^5^>^ xrbe mntverstti? of Cblcaao W- % FOUNDED BY JOHN D. ROCKEFELLER U^'^ STANDARDIZATION OF THE SCHOOLS OF KANSAS A DISSERTATION -^ SUBMITTED TO THE FACULTY OF THE GRADUATE SCHOOL OF ARTS AND LITERATURE IN CANDIDACY FOR THE DEGREE OF DOCTOR OF PHILOSOPHY (department or education) BY JOHN ADDISON CLEMENT THE UNIVERSITY OF CHICAGO PRESS CHICAGO, ILLINOIS THE UNIVERSITY OF CHICAGO PRESS CHICAGO. ILLINOIS Baents THE BAKER & TAYLOR COMPANY NEW YOKE THE CAMBRIDGE UNIVERSITY PRESS LONDON AND EDINBUEQH Ube XIlm\?ersttp ot Cbtcaao FOUNDED BY JOHN D. ROCKEFELLER STANDARDIZATION OF THE SCHOOLS OF KANSAS A DISSERTATION SUBMITTED TO THE FACULTY OF THE GRADUATE SCHOOL OF ARTS AND LITERATURE IN CANDIDACY FOR THE DEGREE OF DOCTOR OF PHILOSOPHY (department of education) BY JOHN ADDISON CLEMENT THE UNIVERSITY OF CHICAGO PRESS CHICAGO, ILLINOIS OV-. ^ Copyright 1912 By The University of Chicago All Rights Reserved Published January 1912 Composed and Printed By Tlie University of Chicago Press Chicago, Illinois, U.S.A. ©CI.A305337 war" TABLE OF CONTENTS CHAPTER PAGE I. General Introduction i II. Statement of the Problem and Methods lo III. Comparison or Relative Standing of Pupils in Grammar Schools, High Schools, and Colleges 19 IV. General Conclusions 126 CHAPTER I GENERAL INTRODUCTION The attempt is being made in many states, and in particular through- out the states- of the Middle West with their state-controlled systems of education extending from the elementary school to the university and college, to relate and unify more closely than has hitherto been the case the primary schools with the secondary, and the secondary with the higher institutions of learning. Some of the forces or factors in this movement have not originated directly in the schools and not all have been equally appreciated or consciously operative toward the unification of state school systems. Obviously one of the first steps to be taken toward the effective direction of these factors is a determination of the actual existing con- ditions as regards the correlation or lack of relation between the different units in the state's educational system. One of the readiest means of estimating the existing relations is through a study of the records made by the students who pass through the three institutions from primary to higher education. If the pupils as a group of individuals maintain about the same relative standing in their work as they pass from one institution to another this is good evidence that the work in the institutions concerned is closely related, and it will be one object of this thesis to try to establish this contention. Several years ago in looking over the high-school certificates kept on file at the University of Kansas the writer became interested in making certain comparisons of the standing of pupils between the high school and university and it occurred to him that it would be interest- ing and worth while to go farther than this, and on the basis of the estimates given by teachers in the form of school marks attempt an evaluation of the relative standing of pupils on a sufiiciently compre- hensive scale to afford a reliable measure of the existing relations of educational institutions throughout the whole state. ^ ^ The problem taken up in this thesis was first suggested several years ago by the appearance of Professor Dearborn's bulletin on- "The Relative Standing of Pupils in the High School and in the University," Bulletin No. 312, High-School Series, No. 6, University of Wisconsin. The writer's investigation during this year has been carried on at the University of Chicago under the inspiration and super- vision of Professor Dearborn. 2 STANDARDIZATION OF SCHOOLS OF KANSAS The problem, then, before us is a state-wide canvass of the existing conditions in respect to the question just raised. Obviously not all the pupils in the schools of a state could be studied, from merely physical limitations of the investigation, but it is believed that a sufficiently wide and discriminating sampling of the school population of the whole state has been made, such that the results to be presented are reliable and representative of the actual conditions. Since we are to study the schools through the individuals who pass through them, the more specific question at issue at the outset is to point out the relation between the scholarship of an individual in his earlier school career and the scholar- ship of his later career. Do pupils who have a good standing based upon their first educational endeavors maintain the same relative stand- ing when they pass on from the elementary institutions of learning to the secondary schools, and also when they pass on into the higher institutions of learning ? And do pupils who begin their school life by doing mediocre and poor work respectively maintain their relative positions throughout their school careers ?^ Granting that this is a question of sufficient concern to justify a careful investigation, it is assumed that one legitimate means of deter- mining the relative standing of pupils from year to year, either within the same institution or the relative standing between different insti- tutions, is through the records which have been preserved. Any insistence upon the importance of keeping records seems almost unnecessary, and yet a few very commonplace analogies may serve to re-emphasize the importance of continuous records over a series of years in any kind of institution whatsoever. Business organizations regularly take account of stock. They make an exact estimate of their profits and losses for the year. On the basis of past records and on the basis of present needs and demands they plan for the advance year's work. Intelligent methods of pro- cedure, and intelligible ways of preserving the facts, and clear means of recording the progress of the business are always of vital concern. Manufacturing plants of all kinds are directed and controlled by persons who know precisely the amount of the output, together with its quality. The efficiency of such plants as this can be determined best through the preservation of the ways and workings of the institutions. And in ^ Since the records only of the pupils who graduated both from the grammar school and high school have been used there was very little opportunity in this study to consider the problem of elimination, and of course in the part of this thesis which deals with the three-institution comparison the very nature of the problem excluded the question of elimination before the first year of college work. GENERAL INTRODUCTION 3 order to get a proper estimate, the records need to be preserved through- out a series of years. The painstaking care of all scientific biologists in the cultivation of the many forms of plant and animal life is suggestive for the modern educationist's procedure. Great progress, too, is being made in scientific agriculture through- out our whole country. The soil is scrupulously analyzed in order to discover what sort of seed will do best when put into a certain quality of ground. Great care is taken to breed up the finest quality of grains, plants, and animals. Records are kept covering a series of years in order to get a scientifically balanced judgment in the midst of varying conditions caused by the great variety of factors entering into the growth of any one single product. Similarly there is sufficient com- plexity in all the affairs of the mental life of individuals to baffie the untutored mind in trying to make analyses of the progress of pupils from year to year. Ignoring for our present purpose many analogous points of interest to be found in these business concerns, manufacturing establishments, plant cultivations, and agricultural pursuits relative to our school system, the analytic care and concern which extends over a period of years is one principle of large significance to be carried over. The intelUgent preservation of intelligible records is a second vital principle, the neglect of which may easily be discovered by any thoroughgoing examination of our present systems of record-keeping in the schools. And yet it has been suggested to the writer a number of times by school people during this investigation that teachers and officers of one institution of learning should not be expected to be held accountable for the progress in the institution to which the pupil passes on. One plausible answer to such a query is: If teachers and school officers are not responsible for keeping continuous records of boys and girls over a period of years, who is responsible for such data ? Everyone grants at once that school systems are institutions which are invaluable assets to any country, state, city, or community. In the United States one hundred years ago the problem of education was simpler and in some ways even primitive. But with the increasing complexity of our social, economic, and industrial and agricultural life there is a new demand made upon everyone who pretends to help guide the educational affairs of our country. When the outside occupations in the home life supplemented so largely the school in the midst of that simple rural community fife 4 STANDARDIZATION OP SCHOOLS OF KANSAS there was probably not so great need of elaborate records. The whole process could be grasped more easily, and consequently results could be evaluated with considerable accuracy. However, at the present time we have accumulated a good deal of experience. We are surely far enough along to ask now with a considerable degree of enthusiasm and seri- ousness, too: What are some of the tangible results of all these years? What is the real efficiency of our present system of education ? When marks are recorded in a complete, accurate, and intelligible manner, and when they cover a series of years, they furnish one impor- tant means among many of evaluating a school system. Let it be clear that it may not be the only means. But since marks have been used and are now used, they furnish one clue to the efficiency both of individuals and of institutions. While this study deals primarily with the relative standing of individuals, its purpose is to throw light upon the efficiency of institutions. Professor Dearborn says: In arguing for the school experiment the writer would not have it for- gotten that in existing school records and reports and in present school prac- tices there is already accumulated or available a body of data which if properly evaluated just as truly represents the results of experimental investigation as new experiments might do. School practices always represent great edu- cational experiments The statistical studies of Thorndike, Ayres, and others have uncovered results which it would take years of new experi- mentation to establish.^ The nature of well-kept records, as already intimated, is a question which is bound to force itself into such a treatise as this. There is little doubt, even upon a most casual investigation, but that we are in need of some clarification on the practical side in record-keeping, and this may or may not necessarily imply a need for absolute uniformity of standard either within a state or city system, but it certainly implies that there ought always to be intelligibility and probably usually trans- ferableness. The great variety of systems used, together with the very frequent incompleteness of records, offered two of the greatest difficulties in collecting the data for this study. Since it is not always easy to trans- late one system of record-keeping into the terms of another, it is often impossible to push comparative results as far as it is really desirable in order to get a wholly satisfactory measure between institutions. Sometimes it was possible to supplement incomplete and inaccurate ^ School Review Monograph, No. i. GENERAL INTRODUCTION 5 data by the knowledge of teachers and principals, but in, many cases such types of data as this simply had to be discarded. It may be of significance on the practical side to mention a few more facts in detail on the side of incompleteness of records. In the first place, many records are literally buried in the dust of an attic, or in the basement, or in the trash room of some school building. Others are scattered either among individual pupils or sometimes among teachers who possess private books containing the marks of pupils. And most exasperating of all, there have been a very considerable number burned or destroyed soon after the pupil leaves an institution and not infrequently before he barely passes out of the hands of the institution. This is not an attempt at rhetorical phraseology but an honest statement of facts experienced during this investigation. On the other hand there was, of course, no school visited but that it had some feature worthy of commendation in some aspect of its mark- ing system at one time or another. It certainly would be a profitable piece of work for someone to collect the strong features of all the sig- m'ficant systems of record-keeping in our various schools, probably including some large business establishments, and then from this work out a more satisfactory plan than now exists in any one separate school system. Probably the one element in such an attempted scheme or plan as this ought to be the element of simplicity, for what we need in a well-devised marking system, and the preservation of it, is simplicity as well as completeness, accuracy, and intelligibility just so far as it can be carried without destroying its significance and workableness. Who is better qualified than administrators with wide experience and enlightened scientific vision to perform such a task as this and thus render a great service to the practical working of our school system ? With respect to the nature of the markings used there is in actual practice great variation. In some schools letters, in others numbers, and in others percentages are used to indicate the standings of pupils in the different subjects, and in other schools fractional numbers as well as integral numbers Where the letters or numbers have been used as estimates, frequently these were accompanied by the plus or minus sign, which in this case widens somewhat the range of estimates within a scale of grading. Whether the range of the scale of grading is wider in many schools in theory than is actually carried out in practice, and whether a merely three-estimate basis is satisfactory, are points that will be furnished with a further basis of judgment in the actual charting of the marks farther along in this treatise. 