chances that real 
 
 difference does 
 
 not eirist are 1 
 
 m: 
 
 Mean 
 
 77.8 
 + 1.39 
 
 75 35 
 ± 519 
 
 2 45 
 
 1.63 
 
 19 
 
 
 
 a 
 
 9.00 
 
 ±.868 
 
 9 42 
 + 334 
 
 -.42 
 
 .46 
 
 3 
 
 (Jases 
 
 42 
 
 330 
 
 
 
 
 
 
 If we consider the relation of mental capacity to the question of 
 the disposition of the cases on first conviction, we find a correlation 
 ratio of .20dz.050 (Table 198). This would seem to indicate that the 
 relationship, though small, is probably valid. The means run in de- 
 creasing order: State Prison, 77.6; Fines, 77.3; Probation, etc., 77.2; 
 Reformatory, 76.9 ; Juvenile Institutions, 76.8 ; Penitentiary and Work- 
 house, 73.1. It will be noticed that the only mean which would seem 
 
i 
 
 Vi 
 CO 
 
 1 
 
 
 CO _4 CH 
 
 S i 
 ^ -s 
 
 1 
 
 a> 
 v 
 
 o 
 
 •■§ 
 
 o 
 O 
 
 1^2 
 
 •3 
 
 S6-S6 
 
 E6-T6 
 16-68 
 
 6S-IS 
 
 IS-9S 
 
 58-68 
 £8-18 
 
 I8-6i 
 61-U 
 
 11-91 
 
 Si-Si 
 
 Sl-li 
 
 ti-69 
 
 69-Z9 
 
 i9-S9 
 
 B9-E9 
 
 T9-6S 
 
 Z9-fiS 
 
 ss-ss 
 
 ES-IS 
 
 19-6^ 
 
 6V-I.f 
 
 NMOOOS^eO 
 
 TjliH U3U3i-i 
 
 COr-l OOlH 
 
 SO Nrtocq 
 
 Tl< CCIONCO 
 
 t^rt WOIOM 
 
 00 rH rH CO »0 i-t 
 
 OO^^CO<M1N 
 
 COIN COrtp- 
 
 M rtl^OO 
 
 rHlOiH O 
 
 iH COCO 
 
 Ot i-t^ 
 
 d CO 
 
 3 o 
 
 o a >^ H fi3 
 
 aaus^nsg ^(saij 
 
 475 
 
 I 
 
 iH d 
 
 
 e3 M 
 £3 
 
 Mia 
 
 -^ h 
 ED ^ 
 V O 
 
 BO 
 
476 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 to differ materially from the others is that for the women committed 
 to the Penitentiary and Workhouse. We feel confident that a part, 
 at least, of the explanation for this lies in the tendency of the judges 
 to send to the Workhouse on their first offense colored women who 
 "had they been white" might have been put on probation or otherwise 
 disposed of. It has been shown in Chapter VIII that a much larger 
 percentage of the colored than of the white had been sent to institu- 
 tions of the Penitentiary and the Workhouse type on their first con- 
 victions. Table 201 shows that the mean score which the colored 
 women as a whole are able to make on Test Aggregate is decidedly 
 lower than that of the white, so that it seems logical that an undue 
 percentage of colored in the Workhouse and Penitentiary group 
 might be the reason for the decided drop in this mean as compared 
 with the others. 
 
 It is evident that our figures do not agree with those which Gor- 
 ing offers as showing that with progressively more' and more severe 
 sentences the mean intelligence of his subjects increases. We print 
 an abbreviation of his table* and quote him as follows : 
 
 Nature of First Sentence Intelligence: Means 
 
 Less than 3 months .59 
 
 3 months 1.47 
 
 6 months 2.3S 
 
 1 year 
 
 5 years and over 3.63 
 
 Birched .58 
 
 Reformatory school .64 
 
 Birched and school .30 
 
 Fined 10 
 
 Nil ; , .25 
 
 Total 1.38 
 
 "It will be observed that the mean intelligence of convicts increases 
 progressively as the sentence, resulting from thir first conviction, for 
 crime becomes progressively more and more severe. Now, in the case of 
 first offenders, a slight sentence presupposes a trivial offense: from which 
 it follows that the more mentally defective a convict may be, the more 
 trivial will his first offense have been. ... It is evident that the condition 
 most closely related to petty crime, the most fruitful source of nearly all 
 that is meant by the term crime in this country, must be mental de- 
 fectiveness." 
 
 It will be remembered, however, that his supposition that a slight 
 sentence presupposes a proportionately slight offense does not strictly 
 
 ' Op. at., p. 283. 
 
MENTAL CAPACITY 
 
 477 
 
 apply in our case, since with our women the same offense, e.g., solicit- 
 ing, may result in fines, probation, reformatory, or Workhouse sen- 
 tence, depending upon the judgment of the court. The only indica- 
 tion of a tendency of our results to support the theory advanced by 
 Goring is that the means for those women sent to state prison on their 
 first offense is slightly higher than any other. 
 
 As was shown in Chapter VI, the number of those who have juve- 
 nile convictions is very small, only 40 cases out of 587. When these 
 numbers have been pruned still further by the elimination of those 
 
 Per 
 
 Cent 
 
 Convicted before 21 yn. 
 • Convicted after 21 yrs. 
 
 93 97 
 
 XHXyf 
 
 Chart XXXV 
 
 Mental Capacity as Measured by Test Aggregate 
 
 Percentage comparison between women convicted first before 21 years of 
 age and women convicted first after 21 years of age. English-speaking cases 
 only. 
 
 who failed to be represented in all parts of Test Aggregate, the num- 
 bers become too small to be suitable for statistical treatment. As a 
 substitute for this we have attempted to group together those cases 
 who were first convicted before 21 years and those whose first con- 
 viction came after they were 21. Chart XXXV shows the distribution 
 for the two. It will be seen that the means are practically identical but 
 the scores of those convicted after 21 years tend to scatter more about 
 the mean than those earlier convicted. Table 199 proves that there 
 is in fact no demonstrable difference between the means but that the 
 amount of difference in the standard deviations is almost certainly 
 valid. We may say, therefore, that, if there is a difference in the 
 average mental capacity of those whose first conviction came at an 
 
478 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 TABLE 199 
 
 FIRST CONVICTED BEFORE 21 YEARS AND FIRST CONVICTED 
 AFTER 21 YEARS OF AGE 
 
 Comparison of the Means and Standard Deviations of Mental Capacity, 
 
 as Measured by Test Aggregate, of those First Convicted before 21 
 
 Years and those First Convicted after 21 Years of Age 
 
 English-speaking Cases Only 
 
 
 Convicted 
 before 21 
 
 Convicted 
 after 21 
 
 Difference 
 
 d 
 0-d 
 
 Chances that real 
 
 difference doea 
 
 not exist are 1 
 
 in; 
 
 Mean 
 
 75.70 
 
 ±.714 
 
 75.65 
 ±.664 
 
 .050 
 
 .0517 
 
 2 
 
 Cm. . . 
 
 
 
 
 (J- 
 
 8.53 
 ±.518 
 
 9.96 
 ±.397 
 
 -1.422 
 
 2.178 
 
 68 
 
 Oa 
 
 
 
 
 Cases 
 
 143 
 
 225 
 
 
 
 
 
 
 early age as compared with those whose first conviction came later 
 in life, 21 years is not the boundary which makes the separation, but 
 that those first convicted late have a greater degree of variability with 
 respect to intelligence. 
 
 Sum/matry 
 
 Summarizing our findings concerning the relationship between the 
 mental capacity of our group and the main aspects of their criminal 
 records, we note : 
 
 ( 1 ) That we find a small but probably significant degree of asso- 
 ciation when the mental capacity of these women as measured by Test 
 Aggregate is related to the nature of the crimes which they commit, 
 classified according to the form employed by the New York City PoUce 
 Department. The mean of the property offenders is sensibly better 
 than that of offenders against chastity and this difference is demon- 
 strated as almost certainly valid. We can say, then, that the women of 
 this group who have been convicted for offenses against property form 
 a group mentally superior to those convicted for offenses against chas- 
 tity. 
 
 (2) Dividing the total group into misdemeanants and felons we 
 find no demonstrable distinction in mental capacity. If the distinction 
 between felons and misdemeanants can be thought of as roughly rep- 
 resenting the seriousness of offenses the indications are that the men- 
 
MENTAL CAPACITY 479 
 
 tal capacity of women offenders varies with the nature of the offense 
 rather than with the degree of the offense. We hesitate to emphasize 
 the point, however, in view of the unsatisfactory nature of the dis- 
 tinction between felonies and misdemeanors. (See Chapter VI.) 
 
 (3) Between the mental capacity of these women as measured by 
 Test Aggregate and the extent of their criminal record as measured by 
 the number of previous convictions, we have found only very slight 
 and more or less ambiguous indications of a negative correlation. 
 
 Using as a measure of criminal tendency the length of time a woman 
 has been imprisoned, we have again found a small relationship which, 
 as before, is not free from considerable ambiguity. In this case the 
 tendency is toward positive correlation. 
 
 In considering both these relationships, we have allowed for the 
 factor of present age and found that it made no appreciable difference 
 in the correlation. 
 
 (4) We have found no demonstrable difference between the men- 
 tal capacity of recidivists and that of those who are first offenders. 
 
 (5) No evidence is available from our data to show the existence 
 of an appreciable correlation between mental capacity and age at first 
 conviction. 
 
 (6) We found a correlation ratio between intellectual capacity and 
 nature of first offense, arranged according to the New York City 
 Police Department classification, very similar to that found between 
 intellectual capacity and the present offense so classified. 
 
 (7) Classifying the first offense as misdemeanors or felonies, we 
 found a slight difference between the two groups in favor of the felons, 
 but a difference not convincingly valid. 
 
 (8) A small relationship was found to exist between the mental 
 capacity of the women and the disposition of their case on first offense. 
 We note that the group of women who, on first conviction, were sen- 
 tenced to the Penitentiary or the Workhouse, is separated from the 
 others by a greater interval than occurs elsewhere. We are inclined 
 to think this may be explained by the large percentage of colored 
 women committed to these institutions, whose mental capacity is 
 found to be inferior to that of the white women. 
 
 (9) We have found no demonstrable difference in the average 
 mental capacity of those first convicted before 21 and those first con- 
 victed at a later age, but have found that those who were convicted 
 later show a greater variability around their mean than do those 
 whose first conviction was earlier. 
 
480 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 MENTAL CAPACITY IN RELATION TO GENERAL FACTORS 
 
 There are certain general factors whose relation to mental capacity 
 one feels called upon to consider and which, although not properly asso- 
 ciated have, for convenience of handling, been grouped together in 
 Chapter VII and will be so treated here. 
 
 (a) Mental Capctcity and Age 
 
 We have considered first the matter of age as it relates to intel- 
 lectual capacity. Table 200 shows the correlation which exists between 
 the two. The coefficient obtained is practically zero, —.01 ±.052. 
 Calculating the ratios shows one to be negligible, .08±.051 and the 
 other to be slight, .18+.050, but considering the irregularity of its line 
 of means, not very significant. We feel safe in saying, then, that our 
 results tend to confirm the belief that, among adults, there is little if 
 any variation in mental capacity with age. 
 
 This fact is especially important since the factor of age is one which 
 seemingly should be allowed for in many correlations, but which, by 
 reason of the small size of its correlation with intelligence, has, as we 
 have seen in the partial obtained between mental capacity and num- 
 ber of convictions with age constant, made only an insignificant change 
 in the size of the coefficient, thereby relieving us of the necessity of 
 making this allowance in many cases. 
 
 (&) Mental Capacity and Classification by Color and Nativity 
 
 In order to consider next the question of color in its relation to 
 mental capacity as measured by Test Aggregate, we have plotted the 
 distribution curves as shown in Chart XXXVI. It requires only the 
 most casual inspection of this chart to prove the decided superiority of 
 the white women. The white women are represented at the extreme 
 lower end of the scale, it is true, but the curve of the colored women rises 
 to its summit and falls away again at an earlier point than that of the 
 white and is moreover unrepresented in the last interval. The means 
 are 77.13 + .442 for the white as against 70.91 ±.839 for the colored. 
 Table 201 shows that this diflference is unquestionably significant. We 
 have no hesitation in saying, then, that there exists a decided differ- 
 ence in mental capacity between the colored and the white women as 
 measured by this standard. As regards the amount of variability dis- 
 played by the two groups, it is interesting to note that they are nearly 
 identical. 
 
a> 
 
 < 
 
 cil 
 bO 
 
 CO 
 
 H 
 
 V 
 
 3 
 "=> ^ 
 
 O ™ 
 
 s 
 
 o 
 
 0) 
 
 d 
 I 
 
 o 
 u 
 
 
 ill 
 
 CO »o •«*< lO m Tji 00 
 
 ■^ CO »0 T)J (D lO U5 
 
 t* t* b- i> i> r^ i> 
 
 
 
 
 
 1 
 
 o 
 
 
 
 »D 
 
 
 O .-< (N irq o lo 00 1- M lo t> -* oi 00 o CO Oi '^ 
 
 iHi-I.HWMCOCDiCOrH 
 
 
 1 
 
 G3 
 
 g 
 
 
 
 < 
 
 1 
 
 Z6-S6 
 
 ,-t r-l iHi-Hi-H 
 
 ID 
 
 1-1 
 
 
 S6-S6 
 
 rHiH iHiH 
 
 ■* 
 
 
 S6-T6 
 
 .H tH iH (N 
 
 >D 
 
 
 
 16-68 
 
 iH «iO«(NTt<N 
 
 S 
 
 
 68-Z8 
 
 rt ^Nrt^ rt rtpHIMCCCl 
 
 CD 
 
 
 
 Z8-S8 
 
 rtrH 1-1 rH N Tji t* N ^ 
 
 g 
 
 
 fiS-E8 
 
 i-HN i-H Tt< tH CD M -^ 
 
 ID 
 C4 
 
 
 
 S8-I8 
 
 rt U5 CO CO cq Tf* CO th i^I 
 
 
 18-6Z 
 
 CO 
 
 ID 
 
 o 
 
 Gl-ll 
 
 1-H .-1 COi-(Tt< lOiHt-CO © 
 
 u-u 
 
 rH,-H -HIM t-i rt N o Tti ■* (N en 
 
 ID 
 
 u 
 
 91-^1 
 
 ^ ^^N-^rHrHCDlO N 
 
 U 
 
 21-11 
 
 ,-( T-i (N .H N •* ^ CC ^ (M CO t> --H N 
 
 00 
 
 bl 
 
 TZ-69 
 
 rH .-H N rH M CO IM CD ■* to 00 
 
 1 
 
 69-Z9 
 
 rn 1-1 (N ^ ^ U3 ID CO •* iH 
 
 S 
 
 
 ? 
 
 Z9-S9 
 
 ^.HWrHiHCq CO 
 
 O 
 
 •s 
 
 S9-S9 
 
 i-H iH (N N r-1 CO M »H 
 
 CO 
 
 C3 
 CO 
 
 PC] 
 
 £9-19 
 
 W W ,H t-*<N .Hrt(M 
 
 o 
 
 T9~6S 
 
 CO CO (N CT 
 
 o 
 
 ID 
 
 
 6S-ZS 
 
 ^ ^r^<^^ 
 
 ID 
 
 
 19-99 
 
 rH NrHW »H 
 
 t> 
 
 to 
 
 'i 
 
 
 fifiTse 
 
 
 o 
 
 
 £5-ie 
 
 -^ 
 
 iH 
 
 
 19-6^ 
 
 - 
 
 -- 
 
 
 6^Z^ 
 
 rH 
 
 iH 
 
 
 If-^f 
 
 
 o 
 
 
 
 
 
 o 
 
 fl'HJ 
 
 Is 
 
 
 83V 
 
 
 a a 
 
 
 
 
 
 
 
 s 
 
 •sf 
 
 en 
 
 oo 
 
 
 
 ml" 
 
 ag 
 
 
 
 
 ^•^ 
 
 
 
 h1 
 
 Sm 
 
 t^ 
 
 
 
 
 
 V0»0 
 
 
 
 
 
 s^ 
 
 ■im 
 
 lis 
 
 sa 
 
 s- 
 
 II II 
 
 s 
 
 3 c-p 
 
 
 . V o 
 
 
 
 
 
 s* 
 
 1-H ^ £ 
 
 
 
 
 
 S' 
 
 1 1S« 
 
 
 Is! 
 
 
 H<t| 
 
 (H Mfl 
 
 
 
 
 fl-'^ OJ 
 
 
 ■eaH*^ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 1^" 
 
 
 -A 
 
 
 
 
 ■a 2 
 
 
 
 
 
 
 
 
 OQ 
 
 481 
 
482 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 Per 
 
 Cent 
 
 White 
 Colored' 
 
 oL 
 
 Chart XXXVI 
 
 Mental Capacity as Measured by Test Aggregate 
 
 Percentage comparison between white (407) and colored (113). 
 speaking cases only. 
 
 English- 
 
 For further interest we have figured here also the percentage dis- 
 tribution curves for the colored and white women on the Yerkes- 
 Bridges Point Scale, the Stanford-Binet, and our group of Perform- 
 ance Tests. Charts XXXVII, XXXVIII, and XXXIX show these. 
 
 An inspection of these makes it clear that they agree in showing the 
 white to be clearly a superior group. It is significant that the tests 
 
 TABLE 201 
 
 COLORED AND WHITE 
 
 Comparison of the Means and Stand^d Deviations of Mental Capacity, 
 as Measured by Test Aggregate, of Colored and White Women 
 
 
 English-speaking Cases Only 
 
 
 
 
 White 
 
 Colored 
 
 Difference 
 
 d 
 
 Chances that real 
 
 difference does 
 
 not exist are 1 
 
 in: 
 
 Mean 
 
 77.03 
 + .442 
 
 70.91 
 ±.839 
 
 6.12 
 
 6.45 
 
 00 
 
 
 
 
 
 /r 
 
 8.91 
 + .305 
 
 8.92 
 ±.654 
 
 -.01 
 
 .083 
 
 2 
 
 <Ta 
 
 
 
 
 Cases 
 
 407 
 
 113 
 
 
 
 
 
 
MENTAL CAPACITY 
 
 483 
 
 Per 
 Cent 
 
 54 60 ee 
 
 POINTS 
 
 Chart XXXVII 
 
 Mental Capacity as Measured by Yerkes-Bridges Point Scale 
 
 Percentage comparison between white (478) and colored (129). English- 
 speaking cases only. 
 
 Per 
 
 Cent 
 
 — . ^ ^ Colored 
 
 M 70 76 82 88 94 100 106 112 118 124 
 
 INTELLIGENCE QUOTIENT Worv... 
 
 Chart XXXVIII 
 
 Mental Capacity as Measured by Stanford-Binet 
 
 Percentage comparison between white (447) and colored (118). English- 
 speaking cases only. 
 
484 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 Per 
 Cent 
 
 *•»"»>» Colored 
 
 . ' 
 
 "I 
 
 
 r 
 
 
 1 
 1 
 
 
 
 1 — ' 1 ' 
 
 ! 
 
 
 
 
 
 
 j 
 
 
 . 
 
 I 
 
 58 
 
 a. 
 
 SCORE 
 
 Chart XXXIX 
 
 Mental Capacity as Measured by Performance Tests 
 
 Percentage comparison between white (453) and colored (107). 
 
 in which language is not involved agree with the other types of tests 
 in these findings. The constants for each are shown below. 
 
 
 Mean 
 
 
 
 
 White 
 
 Colored 
 
 White 
 
 Colored 
 
 Test Aggregate 
 
 77.03 
 + .442 
 
 70.91 
 ±.839 
 
 8.91 
 
 ±.305 
 
 8.92 
 ±.654 
 
 Yerkes-Bridges Scale . 
 
 71.80 
 + .608 
 
 65.4 
 ±1.30 
 
 13.29 
 ±.416 
 
 14.82 
 ±.687 
 
 Stanford-Binet 
 
 73.25 
 ±.661 
 
 65.4 
 ±1.10 
 
 14.01 
 ±.713 
 
 11.90 
 + .461 
 
 Performance Tests. . 
 
