The Relationship between sistence in School and Home Conditions THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY IN EDUCATION IN THE GRADUATE SCHOOL OF THE UNIVERSITY OF ILLINOIS I915 BY CHARLES ELMER HOLLEY A.B. University of Illinois, 1912 A.M. University of Illinois, 1913 Reprinted from The Fifteenth Yearbook of the National Society FOR the Study of Education 1916 J The Relationship between Per- sistence in School and Home Conditions THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY IN EDUCATION IN THE GRADUATE SCHOOL OF THE UNIVERSITY OF ILLINOIS I915 BY CHARLES ELMER HOLLEY A.B. University of Illinois, igi2 A.M. University of Illinois, 1913 Reprinted from The Fifteenth Yearbook of the National Society FOR the Study of Education 1916 Copyright 1916 By Guy M. Whipple All Rights Reserved Published April 1916 «$,S: Composed and Printed By The University of Chicago Press Chicago, Illinois, U.S.A. THE RELATIONSHIP BETWEEN PERSISTENCE IN SCHOOL AND HOME CONDITIONS^ CHARLES ELMER HOLLEY Ohio Wesleyan University PART I INTRODUCTORY STATEMENT THE PROBLEM This study is concerned primarily with the qualitative analysis of the relationships which exist between the schooling of children and their home conditions. It is concerned secondarily with a rough determina- tion of the relative importance of the hereditary and the environmental factors involved in these relationships. ORIGIN AND DEVELOPMENT OF THE STUDY The study is an outgrowth of a social survey of the Decatur, Illinois, high school made by the writer during the school year of 191 2-13. In making this survey a large amount of data was secured, most of which proved to be of relatively little importance, but among the many facts there were a few which suggested family tendencies in the matter of educating children. Some of the famiHes gave all the older children a high-school education, while other families, of similar size and age- composition, did not have one child who had completed the high-school work. All the families having two or more children no longer in the pubHc school were selected and examined. There proved to be 198 such families, containing 642 older children, 334 of whom had secured a high- school education. A further examination showed that 40 per cent of the 198 families furnished 72 per cent of those who had finished the high school, and 30 per cent of the famiHes furnished 57 per cent of those who ^ This study was accepted as the dissertation for the doctorate of philosophy in education by the Graduate School of the University of Illinois. The writer wishes to acknowledge his indebtedness for counsel and suggestions given by Dr. W. C. Bagley and Dr. L. D. Coffman. Further, many useful suggestions were received from Dr. G. M. Whipple, Dr. C. H. Johnston, and the graduate students in education. 9 lO THE FIFTEENTH YEARBOOK had not finished the high school. This difference suggested that there must be corresponding differences in the homes which might be ascer- tained. Data were secured and it was found that these two groups of homes differed markedly with respect to economic, educational, and social conditions. Three years ago Dr. J. K. Van Denburg published the results of an investigation conducted in the New York City schools. He found that "on the whole, the economic status of these pupils (so far as it is shown by monthly rental) seems to be only a slight factor in the determination of length of stay in the high schools. The one most marked influence seems to be that the superior economic status in girls leads to a longer stay in spite of failure to progress at the 'normal' rate."^ At another place Dr. Van Denburg shows^ (Table I) the percentages of the different rental groups^ who graduated from the high school which TABLE I Percentage Graduating, Classified according to Rental Groups Amount Graduates Total Entering Percentage Graduating Boys Not specified $ 8 to $17 22 9 8 4 40 14 10 4 76 34 48 II. 8 $18 to $27 $28 and up 23s 8-3 Girls Not specified $ 8 to $17 99 71 6S 14. 1 $18 to $27 14.0 $28 and up 6.1 they entered four years earlier. He, however, has no record of those who left the pubhc schools and went to private schools, a group mentioned as a factor of some importance. Hence the group " 28 and up," would ' Causes of the Elimination of Pupils in Public Secondary Schools (New York : Published by Teachers College, 19 12), p. 113. = Ibid., p. 134. 3 A rental group is a group of families which paid specified amounts of rent per month. All the families selected were divided by Van Denburg into three rental groups: (i) those paying $8 to $17 per month, (2) those paying $18 to $27 per month, and (3) those paying $28 or more per month. PERSISTENCE IN SCHOOL AND HOME CONDITIONS ii have to be augmented by an unknown quantity to represent the true percentage of those who received the equivalent of four years in the pubHc high school. It is conceivable that this unknown quantity would be large enough to show a definite relationship for the boys between economic status and persistence in school. With the girls the case would not be so clear, for the two smaller groups contain the same percentage of graduates. It may be that the economic factor is of less importance with girls than with boys. To be conservative, it might be said that the economic status of the families in Dr. Van Denburg's study is not of sufficient importance to overshadow or more than counteract other factors which make for per- sistence in, or ehmination from, the public high schools of New York City. He has shown that the presence or absence of younger children in the family, the nationality of the parents, choice or lack of choice of an occupation, and intention with regard to graduation are factors corre- lated with the length of stay in the high school. A more detailed study of home conditions might reveal other factors of far greater influence in this city than economic status. In another study^ Dr. C. H. Keyes showed that acceleration or retar- dation were characteristic of certain families. He found that 6.8 per cent of the families produced 24 per cent of the accelerates, while 7.7 per cent of the famihes produced 24.5 per cent of the arrests. These facts obtained in a New England city tend to support those obtained in Decatur. The apparent disagreement between the conditions found by Dr. Van Denburg in New York City and those found by the writer in Decatur, Illinois, raised the question: "Is Decatur representative qualitatively of the average middle western city?" With this question in mind it was decided to extend the study to other Illinois cities, and information was collected from the high schools of Centralia, Champaign, Gibson City, and Rochelle, Illinois. While these data were being collected, it occurred to the writer that this study dealt with a special class — those whose chil- dren reached the high school — and represented a special situation, and hence that it ought to be extended so as to include statistics from all levels of society. Accordingly the families residing in Urbana who had children between the ages of fourteen and twenty-one were selected, and ' C. H. Keyes, Progress through the Grades of City Schools (New York: Published by Teachers College, 1911). 12 THE FIFTEENTH YEARBOOK a personal canvass was made by the writer which furnished a mass of facts from 234 homes. When these data had been tabulated and evalu- ated, and an interpretation was attempted, it was found that, although important relationships existed between the amounts of schooling that the children received and certain objective home conditions, it was impos- sible to distinguish between environmental and hereditary factors, a dis- tinction that is very important from social and educational points of view. In order more accurately to determine the relative importance of these two types of factors it was decided to secure similar facts about the education and home conditions of adopted children. In outline this presents the origin and development of the study. The presentation of the data will follow the same general order. THE DATA Sources. — The facts presented in Part II were secured from the high- school pupils of Decatur, Illinois, during the fall of 191 2. Those in Part III were collected from the high-school pupils of Centraha, Cham- paign, Gibson City, and Rochelle, Illinois, during the fall of 1913. The main data, those in Part IV, were gathered directly from the homes and from the courthouse records in Urbana, Illinois, during the summer and fall of 1914. The information about the adopted children, given in Part V, was secured from the Urbana courthouse records and from various individuals who resided in Champaign and Urbana during the early months of 1915. Method of collecting. — The original data which uncovered the problem were secured from the high-school pupils of Decatur during the fall of 191 2. One morning in November the writer called at the school with a supply of blanks asking the following questions, as well as a number of others which had no bearing on the present problem: Name Sex Age Country of your mother's birthplace Country of your father's birthplace What language is commonly spoken in your home ? OLDER BROTHERS No. Age Has he finished What is he doing now ? high school? I 2 3 4 5 PERSISTENCE IN SCHOOL AND HOME CONDITIONS 13 OLDER SISTEES No. Age Has she finished What is she doing now ? high school ? The teachers were instructed briefly as to the facts desired and the collection of data was then left in their hands. The first period of the morning was used and each of the pupils attending at that time was required to fill out one of the blanks. Through the assistance given by the room-charge teachers the entire high school furnished the desired information in a short time. After it was discovered that one group of homes educated its children more than the other group, it was thought that an objective description of these homes might be secured from the children who attended high school. For this purpose a blank was prepared asking for the following data: a) Father's occupation b) Father's education mother's education c) What is the family income ? d) What rent does the family pay per month (estimated by the kind of house in which they live) ? e) Church affiliation of father of mother /■) What newspapers does the family take ? What magazines ? g) What is the size of the family library ? h) What clubs or organizations does the father attend ? The mother ? These blanks were given to the pupils from the selected homes and were filled out in conference with the teachers or principal. The results were later checked up by the principal, and reports containing obvious errors were marked so that the erroneous portions could be eliminated. As stated earlier, the facts reported in Part III were secured from the high-school pupils of Centralia, Champaign, Gibson City, and Rochelle, lUinois. A blank asking for the following information was used. 14 THE FIFTEENTH YEARBOOK Name Sex 1. Country of mother's birth 2. Country of father's birth 3. Father's occupation 4. Father's education (in years of schooling) 5. Miother's education (in years of schooling) 6. What monthly rent do your parents pay for the house in which they live ? (If they own their home, estimate the rent by comparing with rented houses in the neighborhood.) 