6 STANDARDIZATION OF SCHOOLS OF KANSAS As to the form in which records have been either temporarily or permanently preserved there is a wide variation in practice, and as to the form in which they should be preserved there is much difference of opinion. In some of the grammar schools records were preserved only in the private books of the individual class teachers; in others these marks had been transferred to large loose sheets on file in the principal's ofi&ce. The most satisfactory ones had been preserved in large bound volumes accessible to all the teachers and school officers, of which t3^e several schools had records extending back for ten or twelve years. In the high schools the temporary records on the whole had been placed upon cards filed in boxes alphabetically arranged and the per- manent records had been preserved in large bound volumes. Some colleges prefer very much the loose or removable leaf to the large bound volume which, too, would no doubt be a satisfactory plan for the high schools. In this discussion an emphasis on the preservation of records that cover a period of years has a large significance, for without this it is scarcely possible to get an estimate of pupils as they pass from one institution to another, and consequently impossible to get a measure of efficiency between various institutions. Now in one large city studied — not in Kansas — it is customary to send up to the high school certificates containing the grades received by the pupil in the different subjects studied during the last year of the grammar school. In the high schools of Kansas it is the practice to send certificates containing the high-school marks to the colleges to which high-school graduates go. It would be a comparatively simple matter to place upon this certificate sent by the high schools to the colleges at least the standing of the pupil in the eighth-grade work, and consequently this would furnish a line on the standing of the individual in the three institutions of learning from primary to higher education. Such data as this kept on file in the vaults of the colleges would not only be the means of furnishing a line on the school career relative to the scholarship of a pupil, but with a sufficient number of these files a more adequate measure of the institutions throughout the state could be obtained. Two features found in the practice of two different high schools tended toward a preservation of complete and intelligible records. In one high school the eighth-grade standing had been recorded in the same bound volume on the same page with the complete high-school standing. In another high school it was possible to tell from the record GENERAL INTRODUCTION 7 just when a pupil did take a certain subject. If a fourth-year pupil took a first-year subject for some reason or another, it was clearly indi- cated; that is, the record showed precisely the order in which the pupil actually took his work. The records used in this thesis were in part secured from the cer- tificates on file at the colleges, and in part from the files in the offices of school principals and school superintendents. Approximately 5,000 records of high-school graduates were collected. But comparatively few of these could be traced back into the elementary school and also up into the college.^ The attempt has been made, as stated, to secure records from as nearly representative schools as possible, attention being given to size, location, organization, and to records covering a series of years sufficient to be significant. The cities represented within the state of Kansas vary approximately from 100,000 to 5,000 in population. Several cities of 50,000 popula- tion are represented, and several of 15,000. Some towns smaller than those of 5,000 population, too, have been used in the comparisons. Two large cities and two smaller towns in other states than Kansas have been included in part. The larger cities here concerned in Kansas are distributed geographi- cally over the north, east, south, and middle west of the state. There are probably not sufficient varying factors in these schools on the social side to affect them very materially. At least in such an agricultural state as this class distinctions as yet play a comparatively minor part in school life. The general method of procedure used in this study has been used chiefly by Professor Dearborn, Professor Thorndike, and Mr. Ay res. In addition to its value in securing accurate results, one of its chief virtues, it is beheved, is its problem-raising power, which has always been regarded a valuable part of any vitalizing philosophy or science. And this method will perform a large service if it succeeds in raising many significant questions, whether it succeeds in furnishing an answer to all the practical difiiculties involved in the problems or not. Since it is frequently true in the field of pedagogy that mere opinion and ofT-hand momentary estimates or snap-shot judgments have been ^ The writer is deeply indebted to the many principals and superintendents and trained helpers who so kindly assisted in collecting the material used in this thesis. It is impossible to be personal in my thanks to all the persons who assisted. While the whole study will be made as impersonal as possible, a private record of all results will be preserved, so that in case any principal or superintendent desires to inquire as to the particular results of his school he may do so. 8 STANDARDIZATION OF SCHOOLS OF KANSAS substituted for the statement of facts well tested through a scientific method of inquiry, this newer method of approach in education ought to have much value. It is the purpose of this treatise to try out many cases before drawing conclusions. In all cases the conclusions are regarded as subject to revision, and they will need to be tried out by other persons. But even such temporary resting-places are much better for educational procedure than are the random and dogmatic judgments too often found in our pedagogical literature. This does not mean to condemn whatever has been good and valuable in our present practice. Investigators who try to use marks as one basis of evaluating some phase of a school system cannot assume -the role primarily either of critics or prophets. It is their business to indicate as accurately as possible the results of the existing practices of our school system. The by-products which come with such an investigation, however, should not be regarded as unimportant. One hears frequent questions of doubt as to whether a scientific evaluation in education is possible. It is commonplace for school people even to ask, "How can mind, being so complex, be estimated upon the basis of marks?" And, "Does not the personal equation of the teacher practically vitiate all comparable results?" "Is not the individuaHty of the pupil suppressed by trying to subject it to any uniform or translatable system of grading?" These questions do suggest obvious difficulties. But is mere com- plexity of mind to bafiie us? An objection of this sort, while appar- ently bafiling, will not stand the test of any thoroughgoing analysis. To admit that the mind activity of the pupil is too complex to be evalu- ated in any sense is partially to admit that we are not worthy of the trust of educating children. Marks in some sense should indicate a real analysis on the part of the teacher of the child's mental ability. It is not the purpose of a scientific study of education to ignore the impor- tance of all the humanizing influences of teachers through their different personalities. But it is our duty at times as educationists also to make even a somewhat cold-blooded analysis of our system on its own account. Any far-reaching system of marking which we may later on evolve will take account of all types of individuality. Both the weak and the strong will be estimated according to their real abilities and will be rated in such a way that results will be comparable throughout any one system or between one system and another. In the midst of our much-debated questions of the relation of primary to secondary education and especially of the relation of the high school GENERAL INTRODUCTION 9 to college, we need to cause these institutions to look intelligently back and forth at each other. Reciprocal action and adjustment is one need, if not the great need, of primary, secondary, and higher education. If we are going to unify these different stages in any adequate sense, one means for bringing it about is through such an investigation, based upon well-tested results, as will show the actual facts resulting from present practices. In order to get any basis for standardization of schools in any state it will be necessary to find out as accurately as possible what are the actual relations existing between various institutions with reference to present practice. After having determined such relations through the relative standing of pupils on the basis of scholarship or marks recorded, it will be somewhat more easy to say what amount of reten- tion we ought to expect to obtain between the different schools. The following discussion will therefore attempt to set forth legitimate means and methods for ascertaining reliable facts relative to present practice, and then on the basis of such results venture a statement as to what we may have a right to expect with reference to the amount of retention within a standardized state school system. CHAPTER II STATEMENT OF THE PROBLEM AND METHODS As has been indicated in the introduction, the specific problem is concerned with the relative standing of pupils in the several institutions; namely, grammar school, high school, and college. The present dis- cussion will deal with a comparison between the standing of pupils in the grammar school and high school; a comparison between the standing of pupils in high school and the same students in college; and lastly a comparison of the standing of the individuals who have attended all three of these school institutions from the lower to the higher.^ The general attitude in the second part of the discussion referred to above (the high school-college comparison) is well illustrated in the University of Wisconsin Bulletin written by Professor Dearborn: The admission to college of students from the accredited schools is deter- mined almost entirely by school records or standing of the applicants, although there is an occasional admission made which is not based wholly on the pre- vious record of the pupil. One purpose of this study is to inquire into the efficiency of this method of admission to college by determining to what extent and how accurately the high-school records forecast what pupils are likely to do in the way of scholarship in the college or university. The main problem is somewhat more general than this and of wider interest; namely, to what extent students maintain in the university the relative rank which they held in the high school. That is, are the best and poorest students in the univer- sity those who stood respectively highest and lowest in their high-school classes? Is the "average" student in the university class identical with the "average" high-school pupil of a few years previous? Or is it true that these relations are to a considerable extent reversed and that many of those who do poorly are quite as likely to lead their classes in the university as those whom the high school considered its better students ?^ It has been stated above that college students in Kansas are admitted through a certificate granted by the various high schools to the grad- uates, although a considerable number of students do enter the colleges ^ These comparisons will be supplemented, in the first two sections of chap, iii, by a comparison of pupils within the grammar school, and also within the high school itself. 