 73.78 
 ±.379 
 
 69.26 
 
 ±.777 
 
 8.08 
 ±.367 
 
 8.04 
 ±.596 
 
 Our only possible basis of comparing native with foreign groups 
 with respect to mentality is the group of Performance Tests, since 
 these can be given to all individuals. In the present comparison we 
 have distinguished the groups in accordance with our classification of 
 English-speaking versus non-English-speaking (see page 59) rather 
 than on an exact basis of nativity. Chart XL shows the relationship 
 between these two groups. The superiority of the English-speaking 
 group, by this measure, is readily apparent. Its mean is sufficiently 
 higher than that of the non-English-speaking group so that there is 
 no question of the validity of the difference (73.67±.363 as against 
 
MENTAL CAPACITY 
 
 485 
 
 67.95 ±.957) and the higher location of the mode may be seen by in- 
 spection. Whether the superiority indicated in favor of the EngHsh- 
 speaking delinquent women over the non-English-speaking can be 
 considered wholly established remains, however, something of a ques- 
 tion, since it is probable that at least a slight handicap persists for the 
 very foreign and frequently wholly uneducated women of the non- 
 
 Cent 
 
 2S . 
 
 '5 . 
 
 EnsHsh-Speaklnir 
 Non Engllsh-SpeBklDE 
 
 Chart XL 
 
 Mental Capacity as Measured by Performance Tests 
 
 Percentage comparison between English-speaking cases (486) and non- 
 English-speaking cases (74). 
 
 English-speaking group. Accordingly, we would not emphasize this 
 distinction overmuch, but merely offer the comparison as the best 
 which our data make possible. 
 
 (c) Mental Capacity and Use of Drugs and Alcohol 
 
 We have attempted to see if the using of drugs or alcohol has had 
 any effect on the mental capacity of these women as measured by Test 
 Aggregate. Chart XLI shows the distribution of those women who 
 have used drugs in comparison with those that have not. As will be 
 seen the curves resemble each other closely. The means are almost iden- 
 tical and Table 202 shows that the slight difference can not be demon- 
 strated as significant. The difference in the variability shown by the 
 two groups is likewise very small and not demonstrably significant. We 
 feel safe in concluding, therefore, that delinquent women who have 
 been addicted to drugs do not differ appreciably in mental capacity 
 from those who have not. 
 
486 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 Per 
 Cent 
 
 Druff-Uaeri 
 Nod Drug-Uiera 
 
 Chart XLI 
 
 Mental Capacity as Measured by Test Aggregate 
 
 Percentage comparison between drug-users and non-drug-users. English- 
 speaking cases only. 
 
 In order to measure the possible relation between mental capacity 
 and the use of alcohol we have calculated a correlation ratio with 
 mental capacity as the quantitative variable and the use of alcohol, 
 classified as excessive, moderate, and non-alcoholic, as the qualitative. 
 We have added to the first group those cases from the special "Intoxi- 
 cation Group" at the Workhouse that were tested by Test Aggregate. 
 Table 203 shows the correlation of the two factors. The correlation 
 
 TABLE 202 
 DRUG-USERS AND NON DRUG-USERS 
 
 Comparison of the Means and Standard Deviations of Mental Capacity, 
 
 as Measured by Test Aggregate, of Drug-users and Non-Drug-users 
 
 English-speaking Cases Only 
 
 
 Drug-users 
 
 Non- Drug- 
 users 
 
 Difference 
 
 d 
 
 Chances that real 
 difference does ^ 
 not exist are 1 ■-: 
 in: 
 
 
 75.71 
 ±.986 
 
 75.63 
 ±.570 
 
 .08 
 
 .076 
 
 2 
 
 
 
 um 
 
 
 a 
 
 9.03 
 ±.668 
 
 9.57 
 ±.361 
 
 -.54 
 
 .703 
 
 4 
 
 ffff , 
 
 
 
 
 Cases 
 
 84 
 
 282 
 
 
 
 
 
 

 g 
 
 
 Hi 
 
 
 ^^ 
 
 
 o 
 
 
 j:^ 
 
 
 o 
 
 
 u 
 
 
 
 
 <! 
 
 
 
 o 
 
 o 
 
 •? 
 
 
 l3 
 
 S) 
 
 ^ 
 
 P 
 
 i 
 
 
 * 
 
 1 
 
 o 
 
 ^ 
 
 c^ 
 
 S> 
 
 a 
 
 c 
 
 i 
 
 a 
 
 a 
 
 4j 
 
 S 
 
 
 S 
 
 H 
 
 
 
 >. 
 
 
 .a 
 
 p< 
 
 TJ 
 
 CO 
 
 4) 
 
 "S 
 
 3 
 
 ■a 
 
 OQ 
 
 d 
 
 tS 
 
 « 
 
 4> 
 
 
 £? 
 
 5 
 
 
 
 o 
 
 ■s) 
 
 m 
 
 -u 
 
 a 
 
 o 
 
 «a 
 
 H 
 
 O 
 
 "S 
 
 ^M 
 
 
 ^ 
 
 <u 
 
 g 
 
 O 
 
 a 
 
 d 
 
 4> 
 
 M 
 
 4> 
 
 c« 
 
 3 
 
 o< 
 
 V 
 
 
 ^ 
 
 M 
 
 
 
 13 
 
 3 
 
 o 
 
 bO 
 
 ■■s 
 
 a 
 
 
 
 <1> 
 
 
 s 
 
 
 o 
 
 
 U 
 
 
 
 
 
 
 1 
 
 
 1 
 
 1 
 
 s 
 
 
 
 
 ^t 
 
 it 
 
 K K 
 
 
 "U 
 
 
 
 
 
 <! 
 
 
 
 
 
 m 
 
 
 
 
 
 ■a 
 
 '^ 
 
 ■* CO 
 
 
 
 
 
 
 
 
 
 
 
 
 
 B 
 
 
 
 
 
 i6-S6 
 
 - 
 
 ^ CO 
 
 U3 
 
 B6-E6 
 
 CO 
 
 fH 
 
 ■<*< 
 
 E6-I6 
 
 PC C4 
 
 lO 
 
 16-68 
 
 CO 
 
 (O »-l 
 
 o 
 
 
 
 
 
 
 
 
 68-Z8 
 
 lO 
 
 lO b- 
 
 t> 
 
 Z8-fi8 
 
 U3 
 
 o o 
 
 g 
 
 S8-E8 
 
 M< 
 
 CO CO 
 
 s 
 
 £8-18 
 
 SO 
 
 lO CO 
 
 
 I8-6i 
 
 W 
 
 GO CO 
 
 CO 
 
 61-11 
 
 o 
 
 ■«J< (N 
 
 S 
 
 e 
 
 
 
 
 
 U-91 
 
 (O 
 
 OJ t^ 
 
 CO 
 
 
 
 
 
 5 
 
 CS 
 C5 
 
 91-S2, 
 
 © 
 
 (0 o 
 
 
 Sl-Xl 
 
 l> 
 
 CO t> 
 
 g? 
 
 
 
 
 
 -»J 
 
 IZ-69 
 
 00 
 
 ^ 2 
 
 s 
 
 
 
 
 
 
 ^ 
 
 69-Z9 
 
 U5 
 
 OS ^ 
 
 s 
 
 
 
 
 
 
 Z9-S9 
 
 CO 
 
 «) IN 
 
 
 
 S9-E9 
 
 lO 
 
 (N I> 
 
 
 £9-19 
 
 ■* 
 
 M lO ■ 
 
 m 
 
 I9-6B 
 
 CO 
 
 CO -<*< 
 
 o 
 
 6S-Z9 
 
 iH 
 
 -* 
 
 o 
 
 ZS-BS 
 
 ^ 
 
 « M< 
 
 t* 
 
 SS-ES 
 
 
 o 
 
 ES-IS 
 
 - 
 
 1-1 
 
 IB-6^ 
 
 - 
 
 iH 
 
 6!"i* 
 
 1-1 
 
 - 
 
 lf-9f 
 
 
 o 
 
 
 
 
 
 cj 
 
 
 
 
 
 
 
 a 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 p- 
 
 ■S ii 
 
 
 
 
 
 s 
 
 SJ 1= 
 
 ja 
 
 
 
 1 
 
 S 1 
 
 1 
 
 a 
 
 iBljoqcoiv 
 
 
 
 } 
 
 ae 
 
 jaaa 
 
 
 1 
 
 rH a 
 
 IS s 
 
 487 
 
488 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 ratio obtained is practically zero, .044±.051. On the chance that this 
 distinction was too fine and the possible diflference might be between 
 those who have used alcohol to excess and those who have not, we 
 have grouped together the moderate users and the non-alcoholics and 
 compared them with those who have used alcohol to excess. Chart XLII 
 shows the relative distributions. The diflference in both the central 
 tendency and the amount of variability is negligible (see Table 204). 
 
 Cent 
 
 
 
 
 J5 
 
 • 
 
 
 ao 
 
 1 1 
 
 
 15 
 
 I — T"" 
 
 h-- 
 
 1 
 
 1 
 
 1 
 
 
 10 
 
 1 
 
 
 5 
 
 1 
 
 
 """l 
 
 
 ■""" 
 
 i_ 
 
 
 , — i___P^ '* 
 
 — '—* 
 
 
 45 49 S3 57 61 65 69 73 77 81 85 
 
 SCORE 
 
 S 93 97 
 
 Chart XLII 
 
 Mental Capacity as Measured by Test Aggregate 
 
 Percentage comparison between women who have been excessively alcoholic 
 and women who have not been excessively alcoholic. English-speaking cases 
 only. 
 
 We feel, therefore, justified in stating that those of our group who 
 have been alcoholic, however this term may be interpreted, do not differ 
 significantly with respect to mental capacity from those that have not. 
 
 Summary 
 
 Summarizing this section, then, we can say : 
 
 (1) Our data confirm the accepted theory that for a group of 
 adults there is no demonstrable correlation between mental capacity 
 and age. 
 
 (2) Our results show also that white women of our delinquent 
 groups are superior in intellectual capacity to colored women. We 
 have found little difference in the degree of variability shown between 
 the white and colored groups. The evidence at hand indicates, further, 
 that the women delinquents, born in English-speaking countries, are 
 somewhat superior to those born in non-English-speaking countries. 
 
MENTAL CAPACITY 
 
 489 
 
 TABLE 204 
 
 EXCESSIVELY ALCOHOLIC AND NOT EXCESSIVELY ALCOHOLIC 
 
 Comparison of the Means and Standard Deviations of Mental Capacity, 
 
 as Measured by Test Aggregate, for those Excessively Alcoholic 
 
 and those not Excessively Alcoholic. Total Group Including 
 
 Special Intoxication Group from Workhouse 
 
 
 English-speaking 
 
 Cases Only 
 
 
 
 
 Excessively 
 Alcoholic 
 
 Not 
 
 Excessively 
 
 Alcoholic 
 
 Difference 
 
 d 
 
 0-a 
 
 Chances that real 
 
 difference does 
 
 not exist are 1 
 
 in: 
 
 Mean 
 
 74.95 
 + .995 
 
 75.69 
 ±.546 
 
 -.74 
 
 .655 
 
 4 
 
 (7n, 
 
 
 
 
 a 
 
 9.64 
 + .703 
 
 9.25 
 + .353 
 
 .39 
 
 .499 
 
 3 
 
 ffa 
 
 
 
 
 
 94 
 
 287 
 
 
 
 
 
 
 (3) We have found no demonstrable difference between the men- 
 tal capacity of those women who have used drugs and those who have 
 not, and no demonstrable difference in the mental capacity of the 
 women who have used alcohol and those who have not. 
 
 MENTAL CAPACITY IN RELATION TO HOME CONDITIONS 
 
 We will consider next the relationship between the mental capacity 
 of these women and (1) the home conditions which surrounded them 
 in childhood and adolescence, and (2) the possible hereditary factors 
 which might have affected their mentality. 
 
 (o) Mental Capacity and Estimate of Home Conditions 
 
 In order to examine the first of these relationships we have plotted 
 the correlation between mental capacity as measured by Test Aggregate 
 and the estimate of home conditions (Table 205). This estimate of 
 home conditions, as was earlier stated, has been based upon a com- 
 bination of the several factors of economic condition, moral standards 
 and parental supervision. This has then been divided into five grades, 
 grade 1 being the lowest in order of excellence. The correlation ratio 
 obtained is .31 + .053, which indicates that there is a clear degree of 
 relationship existent between the mental capacity of these women and 
 
5 
 
 d 
 
 
 Q 
 
 
 'g 
 
 
 
 
 •rt 
 
 
 
 
 Q 
 
 
 u 
 
 
 <u 
 
 
 a 
 
 
 o 
 
 
 » 
 
 
 M-l 
 
 
 o 
 
 
 4) 
 
 
 
 
 
 
 •1 
 
 t 
 
 M 
 
 o 
 
 M 
 
 n 
 
 ■T3 
 
 09 
 
 § 
 
 01 
 
 a> 
 
 bA 
 
 ■t; 
 
 
 & 
 
 ll> 
 
 
 R- 
 
 :^ 
 
 •s 
 
 
 
 ■s 
 
 'So 
 
 ^ 
 
 P^ 
 
 >. 
 
 , 
 
 ^ 
 
 •o 
 
 
 
 'C 
 
 a 
 
 § 
 
 ■a 
 
 o 
 
 cd 
 
 «> 
 
 Q^ 
 
 
 S 
 ^ 
 
 1 
 
 
 
 £<; 
 
 ^ 
 
 o. 
 
 « 
 
 R< 
 
 B. 
 
 o 
 
 
 3 
 
 o 
 
 e 
 
 •Si 
 
 4) 
 
 s 
 
 S 
 
 ^ 
 
 ^ 
 
 
 <D 
 
 
 4^ 
 
 
 B 
 
 
 0) 
 
 
 ^ 
 
 
 a 
 
 
 o 
 
 
 •■0 
 
 
 rt 
 
 
 
 
 
 
 {3 
 
 
 o 
 
 
 U 
 
 
 
 ■a 
 
 i6-S6 
 S6-E6 
 
 E6-I6 
 
 16-68 
 
 6S-IS 
 
 lS-^8 
 
 88-68 
 
 E8-18 
 
 I8-6Z 
 61-2,1 
 
 U-91 
 
 BZ-IZ 
 
 IZ-69 
 
 9-Z9 
 
 i9-99 
 
 S9-E9 
 
 £9-t9 
 I9-6B 
 
 6S-ZS 
 
 Z9-BS 
 
 ES-IS 
 
 6f-i* 
 
 N b- 113 
 
 >H M t^ CO 
 
 ■* t- 00 
 
 N' r* CO 
 
 o> <o e^ 
 
 »-t o OS 
 
 (N IN rt 
 
 & 1 
 
 d> O 01 
 
 > o s 
 
 &< > 
 
 BUOpipuoa 
 
 490 
 
 % 
 
 II <a 
 
 1^ 
 
 (U 9 
 
 o 
 
 HO 
 
MENTAL CAPACITY 491 
 
 the excellence of their home conditions in early life. As has been 
 noted earlier (cf. Chapter IX), it has been necessary to omit the Work- 
 house group from the comparison because of the impossibility of veri- 
 fying such data on many of this group and the evident tendency to ex- 
 aggeration in their own statements on this particular matter. 
 
 We have plotted also (Table 206) the relation between mental 
 capacity and economic status alone, — one of the factors which go to 
 make up the estimate of the home conditions — and find that the cor- 
 relation ratio obtained is identical with the preceding ratio, .31 ±.053. 
 Measured either by the economic status alone or by the estimate of 
 home conditions, there is a clear tendency for those women whose 
 early home condions have been good to make a better showing on 
 Test Aggregate than do those who come from poorer homes. 
 
 (&) Mental Capacity and Hereditary Factors 
 
 We have next considered the possible effect of heredity on the men- 
 tal capacity of these women. Data have been collected by social inves- 
 tigation on many factors, such as alcoholism, insanity, venereal dis- 
 ease, etc., in the family which have been recognized, more or less uni- 
 versally, as possible causes of defective intelligence in the offspring. 
 When, however, we have attempted to compute these separately our 
 numbers were found to be too small to be reliable, and if we try to sub- 
 divide these again into the presence of any one factor in one parent 
 alone we still further reduce our figures. Recognizing the impossibility 
 of making any reliable comparisons in this exact way, we have, there- 
 fore, considered the presence of any one of these factors in either 
 parent, or in the siblings where the defect was present from early 
 childhood, and have called it possible evidence of a defective strain in 
 the heredity. We have disregarded cases beyond the immediaie fam- 
 ily. The factors which we have taken account of are alcoholism, 
 amentia, epilepsy, insanity, neurotic constitution, gonorrhea and 
 syphilis. These may not all be factors universally recognized as either 
 symptomatic of or productive of hereditary defect and there are prob- 
 ably others of importance of which we have taken no account. The 
 varyirig amounts of each of these factors in our group are given in 
 Table 79, Chapter IX. Cases where the field workers had been unable 
 to obtain any data whatever on this point were omitted altogether. 
 There were 129 cases where defective strain was present in father, 
 mother or siblings, and 231 where there was no indication of such. 
 The mean for the former is 75.26+.816 and that for the latter 75.79 
 
S9 
 
 n 
 
 o 
 O 
 u 
 
 
 
 1^1 
 
 
 00 
 
 CO 
 
 
 
 t- 
 
 -* 
 
 g 
 
 
 ! 
 
 « 
 
 ^ 22 3 
 
 " s ::; 
 
 S 
 
 1 
 
 1 
 
 < 
 
 Z6-S6 
 
 -H 
 
 <N ^ 
 
 
 ■* 
 
 B6-E6 
 
 w N 
 
 CO 
 
 E6-T6 
 
 
 tH « 
 
 -H 
 
 ■* 
 
 T6-68 
 
 - 
 
 Til « N 
 
 
 OS 
 
 68-ig 
 
 •^ r^ M 
 
 ;* 
 
 Z8-S8 
 
 ^ t* lO 
 
 «: 
 
 
 S8-SS 
 
 
 CO o o» 
 
 ^ 
 
 CO 
 
 E8-T8 
 
 
 til t- b- 
 
 C4 
 
 o 
 
 C4 
 
 I8-6i 
 
 
 
 N 
 
 S 
 
 61-U 
 
 
 ,-( lO ©2 
 
 (N 
 
 R 
 
 USl 
 
 <N ^ ^ 
 
 S 
 
 9i-Sl 
 
 
 tH CO 1- 
 
 tH 
 
 S 
 
 Sl-U 
 
 
 ■* l> ;2 
 
 CO 
 
 S 
 
 xi-m 
 
 
 <N 00 
 
 -H 
 
 
 69-i9 
 
 
 rH lO «0 
 
 tH 
 
 O 
 
 i9-S9 
 
 
 rH ^ eO 
 
 - 
 
 CO 
 
 S9-E9 
 
 ^ iH lO 
 
 t> 
 
 E9-I9 
 
 rH i-l Tjl 
 
 «o 
 
 I9-6S 
 
 •* N 
 
 CO 
 
 69-ZS 
 
 « fH 
 
 CO 
 
 iS-SS 
 
 
 iH CO 
 
 '^ 
 
 >o 
 
 fig-E9 
 
 
 o 
 
 EB-IS 
 
 »H 
 
 w 
 
 TB-6* 
 
 - 
 
 - 
 
 6f-I.f 
 
 -< 
 
 ^ 
 
 Zf-S* 
 
 
 o 
 
 
 
 ;> 
 
 1 
 
 1 
 
 f- 
 
 1 
 
 1 
 
 
 SI] 
 
 
 OR! 
 irao 
 
 pno 
 [ioo[: 
 
 I 
 
 
 1^ 
 
 h O 
 
 o 
 BO 
 
 492 
 
MENTAL CAPACITY 
 
 493 
 
 ±.621. ■ The standard deviations were respectively 9.26±.577 and 
 9.44±.439. The differences are insignificant and, as Table 207 shows, 
 can not be demonstrated as conditioned by anything other than chance. 
 
 TABLE 207 
 
 DEFECTIVE STRAIN AND NOT DEFECTIVE STRAIN 
 
 Comparison of the Means and Standard Deviations of Mental Capacity, 
 
 as Measured by Test Aggregate, of those whose Immediate Family 
 
 Show Defective Strain and those whose Immediate 
 
 Family Show No Defective Strain 
 
 English-speaking Cases Only 
 
 
 Defective 
 
 Strain 
 
 No Defective 
 Strain 
 
 Difference 
 
 d 
 
 Chances tliat real 
 
 difference does 
 
 not exist are 1 
 
 in: 
 
 
 75.26 
 ±.816 
 
 75.79 
 ±.621 
 
 -.53 
 
 .515 
 
 3 
 
 (Tm 
 
 
 
 
 a 
 
 9.26 
 
 ±.577 
 
 9.44 
 ±.439 
 
 -.18 
 
 .248 
 
 2 
 
 (Ja 
 
 
 
 
 Cases 
 
 129 , 
 
 231 
 
 
 
 
 
 
 As far as our data are significant, therefore, we have nothing to 
 prove that the mental capacity of these women, as measured by Test 
 Aggregate, is in any way affected by the presence of what we have 
 termed defective strain in the immediate family. 
 