7. How many volumes in your home library ? OLDER BROTHERS OLDER SISTERS No. Age Education in years No. Age Education in years of schooling of schooling 2 2. 3 3- 4 4. 5 5. Copies of this were sent to the principals or superintendents of Cen- traha, Gibson City, and Rochelle, and they secured the information from the pupils as best they could. In Gibson City this method resulted in returns from all the pupils attending on the day the information was secured. In Centralia and Rochelle less pressure was put upon the pupils and some failed to furnish any information. In Champaign the writer gathered the data during the Enghsh class periods, personally directing the work of the pupils. By answering any queries which arose because of a misunderstanding of any of the questions and by suggesting ways of estimating some of the items, he secured careful replies from almost all the pupils. They were told that it was not necessary for them to sign their names. Hence it was easy to meet any objections which a pupil might have to answering personal questions, and all the pupils filled out the blanks. In the other three towns the pupils signed the blanks, a fact which made them a little more reserved in their replies. The information which forms the basis of Part IV was secured through a personal canvass made by the writer during June and July, 1914, in Urbana. The university-community portion of the town is a students' residence district and education is a thing uppermost in the minds of those who Uve there. It contains many families who have moved to Urbana to educate their children. Because of this emphasis on education and because of the difficulty of gauging an economic index where there are so many temporary residents, all families who lived west PERSISTENCE IN SCHOOL AND HOME CONDITIONS 15 of Coler Street and south of Springfield Avenue were eliminated from consideration. The families of the university faculty who lived outside of this area were also eliminated. The preliminary list of names was secured from the 19 13 school census records, which gave every home containing an individual under twenty-one years of age. The list finally selected was restricted to those homes which included individuals four- teen to twenty-one years of age, and contained about 550 names. When the actual canvass was made, it was found that a few of these homes contained no children over fourteen (roomers under twenty-one years of age having been found by the school census taker and recorded) and that a few of the listed families had moved out of town. These two factors reduced the list of possible calls to slightly less than 500. The writer called at the homes on all the east and west streets (most homes in Urbana face these streets). Sometimes no one was at home. When convenient a second or even a third call was made to secure the desired information. The canvass resulted in securing information from 234 homes of whites and 5 homes of colored people and gave a random sam- pling of the community. The colored homes are not included in the study because their members belong to a race which is not as yet a homo- geneous element of the population. Their number was too small to be studied separately. As an aid and guide in securing the information the following blank was used: 1. Occupation of father 2. Country of father's birth of mother's birth 3. Father's native language mother's native language 4. Education of father of mother 5. Nmnber of books in the home 6. Number of living-rooms in home 7. Number of people living in house over fourteen years of age Under fourteen years of age 8. Number of members of family living at home 9. Rent per month 10. Children above fourteen years of age Sex Age Years of schooling each has received I 2 3 4 5 6 7 i6 THE FIFTEENTH YEARBOOK In conducting the canvass, the writer, after introducing himself, usually began with an inquiry as to the number of children in the home, their age, and education. Experience showed that parents were quite ready to talk about their children and that, after getting somewhat acquainted with the writer, they were then more free in answering the other questions. By this procedure the facts were secured to question No. lo first and then the blank was filled out in order, beginning with question No. i. The figures for the personal property and real estate assessments were taken from the courthouse records giving the assessments for the 19 15 taxes. In case a name did not appear here, the previous year's records were examined. In a few cases the figures were obtained in the latter way. The data which furnish the basis for the discussion of adopted chil- dren, presented in Part V, were gathered by the writer through a personal canvass. The original list of names was secured from the court records which gave the adoptions made in Champaign County since 187 1 . From these records the sex, date of birth, date of adoption, names of foster- parents with their town addresses, the changed name of the child, and cause of adoption were secured for each child. Excluding all children who would not now be at least fourteen years old, the Hst contained 155 cases of adoption. The present addresses of as many as possible of these foster-parents, of the children, or of someone who could give the desired information were secured from directories and from people who have long resided in Champaign or Urbana. That the results might be com- parable with those presented in Part IV, only those parents who Hved in Champaign or Urbana and reared the children there were included in the study. In securing these data a form quite similar to that used in the earher canvass was employed. It was as follows : Parents' names 1. Occupation of father 2. Nativity of father of mother 3. Schooling of father (in years) of mother 4. Estimated number of books in home 5. Financial status of parents: very poor, poor, average, well-to-do, wealthy (check). 6. Estimated rent of home in which family lived when children were in school 7. Facts about all children living or dead, who reached fourteen years of age PERSISTENCE IN SCHOOL AND HOME CONDITIONS 17 Date of birth Sex Schooling in years I 2 3 4 S 6 7 8 The procedure was approximately the same, after the Hst of names and addresses was secured, as that followed in gathering the data for Part IV. Members of the family or relatives furnished the information for all but one of the children studied. Errors. — The data secured from the pupils through questionnaires which they themselves filled out were probably more inaccurate than those secured by the writer through the personal canvass. The greatest constant error is that of omission. It is thought by the writer that the effect of this is nearly that of pure chance, though this may be proved otherwise if carefully investigated. However, since this is primarily a qualitative study, such errors will be less serious than if it were a purely quantitative investigation. Wilful untruths may have existed in the data, but they were very rare. From the nature of the questions and the conditions under which they were answered, some of the data are estimates, more or less inaccurate. Errors pecuHar to one kind of data will be mentioned during its discussion. Method of treatment. — The statistical method^ will be used in this study. All the important relationships will be expressed through coeffi- cients of correlation. All correlations will be worked according to the "product-moment" method of Pearson where r= — . The reliability n(Ti(r2 of all correlations will be expressed according to the formula P.E.= i—r^ o . 