2 Bulletin of the University of Wisconsin, No. ji2, High-School Series, No. 6, pp. 7, 8. lO STATEMENT OF THE PROBLEM AND METHODS II on conditional terms after having done as much as three years of high- school work. The standings or marks which are found on these certifi- cates, or else the standings as recorded in the offices of the high schools, serve as a basis of the high-school and college comparisons. There is not absolute uniformity in these certificates, though in general they are alike. Some schools make a practice of sending only the grade made during the second semester of the year in any subject. Other schools average the standings of the two semesters' work in any subject for the year and place this estimate upon the certificate sent to the college. Occasionally certificates simply indicate that the student has passed in his high-school subjects but no grades are reported. Some of the schools which use the letters or figures do not interpret these in terms of per cent. Since it is very difficult to get, at present, a large number of pupils who have attended all three of these institutions, the other separate comparisons within the respective institutions of the grammar school and high school, and those between the grammar school and high school, and further those between the high school and college have all been used as a sort of check of investigations. Probably the most original part of this study is the comparison made between the pupils who attend all three institutions, since no one so far has done this particular piece of work. There are many other problems than this one of the relative stand- ing of pupils which could be worked out from this same body of material collected for the present purpose, and for those who care to do it, other comparisons could be made from the charts just as they stand. There is opportunity for comparing various systems of grading. The relative standing of the same individuals in various subjects could be deter- mined. From this same body of data a comparison of small high schools and the relative standing of the pupils of the large high schools could be carried on. One could test the standing of pupils in required and elective subjects. One could compare the standing and scholarship of boys and girls. One could tell accurately out of 5,000 pupils how many of them had pursued Latin, modern languages, or any other subject. But whatever by-products may come out in this discussion, they are all secondary to the main problem dealing with the relative standing of pupils in different institutions. The old-line subjects have been used exclusively. It will readily be observed that the subject of English has been most frequently used in the comparisons. It serves better than any other subject to give a 12 STANDARDIZATION OF SCHOOLS OF KANSAS long line on the pupil's work and school career. The required amount of English in most of these high schools is three years; a few, however, do four years' work. In the main, the comparisons between different institutions are carried on between the same subjects. English, how- ever, in the grammar school is used in a few instances as a basis for comparison with Latin and modern languages, as well as with English in the high school. Manual training was not general enough in the schools concerned as far back as it was necessary to go for some of the records in order to be considered. The method one uses in this sort of investigation is more or less determined by the nature of the data at hand. The practice of keeping school records involves many variable factors. Some schools record only yearly estimates, others record both semester estimates, still others record an average of the two semester estimates. Where the two semesters' marks were available the average of this was always used. The great majority of the first-year college records used represent an average of two semesters of work. There was great difficulty in deter- mining in many high schools and in some colleges in what year certain subjects had been pursued. The averages which have been used are usually averages in different years of the same subject rather than aver- ages of various subjects. One of the most disturbing aspects of this whole study has been the attempt to translate satisfactorily the various systems of marking into comparable forms. The bases of grading represented by the colleges are the letters a, h, c, and the figures i, 2, 3, and the ordinary percentage system. Some of the figures, however, are stated in terms of percent- ages; for example, in one institution i represents 90-100; 2, 80-90, and 3, 70-80. The marks used by high schools vary from the percentage system to the use of the letters a, b, c; e, /, g; e, g, m, p, with the occasional use of the plus and minus. The figures i, 2, 3, 4 are frequently used, too, with the plus and minus. In the grammar schools the per-cent system is used together with the letters a, b, c; e, /, g, together with e', e", e^, e"^, e^, indicating 91, 92, 93, etc. ; i, 2, 3, 4 with frequent use of plus and minus; I, I J, i}, i|, etc. A few schools insisted on not reducing the letters and figures to any percentage. The range of grading in grammar schools varied from 60-100, 70- 100, 75-100; in high schools, from 60-100, 70-100, 75-95, 75~"io*^5 ^°~ 100; in the colleges, 70-100 and 75-100. The range of grading will show clearly in the charts of chap. iii. STATEMENT OF THE PROBLEM AND METHODS 1 3 In interviews with the different school principals, superintendents, teachers, and officers within the same school there was frequently a lack of definiteness of opinion upon the interpretation of marks in actual use. One can scarcely avoid reaching the conclusion that there has been a great deal of ragged, haphazard, and lumping-off work done in this matter of rating individuals. If a marking system is to be of any account at all, it must necessarily be more than a random momen- tary decision. On the other hand, there was no school which did not have some strong point in its marking system, or in the keeping of its records. But it is not infrequently the case that one ward principal or high-school principal is very little in touch with his neighboring school, however good this system may be. There is a sufficiently large body of well-tested material here, if collected into a unified form, to furnish guidance and a working basis for an ideal school, both in practice and theory. One of the outcomes of these investigations ought to be the occasioning of free discussions as to the best methods of rating pupils, and of other questions vital to the progress of any school which pretends to be modernized. The schools will be numbered instead of named, since the study is intended to be as impersonal as possible. It will not, however, be possible to number all the grammar schools separately. They have been charted in composite form in the various cities and all have been given one number as representing the grammar school. The majority of the comparisons in the grammar school include only the eighth grade, and those in the college the first year. In one large city the seventh grade completes the grammar-school work. In this case the seventh grade in place of the eighth is used. A Umited number of cases have been traced through the sixth, seventh, eighth grades, through the high school into the university, and a comparison made between the standing of these pupils in the three d^'fferent insti- tutions. Wherever the standings of individuals have been indicated in terms of percentage the same graphic scheme is used as that found in the bulletin of the University of Wisconsin. "Each student whose marks or grades enter into this study has been assigned an individual number. .... The student's rank is indicated by placing his number above the proper grade in the horizontal scale of marks as arranged in the accompanying chart" (p. 15). When the range of grading is, for example, from 75 to 100 per cent, or from 60 to 100, then an individual's number is placed above the 14 STANDARDIZATION OF SCHOOLS OF KANSAS horizontal line over the grade which indicates his standing. After the distribution of marks is made in the charts, the individual numbers are divided into three equal groups or divisions called ''tertiles." Those numbers which occur in the high third in any chart which is used for a basis of comparison — for example, between the grammar- school English and high-school English — are starred in the high-school chart representing the high-school English. That is, if a certain number occurs in the first chart in the high tertile it will be starred in the second chart with which it is compared, no matterin which tertile it there appears. The numbers appearing in the original chart in the lowest third have a minus sign attached in the second chart, which indicates that originally this number had appeared in the lower tertile, no matter in what tertile it occurs in the second chart. The numbers within the original middle group appear in all cases in the second chart with which the first chart is compared without any signs attached. By this scheme it is possible to trace out any individual pupil as he passes from one institution to another, for illustration of which see sec. I in chap. iii. It is also possible to find his exact place within any tertile at any time. Through this number scheme of charting it is a simple matter to determine the percentage of retention of any group as a whole between one institution and another. The tertile division was used for several reasons. When the divi- sions are too many the perpendicular broken lines are likely to fall on the median or are likely to fall in columns where persons in a higher and lower tertile have really the same standing. The tertile grouping is more economical, and sufficiently accurate for a basis of measurement; and where a three or four or five estimate is used in grading the tertile grouping is large enough and even better than a fine division. It is better to place those who have the highest average at the top of a particular column which indicates the same integral per cent, because in some cases when the groups are divided it is necessary to divide the column by a broken hne. Wherever averages have been used, those having the highest fraction of any one integral percentage are usually placed at the top of the column.^ But where only a final grade, for instance in the first year of the high school, is available, it is not possible ^ That is, where it is necessary to use a broken line in dividing the tertiles it is fairer to put the pupils with the higher average in the higher group. In such a chart as No. 3, the column over 90 per cent was proportionately divided. This was done in instances when there was practically no difference between the standings of the pupils in the column considered. STATEMENT OF THE PROBLEM AND METHODS 15 to differentiate so closely in a column of figures which may appear over 86 per cent, for example. In case letters or figures were used for marking the standings of pupils these have been reduced to a percentage basis and then charted. Where there are only several estimates made in a scale of grading and when the number of cases being considered is large, it makes the columns high. So that it was necessary to break up these columns in the chart- ing and use an accompanying graph to represent the actual distribution of the marks in a given subject. The broken horizontal lines in this case indicate the number of persons receiving any one grade, and the graphs are reduced in size when the original is too large to be printed. A very few charts appear which represent absolute estimates rather than relative standings. A composite of 23 high schools has been used where a small number of pupils from each high school were represented in the same college. It was ascertained from the principals what percentages used by these various high schools would be equal to the i, 2, 3 used in the college. These percentages when translated mto the i, 2, 3, forms, were charted and compared with the standings of these same individuals in the college. The tertile tables used contain a summary of the percentages of retention. The following is a type of those used later on in chap. iii. 7 t ll G r a d e HlBtory School 5 etii Grade 1 2 3 % Tertile Retention 1 22 11 4 59.45 Z 11 15 12 39.49 3 4 12 21 56.75 Total Retention 51.78 A brief general description will here suffice, for these tables will be explained in fuller detail as they appear in the later discussion. The number 22 in the first table indicates that 22 persons who were in the first tertile of the group in the seventh grade in history remained in the first group or tertile in the history work of the eighth grade, or that there was a retention of 59 . 45 per cent of the high group. The number II indicates that 11 of the pupils who originally were contained in the first group fell to the second group in the eighth-grade history work, and the 4 indicates that 4 persons who originally stood in the first tertile in seventh-grade history fell to the third group in the eighth-grade history work. By reading diagonally across the table, the numbers 22, i6 STANDARDIZATION OF SCHOOLS OF KANSAS t •a 2 « s STATEMENT OF THE PROBLEM AND METHODS 1 7 15, and 21, the number of pupils retained in each tertile is ascertained. Or if 22, 15, and 21 are each divided by the equal number representing the three groups, the respective percentages of retention for each tertile are ascertained, namely, 59.45, 39-49, and 56.75. By such a table it is easy to summarize the retention of each tertile when any two charts are compared. This is designated as the ''tertile method." A supplementary device may be used for comparing the retention when three institutions or three years' work are involved. It is the use of a diagram which shows just how many pupils are retained within the original group in which they began, and also shows the nature of a pupil's progress after beginning work in any one year or in any one institution. Diagram I is an actual case, but the charts of this diagram are not included in this study. The diagram has the advantage of showing which way the original group as a whole either progresses or declines. It indicates not only the final classification of the groups, but it indicates the quality of pupils who constitute the groups. The rectangles at the bottom of the diagram show that there were originally 53, 52, and 53 pupils in the high, middle, and low thirds in an averaged sixth-, seventh-, and eighth- grade English. The dotted Hues leading from the first rectangle show that 39 of the 53 pupils who originally were classified in the high third rose to the high third when an average of the four years' high-school EngHsh was used; that 10 pupils fell to the middle third, and that 4 pupils fell to the third group in the high-school work; when the high-school English was compared with the first-year college English, 32 pupils out of the 39 who are retained between the grammar school and high school, are further retained in the high third of the college. By following the heavy dotted line leading from the high third it will be observed that 32 pupils out of the 53 originally in the first tertile held that place throughout the three institutions. By following the heavy continuous line representing the middle group, or third, it may be observed that out of the 52 pupils who originally began in the middle tertile 14 passed straight through in the three institutions. By following the broken line it may be seen that 23 pupils out of the original 53 in the upper tertile have passed through without deviating from the high group. It will be easy to trace the pupils from the second year, or the second institution, as the case may be, to the third year or institution, if it will be noted that the lines fall in groups of three at the top of the diagram. By a glance at any of the diagrams it may be noted that comparatively i8 STANDARDIZATION OF SCHOOLS OF KANSAS few pupils pass from one extreme to another. This fact may be seen also from the tables that accompany diagrams II, III, IV, V. The second general method used in the following comparisons is designated as the ''modified median method." The retention is ascer- tained by finding the average of the percentages of those pupils in the Name -- (Pupil No. 32) 1 1 (•04)E1 3m Set. Ho 5' ('07)H.S.No.5 (•11) Col lege So. 2 Aritii 6th 7th|8tll Math ■ Fr. So Ju. Se LJatn. Fr So J-u.j Se. 95 88 100 67 93 9. 