 Summary 
 
 We can say in summary, then : 
 
 (1) That there is a small but significant degree of association be- 
 tween mental capacity and the kind of home conditions by which these 
 individuals were surrounded in early life. 
 
 (2) We have been unable to demonstrate any relationship between 
 mental capacity and the presence in the immediate family of a defective 
 strain represented by any cases of alcoholism, amentia, epilepsy, in- 
 sanity, neurotic constitution or venereal disease. 
 
 MENTAL CAPACITY IN RELATION TO EDUCATIONAL BACKGROUND 
 
 A twofold interest attaches to the question of the relation between 
 mental capacity as measured by any of our tests and the educational 
 
494 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 background of the women under consideration. We are interested to 
 know, first, whether our data cast any light on the assumption, fre- 
 quently made in Chapter X, to the effect that the grade finished is of 
 itself something of an index of the intelligence of the individual and 
 that this is true of the measures of educational opportunity to much 
 slighter degree. On the other hand, we are concerned to know 
 whether the relationship between schooling and education is so close 
 as to suggest that in our tests we are measuring the influence of educa- 
 tion rather than native ability. 
 
 We have attempted to answer these questions by the use of several 
 
 ■ diflferent combinations of data. First, using the total English-speaking 
 
 portion of our delinquent group, on whom we had both the test results 
 
 and the necessary information on schooling, we obtained the following 
 
 correlations : 
 
 Correlation of Mental Capacity as Measured by Test Aggregate with 
 
 Age at Leaving School Number of Years in School Grade Finished 
 
 r = .27 ± .049 (352 cases) .29 ± .046 (383 cases) .60 ± .032 (383 cases) 
 
 (The correlation array for the first of the above relationships is given 
 in Table 208. The others are not shown.) 
 
 From these figures it is apparent that there is a much higher rela- 
 tionship of mental capacity, thus measured, with grade finished than 
 with either age at leaving school or number of years in school. Ob- 
 taining the partial correlation coefficient of mental capacity with grade, 
 for years in school constant, we find that r = 54. It is very little af- 
 fected by controlling the factor of school opportunity in the sense of 
 the number of years spent in school. On the other hand the partial 
 correlation coefficient of mental capacity with years in school for con- 
 stant grade drops to zero (.003). It appears therefore that length of 
 time spent in school is symptomatic of intelligence only to a very slight 
 degree and only when it shows results in appropriate grade accom- 
 plishment : but that the grade attained has a much higher diagnostic 
 value and that this value is not greatly affected by differences in school 
 opportunity. 
 
 To determine how these relationships were affected by more ac- 
 curate information the same correlations were computed, using only 
 those cases on whom we had verified record of both years in school 
 and grade, with the following results : The correlation of mental ca- 
 pacity, measured by Test Aggregate, with years in school was found to 
 be .32 (159 cases): that of mental capacity with grade finished .75 
 
o 
 o 
 
 u 
 
 I 
 
 IS 
 
 
 bi) 
 
 
 
 a> 
 
 
 
 
 
 
 bJO 
 
 p> 
 
 
 ^ 
 
 a 
 
 
 
 O 
 
 
 0) 
 
 Cfl 
 
 00 
 
 H 
 
 S 
 
 o 
 
 >^ 
 
 U 
 
 (.■1 
 
 rfJ 
 
 
 a 
 
 ■^ 
 
 •i 
 
 PQ 
 
 iLi 
 
 CO 
 
 <1 
 
 3 
 
 0) 
 
 ei 
 
 fft 
 
 M 
 
 o 
 
 1 
 
 
 
 a> 
 
 d 
 o 
 
 «] 
 
 I— « 
 
 o 
 O 
 
 
 1^1 
 
 «oo-*»ooM»motot- -o 
 
 00 
 00 
 
 
 ^ 
 
 ■""'2^ggS§S3'-"°- 
 
 i 
 
 S 
 
 3 
 
 
 Z6-S6 
 
 (N i-Irt 
 
 ■* 
 
 "cD 
 
 S6-S6 
 
 (N^ .H 
 
 •i* 
 
 £6-16 
 
 ^ (N.-< 
 
 tH 
 
 00 
 
 '3 
 
 16-68 
 
 rHtOU5U3M 
 
 s 
 
 68-i8 
 
 COt-COrt 
 
 ■* 
 
 
 i8-S8 
 
 N -H M >n Tt« (N rt ^ 
 
 Oi 
 
 S8-S8 
 
 r-ffMOit*-*^ 
 
 s 
 
 U3 
 
 E8-I8 
 
 -< -^OSt-iOrt 
 
 
 ■* 
 
 I8-6i 
 
 (N « OJ W -* (N w ,H 
 
 g 
 
 (N 
 
 6L-U 
 
 rH COOStOeO^rH rt 
 
 s 
 
 "Tti 
 
 11-91, 
 
 ^^rtrtI>^NN^ 
 
 s 
 
 CO 
 
 SZ-£i 
 
 M i-i'^t^eOM 1-1 
 
 (N 
 
 
 Sl-ll 
 
 t* Tff <0 ■* C<» tJ< tH 
 
 
 M 
 
 IZ-69 
 
 i-H 
 
 00 
 
 "cc 
 
 69-Z9 
 
 <M in M T|< CO <N 
 
 
 lO 
 
 Z9-S9 
 
 NOq^eOrH 
 
 o 
 
 "CO 
 
 S9-S9 
 
 ,-f ,-1 CO rH « r-l rt 
 
 ;3 
 
 © 
 
 E9-T9 
 
 r-l ^(NN.-I (N 
 
 OS 
 
 "CO 
 
 I9-6S 
 
 .-( (N (N i-H (N .H I-H 
 
 o 
 
 (D 
 
 6S-ZS 
 
 1-C WiH rH 
 
 ■* 
 
 
 ZS-S9 
 
 COi-t .H.H 
 
 o 
 
 b- 
 
 B9-ES 
 
 
 o 
 
 
 SS-IB 
 
 
 o 
 
 
 IS-6* 
 
 
 o 
 
 Bf^lf 
 
 
 o 
 
 o 
 
 i^-fi* 
 
 - 
 
 -- 
 
 
 
 OJ CO l> <D »0 Tji CO «' .-H O OS od b^ 
 
 « 
 
 ■a 
 o 
 H 
 
 :1 
 
 ;l 
 
 : bo 
 
 . a 
 
 < 
 
 lOOqOg SUTABS^ IB 93v 
 
 
 CON 
 
 
 II II 
 
 
 a a 
 
 
 
 
 
 
 
 
 S^ 
 
 
 ■c-c 
 
 
 ou 
 
 
 m CO 
 
 
 a a 
 
 
 
 
 
 
 
 
 
 
 03 c3 
 
 
 
 
 «m 
 
 
 t-oo 
 
 
 
 
 oo 
 
 ^ 
 
 
 
 4HI 
 
 g" 
 
 ION 
 
 
 =°m 
 
 II II 
 
 
 
 
 !!• B- 
 
 28 II 
 
 If 
 
 IIS 
 
 
 g^ 
 
 
 ss 
 
 5° 
 ^■5 
 
 ,«S 
 
 ..S 
 
 sia-R 
 
 1^ an 
 
 ■ « o 
 
 P 
 
 
 "^ 
 
 II "S 
 
 SSf, 
 
 
 H<! 
 
 
 3-» -; 
 
 
 
 
 
 
 SH«! 
 
 a.2 
 2-a 
 
 OO 
 
 495 
 
496 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 (164 cases). (The latter correlation is figured in Table 209.) It is 
 evident that these results confirm the evidence of the foregoing cor- 
 relations, showing an even more striking relationship between intelli- 
 gence and grade. 
 
 To test still further the significance of this relationship we com- 
 puted the correlations of mental capacity both as measured by Stanford- 
 Binet and as measured by the Performance Tests, with the measures 
 of schooling-^years in school and grade. These correlations were 
 based on completely verified data for 80 Bedford cases. The inter- 
 relations of the various factors for this particular group are shown 
 below : 
 
 Stanford-Binet Years in School Grade 
 
 Years in School -. .- -23 
 
 Grade 74 .39 
 
 Performance Tests .60 .36 .70 
 
 The partial correlation of Stanford-Binet with grade, for constant 
 number of years in school, is .721 : that of Stanford-Binet with years 
 in school, for constant grade, is —.095. The partial coefficient of the 
 Performance Tests with grade, for constant years in school, is .644: 
 that of the Performance Tests with years in school, for constant grade, 
 is .133. These figures, therefore, are entirely in accord with our pre- 
 vious findings. The Performance Tests, it may be noted, show the 
 contrasts less strikingly than do either Stanford-Binet or Test Aggre- 
 gate. 
 
 Summary 
 
 (1) Whatever measure of intelligence we have used — Test Ag- 
 gregate, Stanford-Binet, or Performance Tests — a high relationship 
 with school grade has appeared, but only a slight relationship with 
 number of years in school. 
 
 (2) This confirms our impression that grade attainment, in itself, 
 affords some index of one's mentality. 
 
 (3) The fact that the correlation of the results of the various 
 intelligence tests with grade is so high, while that of test results with 
 years in school is so low, and that, moreover, the partial correlation 
 of test results with grade, for constant number of years in school, is 
 high, while that of test results with years in school, for constant grade, 
 is zero or nearly zero, is sufficient to allay any fears that the test results 
 measure effect of schooling rather than mentality. Were this the 
 case there is every reason to expect that we should find approximately 
 
o 
 
 o 
 u 
 
 en 
 
 
 0) 
 
 "d 
 
 5 
 
 a> c4 
 
 bfi S 
 < > 
 
 ■a 
 0) 
 H 
 
 >. 
 
 .a 
 
 S 
 
 
 a 
 a> 
 
 a 
 
 a> 
 
 .o 
 
 a 
 o 
 
 b 
 
 o 
 U 
 
 o 
 
 2 " 
 
 m a 
 
 
 i 
 
 
 
 1 
 
 i 
 
 S6-E6 
 
 1-1 
 
 1-H 
 
 86-16 
 
 CQr-liH 
 
 T»( 
 
 16-68 
 
 —1 -^NiH .-1 
 
 a 
 
 68-i8 
 
 ■*rt 
 
 lO 
 
 -S8-S8 
 
 i-H COeOrH 
 
 1— 1 
 
 S8-88 
 
 rHlOrt,-IIM 
 
 o 
 
 1-1 
 
 £8-18 
 
 ,<N(N(N!0(Mrt 
 
 1-1 
 
 t8-6Z 
 
 tO-*i-lr-( 
 
 r-l 
 
 61-11 
 
 T-H 1-1 CONN 
 
 2 
 
 11-91 
 
 T-i ro M 50 OS N 
 
 00 
 
 SZ-EZ 
 
 M'*^ rt 
 
 00 
 
 Sl-U 
 
 N t>>o 
 
 3 
 
 IZ-69 
 
 ■^-^l-ll-l 
 
 o 
 
 69-Z9 
 
 ■*Meo 
 
 o 
 
 I— 1 
 
 Z9-99 
 
 ,-(CO T-l 
 
 Ui 
 
 
 S9-E9 
 
 .-1 ^ (N 
 
 •* 
 
 
 S9-t9 
 
 r-<(N 
 
 CO 
 
 
 I9-6S 
 
 ■-(■* 
 
 lO 
 
 
 6B-ZS 
 
 i-l .-< 
 
 <N 
 
 
 Zfi-SS 
 
 rt (N 
 
 CO 
 
 
 S9-ES 
 
 1—1 
 
 I-< 
 
 
 ES-IS 
 
 il .-1 
 
 IM 
 
 
 - 
 
 O 05 00 1> to >0 ■*' CO IM <-< o 
 
 1 
 
 
 paqsraij apBjQ 
 
 00 
 
 ICCO % 
 
 O'OO > 
 
 CD ■ ■ -w 
 
 t~ II . S 
 
 II Ct- _r 
 
 - * ' S 
 
 DO} O 
 
 i§" I 
 
 ^-r: -a 
 
 03 H I— 1 nj 
 
 Sen (£ £ 
 
 S 5 
 
 SB 
 
 J? "S 
 
 U 
 
 497 
 
498 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 as high a correlation with the time spent in school as with the grade 
 achieved, or possibly even a higher correlation. 
 
 MENTAL CAPACITY IN RELATION TO WORK RECORD 
 
 We will next consider the relation between mental capacity as 
 measured by Test Aggregate and the various phases of the work rec- 
 ord of these women. It would seem a justifiable assumption that, all 
 other considerations being eliminated, the better the mentality of an 
 individual the better would be the work that she could do and, conse- 
 quently, the greater would be her earning capacity. 
 
 (a) Mental Capacity and Age at Starting Work 
 
 Following the order adopted in Chapter XI, we have first considered 
 the age at starting work in its relation to mental capacity, in order 
 to attempt to find out whether there is any tendency for the brighter 
 women to start work at an earlier age than those less bright, or whether 
 this condition is just reversed and it is the stupider women who seek 
 employment younger, or whether ■ neither fact can be demonstrated, 
 and the factor of mentality is in no way concerned with the age at 
 which they begin working. We have, accordingly, plotted the corre- 
 lation which is figured in Table 210. The correlation coefficient is 
 found to be .19±.053. The correlation ratios are only slightly larger: 
 .26±.051 for the ratio of age at first employment on Score in Test 
 Aggregate and .24±.052 for score in Test Aggregate on age at first 
 employment. 
 
 The widest divergence of the means in adjacent groups lies between 
 those starting work from 10 to 14 years and those starting work from 
 14 to 18. If we take the percentage of cases in these two age groups 
 which lie above and below the approximate mean of the group in Test 
 Aggregate, we find that, of those who start work between 10 and 14 
 years of age, 38 per cent are above the average in mental capacity as 
 measured by Test Aggregate and 62 per cent are below, while for those 
 starting work between 14 and 18, 58.7 per cent are above average in- 
 tellectually and 41.3 per cent are below. This fact undoubtedly links 
 up with the correlation between intellectual capacity and age at leaving 
 school, which gives a coefficient of .32 ±.049, indicating a distinct ten- 
 dency for the less intelligent women to leave school at an earlier age 
 (see Chapter X). This in turn would probably be a factor in deter- 
 mining the age at starting work. The significance, in this connection, 
 

 1 
 
 
 
 01 
 
 Ci 
 0) 
 
 a 
 
 I 
 
 .fi 
 
 ci 
 o 
 
 o 
 U 
 
 
 1 
 
 o 
 
 Eh 
 
 Z6-S6 
 
 S6-86 
 
 £6-16 
 
 T6-68 
 
 6S-IS 
 
 Z8-98 
 
 98-88 
 
 E8-I8 
 
 6i-ZZ 
 
 ii-Bi 
 
 Bi-8Z 
 
 £i-TZ 
 
 Ti-69 
 
 Z9-fi9 
 
 S9-S9 
 
 E9-I9 
 
 I9-6S 
 
 69-iS 
 
 ze-es 
 
 SS-S5 
 
 8S-Tff 
 
 IS-6* 
 
 6^Zf 
 
 ^OOrHNC0C0INNC0»0-*QC0 
 
 iHiHrH t-^b-IN 
 
 ^ Tl<t*CO 
 
 iHco^^cq 
 
 l-HiH ON 
 
 .-H tHr-l-^OOCO 
 
 rt COtOCOW 
 
 i-t.-l»HCO« 
 
 „ ^ HOCONOO-^OCO 
 
 lOlOlO-^^MCOCOCQNr-lrHiH 
 
 499 
 
 a a 
 
 •§■§ 
 
 SS 
 
 as 
 ga 
 
 ■^•^ 
 
 3m 
 
 4m 
 
 II II 
 
 Q) o 
 bO ^ 
 
 t II 
 
 lis 
 
 • ■bo ' tj » 
 
 bO bO) bDCt 
 
 0)^^ a o 
 him "^ « 
 
 ■S <D ^- " « 
 
 S.9 
 
 °i 
 
 SB'S 
 
 
500 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 of the age of fourteen as the common upper age limit in compulsory 
 education laws need hardly be emphasized. 
 
 (&) Mental Capacity and Kind of Work 
 
 We considered next the relation of intellectual capacity to the nature 
 of the work, in order to see if women of one grade of intelligence tend 
 to find employment in one line of work while those of another 
 grade gravitate toward another occupation. To this end we have made 
 a correlation table (Table 211) with Test Aggregate as the quantitative 
 variable and the occupations of domestic service, factory work, home 
 work, restaurant and hotel work, work in stores, vaudeville perform- 
 ance, clerical work, professional service, personal service, work of 
 charwoman, and miscellaneous, at which these women were mainlyt 
 employed, as a series of qualitative arrays. From such a table we have 
 obtained a correlation ratio of .46 ±.043. This would seem to be clear 
 indication that there is a decided degree of relationship between the 
 mental capacity so measured and the kind of work at which a woman 
 has been predominantly employed. 
 
 If we compare Table 211 with Table 212, which shows the relation 
 between mental capacity and the nature of the first work in which the 
 women were engaged, we find that the respective means are on the 
 whole very similar. The means for mental capacity (Test Aggregate) 
 run in order of excellence : 
 
 First Work Prevailing Work 
 
 Professional service (4) 86.0 87.5 Professional service (2) 
 
 Clerical work (IS) 83.6 86.6 Personal service (7) 
 
 Vaudeville performers (11) 83.0 82.5 Vaudeville performers (16) 
 
 Personal service (5) 81.6 82.3 Work in stores (37) 
 
 Work in stores (40) 80.7 81.3 Clerical work (9) 
 
 Restaurant and hotel work (24) 76.9 79.6 Restaurant and hotel work (21) 
 
 Home work (7) 76.9 75.4 Home work (12) 
 
 Factory work (102) 75.7 75.1 Factory work 88) 
 
 Miscellaneous (7) 74.3 72.3 Miscellaneous (3) 
 
 Domestic service (141) 72.1 71.8 Domestic service (142) 
 
 Charwomen ( 1 ) 59.0 67.0 Charwomen (2) 
 
 The correlation ratio obtained for mental capacity on the nature 
 of first work is .41hz.044. It will be seen that the arrangement of the 
 means is in general similar for the nature of work in the first job and 
 in the prevailing job, and it seems clearly evident that the mental ca- 
 pacity of these women is a factor in determining both the kind of work 
 which they first enter and that which, subsequently, they continue to 
 
< O 
 
 « Si 
 
 oOiHTjfeoco»0««eooeo 
 
 
 Z6-96 
 
 S6-S6 
 
 g6-T6 
 
 16-68 
 
 <NCO eot>N rt 
 
 68-Z8 
 
 Tt<M MtJh.-hi-h 
 
 Z8-S8 
 
 ■*1N^(NCCMM M 
 
 S8-E8 
 
 OS Tjf IN CO CO i-< i- 
 
 S8-I8 
 
 ■*»cocq»ow 
 
 I8-6i 
 
 61,-11 
 
 il-9l 
 
 ^ b- rH N C^ (N F- 
 
 sz-sz 
 
 at* rHrtF- 
 
 sz-iz 
 
 TZ-69 
 
 ^Oi iHC^r 
 
 eO!0 W'H.-H 
 
 19-99 
 
 S9-S9 
 
 £9-19 
 
 6e-zs 
 
 zs-ss 
 
 TS-6^ 
 
 6*-Z» 
 
 "t3 ni'C 
 
 S.H 
 
 m 
 
 S OS 
 o ^ o 
 
 
 o'ffl " S 9 
 
 ^jd-e.H <u o M S 
 
 s S a s's 2 «.s 
 
 ill 
 
 
 ailOj^ 3UI1IBA3J(J 
 
 
 
 HO 
 
 501 
 
:§ 
 
 CO 
 
 •a 
 
 o 
 
 
 
 Rl 
 
 
 
 U) 
 
 
 
 
 >l 
 
 
 g 
 
 ■a 
 o 
 
 
 <! 
 