6745 ,- . The reliabihty of the difference between two medians wiU \P E" P.E^ be expressed according to the formula P.E.D. = -^ --\ . All cen- ^ til ^2 tral tendencies will be expressed by medians. ' All the formulas used can be found in any standard work on statistical methods. See Thorndike, Mental and Social Measurements; or Whipple, Manual of Mental and Physical Tests, 2d ed., Part I, "Simpler Processes." Whipple gives on p. 35 a table showing the reliability of P.E. according to its relative size. PART II RELATIONSHIPS FOUND IN DECATUR The original data collected in Decatur during the fall of 191 2 revealed 198 children from homes having two or more older children no longer in the public school. These homes when examined could be distributed readily among three groups: (I) those from which all the older children had completed the high-school work; (II) those from which none of the older children had completed the high-school work; (III) those in which some of the older children had graduated from the high school and others had not. In all there were 642 older brothers and sisters, 334 of whom had secured a high-school education. Group I contained 78 families and furnished 72 per cent of the 334 children. Group II contained 59 families and furnished 57 per cent of the 308 who had not finished high school. This section will be devoted to a discussion of the dififerences between home conditions in the first two groups. The replies were most nearly complete with respect to the education of the parents, though a few children failed to give this information. When the replies were checked, it was found that some information was secured concerning 60 homes of Group I and 43 homes of Group II. On some of the blanks there was very Httle information, probably because the pupils, or even the parents in some cases, could not give the facts desired. RESULTS OF THE INVESTIGATION The differences between the two tj^pes of homes are striking. a) Occupations. — The fathers of Group I (the families that gave their children a high-school education) are chiefly engaged in professional and commercial occupations (see Table II). The fathers of Group II (the families that did not provide a high-school education for their children) are chiefly engaged in artisan trades, and in semi-skilled and unskilled occupations (Table II). b) Schooling. — The median number of years of schooHng received by the parents of Group I is twelve; by the parents of Group 11, eight PERSISTENCE IN SCHOOL AND HOME CONDITIONS 19 (see Table III). In Group I, 60 per cent of the fathers and 61 per cent of the mothers have had the equivalent of a high-school education, TABLE II OF Fathers GROUP n Occupation No. Fanner 6 Retired farmer 3 Carpenter 3 Minister 3 Blacksmith 3 Cabinet-maker 2 Night watchman 2 Janitor 2 Railroad engineer Railroad conductor Mail clerk Shoeman Lock-maker Occupations GROUP I Occupation No. Farmer 8 Lawyer 4 Insurance 4 Real estate dealer 3 Retired farmer 2 Physician 2 Public official 2 Jeweler 2 Cashier 2 Minister 2 Implement dealer i Druggist I Millwright i Business i Painter and decorator i Floor-walker i Nurseryman i Mason i Railroader i Music store i Brick business i Bookkeeper i Auto trimmer i Proprietor, machine-shop i Hotel-keeper i Machinist i Cement factory i Carpenter i Secretary and treasurer i Barber i Fumaceman i Railroad engineer i while more than 91 per cent of the fathers and mothers of Group II have had less than four years of high-school work. Indeed, 74 per cent of I I I I I Factory employee i Boiler-maker i Clothier i Gardener i Cement contractor i Conunission dealer i Horse-dealer i Grocer i Miller i Clerk I Passenger engine inspector i 20 THE FIFTEENTH YEARBOOK the fathers and 71 per cent of the mothers of Group II did not go beyond the eighth grade. The mathematical differences between the medians of the two groups, 3.68=^0.38 years for fathers and 3.70=^0.38 years for mothers, have a high degree of reUabihty. TABLE III The Education of Fathers and Mothers Number of Years Group I Group II OF Schooling Fathers Mothers Fathers Mothers 2 I •2 . A. .... C I 6 2 2 II 7 I 24 3 21 8 12 I 6 10 s 4 2 I 4 II 3 2 12 15 4 5 I 3 2 12.33 years 16 2 II IT. 14. 2 I le 16 I 18 Median years of education .... 1 2. 34 years 8 . 65 years 8 . 64 years Difference between median education of Groups I and II, fathers = 3 . 68 ± o . 38 years Difference between median education of Groups I and II, mothers = 3 . 70 ± o . 38 years c) Incomes and rent. — As would readily be inferred from the facts concerning occupation and schooling just presented, the yearly incomes and monthly rentals are higher with those who sent their children through the high school than with the other group. The median yearly income of Group I is $2,000; of Group II, $1,350 (Table IV). Each family studied in this section contained at least three children, and the average is almost five. Thus it seems that the problem of furnish- ing the necessaries of life must be a serious one for many famiUes of Group II. PERSISTENCE IN SCHOOL AND HOME CONDITIONS 21 The differences between the rental values of the two groups of homes are evident to one who simply glances at Table V. Statistically they are shown by the difference in the medians. They are marked, for 8i per TABLE IV Incomes* Below $699 $ 700 to $ 800 to 900 to 1,000 to 1,100 to 1,200 to 1,300 to 1,400 to 1,500 to 1,600 to 1,700 to 799 899 999 1,099 1,199 1,299 1,399 1,499 i,S99 1,699 1,799 Group I Group II 5i,8oo to $1,899 1,900 to 1,999 2,000 to 2,100 to 2,200 to 2,300 to 2,400 to 2,500 to 3,000 to 4,000 to 2,099 2,199 2,299 2,399 2,499 2,599 3,999 4,999 5,000 and above Median income . Group I 3 S 5 5 52,000 Group II $1,35° Difference between medians of Group I and II = $650 ±$242 them. * A number of families had such indefinite incomes that the parents themselves could not estimate cent of the families in Group I pay $25 or more a month while 77 per cent of Group II pay less than this amount. A house with modern improvements, bath, toilet, etc., large enough for a family of six costs TABLE V Rental Values of Homes* Per Month no. II. 12. 13- 14- IS- 16. 17- 18. 19. Group I Group II Per Month 520 22.50 25 30 35 40 50 Median rent No. who own their homes Group I I 10 6 4 S I ho 14 Group II 520.80 9 DijBferences between medians of Groups I and II=$9 . 20=*=$! . 17 *The question which asked for this information was poorly constructed. It was: "What rents does the family pay per month (estimated by the kind of a house in which they live)?" Some replied by merely stating that they owned the home. Others estimated the rent even if they owned the home. 22 THE FIFTEENTH YEARBOOK at least $25 a month in Decatur. Hence a large part of the famiUes of Group II live in somewhat undesirable houses. The number reported TABLE VI Newspapers Taken Group I Group II Decatur papers 77 15 8 ERSISTENCE IN SCHOOL AND HOME CONDITIONS 33 TABLE XXI Correlation between Education of Town Parents and Education of Their Daughters Years of Schooling Average Years of Schoolmg of Parents of Daughters S 6 7 8 g 10 II 12 13 14 IS 16 IQ I 18 17 X 2 3 I 7 I I I 4 I 2 2 I 4 5 2 3 2 I 4 16 I I I I I 3 4 8 I 2 2 2 I 2 I 2 15 I 14 2 2 3 I 4 5 4 4 I I I I 2 2 I la 4 I 2 2 8 I I 17 8 9 8 23 2 I 6 12 I 3 II 10 I 4 8 I I I I I I 7 6 q 4 I r =0.35=1=0.04 n =229 Median education of daughters, 12 years TABLE XXII Correlation between Education of Fathers and Education OF Their Sons Years of Schooling Years of Schooling of Fathers of Sons 3 4 S 6 7 8 9 10 11 12 13 14 IS 16 17 18 20 19 18 4 I 17 I I 3 2 2 2 I 16 4 S 6 2 28 12 24 18 48 6 I I I I 2 2 2 I 2 2 2 2 I 5 3 S I 2 ■ I 16 6 4 I 7 2 2 I I I I? I 14 13 I 5 12 2 I 3 I 6 2 3 4 I 3 5 3 2 I I 2 I II I I 2 I s I 10 Q I I I 8 I 2 7 6 s 4 I r =o.44=*:o.03 n =317 34 THE FIFTEENTH YEARBOOK f) Rent. — It may be rather unfair to combine the figures for the four towns, because rental values vary from town to town for approximately the same accommodations. Such variations tend to reduce the figures TABLE XXIII Correlation between Education of Mothers and Education OF Their Daughters Years of Schooling Years of Schooling of Mothers of Daughters 4 S 6 7 8 9 10 II 12 13 14 IS 16 in 1 l8 17 I 3 3 4 6 27 13 16 13 44 6 I 6 I 6 I 8 9 12 3 5 i6 2 2 3 I 2 1 I I I 3 ic I I I 3 I 2 14 2 I 2 I 3 I 3 I 5 I 2 2 12 1 2 3 4 I 2 4 2 2 1 1 2 13 12 I 3 11 10 I I 2 2 3 .... 1 8 4 2 2 I 7 6 c 4 1 r = o.43=to.o3 « = 300 of relationship obtained, though perhaps not as much as might be expected. There is probably a positive correlation between rental values and the opportunities for education offered by a community. If such be the case, it must counteract the effects of the variations. TABLE XXIV Owners and Renters Owners Renters Centralia 25 85 45 17 3 21 10 Champaign Gibson City Rochelle 7 Total 172 41 PERSISTENCE IN SCHOOL AND HOME CONDITIONS 35 Only 41 out of the 213 families which gave the information pay rent (Table XXIV). Since the pupils were requested to estimate the rental values of their homes when their parents owned them, most of the rental values are estimates. This fact introduces a certain amount of unreHabiHty into the data which would tend to reduce the correlation figures below their probable values. Even if such be the case, the corre- lation coefficients are large enough to indicate a clear relationship TABLE XXV Correlation of Rental Values* and Education of Sons Years of School- Rent of Home per Month, Dollars ing of Sons 10 IS 20 25 30 35 40 45 SO 55 60 6s 70 75 20 I TO .... 18 2 I 3 2 I 17 16 . . . . I I I I 2 I I I 2 2 f 2 I je I 14. . 4 I 2 I 2 f2 I 7 4 3 I 2 ■ ■ 12 S 3 6 7 13 4 8 7 3 4 13 4 9 5 4 3 IX I 6 2 2 4 9 I I 2 2 I I 4 I 3 4 I I II 10 . . 3 7 8 2 3 8 7 ... 6 I e A. I r =0.40= n =241 =0.04 * The rental values were grouped as follows: The $io group includes all living in homes worth $io or less per month, the $is group includes all values between $ii and $15, etc. (Tables, XXV, XXVI). The correlation between rental values and schoohng of sons is 0.40=^=0.04 and between rental values and schooling of daughters it is o. 24=1=0.04. These families were a select group from which those children who never reached high school had been eliminated. Where are those famiUes located in rental distribution whose children never went beyond the elementary school? An answer will be sug- gested by Part IV. 36 THE FIFTEENTH YEARBOOK g) Number of books in the home. — The pupils found it more difficult to estimate the number of books in the home than to estimate the rental TABLE XXVI Correlation of Rental Values and Education of Daughters Years of Schooling of Daughters Rent of Home per Month, Dollars 10 IS 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100 17 2 3 I 4 I 2 I I I il::.... I I 3 3 10 4 5 5 II 3 I 4 I 8 4 8 I 9 I I 4 9 2 5 2 4 2 I 4 7 I 2 4 S I I I I I ic 14 13 12 I I 7 I 4 6 I I II I I I I I 2 I 2 I I I I 2 I II I 10. . . i . . 5 I 10 I 9 8 4 I 7 6 c 4 I ■ r =0.24=1=0.04 n =219 values of the home. The best showing was made by Champaign, where the data were furnished by the pupils while imder the direct supervision of the writer (Table XXVII). Here the pupils were urged to estimate TABLE XXVII Number Who Estimated the Books in the Home Centralia 18 Champaign 108 Gibson City 46 Rochelle 42 and were told that a rough estimate was better than none. As an aid in estimating it was suggested that a sheK three feet long held about twenty-five ordinary books. Chance remarks dropped by some of the pupils later disclosed the fact that some who had many books in their homes made rather wild estimates. In every case reported to the writer. PERSISTENCE IN SCHOOL AND HOME CONDITIONS 37 TABLE XXVIII Correlation of Number of Books in the Home and the Schooling of Sons Number of Books in Home of Sons 10 25 5° 75 lOO ISO 200 250 300 400 500 600 700 I i8 3 I I I 2 I i6 I I I I 2 2 2 5 I 2 3 6 I 2 I 2 I I 2 I I I 2 2 I lO 3 6 3 13 3 I 12 2 2 S 5 7 2 I 2 5 I 5 3 13 6 8 3 .... 