81 87 84 Read. ^ 1 95 1 SO 95 1 89 Lat. j Lat : Gram. 93 1 90 09 56 Eng. 86 88 " s. Eng 84 93 R'i aer. 88 91 78 Ger 91 95 82 87 ?hys . 89 Phys Soi. Hist. 93 90 77 Hi3t. 88 Hist. 89 R4 Eiiyslol Bot. 92 Biol Sci high and low groups of one year's work or in one institution who in the successive year's work or institution remain above or below the median. The final conclusion will be stated in terms of this method. A very simply arranged card was printed for the purpose of pre- serving the records collected. The above card is an exact duplicate of pupil number 32. CHAPTER III COMPARISON OF THE RELATIVE STANDING OF PUPILS IN GRAMMAR SCHOOLS, HIGH SCHOOLS, AND COLLEGES Chap, iii includes an extensive and detailed statement of the com- parisons made on the basis of marks. Because of the length of this chapter and because of the many charts presented, it was thought ad\dsable to give the separate conclusions at the close of each section. The more general conclusions growing out of these comparisons have been brought together in a summary way in chap. iv. For convenience of treatment the comparisons in chap, iii, as has been indicated in chap, ii, have been divided into five sections. Necessarily there is some overlapping in the data used in these different sections. The first one is a comparison of the standing of pupils within the grammar school itself; the second, a comparison of the standing of pupils wdthin the high school itself; the third, a comparison between the standing of pupils in the grammar school and high school; the fourth, a comparison between the standing of pupils in high school and college; and lastly, a comparison of the relative standing of a hmited number of the same pupils who had attended all three institu- tions — namely, grammar school, high school, and college. The reasons for making these separate comparisons are probably self-evident. The writer believed that it would be of some importance to know what is the actual relative standing of pupils within any one institution itself. On the basis of such knowledge as this it would be easier to conclude what ought to be expected to be the relation between different institutions. In connection with each chart presented, certain significant facts will be pointed out, as, for example, the range of the scale of grading, the dif event forms of the distribution of marks over the scale, the shifting of groups as a whole, the interrelation of subjects within the same institution, the relation between grammar school and high school based upon a compari- son of different subjects, the relation also between the high school and college, noting in what cases the percentage of retention is the highest. Not all of these points will necessarily be taken up in each chart and section, but those which have especial significance will be briefly discussed. 19 20 STANDARDIZATION OF SCHOOLS OF KANSAS SEC. I. GRAMMAR-SCHOOL COMPARISONS ONLY Briefly stated, the object of this first section is to determine what is the correlation within the grammar school itself on the basis of comparisons made between single subjects. This has been used as a sort of check experiment for the later comparisons between different institutions. The first part of sec. I is a comparison between the standings of pupils in the same subjects but in successive years. The charts numbered from i to lo in sec. I represent the distribu- tions of marks received by pupils in the seventh and eighth grades in the subjects of English, history, and arithmetic in schools No. 5' and No. 7'. Some references will be made also to charts used in following sections.^ The groups of pupils in the charts are all divided into tertiles as nearly equal as possible. This fact holds throughout this thesis with the exception of a very few charts which have been constructed upon the basis of absolute marks. Chart I shows the distribution of marks given in the eighth-grade history of 112 pupils. There are 37- pupils in the high tertile or group, 38 in the middle tertile, and 37 in the lower tertile, or division. The numbers that are starred in this chart indicate that originally these same numbers represented individuals who stood in the high third of the seventh grade. Those numbers accompanied by minus signs indi- cate that originally these same individuals appeared in the lower third in the seventh-grade history work. Those figures which are not accompanied by any signs indicate individuals who were found in the middle group of the seventh-grade history work. The percentage of retention in the upper tertile is easily ascertained by dividing the number of starred individuals in the second chart who remain in the upper third or tertile by that number of individuals who were in the original chart within the high group. For example, in chart 2 there are 37 pupils within the high group. In chart i, as shown by the starred numbers, 22 of these same pupils remain in the high third of the eighth-grade history. When 22 is divided by 37 the reten- tion is found to be 59 . 45 per cent for the high group. The retentions for the other tertiles may be obtained in a similar way. ^ The records from No. 7' were difficult to secure for both the seventh and eighth grades. In school No. 5' records were available for a number of years back. The records for school No. 5' used here extend from 1902 to 1907; 33 graduates are repre- sented from 1907; 29 from 1906; 30 from 1905; 17 from 1904; 2 from 1903; and i from 1902. COMPARISON OF RELATIVE STANDING OF PUPILS 21 Above each chart the percentages of retention appear in whole numbers. In such summary tables as I, II, for example, the retention is carried out to two decimal points. By reading diagonally across the table, the numbers 22, 15, 21, for example, it is always possible to see at a glance the number of pupils who are retained within the respective tertiles. In case anyone desires to do so, he can readily follow out the career of a particular individual by means of the number which represents that pupil. In charts i and 2, number ^4 retains not only his position within the high group of the eighth- grade history, but he retains the same absolute per cent. Number 60 retains the same absolute grade within the middle group of the eighth-grade history. Number 4 passes from high third to low third, and number 10 from low third to high third. Any individual may in this way be followed out in all, or in any, of the successive charts. Some further details of the charts may be pointed out. Those from I to 10 show that there is some variation in the distribution of marks over the scale used, not only when the two schools. No. 5' and No. 7', are compared but when the distribution of marks within the same school in different subjects are compared. In charts 5 and 9 or charts 3 and 8 it may be seen that the range of the scale varies, being in school No. 5' from 75 to 100 per cent and in school No. Y from 60 to 100 per cent. The effects of probably a too extensive range of marks are illustrated in chart 5 as compared with the more successful grouping in charts 1-4 and 9, 10.^ It may be noted at a glance that although the same pupils are involved in both subjects they are as a group graded distinctly higher in history and in arithmetic than in English. And this holds true of both the seventh- and eighth-grade work. For illustration of this fact refer to charts 1-4, and 9, 10. This same tendency obtains in charts 38, 46, 48, used later on, which include in the 212 pupils there compared the same 112 pupils in these earlier charts. So that the same tendency obtains with an increased number of pupils from this same school. Whether such an arrangement of grouping as this noted above is just to the pupils is a question which will be raised again. While there is a tendency toward a normal distribution of grades in such charts as No. 3 and No. 4, yet in such charts as No. i and No. 9 there is a considerable "skew" in the curve of distribution toward the ^ While 60 per cent is the minimum grade in any one subject in school No. 7', yet an average of 70 per cent in all of the subjects is required for promotion to the high school. In school No. 5' it is not customary to average all the subjects together, and the required grade for promotion must be made in each subject respectively. 22 STANDARDIZATION OF SCHOOLS OF KANSAS top of the scale. All of the charts 5, 6, 7, 8 show a decided skew toward the higher end of the scale of marks. Unused marks in the scale occur quite frequently. Although the actual range of grading is supposed to begin with 60 per cent as the lower limit, there are very few marks appearing below 75 per cent. It might conceivably be answered that if the number of pupils to be considered were much larger than this, then these gaps would be filled up, and this is true in a partial sense. But chart 55 in an advance section includes these same pupils in a group of 270, and although some of the gaps are here filled up, comparatively History School No. 5' gtX Grade . _ 1 ■J English School No. 5' 8th Grade EngliBh School No. 7 §th Grade 9 1 ?. ?> Ter. Ret. 1 2 3 Ter. Ret. 1 2 3 Ter. Ret. ). n 11 4 59.45 ?? 8 6 $2.76 14 8 4 53.84 ? 11 i? 12 39.49 11 3 11 13 10 ai 44.84 5«.75 54.46 10 W 6 35.74 2, 1?, 21 ?6.75 -2 8 J« 61.5? Tot. Ret. 51.78 Tot .R it. Tot.Ret. 51.28 .Arithmetic ^ School No. 7 eth Grade Ter. Ret. 57.69 iL^ Table 1. Showing retention ■between seventh and eighth grade worlc. very few are to be found below 75 per cent. One question which this provokes is: What effect does this have upon the percentage of reten- tion ? This point will be raised later on. Table I is a summary of the percentages of retention in dealing with a comparison of the relative standing of pupils between the seventh- and eighth-grade work, according to the tertile method of grouping. There is some difference in the percentage of retention between the two schools No. 5' and No. 7' in arithmetic. The higher percentage of retention in school No. 7' in arithmetic at first thought argues for a some- what closer correlation between the seventh and eighth grades. But it may be the effect of the wide range of a too detailed scale of grades used. From table I it may be seen that the correlation is the higher between seventh- and eighth-grade English in school No. 5', but that the correla- tion is the higher between seventh- and eighth-grade arithmetic in school No. f. COMPARISON OF RELATIVE STANDING OF PUPILS 23 Another convenient method for ascertaining the relation between the seventh and eighth grades is to determine what percentage of the pupils in the high and low group in the seventh grade remain above or below the median, respectively, in the eighth grade, and then to find the average of these two percentages/ For the subject of English in school No. 5' the retention is 72.97 per cent for the upper group and 81 .08 for the lower group, the average being 77 . 05. In arithmetic it is 72 . 97 per cent for the upper and 62.16 for the lower group, the average being 67 . 56. In history, for the upper third it is 81 . 08 per cent, and 78.37 for the lower third, the average being 79.76. For the subject of English in school No. 7', the retention by this same method of comparison is found to be 69.23 for the upper third and 84.23 for the lower third, the average being 76.73. In arithmetic it is 80 . 74 for the upper and 69 . 23 for the lower third, the average being 74.98 per cent.=^ The percentage of retention is the highest in the subject of history in school No. 5' in terms of method No. 2, namely, 79+ per cent. In school No. 7', it is highest in the subject of English, the average of the high and low third being 76+ per cent. These results indicate that there is a retention of at least 75 per cent in the majority of the subjects compared, in terms of the average of the percentages of those pupils in the upper and lower tertiles who remain above or below the median. By the use of these same charts a brief comparison was made between the standings of the same pupils in different subjects in the same year. Since the subject of English was used as the basis for the majority of comparisons, it seemed worth while to find out by a few comparisons whether pupils have a tendency to be equally good in all subjects, or whether pupils who are good in English might show quite a different amount of capacity in other subjects. If the pupils who take Enghsh do equally well in the other subjects then the frequent use of English as a basis of comparison in this thesis will appear somewhat more justifiable. Chart 4 representing seventh-grade Enghsh is compared with chart 2 representing seventh-grade history. Chart 3, or eighth-grade Enghsh, is compared with chart i, or eighth-grade history. Eighth-grade ' For definition of median, see Professor Dearborn's Bulletin on Relation of High School and College, p. 17. * For convenience the first method of comparison used will be referred to in this thesis as the method No i, or the "tertile method"; the second method used will be referred to as method No. 2, or the "modified median method." 