 
 
 -i-» 
 
 u 
 
 rq 
 
 
 m 
 
 Vh 
 
 <u 
 
 O 
 
 « 
 
 H 
 
 
 H 
 
 >, 
 
 d 
 
 ri 
 
 ^ 
 
 jS 
 
 ^ 
 
 •d 
 
 u 
 & 
 
 
 s 
 
 ,r] 
 
 
 (» 
 
 
 
 01 
 
 »3 
 
 
 ti 
 
 bp 
 
 g 
 
 u 
 
 a 
 
 a 
 o 
 
 o 
 O 
 
 ^ m bO 
 
 Z6-S6 
 
 £6-16 
 
 68-i8 
 
 S8-S8 
 
 E8-I8 
 
 t8-6Z 
 
 61-11 
 
 91-21 
 
 Sl-U 
 
 IZ-69 
 
 i9-S9 
 
 S9-89 
 
 t9-6ff 
 
 6S-iB 
 
 SS-8S 
 
 ES-IS 
 
 IS-6* 
 
 (H-li 
 
 r-(t«003t«OCOOCOO?0 
 
 cqto iHsDiMcq 
 
 M<"* ^<NT-HNr 
 
 iON W-^<M(M N 
 
 ^-^*1-^m»o 
 
 uiomm-^ 
 
 ^OS W^ 'HiH 
 
 Ol^^fMW 
 
 MiH iHCOtH^ 
 
 Ht^ M-* iH 
 
 lOlO iHrHr- 
 
 (>(N rt 
 
 M.N rH iH iH 
 
 K 
 
 fc-s" 
 
 SOS 
 
 iii 
 
 fc» 
 
 > a o 
 
 
 qof ^9Ji^ JO ainq-B^ 
 502 
 
 ffif^ 
 
 iS-s 
 
 oj'g 
 
 ^§ 
 
 s| 
 
 Si- 
 
 bOa 
 
 ■♦3 u 
 
 m u 
 
 03 O 
 
MENTAL CAPACITY 
 
 503 
 
 do. The small number of cases in many of the above groups has de- 
 terred us from making detailed inter-comparisons. 
 
 Since, however, the domestic service group contains over 40 per 
 cent of the cases we have made a comparison between the mental 
 capacity of this group and that of women employed at occupations 
 other than domestic service. The mean for domestic service is 71.83dz 
 .802 and for all other occupations 78.25±.613. Table 213 indicates 
 
 TABLE 213 
 
 DOMESTIC SERVICE AND NOT DOMESTIC SERVICE 
 
 Comparison of the Means and Standard Deviations of Mental Capacity, 
 
 as Measured by Test Aggregate, of those whose Prevailing W^ork 
 
 is Domestic Service and those whose Prevailing Work is 
 
 other than Domestic Service 
 
 English-speaking Cases Only 
 
 
 Domestic 
 
 Service 
 
 Not Domestic 
 Service 
 
 Difference 
 
 d 
 
 Chances that real 
 
 difference does 
 
 not exist are 1 
 
 in: 
 
 Mean 
 
 71.83 
 + .802 
 
 78.25 
 + .613 
 
 -6.42 
 
 6.36 
 
 00 
 
 (Tm 
 
 
 
 
 <T 
 
 9.56 
 + .567 
 
 8.60 
 ±.433 
 
 .96 
 
 1.34 
 
 11 
 
 CTo- 
 
 
 
 . 
 
 Cases 
 
 142 
 
 197 
 
 
 
 
 
 
 Per 
 
 Cent 
 
 ~ ~ ~ " * Other Occupdttoni 
 
 Chart XLIII 
 Mental Capacity as Measured by Test Aggregate 
 Percentage comparison between those whose prevailing work has been 
 domestic service and those whose prevailing work has been other than domestic 
 service. English-speaking cases only. 
 
504 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 the validity of the distinction in favor of those employed at occu- 
 pations other than domestic work. Chart XLIII shows the marked dif- 
 ference in distribution for the twa groups, making evident that the 
 higher standing shown by the mean is based on no eccentric massing 
 but is due both to higher percentages of cases with superior mental 
 ratings and lower percentages with inferior ratings. We are, there- 
 fore, justified in stating that the women whose prevailing work is do- 
 mestic service are less mentally capable on the average than those who 
 are employed at other occupations. 
 
 (c) Mental Capacity and Wage 
 
 We have next considered the possible relationship which one would 
 expect to find between mental capacity and the economic status of these 
 women as represented by the average weekly wage which they have 
 received at their prevailing work. The method of assembling these data 
 on the work record has already been explained in Chapter XI. There 
 has been explained also the necessity of dividing the material on wage 
 earned into two groups, domestic service and allied occupations where 
 board and lodging are in addition to the wage, and occupations other 
 than domestic service which provide no maintenance. If we plot the 
 figures on the former of these, i.e., average wage of women whose pre- 
 vailing work has been domestic service in relation to their mental 
 capacity as measured by Test Aggregate, we obtain Table 214. The 
 correlation coefficient obtained is very low, — .03 ±.089. The correla- 
 tion ratios are .14±.087 for the regression of wage on Test Aggregate 
 and .09±:.088 for the regression of Test Aggregate on wage, both 
 negligible. Our results, therefore, fail to show any relationship be- 
 tween the mental capacity of women whose prevailing work was do- 
 mestic service and their average wage. 
 
 When we come next to compare the scores obtained on this test 
 with the wage obtained at occupations other than domestic service, we 
 find the situation easier of interpretation. Table 215 shows the rela- 
 tion of these factors to each other. The correlation coefficient obtained 
 is .31 ±.070, which would seem to indicate a clear and, in the light of 
 its standard deviation, significant, correlation between the two. We 
 may say in general that, as the mental capacity of these women in- 
 creases, there is a progressive increase in the amount of money they 
 earn at occupations other than domestic service. This is of interest in 
 view of our inability to establish any such relationship between mental 
 capacity and the amount earned at domestic service. 
 
M 
 
 s 
 ^ 
 
 a> 
 
 
 « V 
 
 |8 
 
 <!-g 
 •g S 
 
 a 
 
 "S " 
 
 li 
 
 +J . 
 
 0) 0) 
 
 CO d 
 flj o 
 
 .■&! 
 I 
 
 4) 
 g 
 
 o 
 O 
 
 o 
 
 u 
 U 
 
 0) 
 
 & 
 
 i 
 
 "Si 
 
 1 
 
 ij 
 
 70.9 
 
 73.3 
 73.2 
 72.9 
 71.9 
 
 
 
 5 
 
 '=''^'-«'''sSa'="° 
 
 § 
 
 S 
 
 ii 
 
 
 S6-S6 
 
 (N 
 
 N. 
 
 
 i 
 
 86-16 
 
 N 
 
 M 
 
 16-68 
 
 « ^ 
 
 CO 
 
 J 
 
 68-i8 
 
 1-1 i-t 
 
 n' 
 
 
 i8-S8 
 
 iHMiHrt 
 
 U3 
 
 S8-S8 
 
 (N(NMCO 
 
 O 
 
 £8-18 
 
 rt N rH 
 
 -<* 
 
 I8-6i 
 
 i-H CCiHiH 
 
 o~ 
 
 
 61-U 
 
 ■^aeo IN 
 
 ;=! 
 
 11-91 
 
 CO 00 CO 
 
 03 
 
 SZ-Si 
 
 t-t i-H (N^rt 
 
 CO 
 
 Sl-ll 
 
 (NtJ<<N-^i-I 
 
 Co" 
 
 1-1 
 
 "ii 
 
 Ti-69 
 
 cOi-ieO(N »-* 
 
 s 
 
 69-2,9 
 
 CO CO IN tH 
 
 OS 
 
 Z9-S9 
 
 iH »H CO (N 
 
 l> 
 
 fi9-E9 
 
 iHiHC4« 
 
 eo"" 
 
 
 89-19 
 
 r-tNTpiH 
 
 00 
 
 T9-6S 
 
 i-H^rH N 
 
 U5 
 
 69-19 
 
 i-((N 
 
 CO 
 
 i.9-99 
 
 l-H <N^ 
 
 ■*" 
 
 1-t 
 
 SB-89 
 
 
 O 
 
 ES-TS 
 
 
 o 
 
 15-6* 
 
 »H 
 
 ^ 
 
 
 8SS8S8SSS 
 
 © GO r> CO lOTti CO NrH 
 
 issssssss 
 
 00 1.. to lO tP CO W tH 
 
 ■a 
 1 
 
 1 
 
 IB 
 
 aSBjiV itHaajJl 
 
 9*: 
 
 'HO 
 
 II II 
 
 a a 
 o o 
 
 II 
 
 ss 
 
 cO as 
 
 §§ 
 mm 
 
 oo 
 
 Hl-fl 
 
 , C-B- 
 
 5 MO 
 
 II ll^i 
 "SSI 
 
 „• II II * g 
 
 II ;; 0) " 
 
 mS 0.2 
 <j -.3 J 
 
 4A brm u 
 " rt £ fa 
 
 HPOO 
 
 S 
 g 
 
 I 
 
 .a 
 
 I 
 1^ 
 
 505 
 
SRfc 
 
 Z6-S6 
 
 B6-S6 
 
 E6-I6 
 
 Ts 
 
 16-68 
 
 I— I CO ^ CO I— I 
 
 is-ss 
 
 j-t iHiHrt e^cocqr- 
 
 II II 
 
 E8-I8 
 
 I8-6i 
 
 6i-ll 
 
 11-91 
 
 SZ-SZ 
 
 NWiHt^iH 
 
 CO n-^n 
 
 ^rt»0-*M 
 
 S5 
 
 CD<D 
 
 oo 
 
 •im 
 
 S5 
 
 cii O 
 
 
 Ii-69 
 
 69-i9 
 
 i9-fi9 
 
 S9-S9 
 
 E9-I9 
 
 I9-6S 
 
 6S-iS 
 
 iS-SB 
 
 •-< Mmcq^T-i 
 
 'Hi-(iHD3Ti< 
 
 "N 
 
 oooooooooooooooooooooo 
 oqoqooooooooooooooooooo 
 
 506 
 
 II n 
 
 p-p- 
 
 cow • % g, 
 
 k; II II * S 
 
 mq. O oS <n 
 
 agof 
 s|§g 
 
 HPOO 
 
MENTAL CAPACITY 507 
 
 Although there is no appreciable correlation between mental capac- 
 ity and age, there is a correlation of .264 between age and wage earned 
 at these occupations. We have, therefore, felt called upon to compute 
 a partial correlation between mental capacity and wage with age con- 
 stant. The result obtained is .323 ±.068, a coefficient not appreciably 
 different from the preceding. Since there is practically no correlation 
 between age and wage at domestic service we have not felt it necessary 
 to make this correction there. 
 
 The correlation between mental capacity measured by Test Aggre- 
 gate and the wage earned at first job for the women in domestic 
 service and allied occupations is .03 ±.094. For those in occupations 
 other than domestic service this correlation is .29 ±.069. Since these 
 appear to resemble quite closely the corresponding correlations of men- 
 tal capacity with prevailing wage, we have not presented the scatter 
 tables for these relationships. 
 
 {d) Mental Capacity and Estimates of Work Record 
 
 We have next attempted to find out whether there is any relation 
 between the mental capacity of these women and the regularity with 
 which they have held employment. Table 216 shows such a correla- 
 tion where regularity of employment is divided into five grades 
 (see Chapter XI). The correlation ratio (.13 ± .053) indicates that 
 there is at most only a very slight association between mental ca- 
 pacity and regularity of employment. There are undoubtedly some 
 whose irregular work record is the result of their mental incom- 
 petence, which prevents their holding a job for any length of time, 
 but there are others mentally well equipped whose failure in this 
 respect must be explained by some wholly different factor. 
 
 If we consider last the relation of mental capacity to the estimate 
 of the efficiency of the work record, we obtain Table 217. (The ex- 
 planation of the method of arriving at this estimate is given in Chapter 
 XI.) The correlation ratio obtained is .20±.052, which would seem 
 to indicate that there exists a small but significant degree of relation- 
 ship between mental capacity and excellence of work record, as based 
 on this estimate. 
 
 Swmmary 
 
 Summing up our results on this section we can see, then, that the 
 mental capacity of these women plays some part in the various phases 
 of their work records in that : 
 
•a 
 
 1 
 
 o 
 
 CO 
 
 OS 
 
 CO o 
 
 ^ -^1 
 
 4) 
 
 H 
 
 14 
 
 CO 
 
 I 
 
 I 
 
 a> 
 
 I 
 
 a> 
 
 t; 
 
 o 
 U 
 
 
 
 bll 
 
 §*=! 
 
 a 
 g 
 
 
 Tt 
 
 g 
 
 -* 
 ^ 
 
 
 
 
 B 
 
 t- 
 
 s i 
 
 S 
 
 s? 
 
 iH 
 
 1 
 i 
 
 < 
 
 Z6-fi6 
 
 
 N r- 
 
 
 1-1 
 
 ■* 
 
 66-86 
 
 
 ■H 1- 
 
 
 
 M 
 
 E6-T6 
 
 
 »H 1- 
 
 - 
 
 
 « 
 
 16-68 
 
 
 00 <£ 
 
 ■^ 
 
 iH 
 
 S 
 
 68-Z8 
 
 
 ^ u- 
 
 er 
 
 N 
 
 -«*< 
 
 i8-S8 
 
 IM 
 
 CD -ct 
 
 cc 
 
 
 00 
 
 B8-E8 
 
 
 00 OC 
 
 -* 
 
 <M 
 
 ^ 
 
 E8-I8, 
 
 
 00 c 
 
 •* 
 
 CO 
 
 g 
 
 I8-6Z 
 
 
 l> c 
 
 CC 
 
 N 
 
 s 
 
 61-U 
 
 
 © u- 
 
 c 
 
 -H 
 
 s 
 
 llSl 
 
 
 O) cr 
 
 CO 
 
 CO 
 
 
 91-Sl 
 
 
 CO a- 
 
 IT 
 
 CO 
 
 g 
 
 Sl-ll 
 
 
 <0 OC 
 
 c 
 
 «o 
 
 S 
 
 Ii-69 
 
 
 W t" 
 
 o 
 
 (N 
 
 S 
 
 69-Z9 
 
 
 (N >r 
 
 tc 
 
 - 
 
 S! 
 
 i9-S9 
 
 
 ^ ir 
 
 M 
 
 -^ 
 
 00 
 
 99-E9 
 
 
 CC ■«* 
 
 m 
 
 « 
 
 CO 
 
 1-4 
 
 E9-T9 
 
 
 ■^1< <N 
 
 M 
 
 - 
 
 o 
 
 I9-6S 
 
 
 CO r- 
 
 r^ 
 
 04 
 
 o 
 
 69-ZS 
 
 
 « r- 
 
 
 -H 
 
 ^ 
 
 Z9-SS 
 
 
 « »H 
 
 M 
 
 
 t* 
 
 SB-E9 
 
 
 
 E9-I9 
 
 - 
 
 - 
 
 19-6* 
 
 - 
 
 - 
 
 6f-lf 
 
 .- 
 
 1-1 
 
 l,f-Sf 
 
 
 
 
 
 1 
 
 1 
 
 1 
 
 
 1 
 
 £ 
 
 h 
 
 
 3IJO 
 
 Ml 
 
 a A% 
 
 :i'B[r 
 
 laag 
 
 1 
 
 o 
 
 ,l 
 
 
 t-i o 
 O O 
 
 HO 
 
 508 
 
a 
 
 I 
 
 & 
 
 o 
 
 <0 
 
 m 
 
 I) ^ 
 
 OO CO 
 
 ^ w 
 
 i ^ 
 
 Q> I 
 CO 
 
 cti 
 
 s 
 
 o 
 
 §• 
 O 
 
 I 
 
 v 
 
 (U 
 
 B 
 
 O 
 
 •3 
 
 t! 
 
 o 
 
 
 
 lij 
 
 
 
 to 
 
 CO 
 
 
 
 H 
 
 CO CO C4 CO OC 
 CO rH '^ -3! 
 
 i 
 
 1 
 
 L6-96 
 
 1-1 (M rt 
 
 -* 
 
 96-E6 
 
 iH rH iH 
 
 CO 
 
 E6-I6 
 
 iH T-l tH 
 
 CO 
 
 16-68 
 
 -* 10 ^ 
 
 8 
 
 68-Z8 
 
 (M US lO (M 
 
 Tfl 
 
 Z8-S8 
 
 tH -:J< Tj* 01 
 
 M 
 
 B8-E8 
 
 <N 
 
 cq 
 c^ 
 
 E8-I8 
 
 CO. rt O) 
 
 § 
 
 I8-6i 
 
 rt (N »0 1-1 iC 
 
 1-1 
 
 S! 
 
 61-11 
 
 rt OJ (N CO 
 
 10 
 
 li-9l 
 
 -* o> tH 
 
 s; 
 
 BZ-Ei 
 
 rH tH »0 
 
 8 
 
 Sl-U 
 
 ^ « "* to 
 
 
 ti-69 
 
 CO (O Ttl CO 
 
 S 
 
 69-i9 rt N ttj * 
 
 CO 
 
 ^9-59 « CO N 00 
 
 S9-E9 CO t* M CO 
 
 E9-I9 (N !0 (N 
 
 
 
 I9-6S 
 
 N CO CO (N 
 
 6S-ZS 
 
 .H rH C^l -rP 
 
 19~99 iH CO cq iH r- 
 
 es-ES ° 
 
 ES-IS -^ "' 
 
 ig-gf 
 
 6f-lf -^ '-I 
 
 
 
 > 
 
 i 
 
 (2 
 
 
 P. 
 
 1 
 
 H 
 
 „ 
 
 pioo 
 
 3H3 
 
 [lOjv 
 
 i 
 
 515 
 
 § 
 
 o © 
 
 gH 
 
 
 H(3 
 
 509 
 
510 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 (1) There is a tendency for the less bright women to begin 
 working at an earlier age than those more capable. This is especially 
 to be noticed for those who begin work before they have completed 
 the compulsory school period, i.e., before fourteen years; as contrasted 
 with those starting work after fourteen. 
 
 (2) There is a decided correlation between mental capacity and 
 the kind of work which these women have done, both in their first 
 job and in the prevailing occupation they have followed. Nearly 
 half of these women have been employed at domestic service and we 
 have shown that they are, on the average, unquestionably of lower 
 grade intellectually as measured by our standard than are the women 
 who have been employed in occupations other than housework. 
 
 (3) As regards the average wage which they have earned, we 
 can not demonstrate any unambiguous relationship between this and 
 mental capacity, where the occupation has been domestic service, but 
 we can show a clear correlation between mental capacity and the 
 wage earned for occupations other than domestic service. It would 
 seem, therefore, that, not only are the women of this group who have 
 been employed at domestic service less mentally capable than those 
 otherwise employed, but their varying grades of mental capacity 
 have not been met by a similarly varying earning capacity as have 
 those whose work has been other than domestic service. 
 
 (4) Finally, there exists a small but significant relationship be- 
 tween mental capacity and the degree of excellence indicated by the 
 estimate of the work record, but the degree of regularity of work, 
 which is one of the factors on which the estimate of the work record 
 was based, can not be shown to have played an important part in this 
 relationship. 
 
 MENTAL CAPACITY IN RELATION TO SEX HISTORY 
 
 The relation between the sex history of these women and their 
 mental capacity is of considerable importance in a group such as 
 ours. Table 150, page 386, shows that, out of our total of 551 cases 
 on which we have information, only 78, or 14 per cent, have not at 
 some period of their lives been sexually irregular. 
 
 (a) Mental Capacity cmd Age at First Sex Offense 
 
 Following the order adopted in Chapter XII, we have first consid- 
 ered the relation between the mental capacity of these women and the 
 age at which they committed their first sex offense. (See Table 218.) 
 
a 
 )» 
 
 O 
 
 M 
 CO 
 
 |X4 
 
 s 
 
 
 
 I 
 
 16-96 
 
 S6-E6 
 
 S6-I6 
 
 16-68 
 
 68-i8 
 
 Z8-S8 
 
 S8-S8 
 
 E8-t8 
 
 I8-6Z 
 
 61-U 
 
 11-91 
 
 Si-Si 
 
 Sl-\L 
 
 u- 
 
 69-19 
 
 L9-99 
 
 99-E9 
 
 E9-I9 
 
 T9-6B 
 
 69-19 
 
 19-99 
 
 BS-ES 
 
 ES-IS 
 
 IS-6» 
 
 OOiOffi -^ 
 
 ^M N-*U5 ^ 
 
 »-lrH^O>^ 
 
 NtH WOOOrHrt 
 
 N ««t"^»-<'-t 
 
 i-i iH (O US CO to 
 
 fH NiH t^lOr 
 
 ^MOCDiH 
 
 «Nt»CO rHrt 
 
 rt NIM-* 
 
 Ot-^tHOOU3WOStDMOt- 
 TjeOCOCONNNi-frH^-H 
 
 5" 
 
 -1 
 
 •O 
 
 S2 
 
 Js 
 
 Si 
 
 a a 
 
 o o 
 
 •n-c 
 SS 
 
 oo 
 
 m 09 
 
 a'fl 
 
 oS OS 
 
 a a 
 
 V v 
 
 ^^ 
 
 OS OS 
 
 Sn 
 
 U310 
 OO 
 
 II II 
 B- ft- 
 
 •a g 
 
 d <^ 
 § S 
 
 a«So«° 
 
 M+i Ti o o 
 
 « f II g » 
 
 is 
 
512 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 The coefficient of correlation (.14 ± .060) is so small, as to indicate 
 the presence of at most a very slight degree of relationship. The 
 correlation ratio of mental capacity on age at first offense is .20 ± .059, 
 with a sufficient degree of regularity in the Une of means of test 
 aggregate to indicate that this ratio probably measures fairly the extent 
 of the relationship. On the other hand, the ratio of age at first sex 
 offense on test aggregate (.25 =t: .057) is rendered ambiguous in sig- 
 nificance because of the great irregularity of the lines of means. 
 