7 2 I 2 6 2 3 2 3 2 4 4 2 2 I 4 I S I I I 8 2 3 I 6 I r =0.39=^0.04 « =227 TABLE XXIX Correlation of Number of Books in Home and the Schooling of Daughters Years of Schooling of Daughters Number of Books in Home 10 25 so 75 100 150 200 250 300 400 500 600 700 I I 16 4 I I I I 5 I I I 3 I 4 I 5 5 2 2 3 I 2 4 2 5 2 I 3 I 2 I I 3 4 2 I 4 I 2 2 lA I 2 I 8 I 7 4 6 5 I .... I I I 2 4 9 4 3 2 9 2 I 2 5 2 12 12 3 3 2 I 2 3 4 I 2 10 I I I 8 I 7 6 I r =0.18 ±0.04 n =209 38 THE FIFTEENTH YEARBOOK however, the estimates were low, never high. Those who had few books in their homes made comparatively accurate estimates. The four towns were represented by 214 homes containing 227 sons and 209 daughters. The coefficient of correlation between the number of books in the home and the schooling of the sons is 0.39 =1=0. 04 (Table XXVIII), while the like relationship for the daughters is 0.18 ±0.04 (Table XXIX). SUMMARY AND CONCLUSIONS The coefficients of correlation presented in this section are summed up in Table XXX. TABLE XXX Correlated With Schooling of Sons Schooling of Daughters Average schooling of parents o.43=to.03 0.35 ±0.06 0.30*0.04 o.49=±=o.o3 0.42*0.03 0.47*0.07 0.35*0.04 Average schooling of farm parents .... Average schooling of town parents. . . . Schooling of father Schooling of mother 0.43*0.03 0.24*0.04 0.18*0.04 Rental values 0.40*0.04 0.39*0.04 Number of books in the home These statistics show in a general way the existence of definite rela- tionships between the home conditions of parents of high-school pupils and the amounts of schooling which the children receive. This part supports the general conclusions arrived at in the Decatur study. PART IV PERSISTENCE IN SCHOOL AND HOME CONDITIONS IN URBANA The data presented in Part IV were secured through the personal canvass made by the writer. Only the facts collected from the homes of whites, 234 in number, are used. Some of these homes had no chil- dren who had completed their education. Such homes will not be considered where relationships between schooling and various home conditions are presented. Where the facts are such that it makes no difference whether the children have completed their education or not, the entire group of 234 homes will be used. Any special selection of homes made will be mentioned when the facts are discussed. The method followed in securing the material presented in Part IV is open to the criticism that, since the canvasser knew what he was seek- ing, some of the items may have been more or less unconsciously weighted. Personally, the writer thinks that this criticism need not be taken seri- ously. Throughout the canvass the writer kept as scientific an attitude as possible and faithfully recorded all answers even though they failed to fit his preconceived ideas. As a means of observing this open- mindedness the facts given in Part IV were collected before those pre- sented in Part III had been evaluated. Urbana is composed of a rather homogeneous population. In the few homes which have foreign-born parents all speak the English lan- guage. Out of the total number of homes there were only five in which both parents were foreign born. These were people of German ancestry. Only 23 fathers and 8 mothers were born outside the United States (Table XXXI). A few of the parents born in this country came from homes in which only a foreign language was spoken (Table XXXII). SECTION I. SCHOOLING OF PARENTS AND CHILDREN The relationships existing between the education, as measured by years of schooling, of parents and children wiU be the theme of this sec- tion. In the main the data are approximations, estimates of all of the members of a family fourteen years of age or older given by some member of each family. The age fourteen was taken as the minimum because 39 40 THE FIFTEENTH YEARBOOK the compulsory education law operates until this age is reached, and those under fourteen have not legally completed their education. The local public-school system was used as a standard for comparison and all estimates were made by comparisons with it. An appreciable number of these people were educated in other schools — some in schools of other states. This fact introduces a small degree of unreliabiHty. The writer feels, however, that, if the true amounts of schooling of these indi- viduals could be ascertained, they would not vary from the amounts given here by more than a year or two, except in possibly five or ten TABLE XXXI TABLE XXXII Birthplace Of Fathers Mothers United States , Germany 211 9 4 44 3 2 I 33 226 6 I Ireland Sweden Scotland I Total foreign-bom . 8 Language Commonly Spoken by Parents Of Fathers Mothers English 219 12 I I I 223 10 I German Scotch Swedish Norwegian cases where it was impossible to do more than estimate roughly the edu- cation of the individuals concerned. Such cases were those of dead parents and families where the father had deserted the home. In nearly all cases where there was any doubt, the amount listed is probably an overestimation instead of an underestimation. It was more difficult to estimate the education of those who had never gone beyond the ele- mentary school. The educational level of a home, however, is probably a rather con- stant factor, changing but little after the parents have started to rear their children. RELATIONSHIPS BETWEEN PARENTS AS TO NUMBER OF YEARS OF SCHOOLING Fathers and mothers are much alike with reference to the number of years of schooling they have received. Mothers as a group are sHghtly less variable in the matter of education than fathers (Fig. i). The mode and the median fall at eight years for both mothers and fathers. The PERSISTENCE IN SCHOOL AND HOME CONDITIONS 41 last two years of the elementary school is where a large number of parents finished their schooling, probably because many of them were reared in the country, and rural schools did not extend beyond the eighth grade. Since the high school constitutes another division of the school, we again o 2 4 6 8 10 12 14 16 lb 20 Fig. I. — Education of Urbana Fathers and Mothers: Years of Schooling find, what common-sense has akeady taught us, that the end of the high school was also a stopping-place for a large number. Only a small num- ber of people went to a college or university. This is somewhat sur- prising, until an explanation is sought, for Urbana has been the seat of 42 THE FIFTEENTH YEARBOOK the state university since its foundation in 1869. When it is remembered that university work until quite recently did little except prepare for the professions, this scarcity of college people seems more natural. Further, TABLE XXXIII COEEELATION BETWEEN EDUCATION OF FATHERS AND EDUCATION OF MOTHERS Years of Schooling Years of Schooling of Mothers of Fathers 1 2 3 4 5 6 7 8 9 10 II 12 13 14 IS 20 I IQ I 18 I 17 I 16 3 I 3 I 12 I s 3 3 I I IC I I I I 6 7 3 32 8 2 I 2 I I I I 14 la 12 I 5 3 4 5 2 2 10 I 9 I 2 I 2 2 I 3 II 10 I I I IS 7 2 I I 3 9 2 8 7 2 4 I 2 3 2 6 c 4 3 I I I 2 I 2 I I I I I r ==o.6s=±=o.o3 n =231 Median education of fathers and mothers, both 8 years many of these professional people have been eliminated through the rejection of data from the university residence district. The correla- tion' between the schooling of the father and the schooHng of the mother is high, being 0.65^0.03 (Table XXXIII). ^ It might be well to explain, at this point, what is meant by a coefficient of corre- lation. Coefficients of correlation are measures of resemblance between quantities found coexisting under varying conditions. There may be complete correspondence, +1 . 00 (the + sign is omitted in this study), or the exact opposite, — i . 00. Usually, however, the measures secured contain chance errors and a correlation of i.oo, positive (or negative), is almost never obtained. A coefficient of 0.60 or more, in this study, indicates a high degree of correspondence and becomes quite significant. PERSISTENCE IN SCHOOL AND HOME CONDITIONS 43 RELATIONSHIPS BETWEEN PARENTS AND CHILDREN I. Fathers and sons. — ^The curve (Fig. 3) of this relationship looks as if some factor such as the compulsory education law had modified its general character. At any rate, the coefficient of correlation is low, being 0.47^0.03 (Table XXXIV). r6 14 o 2 4 6 8 10 12 14 Fig. 2. — Correlation between Education of Fathers and Mothers 16 2. Mothers and daughters. — This relationship is much higher than that between fathers and sons and the curve (Fig. 4) lacks the flattened appearance at the lower end which characterizes the other. This may be due to the tendency of girls to stay in school longer than boys, or it may be a mere chance variation. The coefficient of correlation is o . 60 =<= 44 i6 14 THE FIFTEENTH YEARBOOK y^^^f-:^, 2 4 6 8 10 12 14 16 Fig. 3. — Correlation between Education of Fathers and Sons TABLE XXXIV Correlation between Education of Fathers and Education of Sons Years of Schooling of Sons Years of Schooling of Fathers I 2 3 4 s 6 7 8 Q 10 11 12 13 14 IS 16 17 18 19 20 18 I I I 17 I I 16 2 I I IK 2 I I 5 2 3 14 2 2 5 s 4 4 10 7 2 I I 2 I i-z 3 2 I 2 I 12 I 4 3 5 4 14 3 I I I I II I I 2 2 2 2 I 4 14 6 6 4 10 I I I I 9 I 2 2 2 2 I 2 I 3 I I I I S I 8 2 I I I I 3 I 7 5 6 5 2 2 I I 3 4 I 2 3 r =0. 47=1=0. 03 n =224 Median education of sons, 8 years PERSISTENCE IN SCHOOL AND HOME CONDITIONS 45 0.03 (Table XXXV). The daughter who is indicated as illiterate was an epileptic, unable to attend school. TABLE XXXV Correlation between Education of Mothers A>rD Education of Daughters Years of Schooling of Mothers of Daughters - 2 3 4 5 6 7 8 9 10 II 12 13 10 I 18 17 . 16 I 4 2 3 II 2 2 TC ... 