24 STANDARDIZATION OF SCHOOLS OF KANSAS English is compared with eighth-grade arithmetic, and seventh-grade English with eighth-grade arithmetic. These are the same 112 pupils in all cases, but the correlations are not shown in separate charts from those used in the first comparisons. The percentages of retention for this comparison are shown in table II. The retentions for charts 4 and 2, for example, are as follows: 54.05 per cent for those in the high third in the seventh-grade history. The percentage of retention for the lower third is 56.75, and the total is 51.78 per cent. That between the eighth-grade English and eighth- grade history is a total of 53 . 57, and that between eighth-grade English and eighth-grade arithmetic is a total of 52.67 per cent. While that between seventh-grade English and eighth-grade arithmetic is lower than the others, it is high enough to be significant. s 1 Hlst. Sch. K0.5 7th Grade __ ^ I 00 HiBt.J 8th Gj Soh.Ko.S ade ■a Arlth.Sch.No.E 8th Grade 1 Aa:ith.£ch.Fo.5. 8th Grade 1 1 2 3 ler. Ret. . 2 3 Ter. Ret. 1 2 3 Ter. Ret. 1 2 3 Ter. Ret^ 1 20 12 5 94.09 21 13 3 56.75 22 11 4 59^45 1 23 8 6 62.76 2 10 17 11 44.73 11 16 11 42.10 11. 4 2 11 21 42.10 96,76 11 14 13 36,05 2. 7 9 21 S».7» 5 9 23 62,76 3 16 18 48.64 lot. Ret. 51.78 Tot .Ret. 53.57 Tot .Ret. 52.67 Tot. Ret. 49.10 TABLE II Showing the retention of pupils between different subjects in the same year. In table II the total retention between the seventh-grade English and the seventh-grade history is the same as is the total retention in table I between seventh-grade history and eighth-grade history — namely, 51.78 per cent. The total retention when different subjects are com- pared is above 50 per cent, and if the previous method of comparison other than that of the tertile grouping is used, the retention here again would be about 75 per cent. And so from this limited comparison of 112 of the same pupils in different subjects the result is that a large number of those who do well in one subject are likely to do well in another subject, and that those who do poorly in one subject will be likely in large numbers to do poorly in another subject. It may be seen from table II that there is very little difference between the retention in the seventh-grade English and seventh-grade history, between the eighth-grade English and eighth-grade history, and between the eighth-grade English and eighth-grade arithmetic. The total retention for the eighth-grade English and eighth-grade his- tory is a little higher than is the total retention for eighth-grade English of the same pupils and eighth-grade arithmetic. COMPARISON OF RELATIVE STANDING OF PUPILS 25 From both of these comparisons, then, between the same subjects in (lifer- ent years, and between diferent subjects in the same year, within the grammar school, the results show a retention of about j^ per cent in terms of the aver- ages of the percentage of those pupils in the high and low groups who hold their places above or below the median. <£|K hA Sa 11 i,^. I ♦ " 1 II M 10 »<^^ CO 7> « o - obb o s» H »jl 00 M N|tf» ^ ta lO 30 E!e 3S Hnt to CO HO O it 3 r o 3 3 J> ^ 10 » Jl t~ n jf r la a> la S s o n 1 » ll S S: ll O f- g J * " - N«0 «> i> ^^S o N^ 1 r N o o K <0 0» IC e- B- A f » ^ wt- NIC «0 •^l O t :5S (0 « 1- WlOO ^S8 01 lO S ■" s g I ^S2 g s ^SS ssg ^ ft N'.ie- DO .- N« OO le T CQ ooi if n g Ot n » NK CO g II n ^ o U kj. 6- ^^ g:£ «o 6- oo to ( « DO M H r- « -^ ^r a> ' o: n H OO o tctl^ lO H flO«e ole- CO N NtQU «(>«o OO oo r CO lO CO «i :! 1 oo M CO 3,8 5 5 3 s S N H § rl CD 4^ s lO If s »* e- J n S o 6 > !- < t! s ' e 9» c ^ s < DO c 10 6' J j' e^ > > :; H u> S 10 ^. s 4 s 01 £: B* I ^ s ? s s; g 4 1 t c s ,1 i li 1 2 s 1 ) oo H ^•(5 S 1 7 o «3 3 d-lUJr- a o 28 STANDARDIZATION OF SCHOOLS OF KANSAS SEC. II. HIGH-SCHOOL COMPARISONS ONLY The object in making the comparison within the high school was the same analogously as that stated for the comparison of the pupils within the grammar school. This first part of the comparison between the standings of pupils in different years and within the same subject was made in order to determine if possible the retention from year to year within the high school itself. For in order to be able to make any judgments as to what we should expect the retention to be between different institutions — as for example, between the grammar school and high school — ^it will be valuable to know what the retention is between the different years within the high school. A very few comparisons have been made with respect to the relative standing of pupils between different subjects pursued within the high school during the same year. This will enable us to see whether the pupils who stand well, or mediocre, or poor, as they pass from one year to another within the same subject, also stand respectively the same in other subjects. Charts 11-38 represent the marks of pupils in three different high schools in the subjects of English, Latin, and mathematics. The sub- jects are charted separately for the different years. English is traced throughout three years; that is, the standings of pupils in the first- year English are compared with the standings of the same pupils in the second year. The standings then of these pupils in their second year of high-school English are compared with their standings in the third - year English. In mathematics and Latin the comparison is made between only two different years of work, because reliable records were not available for a longer period of time. After the distributions in the different high schools were effected a composite was made for these subjects in the respective schools. It will be noted in charts 11, 12, 13 of school No. 8 that the range of the scale of grading differs from that in schools Nos. 5 and 9. But since so few individuals appear below 75 per cent it was thought unobjectionable to composite the three schools. Some of the results of the first part of the comparison will now be discussed. In looking over charts 11-22 the distribution of the groups throughout the scale of grades is interesting, but one is tempted to ask whether it is justifiable. In general, the pupils are graded as a group somewhat lower in the Sophomore year than they are graded in the Freshman, and again somewhat lower in the Junior than in the Sopho- more year. If the group is to be shifted at all — and it should be kept in mind that this is identically the same group throughout in each COMPARISON OF RELATIVE STANDING OF PUPILS 29 school — might it not better have been done in the opposite direction ? Is this to be accounted for by the fact that pupils do poorer work as they advance; or is it due to a different standard of grading; or is it due to the fact that only a part of all the students who took the work at any one time are here represented; or is it to be accounted for in other ways ? The group in charts 11-13 starts, as is seen in chart 11 with a mode about 90, becomes somewhat bimodal in the second year, with a minor mode at 83; this second mode is then shifted to 78 in the Junior year. The retention of pupils in their relative positions remains high despite the changing in the total appearance of the group, as may be seen in table III. This same sort of shifting of the groups which has been pointed out in the charts 11-22 representing the marks in English also occurs in a general way throughout the Latin and mathematics work, as shown in the charts from 23-38. The distribution of the groups in the subject of Latin toward the lower end of the scale is clearly noticeable in schools Nos. 9 and 5, shown in charts 26 and 28. One-third of the pupils in chart 28 appear between 75-78 percentages inclusive in a scale from 75- 100; and in chart 26, one-third between 75-79 percentages inclusive; and there is a peculiar bunching of grades in charts 23-26. In the sub- ject of mathematics, school No. 8, as was the case in Latin, shows probably the least variation in its grouping of the same pupils in the Freshman and Sophomore years. As may be seen by comparing charts 33-34 with charts 31-32 and 35-36, a bimodal division in general is noticeable in the charts for mathematics, with some tendency to a rec- tangular distribution of grades. Turning to the composite charts in EngHsh represented by Nos. 20, 21, 22, and also to the composite charts in Latin represented by Nos. 29 and 30, or to the composite charts in mathematics represented by Nos. 37 and 38, it may be seen that the grouping in the marks of suc- cessive years of the high school after the first year is always toward the lower end of the scale. But the variation is more extreme in the case of mathematics and Latin than in the case of English. There is much discussion about the pupils of high schools being dis- posed to drop certain subjects because they prefer other subjects. Is it possible that the facts shown in such a chart as 30 or 38 explain some of the tendencies of pupils to drop certain subjects ? What explanation, fair to the pupil, is to be offered in view of the different distributions of the groups in the subject of Latin as shown in charts 38 and 30? 30 STANDARDIZATION OF SCHOOLS OF KANSAS Is this sort of grouping due to the difficulty of the subject itself ? Is it due to the dulness of the pupils ? Is it due to the fact that the subject is not taught as well in the second year ? Is it due to a radically differ- ent standard of grading ? Or is this shifting of the group to be explained apart from such factors as these? The facts, at any rate, justify an insistence on some sort of legitimate explanation. The sort of distribution of grades in charts 25 and 26, and also in 27 and 28, might well enlist the attention of classical teachers who are interested in having the classics maintain their position in the high-school curriculum. What more effective means or ways could be found for discouraging pupils from pursuing further work in Latin than that employed in school No. 5 ? It is very improbable that there is any reasonable justification for handling pupils in such a manner as this. Although the students have kept the same rank in relation to each other to a fair extent, the absolute grade of a large number of pupils has been arbitrarily dropped in the second year's work. I say arbitrarily advis- edly, because on what grounds can it be assumed that a group of this size, of over 200 pupils, as a group is less fitted for the work after a year of prel'minar/ study than at the start? Since as a group their standards of work and effort have not changed, most probably it is merely an arbitrary change in the teachers' standards. The student who receives a considerably lower grade with the same expenditure of effort, and does not appreciate that his rank in the group has not changed materially, might very well conclude that Latin was not his forte and consequently drop it. By referring to charts 40, 47, and 50, which appear in a later section, it may be seen that there is a difference in the distribution of grades in the Sophomore English, history, and mathematics. This group of 212 pupils shifts about from a skewed distribution in English toward the bottom of the scale up to a rather skewed distribution toward the top in the subject of history, and finally with a rather equal distribu- tion of marks over the scale in the subject of mathematics. The modes in chart 50 are noticeably different from those in charts 40 and 47. The very frequent bimodal division occurs in chart 50 with a large number of marks over the lower limit of the scale. The percentage of retention between the first-year and second-year English work in charts 11 and 12 for school No. 8 may be seen by refer- ring to table III. The retention is higher between the second and third year of English than it is between the first and second year. One explanation of this may be that students upon entering the high school COMPARISON OF RELATIVE STANDING OF PUPILS 3 1 need part of the first year to get accustomed to the new order of studies and practices. Charts 14, 15, 16 represent the standings of pupils in a large high school not in Kansas.^ The total retention in school No. 9 between the second and third year is higher than is the total retention between the first and second year of English. This is similar to the case above in school No. 8. The retention between the second and third year is higher, however, in the case of school No. 8 than in school No. 9. This may be due to the fact that the pupils in the former school are likely to have fewer distractions from school work. It may be that a more select group of pupils has been used than were chosen from the other schools, respectively. The retention in school No. 8 is also higher than in school No. 5. One probable reason for the lower retention in school No. 5 may be the crowded and cramped conditions of the schools, and consequently this involves something of the general administration of the school.