 (b) Mental Capacity and Classification as Sex Offenders 
 
 Chart XLIV shows the distribution of those who have been sex of- 
 fenders and those who have not been sex offenders, that is, who have 
 not been sexually promiscuous. It is evident that the sex offenders are 
 a slightly inferior group. Table 219 shows that so great a differ- 
 ence as this could occur by chance only once in 313 times. We are 
 convinced, therefore, that the average mental capacity is a trifle 
 higher for those who have not been sexually promiscuous than for 
 those that have. As far as the variability of the two groups is con- 
 cerned, the sex offenders would seem to have a slightly greater range, 
 but this difference can not be considered established. 
 
 Per 
 
 Cent 
 
 Sex Offenderfl 
 Not Sex Offenders 
 
 Chart XLIV 
 
 Mental Capacity as Measured by Test Aggregate 
 
 Percentage comparison between those who have been sex offenders— sexually 
 promiscuous— and those who have not. English-speaking cases only. 
 
MENTAL CAPACITY 
 
 513 
 
 TABLE 219 
 
 SEX OFFENDERS AND NOT SEX OFFENDERS 
 
 Comparison of the Means and Standard Deviations of Mental Capacity, 
 
 as Measured by Test Aggregate, of Sex Offenders and 
 
 Those Not Sex Offenders* 
 
 English-speaking Cases Only 
 
 
 Sex 
 Offenders 
 
 Not Sex 
 Offenders 
 
 Difference 
 
 
 Chances that real 
 
 difference does 
 
 not exist are 1 
 
 in; 
 
 Mean 
 
 75.28 
 + .507 
 
 79.4 
 ±1.43 
 
 -4.12 
 
 2.73 
 
 313 
 
 ffm 
 
 
 (7 
 
 9.34 
 + .326 
 
 8.49 
 ±.926 
 
 .85 
 
 .87 
 
 5 
 
 (Tff 
 
 
 
 
 Cases 
 
 339 
 
 35 
 
 
 
 
 
 
 *The distinction here is between those who have been sexually promiscuous and those who have not. 
 
 Out of a total of 551 cases there were 365 who were prostitutes 
 at the time of conviction or had been so at some period in their 
 history. They comprise, then, 66.2 per cent of our total group. 
 In Chart XLV we have compared the distribution of the scores on Test 
 Aggregate of those women who have been prostitutes with that of those 
 
 Per 
 
 Cent 
 
 Non Prostitutes 
 
 Chart XLV 
 
 Mental Capacity as Measured by Test Aggregate 
 
 Percentage comparison between those who have been prostitutes and those 
 who have not. English-speaking cases only. 
 
514 STUDY OF WOMEJN UiiLiJNyunJN ib n\ iNtw yuRK 
 
 who have not. The mean of the non-prostitute group is sensibly higher, 
 77.40 ± .960 as against 74.97 ± .556, and Table 220 shows this 
 difference to be almost certainly valid. 
 
 The degree of variability is slightly smaller for the prostitute 
 class, 9.07 ± .350 as against 9.97 ± .641, but Table 220 shows 
 this difference to be only possibly demonstrable, thp chance being one 
 in nine that as great a difference would occur from sampling. 
 
 TABLE 220 
 
 PROSTITUTES AND NON-PROSTITUTES 
 
 Comparison of the Means and Standard Deviations of Mental Capacity, 
 as Measured by Test Aggregate, of Prostitutes and Non-Prostitutes 
 
 
 English-speaMng 
 
 Cases Only 
 
 
 
 Prostitutes 
 
 Non 
 Prostitutes 
 
 Difference 
 
 d 
 Cd 
 
 Chances that real 
 
 difference does 
 
 not exist are 1 
 
 in: 
 
 Mean 
 
 74.97 
 + .556 
 
 77.40 
 ±.960 
 
 -2.43 
 
 2.19 
 
 70 
 
 Cm 
 
 
 
 
 (T 
 
 9.07 
 ±.350 
 
 9.97 
 ±.641 
 
 - .90 
 
 1.23 
 
 9 
 
 (T<r 
 
 
 
 
 Cases 
 
 266 
 
 108 
 
 
 
 
 
 
 In Table 221 we have attempted to make a finer distinction in the 
 matter of sex irregularity and have divided the group into (1) 
 prostitutes, which includes those who have been either regularly or 
 semi-regularly receiving money in return for sex relations, (2) those 
 sexually irregular, which here includes those who have been indis- 
 criminate in their relations but have not made a practice of taking 
 money for it, and (3) those who have never been sexually immoral 
 and those whose delinquency has been restricted to a single offense. 
 With this threefold distinction we have plotted the correlation be- 
 tween mental capacity shown by Test Aggregate and the degree of sex 
 irregularity. 
 
 The correlation ratio obtained is only .14 ± .051, a figure too 
 low to be very significant. The means are seen to run respectively: 
 prostitutes 75, sexually irregular 76.4, and those not sexually irregular 
 79.4. It seems clear, then, from this in connection with Tables 219 and 
 220 that the real distinction in mental capacity comes between those 
 
I 
 
 s 
 ^ 
 
 CO 
 
 
 **-t 
 
 
 o 
 
 
 0) 
 
 
 fld 
 
 
 & 
 
 
 v 
 
 
 Q 
 
 
 'O 
 
 
 § 
 
 
 a> 
 
 
 
 
 «> 
 
 
 ^ 
 
 1 
 
 <i 
 
 lA 
 
 •f-» 
 
 to 
 
 
 
 V 
 
 U 
 
 H 
 
 tin 
 
 >> 
 
 
 •d 
 
 
 
 
 ^ 
 
 •S 
 
 'S 
 
 
 
 "rt 
 
 g .3 
 
 I 
 
 d 
 a> 
 
 I 
 
 (U 
 
 a 
 
 o 
 
 o 
 
 
 "■*' rt 
 
 
 o 
 
 ■* 
 
 ■* 
 
 
 5 
 
 
 fei 
 
 S g S 
 
 K 
 
 
 
 
 
 H 
 
 i K S 
 
 S 
 
 1 
 
 i 
 
 Z6 0% 96 
 
 « w (N 
 
 lO 
 
 56 0? £6 
 
 i-H (M rH 
 
 tH 
 
 86 o* 16 
 
 N IN tH 
 
 U5 
 
 16 0'^ 68 
 
 ^ » CO 
 
 g 
 
 «8 0? IS 
 
 t> »0 Ttt 
 
 
 IS 0% 98 
 
 tH O 
 
 S 
 
 S8 o* 88 
 
 © <© CO 
 
 IN 
 
 £8 0% 18 
 
 g " - 
 
 S; 
 
 IS 0% 61 
 
 « c » 
 
 CO 
 
 61 o? U 
 
 CO 50 CO 
 
 s 
 
 U °% 91 
 
 OS 00 CO 
 
 o 
 
 CO 
 
 Si 0} 21 
 
 g " 
 
 ?5 
 
 SI 0% U 
 
 « ^- CO 
 
 CO 
 
 11 Oi 69 
 
 'OO CO Ttl 
 
 s 
 
 69 o» i9 2 iM « 
 
 s 
 
 19 0» S9 MM 
 
 o 
 
 B9 o» 89 o" ■* 
 
 CO 
 
 £9 o? 19 o> -< 
 
 s 
 
 19 o% 69 
 
 00 C4 
 
 o 
 
 65 °% 19 
 
 Tfi r1 
 
 Ui 
 
 19 Oi ss 
 
 iO O) 
 
 r- 
 
 BS oi 85 
 
 
 
 ES 0% Iff 
 
 I-l 
 
 iH 
 
 IS o» 6* 
 
 t-( 
 
 - 
 
 6f Oi If 
 
 tH 
 
 « 
 
 If o» S^ 
 
 
 
 
 
 1 
 1 
 
 6 
 1 
 g 
 
 1 
 
 > 
 1 
 
 
 1 
 E-i 
 
 
 o 
 
 ■a 
 
 I 
 I 
 
 •a 
 
 s 
 
 
 1 t^ 
 
 0) o 
 
 HO 
 
 SI5 
 
516 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 who have not been sexually irregular and those who have, and that, 
 in the further distinction between professional prostitutes and those 
 who have been promiscuous without being prostitutes, the factor of 
 mentality does not play a part. 
 
 In order to determine whether or not the matter of mental capacity 
 had any relation to the length of time these women have been sexually 
 irregular, we have plotted Table 222 which shows the correlation 
 between mental capacity measured by Test Aggregate and the num- 
 ber of years they have been sexually irregular. This includes both 
 the years they have been prostitutes and the time they have been 
 simply promiscuous. The correlation coefficient obtained is —.13 
 ± .061, which would indicate a negative relationship between the two 
 which is, however, questionable in the ligfit of its relatively large 
 standard deviation. The correlation ratios of .16 ±: .060 and .20 
 ± .059 further indicate a relationship which is at most slight. Con- 
 sideration of the means shows that whatever relationship exists is 
 negative. Our data on this point are, as has been explained in Chapter 
 XII, quite unsatisfactory in view of the fact that there are a goodly 
 number of women who are known to have been prostitutes or sexu- 
 ally irregular for a long time but from whom we have been unable 
 to obtain any statement as to the exact number of years. Nearly 50 
 per cent of these are Workhouse cases, "old rounders" who are 
 quite uniformly characterized by low intelligence, and whose records, if 
 they were obtainable, would have undoubtedly increased the negative 
 correlation. We feel inclined to believe, therefore, that there is prob- 
 ably a valid, though small association, and that there exists a significant 
 negative correlation between mental capacity as shown in this way and 
 the number of years sexually irregular. 
 
 (c) Mental Capacity and Venereal Disease 
 
 If, in considering next the relation between mental capacity of 
 those who are infected with venereal disease, as indicated by the 
 Wassermann and Complement Fixation tests, and those who are free 
 from the same, we plot the distributions graphically, we obtain the 
 curves shown in Chart XLVI. It is at once evident that those free from 
 venereal disease form a superior group. There is a clear difference 
 in the means 77.34 ± .879 as 'against 73.^6 ± .969, and Table 223 
 shows this differejnce ito be almost certainly valid. The difference is 
 very shght as regards the degree of concentration with which the cases 
 
t 
 
 i 
 
 tn 
 
 10 
 
 o 
 
 i 
 
 03 
 
 0) P" 
 
 &^ 
 
 as 
 
 <J 10 
 
 CO ^ e3 
 H "d m 
 
 M ■a 
 
 u 
 
 at 
 
 I 
 
 d 
 
 V 
 0) 
 
 a 
 o 
 
 ■■§ 
 
 I— t 
 
 K 
 
 o 
 O 
 
 ibfe 
 
 S6 o? £6 
 E6 ot 16 
 16 °% 68 
 68 o» i8 
 IS 0% 68 
 S8 o* 88 
 £8 o» 18 
 T8 o» 6i 
 6Z o* Zi 
 
 II 0% 91 
 
 fiZ 01 £Z 
 
 £Z o» IZ 
 
 IZ °% 69 
 
 )0?Z9 
 
 Z9 o* S9 
 
 B9 0% £9 
 
 £9 0% 19 
 
 19 0* 6S 
 
 fo^ZS 
 
 19 0% fiS 
 
 SS 0% £S 
 
 ES o? IS 
 
 IS o? 6t 
 
 »-tOOOiHtJr-tlN^CCC0M40«-*t»«^rtr- 
 
 ^ ,H r-lTHrtlO 
 
 <N rt rH (N iH eO (M ■* 
 
 H^^ tH Tli-* 
 
 HfNrHrHlOWrHt* 
 
 -iTii-<*<eo i-iira 
 
 ^(N r^(N '^rHOO 
 
 (M^i^HiHtN COrlH'<)< 
 
 H.-(t-<rH(MlN«-*'-i 
 
 M T-l iH iHtH 
 
 U^ 4< ci <N ri i O) ot 1^ ti -A -A CC ei ri O dl ot [I at lA 4' CO c!l rH N^ 
 IN M (N W IN (N 1-" 1-1 rH iH i-t i-H 1-t T-( tH ^ H\ 
 
 jBjnaaiJi jCnmxag siBa^ }0 nmg 
 517 
 
 g-s 
 
 
 a a 
 
 
 
 
 
 
 
 
 
 
 
 
 •c-c 
 
 
 uu 
 
 
 n m 
 
 
 sa 
 
 
 as 
 
 
 
 
 
 
 
 
 
 
 wm 
 
 
 OOi 
 
 
 «DU3 
 
 i2 
 
 
 II 
 
 •H-H 
 
 CD(D 
 
 b 
 
 
 
 II li 
 
 
 
 
 
 
 
 
 
 
 
 II 
 
 a-3 
 
 
 
 b 
 
 !! 
 
 
 
 Ss 
 
 3S 
 
 ;2^ 
 
 Mean = 
 sxually irre 
 .061 
 
 y irregular 
 m of years 
 
 r 
 
 H1=ig 
 
 »g 
 
 ss-g 
 
 ^ 
 
 1 p^ 
 
 
 
 ■SS 
 
 "S", 
 
 ^S 
 
 ^•s^ 
 
 
 Ut 
 
 
 S*H 
 
 
 
 
 
 
 1.9 
 
 o o 
 
 00 
 
518 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 Per 
 
 Cent 
 
 Mental Capacity as Measured by Test Aggregate 
 
 Percentage comparison between those having venereal disease and those frfte 
 from venereal disease. English-speaking cases only. 
 
 are grouped around the mean, however, as shown by comparison of the 
 standard deviations. Table 223 shows that so slight a difference is 
 only possibly significant. 
 
 TABLE 223 
 PRESENCE OR ABSENCE OF VENEREAL DISEASE 
 
 Comparison of the Means and Standard Deviations of Mental Capacity, 
 as Measured by Test Aggregate, of those Having Venereal Dis- 
 ease and those Free from Venereal Disease* 
 
 
 English-speaking 
 
 cases only 
 
 
 
 
 Venereal 
 Disease 
 
 V 
 
 Free from 
 Venereal 
 Disease 
 
 Difference 
 
 d 
 
 Chances that real 
 
 difference does 
 
 not exist are 1 
 
 in: 
 
 Mean 
 
 73.96 
 + .969 
 
 77.34 
 ±.879 
 
 -3.38 
 
 2.59 
 
 208 
 
 (Tm 
 
 
 
 
 ff. 
 
 9.19 
 + .602 
 
 8.43 
 ±.630 
 
 .76 
 
 .87 
 
 5 
 
 (T„ 
 
 
 
 
 
 90 
 
 92 
 
 
 
 
 
 
 ♦Reactions of 4+, 3-f and 2-|- to either the Wassermann or Complement Fixation test taken as 
 positive indications and reactions of — or =>= as negative indications. 
 
 As earlier stated in Chapter XII, data for the diagnosis of ven- 
 ereal disease are based upon the findings in the Wassermann and Com- 
 
MENTAL CAPACITY 519 
 
 plement Fixation tests. For purposes of the present comparison 4+, 
 3+ and 2+ were regarded as positive indications of the presence of 
 disease and — and ± as negative. The reports of 1 + were omitted as 
 not being either clearly positive or negative. It will, of course, be 
 objected that we have not selected an infallible means of detecting 
 venereal disease. We do not, however, claim to be doing anything 
 other than comparing the intellectual capacity of those who, by the 
 indications of the blood tests cited, are diagnosed as having, or being 
 free from,' syphilis or gonorrhea. 
 
 (</) Mental Capacity and Age of Entering Prostitution 
 
 Table 224 shows the relationship existing between mental capacity 
 measured by Test Aggregate and the age at which these women 
 entered prostitution. The correlation coefficient is .20 ± .069 and 
 the ratios slightly higher but affected by irregularities of the means. 
 There would, therefore, seem to be a sensible and significant, though 
 slight, correlation between the mental capacity of prostitutes and the 
 age at which they begin prostituting. 
 
 From an examination of the scatter table one notes the interesting 
 fact that, if lines are drawn through the table at the approximate 
 means of mental capacity and age at entering prostitution, thus 
 dividing the table into four quadrants, while the lower quadrants 
 contain almost identical numbers, in the upper quadrants two-thirds 
 as many cases fall in the right hand quadrant as in the left. This 
 would seem to indicate that there are as many bright women as dull 
 women who enter prostitution early but there are a greater percentage 
 of bright than dull women who enter prostitution late. (The terms 
 "bright" and "dull," as used here, are purely relative, designating those 
 above and below the average of this group, respectively.) 
 
 If we compare next the number of years that the women have 
 been in prostitution with mental capacity as measured by Test Aggre- 
 gate, we obtain th^ distribution figured in Table 225. This shows 
 a relationship wholly comparable with that of the preceding table, 
 as indicated both by correlation coefficient and by ratios, although 
 the coefficient is necessarily negative in this case to show the same 
 type of relationship. Separating this sheet into quadrants at the line 
 of the means as in the previous table, we see that here the two upper 
 quadrants have more nearly equal percentages but that there is a 
 greater percentage in the right hand lower quadrant than in the left. 
 
d 
 o 
 
 •43 
 
 I 
 
 o 
 
 a 
 P4 
 
 0) 
 
 bO 
 
 u> o 
 
 < M 
 
 u 
 
 (Si 
 
 O 
 
 d 
 
 I 
 
 d 
 o 
 
 1 
 
 tJ 
 
 o 
 O 
 
 tn 
 n 
 
 S6-S6 
 S6-T6 
 
 16-68 
 
 Z8-S8 
 S8-S8 
 
 TS-6i 
 
 61-U 
 
 iZ-Si 
 
 91-St 
 
 Sl-U 
 
 Ii-69 
 
 i9-B9 
 
 S9-E9 
 
 t9-6S 
 
 6S-ZS 
 
 ZB-fiS 
 
 SS-8B 
 
 es-iB 
 
 TS-6? 
 
 « t- O 
 
 iHOiHiHWiHlOCCNlONOl'O'-"- 
 
 rH fHrH^M i-l »■ 
 
 W i-HM rH 
 
 W.H COCfliH 
 
 Nt-I rHO) iHUJiHCON M 
 
 .-((NiH «^i-Hf-(M, r^ 
 
 iH ^ T)< T-l iH tH 
 
 -*« (N ^^ 
 
 w (N i-< w -* NiH 
 
 rH »-H ^ i-l « i-( ^ 
 
 N tH i-l 
 
 COCOn05CQnCQC>4(NC4CqO)e!|MO)(NC^^>-H^HiHFH^^i:-!rHiH 
 
 nor^n^ii^soj J auuajnia %v aay 
 
 oo 
 
 aa 
 
 oo 
 
 im 
 
 00« "« 
 -l^io PC- 
 OS 0>® « gj 
 
 " II ;Si" 
 "a *§ 
 
 rH^ *' bo 
 
 ■ li " w 
 ll'lao* 
 
 ^alg 
 
 go 
 
 M M O ^ 
 
 o boo o 
 E-i-<!00 
 
 520 
 
o 
 
 I 
 
 CO 
 
 o 
 
 .s 
 
 m 
 
 
 a 
 
 bo I 
 
 coa 
 
 Cd M 
 
 u 
 
 I 
 
 t) 
 
 <u 
 
 d 
 
 o 
 
 I 
 
 V 
 
 11 
 
 u 
 o 
 U 
 
 
 96-E6 
 
 S6-T6 
 
 1&-68 
 
 8-Z8 
 
 i8-S8 
 
 68-88 
 
 £8-18 
 
 T8-6i 
 
 61-U 
 
 U-9i 
 
 9I.-SI 
 
 Sl-U 
 
 11-69 
 
 9-i9 
 
 i9-B9 
 
 99-E9 
 
 E9-I9 
 
 I9-6S 
 
 6S-iB 
 
 zs-ss 
 
 SB-ES 
 
 ES-IS 
 
 TB-6* 
 
 CO X 
 
 ^4 1# 
 
 (N r-< rH rH 1-1 N CO 
 
 H,_| ^ TfiN 
 
 rt oq rt rt Ufa N «-< t> 
 
 HiH i-HCO TjIMCI 1-ltD 
 
 .H iH,-l M (O 
 
 CO rHCOCO 
 
 eDU3-<*ICO(NiHOOSOOI>tt'Ui'*MNrHOaiMt*OiOT(<COIN^ 
 NNININ'NN'-i'-"-"^'-"-"^'^'^'^ 
 
 nonnjusojj nj bM8A jo laqran^ 
 521 
 
 
 II II 
 
 §i 
 
 •as 
 oo 
 
 fl"a 
 
 ea 
 
 (OCO 
 
 Hi ^^ 
 
 ^ ■S.2 
 II ^-ti 
 
 b tA-5 
 
 i| 
 
 53 |S. 
 