1± . . I I 2 2 I 3 4 I 4 II I I I 2 3 IZ I II S II 4 17 10 3 12 I 4 2 3 4 6 6 3 3 7 2 3 2 2 II 2 I 10 S 8 3 3 2 2 I I 2 3 8 9 2 S 2 7 6 3 2 c A. I I I ... I f =0.60=^0.03 tt =234 Median education of daughters, 9 years 14 , / \ / , / J 1/ 2 4 6 8 10 12 14 Fig. 4. — Correlation between Education of Mothers and Daughters 46 THE FIFTEENTH YEARBOOK 3. Fathers and daughters. — This relationship is higher than that between fathers and sons and lower than that between mothers and daughters. The difference is so little in either case that it cannot legitimately be made the basis of any conclusion. The coefficient of correlation is 0.56=^=0.03 (Table XXXVI). TABLE XXXVI Correlation between Education of Daughters and Education of Fathers Years of Schooling Years of Schooling of Fathers of Daugh- ters I 2 3 4 s 6 7 8 Q 10 II 12 13 14 IS 16 17 18 19 20 19 I 18 17 16 I 4 I I 2 I 3 4 2 2 S I 2 I 2 2 4 2 I 15 I 14 I s I I I I 8 I 2 I 2 13 I I 2 I 12 3 8 2 8 6 2 2 I 2 I I 12 2 I 4 2 II 10 4 3 3 2 4 3 12 10 3 I I II I I 10 9 I I 8 3 3 I 2 I 3 2 I I 2 I 7 3 6 5 I 4 3 I I 2 I I r=o. 56=1=0. 03 « = 23I 4. Mothers and sons. — This relationship is almost the same as the preceding, the coefficient of correlation being 0.55=^=0.03 (Table XXXVII). 5. Parental average and children. — When the average schooling of each family is correlated with the schooling of the children, a closer relation- ship is revealed. The coefficients of correlation are o. 65 ± o . 03 for the sons (Table XXXVIII) and 0.62=1=0.03 for the daughters (Table XXXIX), a rather high degree of correspondence. PERSISTENCE IN SCHOOL AND HOME CONDITIONS 47 TABLE XXXVII Correlation between Education of Sons and Education of Mothers Years of Schooling Years of Schooling of Mothers of Sons o I 2 3 4 5 6 7 8 9 ID II 12 13 14 15 i8 I I I 17 I i6 I 3 I 3 2 4 4 5 S i8 lO 3 I 2 IC I I 14, 2 2 5 4 2 I 4 12 I 2 I 4 S 4 I I 5 2 2 2 2 I I I I 12 I I I 3 5 4 3 I I 2 5 II 9 3 2 I 2 II lO I I I I 2 3 6 4 3 5 8 I 7 6 4 I c 4 I 2 r =o.5S±o.o3 n =214 TABLE XXXVIII Correlation between Education of Sons and Average Education of Parents Years of School- Average Years of Schooling of Parents ing of Sons I 2 3 4 s 6 7 8 9 10 II 12 13 14 18 I I 2 I I I I 17 I 2 16 I I IS 2 14 2 2 2 3 3 3 13 6 2 I I 2 2 I 3 2 3 I 2 12 I 4 3 2 2 I I I I 2 5 2 I I 2 2 I I 12 2 I I 2 4 4 2 I I 2 2 13 6 4 5 I 2 3 7 8 13 3 I II I 10 I I I I 4 I I 4 2 I I 8 7 6 4 I I 2 I 5 4 .... 2 = o.6s±o.o3 = 220 48 THE FIFTEENTH YEARBOOK 6. Sons and better-educated parent. — When the relationship which existed between the better-educated parent of each family and the sons in the matter of schooling was evaluated, it furnished a correlation coeflS- cient of 0.60 ±0.03 (Table XL). TABLE XXXIX coreelation between education of daughters and average Education of Parents Years of Schooling of Daughters Average Years of Schooling of Parents I 2 3 4 s 6 7 8 9 10 II 12 13 14 19 I 18 17 16 I 3 2 2 2 s 5 I 2 IC 14 2 3 7 3 la I 2 2 I 2 II 2 4 I 3 I I I 10 I I I 12 2 I 4 3 2 4 5 II 8 2 4 3 8 4 9 7 I I II I 10 9 I 2 2 8 2 I 2 3 3 3 2 2 4 14 12 3 7 6 3 2 5 4 3 ^ I 2 I I r =o.62='=o.o3 n =232 7. 5ow5 awi more poorly educated parent. This relationship proved to be nearly the same as the preceding, being slightly lower, o. 57=1=0. 03 (Table XLI). Comparisons. — All the sons and daughters who have been given in the data thus far presented in this section were reported as having com- pleted their education. A few, perhaps, may reconsider their decisions and continue thek schooling later. On the other hand, the parents passed the customary ages for school attendance long ago. Hence, when the amounts of schooling which the children have received are compared PERSISTENCE IN SCHOOL AND HOME CONDITIONS 49 TABLE XL Correlation between Education of Sons and Education of Better- Educated Parent Years of Schooling Years of Schooling of Better-Educated Parent of Sons 2 3 4 S 6 7 8 9 lO ir 12 13 14 IS i6 17 i8 19 20 i8 ■ I 17 I 2 I i6 I I I le I I I s 3 2 I 3 3 I I 14. 2 I 3 2 4 6 14 lO 3 I I I I I I 3 8 3 2 I 4 2 I I 3 I I 3 3 2 I I I I 12 .... 2 I 2 4 lO lO 3 I I 2 I I II I 2 5 6 5 3 lO 2 I 3 I I I 2 T 8 1 6 4 I 2 2 I e A I r =o.6o='=o.o3 n =2i6 TABLE XLI Correlation between Education of Sons and Education of More Poorly Educated Parent Years of Schoolbg of Sons Years of Schooling of More Poorly Educated Parent o I 2 3 4 s 6 7 8 9 lO II 12 13 14 i8 2 II 2 I 2 2 5 3 2 17 I 3 I 3 3 6 7 6 3 13 6 2 I i6 IC I 14 12 2 2 2 2 3 I I 2 4 I 12 I I I I I 3 6 2 2 5 2 2 2 I I I I 3 12 6 5 3 2 I 6 6 II 2 I I II lO I 3 2 2 2 I I 8 I 2 7 6 5 3 I e A I 2 2 r =0.57=1=0.03 n =216 5° THE FIFTEENTH YEARBOOK with the amounts received by their parents a generation earher, an incomplete quantity is being compared with a complete one. I. Amounts of education received by fathers and sons: The fathers have received almost as much schoohng as their sons. The difference MEDIAN OF FATHERS — 8.45 MEDIAN OF SONS 8.60 DIFFERENCE .15±.19 EDUCATION OF FATHERS' EDUCATION OF SONS PART COVERED BY BOTH CURVES,#^ YRS 2 46 8 10 12 14 16 18 20 Fig. 5. — Education of Fathers and Sons: Years of Schooling between the medians, 8 . 45 years for the fathers and 8 . 60 years for the sons, is only 0.15=^0.19 year (Fig. 5). When these comparative sur- faces of frequency are examined, it is seen that a few more fathers are at the lower end and a few more sons at the upper end. When the char- acter of the school work completed by both groups is taken into consid- PERSISTENCE IN SCHOOL AND HOME CONDITIONS SI eration, it must be admitted that the present generation, although apparently attending school for no more years than its predecessor, has enjoyed a longer school year and a much richer curriculum. 2. Amounts of education received by the mothers and daughters: The mothers have, on the average, received one year less schooHng than Sa. Dd MEDIAN OF MOTHERS 8.60 YRS. MEDIAN OF DAUGHTERS 9,60 ** DIFFERENCE 1.00^.20 ** EDUCATION OF MOTHERS EDUCATION OF DAUGHTERS --- PART COVERED BY BOTH CMWtZ9//// ^z^ o 2 4 6 8 lo 12 14 i6 i8 20 Fig. 6. — Education of Mothers and Daughters: Years of Schooling their daughters. The median number of years of schooling received is 8 . 6 years for the mothers and 9 . 6 years for the daughters. A difference of 1. 00 ±0.20 years (Fig, 6). These sHght differences may be explained partly by the increased educational opportunities offered to the present generation and partly by the desire on the part of parents, especially those poorly educated, 52 THE FIFTEENTH YEARBOOK to give their children a httle better education than they themselves received. The nature of this difference may, perhaps, be seen best in a comparison of the numbers who received more, the same, or less edu- cation than their parents (Tables XLII, XLIII, XLIV). In but few TABLE XLII Comparison of Education of Children with Average Education of Parents Average Years of Schooling, Parents I 2 3 4 s 6 7 8 9 lO II 12 13 14 Sons Received more s 4 3 7 4 2 9 i6 I 13 4 25 4 6 34 3 2 20 13 4 25 8 2 i6 13 8 22 9 9 13 2 4 23 I S II I 6 17 3 2 2 2 3 3 8 7 lO 3 6 I 2 8 I Received same Received less 4 I Daughters Received more 3 2 4 Received same Received less 2 TABLE XLIII Comparison of Education of Children with Education of Fathers Years of Schooling of Fathers o 1 2 3 4 S 6 7 8 9 10 II 12 13 14 IS 16 17 18 19 20 Sons Received more . 7 II 6 3 I 8 I 3 S I I 25 6 4 3° 4 4 17 14 4 2S lO 3 24 lO lO 30 8 6 I 7 6 2 I 8 's' 13 2 8 3 2 6 7 I I 6 S 10 s 8 5 I I I I I 2 I 5 4 2 I 2 2 I 2 Daughters Received more . 3 IS 3 6 I I I I TABLE XLIV Comparison of Education of Children with Education of Mothers Years of Schooling of Mothers I 2 3 4 5 6 7 8 9 10 II 12 13 14 IS Sons Received more .... Received same. . . . I 6 19 5 3 I II 4 3 23 2 2 20 9 8 19 6 6 27 18 14 38 17 14 4 4 10 7 3 S 10 2 5 19 I 7 2 4 7 S II 18 7 9 3 Received less I I Daughters Received more. . . . 6 13 4 3 Received same. . . . Received less I PERSISTENCE IN SCHOOL AND HOME CONDITIONS 53 cases did the children of poorly-educated parents receive less education than their parents. When the education of the children of those parents' who went to the eighth year or beyond is compared with that of their parents, there is no such marked increase. In comparison with the average education of these parents, 49 per cent of their sons and 64 per cent of their daughters received more education and 32 per cent of their sons and 20 per cent of their daughters received less. In comparison with these fathers 39 per cent of the sons and 59 per cent of the daughters received more, while 45 per cent of the sons and 21 per cent of the daugh- ters received less, showing that these sons actually received less educa- tion on the average than their fathers. When the mothers are considered, both the sons and daughters received slightly better average educations, 43 per cent of the sons and 57 per cent of the daughters receiving more than their mothers, and 34 per cent of the sons and 24 per cent of the daughters, less. Schooling of parents and progress of pupils now in school. — The chil- dren fourteen years of age and older who were reported to the writer as intending to continue their schooling were in various grades from the fifth to the last year of the university. An attempt to determine if retardation was greatest among the children of the less educated families was made by comparing each age group with a scale of "ideal progress." According to this scale a boy or girl 14 years of age should have been in the 8th grade 15 " " " " " 9th grade 16 " " " " « loth grade 17 " " " " " nth grade 18 « " « « " i2th grade 19 " " " " " ist year of college 20 « « " " « 2d year of coUege 21-22 " " " " " 3d year of coUege 23-24 " " " " " 4th year of college This scale is entirely arbitrary and is of value only to the extent that it serves as a measure of retardation and acceleration. It assumes, of course, that children enter school at six years of age, which is the general rule in Urbana. This, however, may not have been true of all the cases ' This comparison was limited to these parents because their education extended beyond the age affected by compulsory attendance laws. The children of parents who have less education may be kept in school by law more than through parental influence. 