^ If we compare the table of charts 12 and 13 with the table of charts 18 and 19, however, we can get a measure of the progress from one tertile to another. For example, in" school No. 8 eight pupils go from the lower third to the middle third in the Junior year and one pupil goes to the high third, while in school No. 5 twenty-eight pupils out of the lower third in the Sophomore year go up to the second third in the Junior year and five go up to the high third. So that measured in terms of progress made by pupils, high school No. 5 stands propor- tionately higher than high school No. 8 on the basis of a single subject. Table III shows in the composite charts for English that the reten- tion is higher between the Freshman and Sophomore year than between the Sophomore and Junior year. This is just the opposite of the results found in comparing the schools separately. But in these composite charts a group of pupils were taken from school No. 5 that were not originally included in this separate school comparison. So that this is one probable explanation of this higher retention between the first and second year, namely, that this later-added group of pupils were better adapted to the standards of the school and in particular to those of the individual teacher.^ ^ The graduates from school No. 9 are all from the school year of 1908-9. Some of them are mid-year graduates and some of them June graduates. * The graduates who compose charts 17, 18, 19 are scattered from the years 1905 up to and including 191 1. 3 Such an adaptation as this may sometimes involve the changing of the previous standards, but if it brings more profitable results why would this not be a legitimate procedure ? 32 STANDARDIZATION OF SCHOOLS OF KANSAS From the tabulated results of the comparisons made between the same subjects in the different successive years, as shown in table III, it may be seen that the total retention is for the majority of the schools between 50 and 60 per cent according to the tertile method. The total retention for Latin between the first and second year of the high school Table III, showing relative standings of I 8ame euhjects in different years. I, S. pupils within the i A a I X A 1 H.S.Jo'.8-C!liartB 11.12. i i H.g.Ho.8- ChartB 12,13. 1 n s [i.S.No.9 GJharte l4.ie 6 i H.S.Ho.9 Charts 15,16 . Sbpti. Eng. up. Eng. Saph. Ehg. Jun, Eng. X « 3 Ter. Ret., 1 2 3 Ter. Ret. 1 2 3 Ter. Ret. 1 2 3 Ter. Ret. 1 28 12 2 66.66 33 8 1 78.57 32 16 5 60.39 33 12 8 62.17 2 13 18 11 42.85 8 26 8 61.94 16 21 17 38.88 14 29 11 53.70 a 1 12 29 69.04 1 8 33 78.57 5 17 31 58.49 6 13 34 64.13 Tot. Rct.^ 59.52 Tot. Ret. 73.01 Tot.Ret. 52.50 Tot.Ret, 60.00 H.S.No.S H.S.lfo.5 Charts 18.19 i A Composite of H S-Charts 20.21 i Composite of H ®9h. EnK.. Jun. EnK. Soph. EnK. Jun. Ehk. 1 2 3 Ter. Ret. 1 2 3 Ter. Ret. 1 2 3 Ter. Ret. 1 2 3 Ter. Ret. 2- 3^ 2i 3 22 23 2^ 40 56,f33 35.71 56.33 51 16 4 71.83 B6 55 20 64.45 131 69 21 62.08 16 29 26 41.42 61 84 66 39.76 62 81 68 38.66 5 28 38 53.52 14 72 126 59.24 18 71 12? 57.81 Tot. Ret. 49.52 Tot.Ret. 55.66 Tot.R et. 54.50 Tot.Ret. 52.76 H.S.iIo.8 Charts 23,24 H.S-.1J0.9 Charts 25,26 p=. H.S.No.6 Charts 27,28 a 1 Composite H.Sa Wos. 8,9,5,0 Charts 29,30 S*J? ^,at S-oph. Lat. Saph. L it. 2oi;h.- La-t i- 2. .A, T.p»t I- 19 10 2. 13 ^ T.Ret 61.29 47. 20 2_ 18 28 3_ 25 T.Ret 65.28 31.36 r T- — T.kTt 1- 30 -1 79.47 107 3R 14 68.68 2^ -1 19 12 50.00 7 43.33 47 IP 3fi 45.80 3- 1 11 26 68.42 2 10 19 61.29 5 27 40 55.55 ^ 48 qR 60. S9 58 .'3T :Lot_.. Ret. 65.7^ To..Ret. 54.83 tS" t:^ 52.99 .Ret.- H.S.lic Charts .9 31,32 ■*■'■■ ■ 5 1 S H.S.lTo.8 Charts 33,34 Soph, iiath. 5 ft H.&.1I0.6 Charts 35,36 _Soph. iiath. 5 2 ft Composite H.S". Noe . 9,8, 5 1_ 25 10 j5- T.Ret 64.10 2^ 2.3 2_ 15 4 T.Ret 54.76 1_ 48 2_ 17 A. 6 t.rST 67.69 i?i f.t5T 61.72 2_ ii. ii 10 46.15 13 14 14 34.12 12 34 24 48.57 49 91 57 46.19 J 3_ _3_ ot. 11 26 tT 64.10 6 rot 13 23 t. 54.76 48.00[ 11 Tot 18 42 t. 59.15 58.49 rart 54 .Re 116 t. 59.17 55.68 is higher than for either Enghsh or mathematics. This may be due to the fact that the work done in first-year Latin connects better with the second year's work than in the case of the mathematics and English. Another method besides the tertile method may profitably be used in showing the percentage of retention of pupils between the different years of the high-school work. It is the method, already indicated in sec. I, of finding the average of the percentages of those pupils in the COMPARISON OF RELATIVE STANDING OF PUPILS 33 highest and lowest tertile of a group in one year of the high school who remain above or below the median, respectively, in the advance year's work. For illustration, in chart 11 there are 42 pupils in the higher and lower thirds, respectively. The median in chart 12 occurs at about 89. Of the starred people who come from the high third in chart 11, 37 remain above this median. If 37 is divided by 42, the retention for the upper third is 88 . 09 per cent ; and for the lower third it is also 88 . 09 per cent. The average of the retentions of the upper and lower third in this case is then 88.09 P^^ cent. Now on the basis of this method number two, or the modified median method, there is a retention of about 80 per cent in the comparisons made between the different subjects within high school No. 5; of about 85 per cent in school No. 8 ; of about 75, in school No. 9. For actual percentage see footnote below. ^ When the composite charts for the subjects of Eng- lish, mathematics, and Latin are considered, the percentages are similar.* It was thought that it would be suggestive to try out a few compari- sons between the different subjects within the high school, as was done within the grammar school. In the comparison of the grammar school, high school, and college, later on in this thesis, it will be observed that the subject of English is largely used as the basis, and chiefly because no other subject is likely to be studied so continuously for a period of years. In view of using English as the basis of comparison, it seemed desirable, as indicated above, to make some comparisons between the standing of pupils in English and their standing in other subjects. The comparisons here made are very brief and will need to be carried farther in order to get conclusions that will be valid in any extensive way. However, the groups considered are large enough to be suggestive at any rate. ^ The actual percentage of retention for school No. 5 between the first- and second- year English is 76.05; between the second- and third-year English, 80.84; between first- and second-year Latin, 81 . 24; and between first- and second-year mathematics, 76.80. In school No. 8, between first- and second-year Enghsh, 88.09; between second- and third-year Enghsh, 90.47; between first- and second-year L^tin, 86.83; between first- and second-year mathematics, 71 .42. In school No. 9, between first- and second-year English, 72.63; between second- and third-year English 75.45; between first- and second-year Latin, 70.96, and between first- and second-year mathematics, 87.17. ' In the composite charts of the three schools the actual percentage of retention between the first- and second-year Enghsh is 78.40; between second- and third- year English, 79.38; between first- and second-year Latin, 82 .04, and between first- and second-year mathematics, 75.50. 34 STANDARDIZATION OF SCHOOLS OF KANSAS Table IV shows a summary of some comparisons made m the sub- jects of English, Latin, science, mathematics, and history. Most of the pupils considered are represented in school No. 5. The charts are not presented in this thesis. While it would be more significant to compare the results of science and mathematics to that of English and mathematics within the same school, yet it is of some importance to note that the relation between science and mathematics in school No. o is somewhat closer than that between English and Latin in school No. 5, and also closer than between English and mathematics in school No. 5. AstablelV shows , the rela tion H. 3. Ko. B Ter. i&iSSl 61 f p 7 54.94 H. 3. Ho. 6 gi 8I22I4I 21 Tot .Rat. Ter. Rel, 57^ T^O- ? 21 3£. smm Tot .Ret. Ter. £2^ gOfW 56^ 56.26 TABLE IV Shows relation between different subjects within the high school. between English and history in school No. 5 is higher than that between the English and mathematics. However, the differences in retention between these various subjects are not after all so great as to warrant the conclusion so often made that the majority of pupils who are either good or mediocre or poor are likely to be strongly the reverse in other subjects.^ For when the average of the percentages of those pupils from the higher and lower tertiles who stay above or below the median is secured, the result in the majority of the comparisons made is a reten- tion of over 75 per cent.^ On the basis of the results of the comparisons made between the same subjects in different years within the high school, and on the basis of the brief comparisons made between different subjects within the high school, ^ The above results agree with the conclusion of Walter R. Miles in an article on "A Comparison of Elementary and High-School Grades." "The rank which" a pupil "receives in any one subject will represent the rank which he receives in all subjects" (p. 22). 2 The actual percentage of retention expressed in terms of the median method is for school No. 5, between Freshman English and Freshman Latin, 78.68; between Freshman English and Sophomore history, 77.46; for school No. 9, between Fresh- man science and Freshman mathematics, 72.62. COMPARISON OF RELATIVE STANDING OF PUPILS 35 it is fair to conclude that there is in actual practice a retention within the high school of approximately 80 per cent. The above result is stated in terms of the modified median method. On the other hand, based on the results of the comparisons made between the same subjects and between different subjects, according to the "tertile group" method, there is a retention of at least between 50 and 60 per cent within the high school itself. By means of a diagram it is possible to get another measure of reten- tion not necessarily in terms of percentage. Diagram II shows not only how many persons are retained straight through three years of English work, but it also shows the amount of shifting and retention that has occurred within the' groups — high, middle, or low, respectively. Diagram II for school No. 8 shows that 24 pupils went through three years' work without going out of the high tertile and also 24 other pupils went through the same number of years without going out of the lower tertile group; 28 pupils out of the 42 in the high tertile of the first year's work remain in the high tertile of the second year's work; 12 of these same pupils pass down to the. second terale; and 2 of them, to the third tertile in the second year's work. The diagram is simple, providing the reader keeps in mind that by a glance it may be seen that the arrangement of the numbers in the second year's work, namely, 29, 11, and 2, in the lower tertile indicates that 2 pupils have come from the high third, 11 from the middle third, and 29 from the lower third of the first year's work. The tertiles at the top of the diagram are divided into three sections, which may be seen and interpreted at a glance. For illustration, the numbers 24, 9, and i, representing pupils in the third year, indicate that these pupils have come from the lower tertile of the second year's work; 4, 3, and i, or a total of 8, have come from the middle group of the second year. A convenient means of indicating what happens to any one pupil is by the use of figures placed opposite the numbers that represent different pupils. For example, let 46-1, 46-1, 46-1 indicate the fact that this pupil has maintained his position within the high third of the group throughout the three years of work. Similarly, let 24-2, 24-2, 24-2, or 27-3, 27-3, 25-3, indicate the position of two other pupils in the middle and lower groups throughout three years of English work. The following is a summary way of indicating the positions of 126 pupils throughout the three years of high-school EngHsh. From such a table as this it is a simple matter to construct the above-mentioned diagram. 36 STANDARDIZATION OF SCHOOLS OF KANSAS ~4 «» 1T1 Lr V--1 'o 1— 1 bf) -i g 2 < M p< o -in < ^ o COMPARISON OF RELATIVE STANDING OF PUPILS 37 This tabulated scheme shows not only the standing of the class as a whole, but any one individual's relative standing in the three years' work can be seen at a glance. Both the diagram and this scheme show clearly that proportionately more of the upper- and lower-third pupils pass straight through than Relation of the low-l Relation of the =>id- Relation of the high er tertile of first | die tertile of first tertile of thrf first year English to eec- English to second year English to sec- ond and third year | and third year Eng- ond and third year English lish English 1 24 3 3 3 29 2 2 2 46 5 3 3 3 31 2 2 2 72 103 3 3 3 50 2 2 2 74 1 112 3 3 3 70 2 2 2 79 1 120 3 3 3 101 2 2 2 47 125 3 3 3 106 2 2 2 49 4 3 3 3 114 2 2 2 71 105 3 3 3 68 2 2 2 80 109 3 3 3 94 2 2 2 51 58 3 3 3 107 2 2 2 76 13 118 3 3 3 3 3 3 35 ^ 2 77 2 90 2 2 1 1 3 3 3 27 2 2 1 22 36 3 3 3 19 2 2 1 87 116 3 3 3 96 2 2 1 93 15 3 3 3 14 2 2 3 59 25 3 3 3 43 2 2 3 67 104 34 3 3 3 3 3 3 23 100 2 2 3 1 69 48 ^- 1 84 3 3 3 30 2 1 1 88 55 3 3 3 33 2 1 1 69 121 3 3 3 82 2 1 1 83 28 3 3 3 97 2 1 1 40 61 91 3 3 3 113 92 2 2 1 1 1 1 26 111 3 3 1 9 3 3 2 95 2 1 1 117 21 3 3 2 119 2 1 2 110 65 53 3 3 3 3 2 2 39 78 2 2 1 1 2 2 86 54 1 2 56 98 3 3 2 1 64 115 2 2 1 1 2 2 52 75 2 2 3 1 99 3 2 2 45 2 1 2 81 2 102 3 2 2 17 2 3 2 85 2 2 6 3 2 2 124 2 3 3 60 2 2 10 3 2 2 73 2 3 3 66 2 2 108 3 2 2 122 2 3 3 32 2 2 37 3 2 2 12 2 3 3 62 2 2 57 3 2 3 123 2 3 3 63 2 2 8 3 2 3 126 2 3 3 11 2 2 16 3 2 3 38 2 3 3 20 2 2 41 3 2 3 44 2 3 3 42 2 3 3 3 2 3 18 2 3 3 7 3 3 is the case in the middle group. It is possible, therefore, to determine in many cases what pupils are likely to do after their first year of high- school work by such a graphic scheme as this. For it has been seen that a large proportion of the pupils who do well, mediocre, or poor in the first year of high-school English are likely to be similarly grouped in the other years of their high-school English. 