 ""°!« 
 
 S<^S ° 
 ■=■ ^3 0) 
 
 II '-3 • ■ (o * 
 
 ^ O O « n 
 
 a.a « - 
 ■■ES.S 
 
 Eh 1300 
 
522 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 This would seem to indicate that while there are as many bright 
 as dull women who have been for long periods in prostitution, there 
 are more bright than dull women who have been in it only a short 
 time. 
 
 Summary 
 
 To summarize this section, we feel that we can say regarding 
 the relations which have been found to exist between the mental 
 capacity of these women as measured by Test Aggregate and the 
 phases of their sex history considered : 
 
 (1) That there is a slight positive relationship between mental 
 capacity measured by Test Aggregate and the age at which these 
 women committed their first sex offense. 
 
 (2) We have found a small, and perhaps questionable, negative 
 correlation of — .13 ± .061 between mental capacity and number 
 of years sexually irregular. We are inclined to believe that this would 
 be increased if we could include the hundred-odd cases whom we 
 know had been sexually irregular for a long time, but on whom 
 we have been unable to get a record of the exact length of time. 
 
 (3) Comparing the means of the mental capacity of those women 
 who are infected with venereal diseases, as evidenced by the reactions 
 to the Wassermann and Complement Fixation Tests, with those whose 
 reactions to these tests were negative, we found an appreciable differ- 
 ence in favor of the latter. The difference in the amount of variability 
 shown by the two groups was small and only possibly significant. 
 
 (4) When we compare the mental capacity of the women who 
 have been either prostitutes or sexually irregular with those who 
 have not been sexually promiscuous, we find that there is a difference 
 in the central tendency of the two groups in favor of those who have 
 not been sexually irregular. The amount of scatter about the mean 
 which the distributions show is not noticeably or significantly different. 
 
 Making the division, however, between prostitutes and non-pros- 
 titutes, i.e., by combining those who have been sexually irregular and 
 those who have not been sexually irregular, and comparing them 
 with the prostitutes, we obtain a sensible difference in the means 
 which is almost certainly significant. This difference is, it will be 
 noted, not as great as that between the combination of prostitutes 
 and those sexually irregular compared with those not sexually irregular. 
 It would seem, therefore, that, as far as mental capacity is concerned 
 the difference between those who have prostituted and those who 
 
MENTAL CAPACITY 523 
 
 have simply been sexually irregular is insignificant and that it is 
 the presence of those not sexually irregular in the comparison be- 
 tween prostitutes and non-prostitutes which gives it its significance. 
 
 A classification of the group into three divisions — prostitutes, those 
 sexually, irregular who have not prostituted, and those hot sexually 
 irregular — shows that the above assumption is correct. 
 
 (5) There is a sensible though small correlation between the 
 mental capacity and the age at entering prostitution, and an in- 
 spection of the scatter table would indicate that this seems to be 
 brought about by a lack of representation of the stupider women 
 among those who enter prostitution late rather than a concentration 
 of the brighter women there. 
 
 (6) A negative correlation of —.21 ± .069 was found between 
 mental capacity and the number of years in prostitution. Here an 
 inspection of the scatter table would seem to indicate that the women 
 who stay for a long time in prostitution are about equally divided 
 into those above and below the average mentally but that there is 
 a greater concentration of the brighter women among those who 
 remain in it less time. 
 
CHAPTER XVI 
 SUMMARY AND CONCLUSIONS 
 
 THE purpose of this study has been to achieve a more adequate 
 understanding of women delinquents. To some extent such 
 an understanding is furthered by the mere assembling of facts about 
 the group. Such information may at least help to remove and to 
 prevent certain misconceptions regarding their characteristics. For 
 a more basic understanding, however, these facts must be brought 
 into relation with other facts. That is, comparisons must be made 
 of one portion of the delinquent group with others, and of the group of 
 delinquent women both with groups of men delinquents and with the 
 general population. These comparisons we have attempted to carry 
 through, wherever the necessary data were available. We have 
 had occasion constantly throughout this study to comment on the 
 dearth of such information with reference to the general population. 
 This lack is responsible, more than any other one thing, for a general 
 want of conclusiveness about our findings. 
 
 In the present brief statement it is not our intention tO' indicate 
 the results of our study in any detail. Such material has been given 
 in summarized form, in the various chapters concerned. Here, 
 we are interested only in showing major trends and the more general 
 indications of our findings. Any discussion of the more detailed 
 relationships and distinctions between different portions of the de- 
 linquent group would involve unnecessary duplication. For such 
 information the reader is, therefore, referred to the earlier state- 
 ments. 
 
 It should be remembered that the women delinquents regarding 
 whom we offer information are those convicted of legal offenses, 
 through due process of law: and that they do not include those 
 closely related cases who offend in essentially the same ways but 
 escape arrest and conviction. 
 
 It may be desirable, also, to remind the reader that we make no 
 claim to comprehensiveness in our survey of the factors of im- 
 
 524 
 
SUMMARY AND CONCLUSIONS 525 
 
 portance in connection with delinquency. Certain very obvious ones 
 have been omitted, — some because we did not find it practicable to 
 include them in this investigation, others because of difficulty in 
 j-educing our observations to definite objective form, and still others 
 because they appeared so general as to elude exact observation, re- 
 quiring perhaps the perspective of time or of the knowledge of 
 many localities for even an attempt at precise formulation. Note- 
 worthy omissions to be accounted for from these various points of 
 view are the following: information regarding the physical condition 
 of the women of our group ; data regarding their emotional organiza- 
 tion and the more general characteristics of the personality as a 
 whole; data regarding the general environmental background, in- 
 cluding especially the quality and the number of the recreational 
 opportunities; facts about the public opinion of the community with 
 reference to moral standards, particularly as these concern sex mat- 
 ters; these and other factors perhaps equal in importance. We 
 shall not be drawn aside into a discussion of the significance of 
 these factors, important though we believe them to be, since we 
 lack the fa:ctual material to support any views which we might offer. 
 
 Turning to the positive aspects of our investigation, we note 
 two lines of influence which seem to have a bearing on the problem 
 of delinquency among women, namely: (1) poor economic back- 
 ground with few advantages or opportunities, including such conditions 
 as poor homes, very limited school opportunity, early age at start- 
 ing work and meager industrial training; and (2) a somewhat in- 
 ferior mentality. 
 
 A survey of the early home conditions in which these women 
 grew up shows that a large percentage came from the poorest homes 
 included in our estimate of home conditions, and that less than 
 ten per cent came from homes considered better than mediocre in 
 respect to economic status, moral standards and parental super- 
 vision. Though there is no possibility of determining what percentage 
 of the homes of non-delinquents would fall in the lower grouping, 
 we are able to show within the delinquent group that there is a 
 tendency for those brought up in the poorer homes to enter prostitu- 
 tion when younger, and to be convicted at an earlier age, than those 
 who were brought up in better homes. 
 
 With reference to educational background we find, again, that, 
 without reliable standards for the general population, we have no 
 way of knowing how far .the delinquent group lags behind the gen- 
 
526 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 eral population in this respect. We have found, however, that th? 
 delinquent group falls decidedly behind the requirements of present 
 opinion regarding the minimum amount of schooUng which should 
 be accepted as furnishing the rudiments of preparation for meeting 
 the demands of adult life. This is especially true as regards school 
 attainment. The record for the total group, with regard to school 
 grade completed, shows that on the average they have finished less 
 than the fifth grade. We saw reason for believing that this was 
 to some extent a secondary effect of the somewhat inferior intelligence 
 of the group. We were not able to discover more than very slight 
 relationships between the educational factors and the various aspects 
 of the criminal career. 
 
 Referring to the work records of the delinquent group, we note 
 many indications of general ipdustrial inefficiency. Nqt only is the 
 average prevailing wage low, but the records show much irregularity 
 of work, and a massing in the occupations which demand least in 
 the way of general educational background or special industrial 
 training.. That the situation in these respects is any worse than 
 among women wage earners as a whole it is impossible to say. It 
 is necessary to remember that the community has been very slow 
 to apply the same standards of industrial efficiency and remuneration 
 to women that it has applied to men, and that inability to earn a 
 decent living for herself and family is not considered an indicatioij 
 of incompetency in a woman to the same degree as in a man. Further, 
 the occupations which demand least in the way of background and 
 special training— domestic service and factory work-;— are the occu- 
 pations which include the largest numbers of women wage-earners 
 in general. The available data tended to show, however, that the 
 proportions in these occupations were somewhat overweighted for the 
 delinquent group as compared with the whole body of working 
 women. 
 
 It seems worthy of note that the women of this group, taken as 
 a whole, have evidently had demands upon them, during at least cer- 
 tain periods in their Ijves, to be financially productive. Only about 
 3 per cent of the total group have never been gainfully employed. 
 It appears, further, that over half of the group were idle, that is, 
 not engaged in a legitimate occupation, at the time of the present 
 conviction. It does not follow, however, that they were not earning 
 at this time, for it must be remembered that in general the offenses 
 committed by this group bring in remuneration, often on a scale 
 
SUMMARY AND CONCLUSIONS 527 
 
 decidedly above that which the same individuals can reach in legitimate 
 occupations. If, then, we urge the importance of economic factors in 
 relation to delinquency among women, it is not because we contend 
 either that the delinquent women, as a whole, could not earn a living 
 wage in some occupation, or that the wages which they do earn are 
 conspicuously lower than those of working women who have not 
 become delinquent. It is rather, in the first place, because some 
 degree of economic necessity is operative in the case of this group, 
 and, in the second place, because the offenses of most frequent occur- 
 rence offer financial attractions sufficiently great so that they may 
 at least compete with the legitimate occupations as a source of in- 
 ,come. If inhibitions of one sort or another are not sufficient to 
 deter a woman, it may well be that the greater probable lucrativeness 
 of prostitution, shop-lifting, or other forms of delinquency may prove 
 the adequate explanation for her entrance upon this career in at 
 least a fair proportion of the cases. 
 
 In our reference to inferior mentality we have in mind espe- 
 cially the general intelligence of the group. Had we adequate data, 
 in sufficiently exact form, we should make this category more com- 
 prehensive and include cer tain char actedstics- of temperament- and 
 emotional organization^which we believe to be^quite^ as important 
 .-as-4-he- factor of intelligence but which are far more difficult of 
 measurement. With regard to intelligence, all indications are that 
 the group of delinquent women is somewhat inferior to the general 
 population, though the difference is slight and the overlapping large. 
 This statement holds, though in varying degrees, whether we consider 
 the separate sub-groups or the composite group. It should be noted, 
 however, that our data have indicated a less extreme distinction, with 
 respect to intelligence, between the delinquent and the non-delinquent 
 groups than that urged by many recent investigators, notably Goring. 
 The most that we are prepared to say is that, other things being 
 equal, there is apparently a greater presumption in favor of de- 
 linquency in a group of women who are below the average in in- 
 telligence than in a group above the average. In view of this apparent 
 overweighting of the delinquent group in the direction of mental 
 inferiority it is somewhat surprising to find how slight are the in- 
 dications of relationship between intelligence and the various aspects 
 of the criminal career, such a^ degree of recidivism, age at first con- 
 viction, nature of offense, etc. Apparently low intelligence has more 
 weight in swinging the balanee-so that the woman will appear within 
 
528 STUDY OF WOMEN DELINQUENTS IN NEW YORK 
 
 the delinquent group than in deterrnining what type of delinquent she 
 will be or how serious an offender^/ 
 
 The factors mentioned above/Kave impressed themselves upon us 
 as important general influences associated with dehnquency. Never- 
 theless, when we turn to any specific comparison, involving these, 
 factors in relation to some aspect of delinquency, we are even more 
 impressed by the smallness of the relationships than by the fact of 
 their existence. Even when we compare the delinquent' group with 
 the general population we find relatively slight distinctions and much 
 overlapping. This fact may be exaggerated, to some degree, by the 
 inadequacy of our information regarding the general population, as 
 well as by the small numbers in the delinquent group. The evidence 
 available indicates very strongly,, however, that even with fuller 
 data we should still be dealing with small differences. This sug- 
 gests, further, that any seaj-ch for, a well-defined type of individual, 
 appearing as the delinquent woman, will- probably be fruitless. Ap- 
 parently the concept of such a type can not be saved even by ex- 
 panding it beyond Lombroso's anthropological criminal type and 
 pruning off certain of the absurdities incorporated in his idea. Within 
 all groups and all classes there are doubtless individuals whose 
 adjustment to the demands of society is more or less precarious. 
 Whether or not they become delinquent will depend, not so much 
 upon the appearance of a single decisive factor, as upon the massing 
 of factors in such a way as to disturb a more or less unstable initial, 
 adjustment. That certain factors, notably poor economic conditions, 
 variously operative, and inferior intelligence, are particularly likely 
 so to disturb the balance is the main point which we should urge in 
 emphasizing these conditions. 
 
 In further support of this point of view we would call attention 
 to our findings regarding inter-comparisons within the group of de- 
 linquents. Reference to the detailed discussions of the foregoing 
 chapters makes it clear that even those factors which we consider 
 most important in their bearing on delinquency have, apparently, not 
 been very effective in determining jhe^Ela-c e of an indi v idual__ sdlhin 
 the,delin^entjrou£ with respec^to the„serjousness_qt, the -f-reqaenty 
 of.oflfensesorotfiersimirar categories. It is interesting to note, 
 for example, with reference to intelligence, that there is more differ- 
 ence between the colored and the white of our group, or between 
 those who have been in domestic service and those who have not, 
 thaii between recidivists and first offenders, or between felons and 
 
SUMMARY AND CONCLUSIONS 529 
 
 misdemeanants, or between offenders against chastity and offenders 
 ■against property. Instances could be multiplied to show that a dichot- 
 omy of the delinquent group, based on factors not intrinsic to the 
 problem of delinquency, frequently yields more striking distinctions 
 than does one based on differences which are fundamental from point 
 of view of delinquency. 
 
 On the basis of our findings we can not agree with Goring in the 
 importance that he attaches to the constitutional factors in contrast 
 with the environmental factors, as determinants of crime. He states, 
 for example:^ "While with many of the former, high degrees of 
 association have been revealed, with practically none of the latter 
 do we discover any definite degree of relationship." And again: 
 "Crime is only to a trifling extent (if to any) the product of social 
 inequalities, of adverse environment, or of other manifestations of 
 what may comprehensively be termed the force of circumstances." 
 To the extent of insisting upon the slightness of observable rela- 
 tionships we are in agreement with Goring. We disagree, however, 
 in the preeminence attached to such a constitutional factor as defective 
 intelligence in contrast with economic factors. The relationships 
 which we have observed have been, if anything, more slight in case 
 of the measures of intelligence than in that of the indices of social 
 and economic factors. Fundamental differences between Goring's 
 group and ours may be offered in partial explanation of the dis- 
 agreement. We would suggest, as further possible explanations of his 
 findings, on the one hand, the fact that he dealt only with estimates 
 of intelligence, which are markedly subject to the influence of the 
 personal equation, and on the other hand, the fact that his data 
 regarding environmental factors are less complete and well controlled 
 than are his other data. It is at least possible that more thorough- 
 going investigation into the social and economic factors would have 
 disclosed quite as much reason for ascribing significance to the 
 environmental factors as to the constitutional, with no one factor of 
 either sort appearing as .extreme in importance. Our own findings 
 would suggest such an outcome. 
 ' Op. cit., p. 371. 
 
BIBLIOGRAPHY 
 
 Abbott, Grace: "Immigration and Crime." Journal of the American 
 Institute of Criminal Law and Criminology; November, 1915. 
 
 Annual Report of the Commissioner General of Immigration, 1914. 
 
 Annual Report of the State Commission of Prisons, 1916. 
 
 Annual Report, Police Department, City of New York, 1916. 
 
 AscHAFFENBURG, G. : "Crime and Its Repression." Translated by A. 
 Albrecht; 1913. 
 
 Ayres, L. p.: "A Scale for Measuring the Quality of Handwriting of 
 School Children." Rmsell Sage Foundation; 1912. 
 
 "Child Accounting in the Public Schools." Cleveland Education 
 
 Survey; 1915. 
 
 "Laggards in Our Schools," 1909. 
 
 BiNET, A., and Th. Simon: "Method of Measuring the Development of 
 the Intelligence of Young Children." Translated by C. H. Town. 
 
 Bogardus, E. S. : "The Relation of Fatigue to Industrial Accidents." 
 American Journal of Sociology; 1911-1912. 
 
 BoNSER, F. G. : "The Reasoning Ability of Children of the Fourth, Fifth 
 and Sixth School Grades." Teachers College, Columbia University, 
 Contributions to Education; No. 37, 1910. 
 
 Breckinridge, Sophonisba P., and Edith Abbott: "The Delinquent 
 Child and the Home." Russell Sage Foundation; 1912. 
 
 Buckingham, B. R. : "Spelling Ability, Its Measurement and Distribu- 
 tion." Teachers College, Columbia University, Contributions to 
 Education; No. 59, 1913. 
 
 Code of Criminal Procedure, State of New York, Title VI. 
 
 CoNYNGTON, Mary: "Relation Between Occupation and Criminality of 
 Women." Report on Condition of Woman and Child Wage-Earners 
 in the United States, Vol. XV : Government Printing Office, Wash- 
 ington, 1911. 
 
 Courtis, S. A. : "Manual of Instruction for Giving and Scoring the Cour- 
 tis Standard Tests." 
 
 Davenport, Charles B.: "The Family History Book." Bulletin No. 7, 
 Eugenics Record Office: Cold Spring Harbor, N. Y., September, 
 1912. 
 
 531 
 
532 BIBLIOGRAPHY 
 
 Davis, Katharine Bement: "The Rational Treatment of Women Con- 
 victed of Crime in the Courts of New York City." 
 
 "A Study of Prostitutes Committed to the State Reformatory for 
 
 Women at Bedford Hills," Chapter VIII of "Commercialized Pros- 
 titution in New York City," by George J. Kneeland; 1913. 
 
 Ellis, Havelock: "The Criminal"; 1895. 
 
 Franz, S. I.: "Mental Examination Methods"; 1912. 
 
 GiLLiLAND, A. R. : "The Mental Ability of 100 Inmates of the Columbus 
 
 (O.) Workhouse." Journal of the American Institute of Criminal 
 
 Law and Criminology ; Vol. 7,' No. 6; 1917. 
 Glueck, Bernard: "A Study of 608 Admissions to Sing Sing Prison." 
 
 Mental Hygiene, Vol. 2, No. 1 ; 1918. 
 GoDDARD, H. H.: The Standard Method for Giving the Binet Test." 
 
 Training School, Vol. X, No. 2; 1913. 
 
 "The Binet-Simon Measuring Scale for Intelligence." Training 
 
 School, Vol. VIII, No. 4; 1911. 
 
 "The Form Board as a Measure of Intellectual Development in 
 
 Children." Training School, Vol. IX; 1912. 
 
 i "Feeble-mindedness. Its Causes and Consequences"; 1914. 
 
 Goring, Charles: "The English Convict. A Statistical Study." T. 
 Fisher Unwin: London, W. C, 1913. 
 
 Guibord, Alberta S. B. : "Physical States of Criminal Women." Jour- 
 nal of the American Institute of Criminal Law and Criminology, 
 Vol. VIII, No. 1; 1917. 
 
 Haines, Thomas ,H. : "Feeble-mindedness Among Adult Delinquents." 
 Journal of the American Institute of Criminal Lazv and Criminology, 
 Vol. VII, No. 5; 1917. 
 
 Healy, W. : "A Pictorial Completion Test." Psychological Review, Vol. 
 21; 1914. 
 