54 THE FIFTEENTH YEARBOOK recorded in this study. Some may have entered at eight or nine and have progressed through the grades in the normal number of years. When the resulting comparisons are examined, it is seen that there is a positive relationship between home conditions and the progress of the pupils. With the girls this is only o . 2 2 =»= o . 06 (Table XL VI) , while TABLE XLV Correlation between Average Education of Parents and Progress of Sons Yet in School relation to progress, boys Years Average Years of Schooling of Parents 5 6 7 8 9 10 II 12 13 14 +2 + 1 2 4 5 I 2 7 3 2 2 4 7 S 3 4 I 2 2 2 2 I — I 4 2 2 3 2 I I — 2 I — 3 — 4 — 5 I r — o.37±o.o7 n =79 Average retardation, o . 96 year TABLE XLVI Correlation between Average Education of Parents and Progress of Daughters Yet in School relation to progress, girls Years Average Years of Schooling of Parents S 6 7 8 9 10 II 12 13 14 IS 16 + 1 I 4 2 2 S 5 S I 3 4 I 2 4 6 7 2 I 4 2 2 S 2 I I — I 2 2 I 2 2 7 I 2 I I — 2 — 3 I — 4 I r =0.22=1=0.06 « =97 Average retardation, 0.63 year PERSISTENCE IN SCHOOL AND HOME CONDITIONS 55 it is o . 37 ± o . 07 with the boys. The boys, with an average of o. 96 year retardation, were retarded more than the girls, who averaged o . 63 year. SUMMARY AND CONCLUSIONS The relationships presented in this section may be summed up as in Table XLVII. TABLE XLVII Education of fathers correlated with education of mothers o.65±o.o3 " " fathers " " " " sons 0.47^0.03 " "mothers " " " "daughters 0.60^0.03 " "fathers " " " "daughters 0.56^0.03 " " mothers " " " " sons 0.55^0.03 " " parents " " " " sons 0.65 ±0.03 " " parents " " " " daughters 0.62^0.03 " " better-educated parent correlated with education of sons 0.60=^0.03 " " more poorly educated parent correlated with educa- tion of sons 0.57^0.03 " " parents correlated with progress of sons 0.37=^=0.07 " " parents " " " " daughters 0.22=^0. 06 Fathers are slightly more variable with respect to number of years of schooling received than are the mothers. The median amounts of schooling of parents and children are as follows : Fathers, 8 . 45 years Mothers, 8 . 60 years Sons, 8 . 60 years Daughters, 9 . 60 years Difference, o. 15=1=0. 19 years Difference, i . 00=1=0. 20 years The boys now in school are retarded more than the girls, as indicated by an age-grade distribution. There is a close relationship between the educational level of a home and the length of time children remain in school. SECTION II. ECONOMIC HOME CONDITIONS This section deals with the economic status of the families under con- sideration. The economic status of a family is not always apparent to a visitor. Nor can one receive a wholly reliable estimate of it from an examination of the assessor's sheets. Since this study includes families all of whose children are grown, other famihes with infants 56 TEE FIFTEENTH YEARBOOK taxing their resources, and all sorts of intermediate types, it is quite apparent that an index which adequately represents the economic status of each family is not easily obtainable. Three indices — rental value of home, personal property assessment, and real estate assess- ment — were selected as criteria, and the results bearing upon them are presented for what they are worth. RENTAL VALtTES AND SCHOOLING OF CHILDREN Every home was assigned a rental value at the time the data were collected. This was a comparatively easy matter, for in most cases where the home was owned by the family the member who furnished the information to the writer was fairly well acquainted with rental values in the neighborhood. A little difficulty was experienced in determining rental indices for a few of the better homes which were built by their present occupants for their own use and which far sur- passed all rented homes in the neighborhood in beauty and conveniences. In such cases the writer usually offered a conservative figure to some responsible member of the family for approval. Hence, nearly all the homes with rental indices of $40 a month or more are probably under- estimated. Since rental values are subject to fluctuation, the approxi- mations given here cannot be considered as valid or representative for any considerable period of time. A further complication was due to the presence of roomers in a few homes. This tended to reduce the real rents below the values assigned to these homes. Such famihes were included in the group given here, although such a procedure may be open to criticism. In spite of all the disturbing influences mentioned, it is felt by the writer that the rental index is a fairly good measure of the economic status of families. When the rental values were correlated with the amounts of schooling which the children have received, the coefficients of correlation, 0.63+ 0.03 for the sons (Table XL VIII) and o.64-fo.o3 for the daughters (Table XLIX), were obtained. If the large number of disturbing factors which have affected the indices are taken into consideration, these cor- relations seem high. PERSONAL PROPERTY ASSESSMENTS AND SCHOOLING OF CHILDREN The personal property indices were taken from the 19 15 tax books at the courthouse in Urbana. These assessments were made during the PERSISTENCE IN SCHOOL AND HOME CONDITIONS $7 TABLE XLVIII Correlation between Rental Values and Education of Sons Years of Schooling Rental Values of Home, Dollars per Month of Sons 10 12.50 IS 17-50 20 22.50 25 27.50 30 35 40 45 50 55 60 i8. . . . I I I 17 ... . I i6 I s IK. . . . 2 I I 14 I 2 I 4 2 2 3 2 I I I I I I I 12 I 3 2 2 3 2 I I 12 I 2 6 3 3 9 16 14 II I I 2 I 3 I 2 2 I 3 2 7 3 2 II I 2 I I 10 9 8 I 7 7 I 4 6 3 7 9 6 7 2 I 3 2 7 . 2 I 6. . 5 4. 2 r =0.63=1=0.03 « =224 14 O^O ■^— — — — 10 20 30 40 50 60 Fig. 7. — Correlation between Education of Sons and Rental Values 58 THE FIFTEENTH YEARBOOK summer of 19 14 and the figures are supposed to represent one-third of the actual valuation that the properties would have at a forced sale. A few families that were overlooked by the assessor were given the values of the 1 9 13 assessment. A few families that have more personal property than the average were missed by the assessor both times. Owing to the almost universal practice of "tax-dodging," the values given here con- tain a large element of unreliability. How large this is, cannot be TABLE XLIX Correlation between Rental Values and Education of Daughters Years of Schooling Rental Values of Home, Dollars per Month of Daughters 10 12. so IS 17 so 20 22. so 2S 27.50 30 35 40 45 50 55 60 10 I 18 17 16 3 2 I 6 3 2 2 3 2 ic 14. I I I? I I I 2 2 6 3 I I 5 I I I I I 2 2 I 2 I I 4 12. . . 10 2 7 3 14 12 7 3 I I 2 4 2 2 9 2 3 4 5 I I 3 I I I II . 10. 2 I 6 9 3 3 5 "h' 6 5 3 I 2 8 2 7 6 "?.... I 4 ? I I 2 I I r =0.64=1=0.03 n =226 determined. If it is a constant factor afifecting all classes alike, it reduces the indices but does not shift them from their true order. Taking these errors into consideration, it is surprising that the correlations between the schooling of the children and the personal property assessment indices are as large as they are. They are 0.47=1=0.04 for the sons (Table L) and o. 52=1=0 . 04 for the daughters (Table LI). These figures were calculated for the group who were assessed. PERSISTENCE IN SCHOOL AND HOME CONDITIONS 59 TABLE L Correlation between Personal Property Values and Education of Sons Years of Personal Property Assessment of Home, Dollars of Sons o lO 20 30 40 so 60 70 80 go 100 125 150 175 200 250 300 400 500 600 l8 I I I I 2 I 17 i6 I I T C I I I 2 I? I I I 3 2 2 2 I I I 2 I I I I I I 2 I 2 I 2 12 4 I 2 3 I 4 2 4 I I 2 2 4 I 8 I I I T II 2 2 2 8 10 4 I 4 I I I 7 I 2 2 I 3 4 4 I I 2 7 4 I 5 lO 12 6 2 2 I 2 I 2 8 T 7 I I 6 e I I 2 r =0. 47=^0.04 n =198 A / \ ^^ .h ^ ^ y \ \ ^ r \ ^ / o 50 100 150 200 250 300 350 400 450 500 550 600 Fig. 8. — Correlation between Education of Sons and Personal Property Values 6o THE FIFTEENTH YEARBOOK TABLE LI Correlation between Personal Property Values and Education of Daughters Years of Schooling of Daughters Personal Property Assessment of Home, Dollars 10 20 30 40 so 60 70 80 90 100 I2S 150 175 200 250 300 400 500 600 10 I i8 17 i6 I 2 I 2 2 2 I I I 3 2 2 4 ic 14 13 12 I 2 2 I I 2 2 I 4 10 II I 2 I I i 3 2 2 4 3 2 6 4 I I 7 2 3 3 4 I 2 4 2 2 I 7 2 2 I 6 5 2 I I I 3 I 3 I I I S I I I I I II 10 I 2 2 3 3 10 7 3 I 8 7 5 I 3 I 7 6 c 4 ■3. I I 2 I I f =0.52=1=0.04 n =212 values of home and schooling of children The real estate assessment indices were taken from the 1915 tax books just as the personal property indices were. Owing to the unal- phabetical arrangement of the books, it would have been an extremely laborious and probably unprofitable task to ascertain the total values of the real property owned by the different individuals represented in our investigation. Because of this fact it was decided to take the value of the home in which the family lived, if owned by one of its members, as the real estate index. The assessed valuation was one-third of the actual valuation. The correlation of the real estate assessment indices with the schooling of the sons is 0.63=1=0.04 (Table LII), and with the schooling of the daughters it is 0.58=^0.04 (Table LIII). These figures are calculated from the group of those who owned their homes. PERSISTENCE IN SCHOOL AND HOME CONDITIONS 6l TABLE LII Correlation between Real Estate Values and Schooling of Sons Years of Schooling Real Estate Assessment of Home, Hundreds of Dollars of Sons I 2 3 4 5 6 7 8 9 10 II 12 13 14 IS 16 i8 I I 17 I 16 I 2 2 I 4 2 2 6 23 20 13 6 5 2 I I 2 I 15 I 14 I I I I I 2 I I !•? I I I 4 2 12 2 I I 4 5 s 2 I I I 3 3 2 I I 2 I 4 3 I 2 3 I I I 3 I I II I S I I 2 2 s 4 5 3 3 10 I 9 2 I X I 8 I I I I 7 6 I 5 I 4 3 r =0.63=1=0.04 « =129 18 16 14 / \ /S , / \ M r V V A ^ i ' V 200 400 600 800 1000 1200 1400 1600 Fig. 9. — Correlation between Education of Sons and Real Estate Values 62 THE FIFTEENTH YEARBOOK TABLE LIII Correlation between Real Estate Values and Schooling of Daughters Years of Schooling Real Estate Assessment of Home, Himdreds of Dollars of Daughters I 2 3 4 s 6 7 8 9 10 II 12 13 14 IS 16 10 I i8 17 i6 4 I I I I 4 2 I I 4 IC 2 14. I 2 I 13 2 8 3 10 8 16 16 II 7 I I I 5 4 5 I 2 I 4 2 4 2 2 I I 2 I 12 s I 2 4 I 2 2 2 I I I I 5 II 10 2 I I 4 9 8 I 2 2 3 3 2 12 7 3 I I I I 2 2 7 6 5 4 ■J. I I 2 I I r =0.58=1=0.04 n =144 Summary and Conclusions The relationships presented in this section may be summed up as follows: TABLE LIV Rental value of home correlated with schooling of sons 0.63=^0.03 Rental value of home correlated with schooling of daughters .... o. 64=^=0. 