38 STANDARDIZATION OF SCHOOLS OF KANSAS ~ ^ F ■ — :l!S|^ !o > . lo* .-,» '^ s «n » » r-^ s> D aco B-OH CO » «l>>0> •■^^«- 0» y^ (F » » • » o. oa to t> SS^S: ss: '« la ooa o> » ^ 10 r •(►■ <0 r- 0> cn u fre> o» "^ OICV ca iO ■«. > » » » » "•s^assss g »- l» u Mtfiar-ln a » » » lO a to (DC 10 oo 1- g!SSS§SS I ^s^i^g^ss!; t- c- 00 COO m '«>C 1^ f » f » M D ss S s :^eSSS3g c ^ ssisggs * t. » 1 » » ► V f » ( F » » •5 H e- A loa o ■vj <0<»r- <0 to e- o to t- CM ■o 00 to 00 en 5 flOO o ■h 9> r« in MlOf §SS3 ■^"'^r^plpl'' kksssss o> o Hi-t -{ b o « CM f r >o <^ 3 cH rHr 0» o t~ e- 00 o m -DIJ>COri2HO» H Otf I"* » 5SS^gSS .- Hn r H ss lOd m s 10 ro g g co to oo rt'Hx, to 00 CO *suss O S3 lO t. "*" >00 \a cio>Ht!- O ;sse|^§ (0 to U3C o oo CMtOIOOO ■^ V c ^ X} . 1 H >eM f- 4 D 3 ?■ sf r-4 H H i S ti f « Ao I on MO a n Ol o > O -1 M M e> H ^ H H Ct| I tI ifi ici AJ^ toot H ST" H <0 9> ^i .^10 lo a O 9t icic «a s »%!S;2 H 101 Ok CO o* «l|iO ^^ ^s§ as 5 Hi-«HH ZZ9 as?! sir 40 STANDARDIZATION OF SCHOOLS OF KANSAS COMPARISON OF RELATIVE STANDING OF PUPILS 41 ^3 SSS^ 10 CO moo oto^ CD ^CtO ss§ o»oe» WWiN la 10 u) 10 1 >Hi-iM ao> wco< ! OOCDOIWI Jji. oico'oou 00 woe «k IJ MM 00iO>M,N«l^l ^[idIiaIu! CO 00! U>lo»0, jJJJhIJhIJJo '^1 ^ u>; b-; a>' o^ O o n c^ eQ;tO{ N ^, la; to 10 lO 42 STANDARDIZATION OF SCHOOLS OF KANSAS . 4oi r- ed« SSSi t t H as 5? e- M» On ss?i: ^^ 2^ ■Si a SOI H C4 CvlM Ol ^(i .&»s oc a sgs SSS' s^a S.tA IQ gE ^li SkO ^^ g N to r( IQ w N tonoo ii C M J_U iJ>H'c 8i> SI ?31 ss i^ra 4*« 53 iQco r 00 H J» » *» 9 to HI I- M - VC «0 ;2S ^ s H M 2^ COMPARISON OF RELATIVE STANDING OF PUPILS 43 44 STANDARDIZATION OF SCHOOLS OF KANSAS la Si u\ S3 SS SS M Ss^ I s 3 H ssss QM iSioio SI8|a§* KON>aiO §§ls SI 3"* o s as 'din »■ £iS CM S s s? Ot«0 M as'- COMPARISON OF RELATIVE STANDING OF PUPILS 45 I 41 CEO ilj. Nl Ttl lOjlO lO^I wSw ^ to H Id H H r^ 52 COM /) oitr.t- i2« 3m HWNHW 35 <*H- 46 STANDARDIZATION OF SCHOOLS OF KANSAS sss I3g ^tt Si^S as t^ CO 92: ;?a ss g&i ^1010 ss 0>0>OOHO <0 CO lO i T* CC ^ r* 1 H tC I lO lOoo to ^< o» cceoa>>oc-oit» m I lO e-j to cr> t» < cMOtCvlcv)kO(OOoa>r-i>0 00 H tO'^O H t J us «3 IT- to M lO I Jo»ou) wtoi- 1^^ T^^ia loi I ijio iA> fHCvINbOUSiauXOtOt HM'*C w t^ in I , , liAdOiAt lev] ^- 00 c- o> < HOO^3CJ^-00>*C )00>ONH«OU>HQO I ■* ^ •»!< >* ^ in 1 J>^ •ui Jt^ i-^-^'^ioiotaiO'OK 0> H CON rN4.. u^ <4< ' Pi HHHh i>4 ^S H W < ss ■"CIO n «He ^0> 3 COMPARISON OF RELATIVE STANDING OF PUPILS 49 AJ> SS ^4 01 Jb o «< ss 5i5 4i ass n H H HHHN ii si c^J. ^?: a ;2 33 SS: ^3 HMto 50 STANDARDIZATION OF SCHOOLS OF KANSAS ;?^ mS 22 S2i£ S^3 Slg COMPARISON OF RELATIVE STANDING OF PUPILS 51 i4 S2 10 M SSI U3 CJ> O CJ a Jici ^s cit ss cici.i44i c^A^ J,ci to « J: 16^ U 4.4ti .4 J; 6A 6>h 52 STANDARDIZATION OF SCHOOLS OF KANSAS SEC. III. THE RELATIVE STANDING OF PUPILS BETWEEN GRAMMAR SCHOOLS AND HIGH SCHOOLS These comparisons are made largely within each school separately, since it did not seem plausible to bring them together clearly in one general comparison. While the main purpose is to find out the corre- lation between the grammar school and high school through a compari- son of pupils in single subjects, yet any facts that may corroborate or weaken points already brought out in previous sections relative to dis- tribution of marks will also be discussed. Charts 38^-88 inclusive represent the distributions of the marks of pupils in the grammar school and high school.^ Some of the schools here involved did not use the percentage system of marking. In such cases, where the number of pupils is large, the actual distribution of marks is shown by accompanying graphs, and the columns as orginally charted are broken up into convenient forms for printing. It was impossible to chart separately in tertile groups each of the ward schools of the various cities. Consequently a composite chart for the ward-school marks is used to represent the grammar school. For example, in the first comparison which follows, the high school is represented by 5, and the composite of the ward schools of this city is numbered 5'. The eighth-grade work has been used as a basis of comparison to represent the grammar school. But in one city included in this section of the discussion there is no eighth grade. Pupils are sent on to high school, therefore, after successful completion of the seventh-grade work. Some of the marks secured from the high-school records are averages of the two semesters; others are marks representing the final standing at the close of the year. The marks used in the eighth grade in case of school No. 5' which follow are averages of the estimates made by teach- ers throughout the year and of the final examination given at the end of the year.2 a) Comparisons between grammar school No. 5' and high school No. 5. — Sometimes the question has been raised as to whether the relation between the eighth-grade work and the high school varies to any con- siderable degree in case the comparison is made beyond the Freshman ^ A repetition of 38 in numbering the charts made it necessary to number the above one 38^ " The grades of all the ward schools of this city are kept in a centrally located building. They cover a period of over ten years. They are preserved in large bound volumes, and would furnish a large amount of data for further investigation. COMPARISON OF RELATIVE STANDING OF PUPILS 53 year in the high school. A comparison has here been made between the eighth-grade EngHsh and between each of the four years separately, together with a comparison between the eighth grade and the average of the four years' work in high-school English, in order to determine, if possible, whether any one year, or whether the average of four years, should better be used in trying to measure the efficiency of the relation between the two institutions. The previous discussions have called attention to the variations in the distributions of marks within the same institution, and it may be noted that in charts 38^-43 there are frequent variations in the curve of distribution. The eighth-grade English in chart 38^ tends toward a normal dis- tribution with the mode about 90 per cent. Ignoring for a moment the characters which accompany the numbers in high-school charts 39-43, a considerable fluctuation of the groups may be observed. The first year of high school in chart 39 has a rather rectangular distribution, with the fewer number of marks toward the top of the scale and with several modes. The skew toward the lower end of the scale is even more marked in the Sophomore and Junior years. And in chart 42 there is apparently a somewhat capricious change in the grouping. An average of four years' work of English naturally smooths out the irregularities in the distribution, as may be seen in chart 43. Is the low skew in chart 39 justifiable when compared with the same pupils in the eighth-grade English in chart 38^? In view of the high- school English marks, as charted in the remaining years, one can scarcely avoid the conclusion that either this skewing toward the bottom of the scale is capriciously done, or that there is some lack of co-ordination within the high school itself. Further scientific evidence bearing upon the above conclusion could be obtained by making such comparisons between the sixth, seventh, and eighth grades as have been made between the seventh and eighth grades in sec. I of the previous discussion. The distribution of the marks of 181 pupils in eighth-grade EngHsh, as shown in chart 44, is somewhat similar in its grouping of students to that of chart 38^, from which group of 212 pupils these 181 are taken. A more rectangular equalization of marks over the scale occurs in the case of the Latin, chart 45. The retention shown in table V indicates a closer relation than was found to be the case between the eighth-grade English and the Freshman high-school English. A legitimate question to raise here is, To what extent may this closer correlation between Latin and English be due to the influence of formal grammar work in the eighth grade ? 54 STANDARDIZATION OF SCHOOLS OF KANSAS Charts 46 and 47, comparing eighth-grade history with Sophomore history, show a skewing of the curves of distribution in the opposite directions. This same tendency occurred in the previous charts, 38^ and 39, comparing eighth-grade English with the Freshman year. But the total retention in case of history is several per cent higher than in the case of English, either for the Freshman or Sophomore years. There is more similarity in the distribution of marks between charts 48 and 49 than there is between charts 48 and 50. It may be noted, however, that the higher end of the scale is used in the eighth-grade arithmetic, but not in the Freshman mathematics. Three-fourths of the marks of the pupils occur in the upper half of a range of twenty-six points in the scale used in arithmetic. What explanation or justification is to be offered for omitting wholly the five points at the upper end of the scale in the Freshman year, and for the rather equal distribution over the scale, with a weighting at the bottom In the Sophomore year, and in spite of the fact that these are the same pupils ?^ Table V, showing a summary of the comparisons made between the grammar school and high school, indicates that the percentages of retention are lower between the two institutions than was found to be the case earlier within the same institutions.^ While standards and practices are no doubt more likely to differ between different institutions than in the same institution, yet when there is a low retention between primary and secondary school work is there not something to be done to remedy matters, either from one side or the other, and probably in most cases from both sides? In such large groups as we have here been considering would it be too much to ^ Since there are 2^ years of mathematics required for entrance to college, students of course cannot drop it at the end of the first year if they are expecting to go on to college. * While the comparisons made in English show some variations in retention, yet a tentative conclusion is justifiable. Either the retention for the Freshman year or for the four years' average would give a fair indication of the relation between the two institutions on the basis of a single subject. One advantage in using the Freshman year, providing the two institutions are working in co-operation, would of course be that there is not likely to be so much difference between the eighth-grade English and the Freshman English as there is between the eighth-grade and some of the later years of high-school English, and hence in some sense this would result in a fairer statement of the correlation. Further- more, real articulation of the grammar school and high school depends more upon the first year of high school than on any other, and so the matter needs to be thought of in terms of expediency. An advantage in using the four years' average is that it includes all the variable factors entering into the three or four years of English taken. COMPARISON OF RELATIVE STANDING OF PUPILS 55 expect that the retention between the eighth-grade work in a single subject should be about as high as that between representative years within the same institution ? When table V is compared with the previous table, that summarizes the retentions within the institutions themselves, it will be found on the whole that the results are lower between different institutions than within the institutions. According to the tertile method, the retention is below 50 per cent except in the comparison between eighth-grade English c m u si Eng.of 212 pupils H.S.N0.5 Fresh. Yr. SODh.Yr. W u si •M CO Jun, Yr. 6 Si Sen. Yr. 1 2 3 rtr. Bet. 1 2 3 Ter. Ret. 1 2 3 Ter. Ret. 1 2 3 Tsr, Ret. J. 2 _33 25 25 13 46.47 41 17 13 57.74 40 19 12 56.33 36 23 12 50.70 26 „ 37.14 19 28 23 40.00 21 30 19 42.85 24 22 24 31.42 L^ 13 ., 39 54.84 11 25 35 42.29 10 21 40 56.33 11 25 35 43.39 T-Ot.r.et. 46.17 Tot.Ret. 44.33 Tot.Ret. 53.17 Tot.Ret. 43.39 5 < si CO H.S.7r.lfath. Si <0 ^.S.So.Uath. si H.S.So.HiBt. -A H.S.Fr.Lat. . 1 1 2 3 Ter. Ret. 1 2 3 Ter. Ret. 1 2 3 Ter. Ret. 1 2 , Ret! 1 40 16 15 56.32 38 23 10 53.52 36 27 8 50.71 32 16 12 53.33 2 17 25 27 37.14 17 21 32 29.57 23 25 22 35.71 17 28 16 45.91 S 14 28 29 40. R4 15 26 29 40.84 12 18 41 57.84 11 17 32 53.33 Tot .Ret. 44,81 tot. Ret. 41.50 Tot.Ret. 48.11 Tot.Ret. 50.82 ! a si Aver, of 4 yrs Table V ah graaan Ho. 5. owB the retention ar school (No. 5') betw and 1 2 3 Ter. Ret. 1 38 24 9 57.74 een the 2 25 25 20 35.73 high school 3 6 21 42 59 .le Tot.Ret. 49.52 and Junior English and the comparison between eighth-grade English and Freshman Latin for high school No. 5 and grammar school No. 5'. When these same comparisons are performed according to the modified median method the general result is that there is a retention of about 70 per cent between grammar school No. 5' and high school No. 5. The exact percentages of retention appear in the footnote below. ^ ^ For eighth-grade and Freshman English, 66 . 