 "The Individual Delinquent." Boston: Little, Brown & Co.; 1915. 
 
 and G. M. Fernald: "Tests for Practical Mental Classification." 
 
 Psychol. Monog., Vol. 13, No. 54; 1911. 
 
 Kent, G. H., and A. J. Rosanoff : "A Study of Association in Insanity." 
 
 American Journal of Insanity, Vol. 47; 1910. 
 Kneeland, George J. : "Commercialized Prostitution in New York City." 
 
 The Century Co.; 1913. 
 Knox, H. A. : "A Scale Based on the Work at Ellis Island for Estimating 
 
BIBLIOGRAPHY S33 
 
 Mental Defect." Journal of the American Medical Association, Vol. 
 62; 1914. 
 KoHS, Samuel C. : "The Practicability of the Binet Scale and the Ques- 
 tion of the Borderline Case." Bulletin No. 3, Psychopathic De- 
 partment, Series No. 2. Publications of the Research Department, 
 Chicago House of Correction; 1915. 
 
 Lauck, W. Jett and Sydenstricker, Edgar: "Condition of Labor in 
 American Industries." Funk & Wagnalls Co., New York; 1917. 
 
 LoMBROSO, CiESARE : "Crime : Its Causes and Remedies." Translated by 
 Henry P. Horton. Boston: Little, Brown & Co.; 1911. 
 
 Marx, Karl: "A Contribution to the Critique of Political Economy." 
 Translated from the Second German Edition by N. I. Stone. New 
 York: The International Library Publishing Co.; 1904. 
 
 NoRSWORTHY, N. : "The Psychology of the Mentally Deficient." 
 
 Ordahl, Louise E., and George Ordahl: "A Study of 49 Female Con- 
 victs." Journal of Delinquency, Vol. II, No. 6; 1917. 
 
 Osborne, Thomas Mott : "Society and Prisons." Yale University Press ; 
 1916. 
 
 Pearson, Karl: "On the Relationship of Intelligence to the Size and 
 ♦Shape of the Head, and to Other Physical and Mental Characters." 
 Biometrika, Vol. 5. 
 Pintner, R., and M. M. Anderson: "The Picture Completion Test." 
 
 Educ. Psychol. Monog., No. 20. 
 and D. G. Paterson: "A Scale of Performance Tests"; 1917. 
 
 Report of the City Council Committee on Crime of the City of Chicago, 
 
 March 22, 1915. 
 Report of the Superintendent of State Prisons. New York, 1917. 
 Richmond, Mary E.: "Social Diagnosis." Russell Sage Foundation; 
 
 New York, 1917. 
 Rossolimo, G. : "Die psychologische Profile zur Methodik der quantitativen 
 
 Untersuchung der psychische Vorgange in normale und pathologische 
 
 Fallen." Klinik d. psych, und nerv. Krankh. Vol. 6; 1911. 
 Rossy C S.: "Report of the First Three Hundred Cases Examined at 
 
 'the Massachusetts State Prison." Bulletin No. 17 of the Mass. 
 
 State Board of Insanity; January, 1916. 
 RUML, Beardsley: "The Reliability of Mental Tests in the Division of an 
 
 Academic Group." Psychol. Monog., Vol. 24, No. 4; 1917. 
 
534 BIBLIOGRAPHY 
 
 Stenquist, J. L., E. L. Thorndike, and M. R. Trabue: "The Intellectual 
 Status of Children Who Are Public Charges." Arch, of Psychol., 
 Vol. S, No. 33; 1915. 
 
 Sylvester, R. H. : "The Form Board Test." Psychol. Monog., Vol. IS, 
 No. 65; 1913. 
 
 Terman, L. M. : "The Measurement of Intelligence" ; 1916. 
 
 and others : "The Stanford Revision and Extension of the Binet- 
 
 Simon Scale for Measuring Intelligence." Educ. Psychol. Monog., 
 No. 18; 1917. 
 
 Thompson, H. B.: "Psychological Norms in Men and Women"; 1903. 
 Thorndike, E. L. : "Handwriting." Teachers College Record, Vol. 11, 
 No. 2; 1910. 
 
 "Reading Scale and Visual Vocabulary, Scale A," and "Administra- 
 tion Scale A for Visual Vocabulary." Teachers College Record, 
 September, 1914. 
 
 "The Elimination of Pupils from School." Bulletin No. 379, De- 
 partment of the Interior, Bureau of Education; 1907. 
 
 Trabue, M. R. : "Completion-Test Language Scales." Teachers College, 
 
 Columbia University, Contributions to Education, No. 77; 1916. 
 Tredgold, a. F. : "Mental Deficiency." Second Edition, 1914. 
 
 United States Census: Abstract "Thirteenth United States Census Re- 
 port," 1910. 
 ■ "Occupational Statistics," 1910. 
 
 "Prisoners and Juvenile Delinquents," 1910. 
 
 "Special Report of the Census Office: Prisoners and Juvenile Delin- 
 quents in Institutions," 1904. 
 
 "Thirteenth Census of the United States," Vol. Ill, 1910. 
 
 Weidensall, Jean: "The Mentality of the Criminal Woman"; 1916. 
 Whipple, G. M.: "Manual of Mental and Physical Tests"; 1915. 
 WooDWORTH, R. S., and F. L. Wells: "Association Tests." Psychol. 
 
 Monog., Vol. 13, No. 57; 1911. 
 Woody, C. : "Measurements of Some Achievements in Arithmetic." 
 
 Teachers College, Columbia University, Contributions to Education, 
 
 No. 80; 1916. 
 Woolley, Helen Thompson, and Charlotte R. Fischer: "Mental and 
 
 Physical Measurements of Working Children." Psychol. Monog., 
 
 Vol. 18, No. 1; 1914. 
 
 Yerkes, R. M., J. W. Bridges, and R. Hardwick: "A Point Scale for 
 Measuring Mental Ability"; 1915. 
 
INDEX 
 
 'A 
 
 Abbott, Edith, 209 
 
 Abbott, Grace, 195 
 
 Age, distinction between "legal" and 
 "actual," of delinquents, 13 note; in 
 relation to number of convictions, 
 105 ff.; at first conviction, 102 ff., 
 128 fif., 138 f., 196 ff., 227 f., 261 ff., 
 288 f., 309 ff., 447 f.; of delinquent 
 women by institutional groups and 
 by total, 147 ff. ; definition of unit 
 for measuring, 148; of delinquent 
 women at first marriage, 162; of 
 first husband at marriage, 162 f. ; 
 of delinquent women classified by 
 nativity and color, 174 ff. ; of parents 
 at birth of subject, 231 ff. ; of sub- 
 ject at death of parents, 232 ff. ; at 
 leaving school, 250, 256 ff., 494 ff.; 
 at starting work, 306 ff., 498 f.; in 
 relation to wage, 353 f . ; at first sex 
 offense, 380 ff., 510 ff.; at entering 
 prostitution, 391 ff., 519 f. 
 
 Albion, Western House of Refuge at, 
 15; mental capacity of, and Bedford 
 groups compared, 447 f., 458. 
 
 Alcohol, use of, 147, 153 ff.; mental 
 capacity and use of, 485 ff. 
 
 Alcoholic group. See Intoxication 
 group (Workhouse). 
 
 Aliens. See Foreign born. 
 
 Aschaffenburg, G., 245, 413 note. 
 
 Auburn prison, study of men inmates 
 of, by Heacox, 7; comparison of 
 mental capacity of men inmates of, 
 and women inmates of, 450 ff., 458. 
 See also State Prison for Women 
 at Auburn, N. Y. 
 
 Ayres, L. P., 46, 259 ff. 
 
 Ayres handwriting scale, 46, 52, 291, 
 294, 297 f., 299. 
 
 B 
 
 Bedford Hills, State Reformatory for 
 Women at, passim. See especially 
 8 10 IS ff., 39 ff., 54 ff., 63, 70, 86 
 ff., 107 ff., 110, 112 ff., ,121 ff., 128 
 
 ff., 135 ff., 139 ff., 143, 148 ff., 152 f., 
 157 ff., 169 ff., 208 ff., 230 ff., 253, 
 254 ff., 258 f., 264 ff., 271, 273, 275, 
 291 ff., 307, 313 ff., 330 f., 335 ff., 
 363, 368 f., 381 f., 384 ff., 400, 408 f., 
 421 ff., 429 ff., 437 f., 440 ff., 447 ff. 
 
 Blakeman's criterion, 35. 
 
 Bogardus, E. S., 45. 
 
 Bonser, F. G., 45. 
 
 Breckinridge, S. P., 209. 
 
 Bridges, J. W., 43, note. 
 
 British Royal Commission for the 
 Care and Control of the Feeble- 
 minded, 414 
 
 Buckingham, B. R., 45. 
 
 Buckingham spelling scale, 45, 52, 291, 
 294, 296. 
 
 Case history method, in cotnparison 
 with 'mass methods, 11 ff. 
 
 Catholics. See Religion. 
 
 Central tendency, defined, 32. 
 
 Charwomen, 312 ff. 
 
 Chastity, offenses against, 119 f., 131 
 ff., 177 ff., 460 ff., 472 ff., 478. See also 
 Prostitution, Sex irregularities. New 
 York City Police Department Class- 
 ification of Offenses, Offense. 
 
 Chicago House of Correction, com- 
 parison of mental capacity of, in- 
 mates with that of Bedford women, 
 449, 458. 
 
 Child Labor Law, provisions of, 306. 
 
 Cincinnati Bureau of Vocational Guid- 
 ance, 415, 425 ff. 
 
 Civil condition, of delinquent women, 
 161 f. 
 
 Clerical work, 312 ff. 
 
 Color, in relation to delinquency, 167 
 ff. ; of parents, 230; in relation to 
 educational factors, 265, 268, 275 ff., 
 290 ff. ; in relation to kind of work, 
 324 f. ; in relation to sex offenses, 
 382 f., 388 f., 393 ff.; in relation to 
 mental capacity, 480, 482 ff. 
 
 Commitments, to hospitals for insane, 
 159 f. 
 
 535 
 
536 
 
 INDEX 
 
 Compulsory education law, 257. 
 
 Complement fixation test, 522. 
 
 Contemporary group (Bedford), 17., 
 
 Contingency coefficient, defined, 33. See 
 RelationsHips. 
 
 Conviction, age at first, 102 fif., 128 ff., 
 138 f., 196 ff., 227 f., 261 ff., 288 f., 
 309 ff., 477; defined, 103; kind of 
 work at time of first, 317; kind of 
 work at time of present, 318; wage 
 at time of first, 338; wage at time 
 of present, 339 f. 
 
 Convictions, number of previous, 102 
 ff., 105 ff., 123, 132, 156 f., 182 ff., 
 191 ff., 228 f., 269 f., 286 ff., 351, 355 
 ff., 396 f., 464 ff. See also Recidiv- 
 ists, First Offenders, Offense. 
 
 Convicts, comparison of women de- 
 linquents with English male, 129 ff. 
 See also Criminals, Delinquents, 
 Felons. 
 
 Conyngton, Mary, 85, 322 f., 326, 361, 
 377 f. 
 
 Correlation coefficient, defined, 33. See 
 Relationships. 
 
 Correlation ratio, defined, 33. See 
 Relationships. 
 
 Counties, in which delinquent women 
 were convicted, 86. 
 
 Courtis, S. A., 45, 51. 
 
 Courtis . arithmetic tests, 45, 51, 291, 
 298, 300 f. _ 
 
 Courts, in which convictions were 
 made, classified, 87. 
 
 Criminality, relation of education to, 
 245; attitude of the Italian School 
 and Socialists toward, 305. See also 
 Delinquency. 
 
 Criminals, importance of understand- 
 ing of, 3 ff. ; defined, 13. See also 
 Convicts, Delinquents, Felons, Of- 
 fenders. 
 
 Crime, importance of consideration of 
 criminal as well as of, 3 ff. See 
 also Offense. 
 
 D 
 
 Davenport, Charles B., 76, 241. 
 
 Davis, Katharine B., 16, 160, 235, 241, 
 243, 324, 400 f. 
 
 Delinquency, nature of, 84 ff. ; extent 
 of, 102 ff., nativity and color in re- 
 lation to, 167 ff., relations of early 
 home conditions to, 205 ff., relation 
 of facts of family history to, 228 ff. ; 
 relation of education to, 245 ff. ; sex 
 irregularities and, 380 ff. ; mental ca- 
 
 pacity in relation to, _ 413 ff. See 
 also Offense, Convictions, Crimin- 
 ality. 
 
 Delinquents, passim. See especially: 
 comparison of women, with -general 
 population, S ff., 171 ff., 259 ff., 288 
 290_ff., 320 ff., 358 ff., 413 ff.; com- 
 parison of women, with men, 7 ff, 
 100 ff., 127,129 ff., 159, 449 ff.; use 
 of term, in preference to term 
 "criminals," 13; groups of, studied 
 in present investigation, 13 ff. ; class- 
 ification of offenses of, 84 ff. ; juve- 
 nilei 140 ff., 200. ff., 477 f. See also 
 Convicts, Criminals, Felons, Offend- 
 ers, Misdemeanants. 
 
 Domestic Service, - percentages of 
 women delinquents employed in, 311 
 ff. ; comparison of delinquents with 
 general population with reference to 
 percentages employed in, 322 ff.; 
 comparison of mentality of delin- 
 quents employed in, with that of de- 
 linquents in other occupations, 
 503 f. 
 
 Drugs, addiction to, 147, 153, 157 ff.; 
 addiction to, in relation to mental 
 capacity, 485 f. 
 
 Economic efficiency, relation of, to 
 delinquency, 304 ff., 526 f. See also 
 Wage, Work. 
 
 Education, relation of, to crime, 245 
 ff. See also School. 
 
 Educational attainment, indices of, 250 
 ff. ; relation of mental capacity to, 
 493 ff. See also Educational tests. 
 Grade finished. School. 
 
 Educational opportunity, indices of, 
 250 f . ; relation of mental capacity 
 to, 493 ff. See also School. 
 
 Educational tests, list of, 45 f., 51 f.; 
 results of, 292 ff. 
 
 Ellis, Havelock, 413, note. 
 
 Elmira, New York State Reformatory 
 at,_ 100 ff. 
 
 English-speaking, necessity of restrict- 
 ing groups to, cases for majority of 
 mental tests, 58 f., 416; meaning of 
 term, as used in present study, 58 f. ; 
 comparison of mental capacity of, 
 cases with that of non-English- 
 speaking cases, 484 f. 
 
 Examination, methods of psycholog- 
 ical, of delinquent groups, Z6 ff. See 
 also Tests. 
 
INDEX 
 
 537 
 
 Factory work, 311 ff. 
 
 Family, relation of, status to delin- 
 quency, 205 ff., 228 ff.; size of, in 
 relation to delinquency, 235 f. See 
 also Home. 
 
 Feeble-mindedness, relation of delin- 
 quency to, 413 ff. See also Mental 
 capacity. 
 
 Felonies, most criminological studies 
 concerned with men convicted of, 7 
 f . ; institutions receiving commit- 
 ments for, 15, 17, 20; distinction be- 
 tween, and misdemeanors, 89; per- 
 centages convicted of, by institu- 
 tions, 97 ff. ; percentages convicted 
 of, on iirst offense, 136 ff. ; percent- 
 ages convicted of, by nativity groups, 
 180 f., 200. See also Felons, Offense. 
 
 Felons, comparison of male, and fe- 
 male, lOO ff., 127, 129 ff., 159, 450 ff.; 
 number of previous convictions 
 among, 116 ff., 185 ff. ; age at first 
 conviction among, 131 ; recidivists 
 and first offenders among, 126 f . ; 
 comparison of mental capacity of 
 misdemeanants and, 463, 474. See 
 also Convicts, Criminals, Delin- 
 quents, Felonies, Offenders. 
 
 Fernald, Grace M., 44, 416. 
 
 Finger-print records, 7l f. 
 
 First offenders, percentages of recid- 
 ivists and, 126, f., _ 188 ff. ; mental 
 capacity of recidivists and, 468 ff. 
 
 Fischer, Charlotte R., 42 f. 
 
 Foreign born, comparison of percent- 
 ages of, among delinquents and in 
 general population, 167 f., 171 ff. See 
 also Nativity. 
 
 Franz, S. I.. 44, 50, 57 
 
 Frye, Elisabeth, 4 note. 
 
 Gilliland, A. R., 456. 
 
 Glueck, Bernard, 7, 88 f., 191 note, 243, 
 325, 378. 
 
 Goddard, H. H., 43 note, 45 note, 413 
 note, 449. 
 
 Gonorrhea. See Venereal disease. 
 
 Goring, Charles, 7, 12, 36 f., 104 f., 
 130 f., 236, 244, 247 ff., 413 f., 466, 
 468, 472, 527, 529. 
 
 Grade finished, as index of educational 
 attainment, 251, 493 ff. ; comparison 
 of verified and unverified data re- 
 garding, 253 f.; data on, by institu- 
 
 tions and by nativity and color 
 groups, 270 ff.; relation of, to age, 
 275 ff. ; relation of, to age at leaving 
 school, 278 ff. ; relation of, to years 
 in school, 280 ff. ; relation of nature 
 of present offense to, 284 ff. ; rela- 
 tion of number of previous convic- 
 tions to, 286 ff. ; relation of age at 
 first conviction to, 288; relation of 
 occupational factors to, 326 ff., 332 
 ff., 348, 350 ff., 373, 375 ; relation of 
 age at entering prostitution to, 397 
 f. ; relation of mental capacity to, 
 494 ff. 
 
 Groups, descriptive account of delin- 
 quent, 15 ff. ; significance of classi- 
 fication by institutional, 29 f . ; statis- 
 tical methods of describing and com- 
 paring, 32 ff. 
 
 Guibord, Alberta S. B., 159. 
 
 H 
 
 Habits, 153 ff. See also Alcohol, 
 Drugs, Tobacco. 
 
 Haines, Thomas H., 416, 455. 
 
 Hardzmck, R., 43 note. 
 
 Heacox, Frank, 7, 18 note, 55, 127 
 note, 159. 
 
 Hebrews. See Religion. 
 
 Healy, William, 44, 44 note, 51, 140 
 ff., 416. 
 
 Hereditary factors, in relation to de- 
 linquency, 240 ff. ; in relation to 
 mental capacity, 491 ff. 
 
 Home, relation of early, conditions to 
 delinquency, 205 ff. ; methods of es- 
 timating, conditions, 205 ff. ; estimate 
 of economic status in the, 207 ff. ; 
 estimate of moral standards in the, 
 210 ff. ; estimate of parental super- 
 vision in the, 212 f. ; total estimate 
 of the, 214 ff. ; cases representing 
 different classes of, conditions, 219 
 ff. ; relation between, conditions and 
 age at first conviction, 227 f . ; rela- 
 tion between, conditions and num- 
 ber of convictions, 228 f. ; relation of 
 mental capacity to early, conditions, 
 489 ff. 
 
 Hotel work, 312 ff. 
 
 Illiteracy, amount of, in relation to 
 extent of criminality, 246 f. ; extent 
 of, among delinquent women com- 
 pared with figures for general popu- 
 lation, 288 ff. 
 
538 
 
 INDEX 
 
 Immigrants. See Foreign born. 
 
 Immigration Act of February 20, 1907, 
 quoted, 173 f. 
 
 Intelligence, 527 f. See also Mental 
 capacity. 
 
 Intoxication, sentences given cases 
 convicted of, 22; difficulty in study 
 of cases committed to Workhouse 
 for, 23 f. ; omission from general 
 Workhouse group of individuals 
 committed for, 24 f. See also 
 Alcohol. 
 
 Intoxication group (Workhouse), de- 
 scribed, 24; data regarding, 1S4 ff. ; 
 included in group of those "exces- 
 sively alcoholic" for measurement of 
 mental capacity, 486 flf. 
 
 Investigation, methods of social, em- 
 ployed in present study, 60 ff. ; at- 
 titude of different delinquent grovips 
 toward, of record, 63 ff. 
 
 Inwood House, 15 ; account of, group, 
 18 ff. See also Magdalen Home. 
 
 J 
 
 Judgments of intelligence, as criterion 
 of reliability of tests, 52 ff. ; Pearson 
 scale for, S3; relation of test results 
 to, 55 ff. 
 
 Juvenile delinquents, number of women 
 of present group having record of 
 commitment as, 140 ff. ; contention 
 that majority of criminals begin 
 careers as, 141 ; summary of facts 
 regarding women who have been, 
 143 f . ; smaller proportions of, 
 among foreign born of delinquent 
 group than among native born, 196; 
 study of home conditions of, in 
 Chicago, 209 f. 
 