03 Personal property assessment correlated with schooling of sons. . 0.47=1=0.04 Personal property assessment correlated with schooling of daugh- ters o. 52=1=0.04 Real estate assessment correlated with schooling of sons o. 63=1=0. 04 Real estate assessment correlated with schooling of daughters ... o. 58=1=0.04 Allowing for the approximate character of the indices, it may be said that economic home conditions in Urbana are closely correlated with the amounts of schooling which the children receive. PERSISTENCE IN SCHOOL AND HOME CONDITIONS 63 SECTION III. SOCIAL AND QUASI-SOCIAL RELATIONSHIPS NITMBER OF BOOKS IN THE HOME AND SCHOOLING OF THE CHILDREN The number of books in a home is a rough index of the culture of the home. It does not take into consideration the possibiHty of using the free pubHc hbrary, an opportunity which has been open to all Urbana homes during recent years.^ It disregards the differences in the quality TABLE LV Correlation between Number of Books in Home and Education of Sons Years of Schooling Number of Books in Home of Sons 10 25 5° 75 100 ISO 200 250 300 350 400 500 18 2 I 17 I I 2 2 I 16 2 I 2 IC 2 2 I I I I I 14 I I 3 4 5 I 6 I 2 I I I 3 3 2 I 13 I 2 I 2 I 5 2 I I 12 3 2 6 2 7 II 19 8 6 3 I 2 I I II 10 3 7 17 5 3 I 2 I I 2 8 4 8 9 4 5 7 6 I I 5 4 3 r =0.67*0.03 n =222 and character of the books, which were probably marked in some cases. Yet, in spite of these limitations, it bears a closer relationship to the number of years of schooling children receive than any other measure used in this study. For the sons the coefficient of correlation between the books in the home and the number of years of schooHng is 0.67=^=0.03 (Table LV); for the daughters it is 0.68 ±0.02 (Table LVI). ^ The public Kbrary in Urbana has been in a position where it could be of service to the community for more than thirty years. 64 THE FIFTEENTH YEARBOOK 16 i6 \ J Y 14 12 10 ^ .^^ \ y / —/ / \ / 8 / J- 6 / o 50 100 150 200 250 300 350 400 450 500 Fig. 10. — Correlation between Education of Sons and Size of Home Libraries TABLE LVI Correlation between Number of Books in Home and Education OF Daughters Years of Schooling Number of Books in Home of Daughters 10 2S 50 75 100 150 200 250 300 3SO 400 500 600 10 I 18 17 16 I 2 I 3 I I 2 7 I 2 I ic 2 14, I 2 9 4 I 2 I I I I 12 I 13 3 8 I 5 I 2 2 2 I 5 I I 3 I 12 I 4 3 S 2 5 16 6 3 3 4 I 6 I II 10 I 4 4 13 17 6 5 Q 8 2 3 7 4 2 7 6 c I 4 •3. I I 2 I I r =o.68='=o.02 « =231 PERSISTENCE IN SCHOOL AND HOME CONDITIONS 65 HOUSING AND SCHOOLING OP THE CHILDREN Out of a total of 234 families 34 reported one or more grown indi- viduals not members of the family but living in the home. Housing conditions are measured by number of rooms per individual. In finding this index no distinction was made between children and adults. In general, the housing conditions found in this investigation were quite good. Very little overcrowding existed and, in an appreciable number of cases, it seemed as though the people had more room than they could use conveniently. Housing conditions are probably a reflection of economic status. Measured merely by the number of rooms per indi- vidual the relationships which exist between housing conditions and education of sons and daughters are 0.50=^=0.03 and 0.48 ='=0.03, respectively (Tables LVII, LVIII). If the size of the rooms and the presence or absence of modern conveniences, such as bath and toilet, had been taken into consideration, the correlation would probably have been higher. TABLE LVII Correlation between Housing Conditions and Education of Sons Years of Schooling of Sons Rooms 3er Individual in Home i I li 2 3 4 18 2 I I I 3 I 2 I 17 I 2 16 I I I 2 I I 2 IC I 14. 4 I 12 3 7 12 16 22 8 4 4 2 3 7 8 S 4 14 2 2 12 12 II 10 I I 3 2 4 I 4 16 9 10 3 I 8 2 7 6 s A I 2 r =0.50^0.03 n =223 THE FIFTEENTH YEARBOOK 01234 Fig. II. — Correlation between Education of Sons and Housing Conditions TABLE LVIII Correlation between Housing Conditions and Education of Daughters Years of Schooling of Daughters Rooms per Individual in Home i I li 2 3 4 10 I 18 17 16 I 4 8 I 2 2 18 6 6 2 9 4 2 2 6 I 2 3 I I I 3 15 2 14 ; . I 3 II 7 II 5 18 12 7 6 13 I 5 I 4 4 IS 12 7 2 12 2 II 10 8 2 3 I 7 6 c 4 7. 2 2 I I r =0.4? w =231 = 0.03 PERSISTENCE IN SCHOOL AND HOME CONDITIONS 67 INTERRELATIONSHIPS Thus far in Part IV the various factors have been considered sepa- rately. In reaUty, they are all interrelated. A few of these inter- relationships will be given to show the fallacy which results when conclusions overlook the complex character of social phenomena. a) Schooling of parents and number of hooks in the home. — As might be forecasted, there is a close relationship between the schooling of the parents and the number of books found in the home. This correlation, o. 60 =•= o. 03 for the fathers (Table LIX) and o . 6 1 ± o. 03 f or the mothers TABLE LIX Correlation between Number of Books in Home and Education of Fathers Years of Schooling Number of Books in Home of Fathers 10 25 so 75 100 ISO 200 250 300 3S0 400 Soo 600 I I 18 I T7 I 16 I I I I I I 2 I I TC I I I 2 5 7 3 12 3 2 14. . . . 2 12 . . 12 4 3 I 4 4 5 9 II 3 I I I I I I "'6' I 2 I I 3 4 2 5 4 7 2 4 I 3 I 3 3 I I I I 2 13 9 8 3 2 I I 4 4 6 I 2 I I I 2 8 I 6 I I I I I I I r =0.60=1=0.03 n =230 (Table LX), is not so high, however, as that previously noticed between the number of books in the home and the schooling of the children. The difference is not enough to be very significant, however. h) Number of books in the home and size of family. — The relationship which exists between the number of books in the home and the number of children in that home is slightly negative, —o. 10 ±0.04 (Table LXI). 68 THE FIFTEENTH YEARBOOK This shows that the number of books owned by a family is not at all dependent upon the number of people there are to read them. TABLE LX Correlation between Number of Books in Home and Education of Mothers Years of Schooling Number of Books in Home of Mothers lO 25 so 75 100 ISO 200 2SO 300 350 400 Soo 600 ic I 14. I I 2 17, I 5 3 7 3 IS I I I I 12 4 I 5 6 9 12 4 I I 4 S I I I 2 I 6 2 I I 9 I 6 9 2 4 2 4 II lO 3 2 2 I I I I I .... 2 6 3 I I 5 5 13 lO 7 3 3 8 I 7 6 c 4 . 4 I 2 2 2 I I O I r =0.61=^0.03 n =230 TABLE XLI Correlation between Size of Family and Number of Books in Home No. of Children Number of Books in Home in Family 10 25 so 75 100 ISO 200 250 300 350 400 SOO 600 10 2 I I 7 3 7 10 7 4 I I 2 4 6 3 9 9 6 5 I I 4 I 3 2 4 2 3 8 2 I 2 2 3 4 2 7 I I 2 6 S 6 4 I I I I I 6 I 3 3 2 2 2 4 8 7 12 7 I e I 2 I 4 I I I 3 4 4 I I 3 7, 2 2 I I f =— o.io=to.o4 n =233 PERSISTENCE IN SCHOOL AND HOME CONDITIONS 69 c) Rent and size of family. — To a slight extent the better homes are occupied by the smaller families. The coefficient of correlation between size of family and rental values is also slightly negative, being — o. io± 0.04 (Table LXII). TABLE LXII COREELATION BETWEEN SiZE OF FAMILY AND RENTAL VaLXJES No. of Children Rental Values of Home per Month, Dollars 10 15 20 25 30 35 40 45 SO 55 60 10 3 2 2 7 9 5 16 14 13 6 3 I I 8 I 2 2 3 4 4 7 4 7 I I 2 3 2 7 I 6 4 5 II S 8 4 2 3 3 4 2 I 4 4 S 7 6 2 c I I 4 ■J 3 3 4 I I 2 3 2 I I r =—0.10=1=0.04 n =234 ^) Schooling of parents and size of family. — That educated parents have smaller families has been observed so often that it has become a matter of common knowledge. When expressed by a coefficient of correlation, this relationship is — o.2o=±=o.o4 (Table LXIII). Of TABLE LXIII Correlation between Size of Family and Education or Parents No. of Children Average Schooling of Parents 1 2 3 4 5 6 7 8 9 10 II 12 13 14 IS 16 10 I I I I I 8 I I 4 I 7 4 6 9 8 7 I I 3 S S 8 7 I 2 4 3 3 4 8 5 7 I I 2 I 4 7 I I 3 5 I 7 3 5 2 2 2 I 7 8 2 3 I 2 4 2 I 2 2 6 5 2 I 6 c I I I 3 4. 2 a I 2 I I 4 I I r =— o.20±o.o4 n =227 70 THE FIFTEENTH YEARBOOK course, it must be kept in mind that only families that had children were included in this group. It may be that there are more families without children among the better educated. If so, a selection of homes which included such homes in addition to those studied here would reveal a larger negative correlation. e) Education of children and size of family J- — When the entire group is examined, it is seen that the children who came from large families did not go to school so long as those who came from small famiUes. This fact is expressed by the coefficient of correlation, — o.2o±o.o5 (Table LXIV). This is the same as the relationship which exists TABLE LXIV Correlation between Size of Family and Average Education OF Children No. of Children Average Education of Children No Longer in School 4 S 6 7 8 9 10 II 12 13 14 IS i6 17 i8 lO I I I I I 4 5 2 8 5 2 2 9 I 2 5 2 4 2 5 4 8 I 2 I I 2 2 2 2 I I 7 2 I 3 I 7 3 I I I I 6 3 3 8 2 4 I 2 6 2 2 2 I 3 4 S 2 s I I I 3 2 I 3 4 4 I 3 I I 2 2 I I S I I I r =— o.2o±o.o5 n =i8o between the schoohng of the parents and the size of the family. It has already been shown^ that there is a decided relationship between the schooling of the parents and the schoohng of the children. The fore- going coefficient of correlation, then, may be merely another way of ex- pressing the relationship which exists between the schooling of the parents and the size of the family. ' In these tables the education of the children was averaged for each family. This gives each family a single index and does not over-weight the large famiUes. = Pp. 43-48. PERSISTENCE IN SCHOOL AND HOME CONDITIONS 71 If the influence of the education of the parents could be eliminated, it might be possible to ascertain the presence or absence of a true rela- tionship between the size of family and the schooling of the children. An attempt to do this was made as follows: The median schooling of parents is eight years for the entire group. The average schooling of the children of each family was increased or decreased by the same number of years that the average schoohng of the parents varied from this median. Thus, if the parents averaged seven years and the chil- dren averaged six years, the parents would be one year below the median and the index of the children would be increased by one year. Similarly, if the parents averaged twelve years and the children fifteen years, the parents would be four years above the median and the index of the children would be decreased four years. These revised educational averages of the schooling of the children were then correlated with the number of children in each home. This procedure eliminates the influence of the schoohng of the parents. It does not counteract other factors which may act somewhat independently of the education of the parents, such as economic status or number of books in the home. Further, compulsory education influences affect the level of some of the homes of the poorly educated which have large families and tend to counterbalance any negative relationship which may exist. The results do not show any decided cor- relation. The shght negative relationship, —0.06=^0.05 (Table LXV), which was found, is virtually a zero correlation. A FAMILY nSTDEX The fact that the factors thus far considered probably acted con- jointly instead of independently in determining the amounts of schoohng which the children received suggested that it might be possible to weight the various items in such a way as to give each family an index and then find the relationship which existed between this index and the schoohng of the children. This was done as follows: The 25 percentile deviation from the median was found for each of the three items, average education of the parents, number of books in the home, and monthly rental. These figures, which were approximately 2 years, 62^ volumes, and $7 . 50, respectively, were then divided by five to give more convenient divisions. Each of these divisors, 0.4 year, 12I volumes, and $1.50, was given a value of one unit. The number of times the respective divisors were 72 THE FIFTEENTH YEARBOOK contained in the quantities which represented the average education of the parents, the number of books in the home, and the monthly rental of a family gave the number of units credited to each of these items. The figure representing the units given a family for an item was squared and the sum of the squares for the three items gave the family index. This can be made clear best by a concrete example. A family whose parents have an average education of 8 years, which has one hundred books in the home, and pays $15 a month rent will serve as an illustration of the TABLE LXV Correlation between Size of Family and Schooling of Children, Effect OF Schooling of Parents Having Been Eliminated Average Years of Schooling Number of Children in Family of Children I 2 3 4 s 6 7 8 9 10 16 I I I I 7 2 8 7 I S I I ic I 2 I 5 S 8 6 8 I I I I I 2 2 I 5 S 5 I 6 2 2 14 2 I I? I I 12 2 4 I 3 I 4 I 2 3 3 4 4 II 6 2 I 3 2 2 I 4 I I 10 9 8 2 2 I I I 7 6 5 4 3 I r =—0.06=1=0.05 n =178 method. Dividing 8 years by the educational divisor, 0.4 year, gives 20 units, which is 400 when squared. Similarly, one hundred books when divided by the Hbrary divisor, 12I volumes, gives 8 units, which equals 64 when squared. The rental index, $15, divided by the rental divisor, $1 . 50, gives 10 units, which, when squared, furnishes 100 more. The sum of 400, 64, and 100, or 564, is the index of this family. This procedure is purely arbitrary, but the writer thinks that the resulting indices are quantitatively representative of the differences in PERSISTENCE IN SCHOOL AND HOME CONDITIONS n the opportunities presented to the children by their respective homes. This method gave the best home an index of 4,289, while the poorest received but 32. The possibihties of the best in contrast with the poorest are, according to the opinion of several people acquainted with both homes, as different as these indices imply. There is a gulf between them. The coefi&cients of correlation between this family index and the education of the children are higher than those expressing any single relationship. They are the same, 0.73=1=0.02 (Tables LXVI, LXVII) for both sons and daughters. TABLE LXVI Correlation between Family Index and Schooling of Sons Years of School- ing Family Index in Hundreds I 2 3 4 S 6 2 4 I S 2 2 7 3 2 I 2 7 S 8 I I 9 1 2 2 I 2 I 10 2 I 3 I II I 2 S 2 I I I 12 I I 13 2 3 \ ■ 14 I I IS I 2 I I I I 16 I I 17 18 I I I I 19 2 I I I 20 I 21 22 23 24 I I I 25 26 '. '. I 27 28 I I I 29 1 30 31 32 33 2 34 33 18 17 i6 IS 14 13 12 2 I 2 6 14 13 4 2 I II I 2 2 S I 4 3 6 S 4 4 1 2 4 6 2 I 8 7 6 s 4 3 I 3 I r=o.73±o.o2 « = 21S 74 THE FIFTEENTH YEARBOOK W g I-) M < s CO M vo Clvo fO c •H fO OOO t-^vO lo Tf CO w t-t O OOO l>vO to ^ CO w M O II u PERSISTENCE IN SCHOOL AND HOME CONDITIONS 75 SUMMARY AND CONCLUSIONS The relationships presented in this chapter are shown in Table LXVIII. TABLE LXVIII Number of books in home correlated with schooling of sons. . . 0.67=1=0.03 Number of books in home correlated with schooling of daughters o. 68 =»= o . 02 Nimiber of rooms per individual correlated with schooling of sons o. 50=4=0.03 Number of rooms per individual correlated with schooling of daughters 0.48=1=0.03 Number of books in home correlated with schooling of father. . o. 6o=*=o. 03 Number of books in home correlated with schooling of mother . . o . 6 1 =t= o . 03 Ninnber of books in home correlated with size of family — o. io=*=o.04 Rental values correlated with size of family — o. 10=1=0. 04 Schooling of parents correlated with size of fanuly —o. 20=^=0.04 Schooling of children, uncorrected, correlated with size of family —o. 20=^0.04 Schooling of children, corrected, correlated with size of family . . — o . 06 =t o . 05 Schooling of sons correlated with famUy index o. 73=^=0. 02 Schooling of daughters correlated with family index o. 73=^0. 02 The number of books in a home is the best single index of the probable educational level which the children may expect to reach. The number of books in a home is closely correlated with the school- ing of the parents. The various indices used in this part of the study are more or less interrelated. As measured by the method used here, size of family has only a slight negative correlation with the schooling of the children. SECTION IV. OCCUPATIONAL AND OTHER GROUP RELATIONSHIPS OCCUPATIONS OF THE FATHER The occupations of the fathers (Table LXIX) show that this group contains representatives from almost every stratum of the economic life of the community. Most of the occupations are represented by too small a number, however, to furnish comparisons. The first thirteen occupations will be compared with respect to the schooling of the fathers, the rent of the homes, the number of books in the homes, and the school- ing of the children.^ » The group "Farmers" is not on a par with the others. Six of the 13 fathers are dead, having been deceased in some cases for fifteen years. All of these families are living in town. Most of these farmers have retired as far as any active farm Ufe is concerned. 76 THE FIFTEENTH YEARBOOK TABLE LXIX Occupations of Fathers Occupation No. Reported Laborer 24 Carpenter 15 Retired farmer 15 Farmer 13 Painter and paper-hanger .... 8 Real estate and insurance .... 8 Machinist 6 Stationary engineer 6 Blacksmith 5 Grocer 5 Janitor 5 Evangelist and minister 5 Merchant 5 Druggist 4 Railroad conductor 3 Salesman 3 Physician 3 Driver of ice wagon 3 Grain-buyer 2 Car-repairer 2 Contractor 2 Tinner 2 Railroad engineer 2 City fireman 2 Printer 2 Banker 2 Policeman 2 Laundryman 2 Jeweler 2 Agent and solicitor 2 Carpenter contractor Teamster Teacher Cement contractor Roundhouse foreman Bank cashier Mine-owner Foundry-owner Barber ... 2 ... 2 .... 2 .... 2 .... I .... I ... I .... I .... I Ticket agent i Butcher i Section foreman i County superintendent of schools I Musician i Occupation No. Reported Furnace contractor i Lumber dealer i Pump-dealer i Tool-polisher i Plumber i Roadster i Shop foreman i Coal-dealer i RaUroad official i Grocery clerk i Postmaster i Foreman for brick company . . i Clothier and dry goods merchant i Manufacturer i Optician i Undertaker i Road boss on Big Four i Horseshoer i Jailor I Superintendent of signals and water service, Big Four .... i Roofing business i Carpenter superintendent .... i Tailor i Ditcher i Overseer of water-main laying i Implement dealer i Contracting excavator i Dentist i Sheriff i Veterinary surgeon i Foreman of water service on Big Four I Feed-store clerk i Manager of cold storage plant i Engine inspector i Drayman Retired minister Bookkeeper , Night watchman Railroad fireman Hostler , Brickmason Mail-carrier PERSISTENCE IN SCHOOL AND HOME CONDITIONS 77 Occupation TABLE LXIX— Continued Occupation No. Reported No. Reported Mail clerk i Restaurant keeper i House-moving contractor .... i Deliveryman i Postal clerk i Horse business i Runs ice-cream wagon i Itinerant photographer i Justice of peace i Foreman for contractor i Cigar-factory foreman i Bricksetter i a) Occupations and education of fathers. — The number of individuals (Table LXX) in several of the groups is too small to furnish any very TABLE LXX Education of Years of Schooling 3 c & ni U 2 S % fa to II ca c 'c 1 g.S 2 a 5: 1 1 5 2 O 00 -"t^ fO iO»^cot-» ^W co»H O roO 0**0 r* O f* O* t J \0 tT)^ O O b 00 iK^O^mOW^iriM coi-t cOt^ O rOO On »0 0>00 O o O* O ' cv^ M rt M "O r^ fefefefoSfefeSfefofefofoSSfeSSfefc^feSSfelSSfoStEHfofefe 00-CO 00 O<00 O" r^oO OiCSO-OOoOOOoO OiOO r^ Ov r- Oi 0> O cooooooooooooooooooooooooooooooooooocooo ooo OOOOOOOOOOOOOOOOOOOO rf W ■00>0000 000000000000"»0"50000000 00 00 O O O O "^00 1^00 N 00 M ooo O>00 M t^OOO t^N -^N ^N OOO OOOO 00 00 0>O o O. lo c ■ o 00 00 00 00 O OCO OOr-ONOtNOOOOOOt"- C/5 C/) C/2 C/2 C/^ C/2 Cfi M O) . CAl (/5 CAl C/2 Cfi(/3 C/2 C/2 Jj en C/2 CO C/3 C/2 c/:i M IZl mc/J