 K 
 
 Kelly, T. L., 416. 
 
 Kent, G. H., 44. 
 
 Kneeland, George J., 16 note, 160 note, 
 
 235 
 Knox, H. A., 44, 57. 
 Kohs, Samuel C, 449, 458. 
 
 Laboratory of Social Hygiene, genesis 
 
 of, 4. 
 Language difficulty, factor of, as 
 
 affecting application of mental tests, 
 
 S7 ff., 416. 
 
 Lauckj W. Jett, 359 note. 
 Lombroso, Caesare, 3, 12, 36, 141 f 
 246 f., 304. 
 
 M 
 
 Magdalen Home, passim. See espe- 
 cially 8, 15, 18 ff., 55, 63, 86 ff., 107 
 ff., 110, 112 ff., 121 ff., 128 ff., 135 ff., 
 139 ff., 143, 148 ff., 152 f., 157 ff 
 169 ff., 208 ff., 230 ff., 253, 255 «".. 
 258 f., 264 ff., 271, 273, 291 ff., 307 
 313 ff., 324, 330 f., 335 ff., Z63, 368 f 
 381 f., 384 ff., 408, 421 ff., 424 f., 429 
 ff., 437 f., 440 ff. 
 
 Marriage, age at first, 162 ; age of first 
 husband at, 162 f. 
 
 Marx, Karl, 304. 
 
 Massachusetts Bureau of Statistics 
 359 note. 
 
 Men, comparison of number of, with 
 number of women committed to 
 state prisons in New York, 7 ff., 17; 
 comparison of women delinquents 
 with, delinquents, 100 ff., 127, 129 ff 
 159, 449 ff. 
 
 Mentaces, use of term by Pearson, 54. 
 
 Mental capacity, methods of measur- 
 ing, 36 ff. ; judgments of, by Pearson 
 scale, 52 ff. ; selection of tests for 
 measuring, 55 ff. ; "Test Aggregate" 
 best single measure of, available for 
 present group, 58; relation of, to 
 educational factors, 250 ff., 282 f., 
 493 ff.; of delinquent women in com- 
 parison with that of general popula- 
 tion, 413 ff., data on, of draft group 
 in army, 416 ff. ; as measured by 
 Stanford-Binet Scale, 416 ff., 447 ff., 
 452 f., 483 f., 496; as measured by 
 Yerkes-Bridges Point Scale, 424 ff., 
 450 ff., 455 ff., 482 f. ; of non-delin- 
 quent girls examined at Cincinnati 
 Bureau of Vocational Guidance, 425 
 ff. ; inter-comparison of delinquent 
 groups with respect to, 435 ff.; as 
 measured by "Test Aggregate," 437 
 ff., 460 ff., 485 ff. ; is measured by 
 Binet-Simon Scale (Goddard revis- 
 ion), 449 f. ; relation of, to nature of 
 offense, 460 ff., 472 ff.; relation of, 
 to extent of criminal record, 464 ff. ; 
 relation of, to age at first conviction, 
 470 f., 477 f.; relation of, to nature 
 of first sentence, 474 ff. ; relation of, 
 to age, 480 f . ; of women delinquents 
 classified as to color, 480, 482 ff. ; of 
 women delinquents classified as Eng- 
 lish-speaking and non-English-Speak- 
 
INDEX 
 
 539 
 
 ing, 484 f.; as measured by Per- 
 formance Tests, 484 f. ; relation of, 
 to use of drugs and alcohol, 485 ff. ; 
 relation of, to home conditions and 
 hereditary factors, 489 ff. ; relation 
 of, to work record, 498 ff.; relation 
 of, to sex history, SIO ff.; indi- 
 cation of significance to be attached 
 to, in relation to delinquency, 52S, 
 S27ff. 
 
 Mentality. See Mental capacity. 
 
 Misdemeanants, comparison of women, 
 with men, 456 f . ; number of previ- 
 ous convictions among, 116 ff., 185 
 ff. ; recidivists and first offenders 
 among, 126 f. ; age at first convic- 
 tion among, 131 ; comparison of men- 
 tal capacity of felons and, 463, 474. 
 See also Criminals, Delinquents, 
 Misdemeanors, Offenders. 
 
 Misdemeanors, groups of present study 
 receiving commitments for, IS f., 18 
 f., 20, 21, 25; distinction between 
 felonies and, 89; percentages con- 
 victed of, by institutions, 97 ff. ; per- 
 centages convicted of, on first of- 
 fense, 136 ff. ; percentages convicted 
 of, by nativity groups, 180 f., 200. 
 See also Misdemeanants, Offense. 
 
 N 
 
 Nativity, in relation to delinquency, 
 167 ff. ; of parents, 228 ff.; in rela- 
 tion to educational factors, 265, 268, 
 275 ff., 290 ff. ; in relation to kind of 
 work, 324 f. ; in ^relation to sex 
 offenses, 382 f., 388 f., 393 ff.; in 
 relation to mental capacity, 480, 484 
 f. See also English-speaking, For- 
 eign born, Non-English-speaking. 
 
 Negroes. See Color. 
 
 New Jersey Bureau of Statistics, 359 
 note. 
 
 New York City, interpretation of term 
 "disorderly conduct" in, 85; majority 
 of present group convicted through 
 courts of, 86, 305 ; comparison of de- 
 linquent women with general female 
 population of, with respect to nativ- 
 ity and color, 171 ff. ; lack of school 
 record system in, prior to 1909, 
 252. 
 
 New York City Police Department, 
 classification of offenses by, 85 f., 
 88 ff., 100 ff.. 135 ff., 143, 177 ff., 198 
 ff., 46Q ff., 472 ff., 478 f. See also 
 Chastity, Property. 
 
 New York City Workhouse, passim. 
 See especially 10, 15, 21 ff., 55, 63, 
 86 ff 107 ff., Ill, 112 ff., 121 ff ; 128 
 ff., 135 ff., 139 ff., 143, 148 f., 1^1 ff., 
 157 ff., 169 ff., 206, 218 f., 230 ff., 
 253, 255 ff., 258 f., 264 f., 267 
 271, 274 f., 291 ff., 307 f., 313 ff. 
 330 f., 335 ff., 363, 368 f. 381 f. 
 384 ff., 408 f., 421, 429, 437, 439 ff. 
 456 f. 
 
 New York County Penitentiary, 
 passim. See especially 8, 15, 20 f 
 63,70, 86ff., lO^ff., Ill,'ll2'ff.,m' 
 ff., 128 ff., 135 ff., 139 ff., 143, 148 f , 
 151 ff., 157 ff., 168 ff., 208 ff., 230 ff. 
 253, 254 ff., 258 f., 264 f., 267 f., 271 
 274, 291 ff., 307 f., 313 ff., 330 f' 335 
 ff., 363, 368 f., 381 f., 384 ff., 408 f., 
 421, 429, 437, 439 ff., 454 f. 
 
 New York State, present study limited 
 to cases committed from, 5, 13, 305 ; 
 relative numbers of men and women 
 committed to state prison in, 7 f.; 
 enumeration of groups of delinquent 
 women of, not included in present 
 study, 27 f. ; meanings for term 
 "Vagrants" in, 86; number of per- 
 sons per family in, 235 ; provisions 
 of Compulsory Education Law in, 
 257; amount of illiteracy in general 
 female population over ten years of 
 age in, 290 ff.; provisions of Child 
 Labor Law in, 306; percentage of 
 general population of, working be- 
 tween ten and fourteen years of age, 
 306 f . ; comparison of females over 
 fifteen years of age in, with women 
 delinquents with respect to kind of 
 work done, 320 ff. ; wage statistics 
 for certain industries in, 358 f. 
 
 New York State Prisons, relative 
 numbers of men and women com- 
 mitted to, 7 f. See also Auburn, 
 Sing Sing; State Prison for Women 
 at Auburn, N. Y. 
 
 New York State Reformatories. See 
 Albion, Bedford Hills, Elmira. 
 
 Night Court for Women for Boroughs 
 of Manhattan and Bronx, probation 
 group made up of women placed on 
 probation through, IS, 25 f. 
 
 Ninth District Court, IS. 
 
 Non - English - speaking, performance 
 tests for, cases, 57, 59; necessity of 
 excluding, cases from majority of 
 mental tests, 58 f. ; comparison of 
 mental capacity of, cases with that 
 of English-speaking cases, 484 f. 
 
 Norsworthy, N., 44 note. 
 
540 
 
 INDEX 
 
 Occupational history, relation of, to 
 delinquency, 304 S., 526 f. See also 
 Wage, Work. 
 
 Occupations, classification of, 311 f., 
 324 f. ; distribution of delinquent 
 women with reference to, 313 ff. ; 
 comparison of delinquent women 
 with general female population with 
 respect to, 320 fl. ; relation between, 
 and school grade, 326 ff. ; relation 
 between, and excellence of work 
 record, 370 ff. ; relation between, and 
 mental capacity, SCO ff. See also 
 Work. 
 
 Offenders, use of term, 13, 15 ; women, 
 who escape detection, 28 f., 417, 524. 
 See also Criminals, Convicts, Delin- 
 quents, Felons, Misdemeanants. 
 
 Offense, nature of present, for institu- 
 tional groups and for total, 88 ff., 145 ; 
 comparison of female felons with 
 male felons with respect to nature 
 of, lOO ff. ; relation between number 
 of convictions and nature of, 116 ff. ; 
 relation between age at first convic- 
 tion and nature of, 131 ff. ; nature of 
 first, for institutional groups and for 
 total, 135 ff., 146; nature of, for 
 women who have been juvenile de- 
 linquents, 143; comparison of nativ- 
 ity groups with respect to nature of 
 present, 177 ff. ; comparison of nativ- 
 ity groups with respect to nature of 
 first, 198 ff. ; data regarding first sex, 
 380 ff. ; nature of present, in rela- 
 tion to mental capacity, 460 ff. ; 
 nature of first, in relation to mental 
 capacity, 472 ff. See also Felonies, 
 Misdemeanors, New York City Po- 
 lice Department Classification. 
 
 Ohio Penitentiary, comparison of, men 
 with New York Penitentiary women 
 with respect to mental capacity, 455 
 ff. 
 
 Ordahl, George, 456. 
 
 Ordahl, Louise E., 456. 
 
 Osborne, Thomas Mott, 4. 
 
 Partial correlation coefficient, defined, 
 34. See also Relationships. 
 
 Paterson, D. G.. 44 note. 
 
 Pearson, Karl, 52 ff., 114 note. 
 
 Penitentiaries. See New York County 
 Penitentiary, Ohio Penitentiary. 
 
 Penology, evolution of present-day, 3 
 ff. 
 
 Performance tests, selection of group 
 of, for use where language factor is 
 important, 57; correlation of, with 
 other series of tests, 58. 
 
 Personal service, 312 ff. 
 
 Pintner, R., 44, 51, 57. 
 
 Police Department, _ New York City. 
 See New York City Police Depart- 
 ment. 
 
 Pregnancies, number of, 238. 
 
 Prisons. See New York State prisons. 
 
 Private institutions, 27. 
 
 Probation group, passim. See especial- 
 ly 1-5, 25 ff., 63, 86 ff., 107 ff., Ill, 
 112 ff., 121 ff., 128 ff., 135 ff., 139 ff., 
 143, 148 f., 151 ff., 157 ff., 168 ff., 
 208 ff., 230 ff., 253, 255 ff., 258 f., 264 
 J., 267, 271, 274 f., 291 ff., 313 ff., 
 324, 330 f., 335 ff., 363, 368 f., 381 f., 
 384 ff., 408, 421, 429, 437, 439 ff. 
 
 Professional service, 312 ff. 
 
 Property rights, offenses against, 119 
 ff., 131 ff., 177 ff., 460 ff., 472 ff., 
 478. See also Offense, New York 
 City Police Department Classifica- 
 tion of Offenses. 
 
 Prostitutes, study of, by Dr. Davis, 
 160, 324; definition of term, 385; 
 percentages of, by institutions and 
 by nativity groups, 385 ff.; special 
 study of, 391 ff. ; comparison of, with 
 non-prostitutes, with respect to men- 
 tal capacity, 513 f. 
 
 Prostitution, length of time in, 389 f., 
 521 f . ; age at entering, 391 ff., 519 f. 
 See also Chastity, Offense, Sex. 
 
 Protestants. See Religion. 
 
 Psychological examination, methods 
 of, 36 ff. 
 
 R 
 
 Rape, or consent at first sex offense, 
 381 ff. 
 
 Recidivists, defined, 126, 191 note; per- 
 centages of, and first offenders by 
 institutional groups, 126 f. ; percent- 
 ages of, and first offenders by na- 
 tivity and color groups, 188 ff. ; men- 
 tal capacity of, and first offenders, 
 468 ff. 
 
 Relationships, methods of expressing, 
 33 f. 
 
 Religion, 163 ff. 
 
 Restaurant and hotel work, 312 ff. 
 
 Richmond, Mary E., 60. 
 
 Rosanoff, A., 44. 
 
INDEX 
 
 541 
 
 Rossolimo, G., 51. 
 Rossy, C. S., 454. 
 Ruml, Beardsley, S3. 
 Russell Sage Foundation, 60. 
 
 School, verification of, record, 252 flf. ; 
 kind of, attended, 254 f. ; regularity 
 of, attendance, 255 f . ; age at leav- 
 ing, 250, 256 ff., 494 ff.; number of 
 years in, 264 ff., 494, 496; grade 
 finished in, 270 ff., 326 ff., 348, 351 
 ff., 373 375, 494, 496 f. See also 
 Education. 
 
 Sex, history of, irregularities, 380 ff. ; 
 conditions relating to first, offense, 
 380 ff. ; source of first, instruction, 
 
 383 ff. ; classification of women de- 
 linquents by degree of, irregularity, 
 
 . 385 ff. ; mental capacity and classi- 
 fication as, offenders, 512 ff. See 
 also Chastity, Prostitution. 
 
 Simon, Tk, 43 note. 
 
 Sing Sing Prison, 7 f., 8 note, 17, 127, 
 146, 243, 325. 
 
 Socialists, 304 f., 379. 
 
 Standard deviation (a), as measure of 
 variability, 32; of the average as in- 
 dex of reservation, 34; of difference 
 in relation to observed difference, 34 
 f., 109, 112, 114; use of, in present 
 study instead of P. E., 35; device 
 for indicating, graphically, 35 ; use 
 of, in obtaining "standings," 56 f. 
 
 Stanford-Binet Scale, use of, as meas- 
 ure of mental capacity, 39, 42, 43, 47 
 f. ; correlation of, with judgments of 
 mental capacity, 56; correlation of, 
 with other tests, 58; comparison of 
 mental capacity of groups on basis 
 of, 416 ff., 447 ff., 452 f., 483 f., 496. 
 
 State Prison for Women at Auburn, 
 N. Y., passim. See especially 15, 
 17 f., 55, 63, 70, 72, 86 ff., 107 ff., 
 110, 112 ff., 121 ff., 128 ff., 135 ff., 
 139 ff., 143, 148 ff., 152 f., 157 ff., 168 
 ff., 208 ff., 230 ff., 253, 254 ff., 258 f., 
 2M ff., 271, 273, 275, 291 ff., 307, 313 
 ff., 330 f., 335 ff., 363, 368 f., 381 f., 
 
 384 ff., 408 f., 421, 429, 437 f., 440 ff., 
 450 ff. 
 
 State prisons. See New York State 
 prisons. 
 
 State reformatories. See Albion, Bed- 
 ford Hills, Elmira. 
 
 "Statement of Girl," description and 
 illustrative copy of folder for re- 
 cording, 64 ff. 
 
 Statistical reservations, discussed, 34 f. 
 
 Stenquist, I. L., 45. 
 
 Stores, work in, 312 ff. 
 
 Sydenstricker, Edgar, 359 f. 
 
 Sylvester, R. H., 45. 
 
 Syphilis. See Venereal disease. 
 
 Terman, L. M., 41, 43, 417 ff., 424. 
 
 Test Aggregate, obtained by combin- 
 ing other series of tests, 58; restric- 
 tion of use of, to intercomparisons 
 within present group, 58, 415; re- 
 striction of use of, to English-speak- 
 ing cases, 58 f . ; relationship of men- 
 tal capacity as measured by, to other 
 factors, 437 ff., 460 ff., 485 ff. 
 
 Tests, factors affecting choice of, for 
 measuring mental capacity, 38 f . ; pre- 
 liminary series of, 41 ff. ; selection 
 of, for present study, 46 ff., 55 ff. ; 
 reliability of, 52 ff. See also Educa- 
 tional tests. 
 
 Thompson, H. B., 45. 
 
 Thorndike, E. L., 46, 52, 259, 260 
 note, 261. 
 
 Tobacco, use of, 147, 153, 157 ff. 
 
 Town, C. H., 43 note. 
 
 Trabue, M. R., 44. 
 
 Trabue completion test (Scale A), 51, 
 56, 291, 294 f. 
 
 Tredgold, A. P., 413 note. 
 
 U 
 
 United States, years in, and number of 
 convictions, 190 f . ; age at coming to, 
 and number of convictions, 190, 192 
 ff. 
 
 Variability, importance of measuring, 
 of groups. See also Standard devia- 
 tion. 
 
 Vaudeville performers, 312 ff. 
 
 Venereal disease, 152, 241 f., 408 ff., 
 516, 518 f. 
 
 Verified History blank, 76 ff. 
 
 W 
 
 Wage, difficulties in study of, 328, 330 ; 
 in first job, 330 ff., 343 ff., 507; in 
 latest job, 335 f., 343 ff. ; at first con- 
 viction, 337 f., 343 ff.; at present 
 
542 
 
 INDEX 
 
 conviction, 339 f., 343 fif. ; average, 
 for prevailing work, 341 S. ; average, 
 in relation to other factors, 348 ff., 
 504 flf. ; comparison of average, 
 of delinquent women with that 
 of women of general population, 
 3S8 ff. 
 
 Wassermann test, 408 ff., 412, S22. 
 
 Weidensall, Jean, 16, 42, 415. 
 
 Wells, P. L., 44, 45, 49, SO, 51. 
 
 Western House of Refuge, Albion, 
 N. Y. See Albion. 
 
 Whipple, G. M., 44, 45. 
 
 Williams, J. H., 416. 
 
 Women, present study concerned with, 
 delinquents, 5 ff. ; comparison of, de- 
 linquents with, of general popula- 
 tion, 5 ff., 171 ff., 259 ff., 288, 290 
 ff., 320 ff., 358 ff., 413 ff. ; comparison 
 of number of men with number of, 
 committed to state prisons" in New 
 York, 7 ff., 17; comparison of, de- 
 linquents with men delinquents, 7 ff., 
 100 ff., 127, 129 ff., 159, 449 ff.; 
 groups of delinquent, studied, 13 ff. 
 See also Delinquents. 
 
 Woodworth, R. S., 44, 45, SO, 56. 
 
 Woody, C, 51. 
 
 Woolley, Helen T., 42 f., 48 f., 57, 415. 
 
 WooUey series of tests, 39, 48 ff., 432 f. 
 
 Work, age at starting, 30i6 ff. ; classi- 
 fication of kinds of, 311 f.; kind of. 
 
 in first job, 313 f., 325, 327; kind of, 
 in latest job, 315 f., 325, 327; kind 
 of, at first conviction, 316 f._, 325, 
 327; kind of, at present conviction, 
 317 f ., 325, 327 ; prevailing, of women 
 delinquents, 318 ff., 324 ff. ; compari- 
 son of prevailing, of women delin- 
 quents with that of women of gen- 
 eral population, 320 ff. ; relationship 
 between kind of prevailing, and 
 grade finished, 326 ff. ; average wage 
 for prevailing, 341 ff. ; regularity of, 
 361 ff. ; total estimate of, record, 
 364 ff. ; summary regarding, records, 
 377 ff., 526; mental capacity and kind 
 of, 500 ff. ; mental capacity and esti- 
 mates of, record, 507 ff. See also 
 Wage. 
 Workhouse. See New York City 
 Workhouse. 
 
 Yerkes, R. M., 43, 424 
 Yerkes-Bridges Point 
 measure of mental 
 47 f . ; correlation of, 
 of mental capacity, 
 of, with other tests, 
 of mental capacity 
 basis of, 415, 424 ff. 
 482 ff. 
 
 Scale, use of, as 
 capacity, 42, 43, 
 , with judgments 
 56 ; correlation 
 58; comparison 
 of groups on 
 450 ff., 455 ff..