18; for eighth-grade and Sophomore English, 64.78; for eighth-grade and Junior English, 71.83; for eighth-grade and Senior English, 67.60; for eighth-grade and the four years' average, 76.05 per cent; for eighth-grade and Freshman mathematics, 61.96; for eighth-grade and Sophomore mathematics, 68.30; for eighth-grade and Sophomore history, 73 .93, and fo- eighth-grade English and Freshman Latin, 71.66 per cent. It will be observed that the percentages of retention according to this method change a little relatively among the different years and subject themselves from that shown in table V. This is to be accounted for by the fact that in the tertile divisions the middle third is such a variable factor. 56 STAND AEDIZATION OF SCHOOLS OF KANSAS b) Comparison between grammar school No. 8' and high school No. 8 in English only. — -Charts 51 and 52 represent the distributions of the marks in eighth-grade EngHsh and an average of the three years of work done in the high school.^ In chart 51 the marks of the eighth-grade pupils are numerous toward the higher end of the scale. This was found to be characteristic not only of grammar school No. 8', but in the majority of the grammar schools studied it was found to be a general tendency to load the scale of grading toward the top. Such tendencies as these have already been observed in charts 38^, 46, 48. It may also be noted in the advance charts 55, 57, 83, 85, and in such graphs as accompany charts 60, 62, 64. Several reasons were offered for this by various principals and super- intendents. Some said it was because there was more inclination to lump off grades in the grammar school than in the high school; others, that it was partly due to the fact that parents influenced either directly or indirectly the estimates made by teachers; others, that it is not possible to make fine discriminations in the ratings of pupils in the grammar-school work; others, that it was an attempt on the part of teachers so to encourage pupils that they would continue their work and go on to high school. Whatever the explanation may be of this tend- ency to skew toward the top of the scale, the tendency obviously exists. An exception to this occurs in chart 53.^ The result of the comparison in grammar school No. 8' and high school No. 8, as indicated in table VI later on, shows that the total reten- tion between the eighth-grade English and the three years' average of the high-school English is about the same as that for the eighth-grade English and Freshman year in schools Nos. 5' and 5, but it is lower than the total retention for the eighth grade, and the average of the four years' EngHsh in schools Nos. 5' and 5, respectively. In terms of the modified median method the retention is 71 .43 per cent. c) Comparison between grammar school No. 10' and high school No. 10 in English only. — High school No. 10 is of the older type of the county high schools of Kansas to which country-school pupils are admitted upon the satisfactory completion of the eighth-grade work. Some of '^ In high school No, 8 the eighth-grade marks here used were recorded in the same book on the same pages with the marks made by pupils in the high school. Such an arrangement would make it an easy matter to send to the college or university a statement of the pupil's previous school career. ^ Some of the marks used in chart 53 represent standings in the city school; some of the marks are those received by pupils from the country, who upon entering the high school are given an entrance examination by a board of examiners. COMPARISON OF RELATIVE STANDING OF PUPILS 57 the pupils in this high school consequently have come directly from the country; some of them have completed their eighth-grade work in the city schools; many of the parents of these latter children have moved from the farm to the city. This much is said because while the high school of this city and the grammar schools are carried on somewhat separately, yet in reality the previous conditions of both the high-school pupils and grammar-school pupils have been very similar. 1 00 H.S.50.8 Charts 51, 52 1 u H.S.Ko.lO Charts 53.54 ■S fi 00 H.S.No.7 Charts 55.56 H.S.No.7 Charts 57.58 ATer.3 yi ■s.EriK ATer.3 yi -B.ECf? Fresh. 1 ath. Fres h. Ena. 1 1 2 3 Ter. Ret. 1 2 3 Ret. 1 2 3 Ter. Ret. c 1 O 1 2 ■3 fer. Ret. 1 22 10 10 52.38 30 15 15 60.00 48 27 15 53.33 46 28 16 51.11 2 14 16 12 38.09 11 18 21 36.00 30 34 26 37.77 30 29 16 32.22 3 6 16 20 48.61 9 17. 24 48.00 12 29 49 54.44 14 33 43 47.77 Tot.Eet. 4-8.03 Tot .Ret. 48_.00 Tot .Ret. 48.51 rot.. Ret . 43.70I i a to H.S.No.7 Chart a 57.59 u < 1 a CO H.S.H0.6 Charts 60.61 1 CO H.S.N0.6 Charts 62.63 ti c ■% I. o H.S.IIO.6 Charts 64.65 Soph. En(?. Aver. of Freeh. & SflDh.Kath. Aver. of Fresh. & Soph. Lat. Aver, of Fresh. Sc Soph. Enp. 1 2 3 Ter. Ret. 1 2 3 Ter. Ret. 1 2 3 Ter. Ret. X 2 3 Ter. Ret. 1 50 22 la 55,55 55 36 22 47.78 65 32 4 64.35 ,= 29 9 66.37 -2 1°. 14 33_ 35 31 41 36.66 45.65 42 40 30 35.71 27 36 37 36.00 25 49 38 43. 7e 16 36 61 53.98 9 32 60 59.40 53.31 13 34 66 58. 4C Tot,pfit. 45.74 Tot. Ret. 46.15 Tot.Eet. Tot. Ret. 56.21 TABLE VI Shows the retention between eighth-grade and high-school work in schools Nos. 10, 7, and 6. In the high school the range of the scale of marking is from 80-100 per cent. And as noted before, when an average of three or four years is taken the curve of distribution is likely to be more normal than when any one year is considered separately. The percentage of retention is somewhat higher than in school No. 8, where the average for three years is used. But as is shown in table VI, school No. 10 is nearer the retention in school No. 5, where the average of the four years is used. According to the tertile method the retention between school No. 10' and school No. 10 is a little below 50 per cent when the single subject of EngHsh is considered; according to the modified median method it is 73 per cent. d) Comparison between grammar school No. f and high school No. y in English and arithmetic. — Where different systems of markings are used it is somewhat difiicult to make absolutely accurate comparisons, either between different subjects within the same institution or between the same subjects in different institutions. In grammar school No. 7' 58 STANDARDIZATION OF SCHOOLS OF KANSAS the percentage system is used; in high school No. 7, numbers i, 2, 3, 4. In the high school, numbers i, 2, 3 represent the passing-marks, respec- tively, from high to low standing; number 4 indicates failure. As previously pointed out, when only a three-estimate system of rating pupils is used it is convenient to represent the distribution by graphs accompanying the charts whose columns are broken up into convenient forms for printing. When the graphs are large they have been reduced in size. If in the case of the high-school marks we let i, 2, 3 represent respec- tively 95, 85, 75 per cent, it is more easy to find an average of the two semesters' work done in any one year.^ In case a pupil receives a mark of i for the first semester and 2 for the second semester, by the above translation his standing would be 90 per cent. It is difficult for an investigator to find an average of the pupil's standmg by means of merely the marks i and 2, for example, and so the high-school marks of school No. 7 have all been reduced to percentages based upon semester marks.^ The range of marking used in charts 55 and 57 is unusually wide, and as pointed out in a previous chart, 48, there is a great non-use of the points toward the lower end of the scale. The number of marks occur- ring below 75 do not evidently represent as distinct steps or gradations as those of the next fifteen points above 75 per cent. From the graphs which accompany the charts it is easy to see at a glance the tendency, in the distribution of marks, to skew toward the top of the scale. This appears in a large number of the high schools, as may be seen in charts 56, 58, 59, representing school No. 7, and in such later charts as 89, loi, 103, 105, and in the charts representing the 23 different high schools which are compared with the college. This, of course, is not necessarily a criticism. Some exceptions to this upward tendency will appear farther along. ^ This is not so accurate a method as where the percentages are given all along in a wider scale. Since i really stands for a range of grade about 95, 2 for a range of about 85, and 3 for a range of about 75, there may be some objections raised against the above translations. But if all of the three-estimate systems are treated in the same way the facts will not be distorted. ^ The records in the high school are temporarily preserved on cards, and per- manently in bound volumes, but it was very difi&cult to secure records for the ele- mentary schools covering any number of years. Since only high-school graduates were considered, it was necessary to begin as far back as 1907 for the first elementary records. Many of the records before this were not available. It was only through the assistance of trained helpers and persistent ward-school and high-school princi- pals that the records which had not been destroyed were secured. COMPARISON OF RELATIVE STANDING OF PUPILS 59 The percentage of retention for schools Nos. f and 7, as shown in table VI, for EngHsh is lower than in the other schools so far compared, but this may be partly due to the fact that such a different scale of marks is used in the two institutions. HowTver, the retention for mathematics is higher than was the case in school No. 5. The result stated in terms of the modified median method is a retention of about 70 per cent.^ d) Comparison between grammar school No. 6' and high school No. 6 in arithmetic, Latin, and English. — A few general explanations will be of assistance in making clear the comparisons in schools Nos. 6' and 6. The marks used are in terms of e, g, f; e stands for excellent, g for good, and / for fair. The records for many years showed that exponents had been used with the letters in order to make finer discriminations. For example, in a scale from 70-100, p p or g^ g^ or e^ e^, indicated 71, 72, 83, 85, 96, and 97, respectively.^ Since there were so many available records of the later years which did not include exponents, these were dropped from the letters in the earlier marks. But the exponents made it possible to number the pupils in the eighth-grade English in chart 64 approximately in order of their standing. Consequently, number i begins by representing one of the pupils among the very best, and 338, the final number of the list, repre- sents one of the pupils among the poorest in the whole group. This means in such a chart as 64 that individual 102 has a higher standing than has no or 114; or that 218 has a higher standing than 223 or 229, for example. The pupils in chart 60 are represented by these same numbers. The letters e, g, f are used also in the high-school markings.^ These are reduced to percentages, as was done previously in the case of numbers. If we let e = 95, ^ = 85, and/=75, then when a pupil has a standing of e during one year and a standing of g during another, the average for these two years' work is eg, or 90 per cent; in like manner ef is 85 and/g is 80 per cent. The base lines are broken simply for the purpose of assisting the reader at a glance in seeing the relative number of pupils who have received the various standings. ^ The exact retention according to method No. 2 is 67 . 21 for eighth-grade English and Freshman Enghsh; 69.94 for eighth-grade English and Sophomore English; and 69.99 for eighth-grade arithmetic and Freshman mathematics. ' The records for the ward schools of this city were the most elaborate of any school investigated. The marks of the elementary-school pupils have been preserved for ten or fifteen years in large bound volumes, and are kept on file in the superin- tendent's office. 3 The high-school records are kept on cards filed in boxes alphabetically arranged. 6o STANDARDIZATION OF SCHOOLS OF KANSAS The distribution of marks in the elementary school, as indicated by the graphs, re-emphasizes the previously mentioned tendency to skew toward the upper end of the scale. However, graph 60 shows a some- what more normal distribution than 62 or 64. Chart 63 shows a peculiar equalization of marks over the scale with about as many excellent and poor as there are mediocre pupils. This rectangular distribution pro- vokes the question as to whether in a group of over 300 pupils capacities are really so equally divided as this chart would indicate. Graph 60 shows that the absolute marks of the pupils as a group are higher than was the case in high school No. 5, where the groups were shifted toward the lower end of the scale. But when the graphs repre- senting 60 and 61 are compared with each other there is considerable similarity, which probably indicates that the two institutions are using approximately comparable systems of marking, at any rate in this particular subject. The general result of the comparison in grammar school No. 6' and high school No. 6, as shown in table VI, indicates that the retention in mathematics is below 50 per cent, as has been the case in schools Nos. 5 and 7. The retention in English in school No. 6 is higher than in any of the previous schools, which is a probable indication of the closer correlation between the eighth grade and high school in this single subject. The retention between English and Latin is higher than the retention in schools Nos. 5' and 5 in this same subject. The result of the comparisons in schools Nos. 6' and 6 upon the basis of the modified median method is a retention of over 75 per cent.^ The result of the comparison of these different grammar schools and high schools of Kansas is that there is a retention of about yo per cent; that for schools Nos. 5' and 5 being about 70 per cent; that for schools Nos. 8' and 8 in English only, 71+ per cent; that for schools Nos. 10' and 10 in English only, 73 per cent; that for schools Nos. f and 7 about 70 per cent, and that for schools Nos. 6' and 6 above y^ per cent.^ ^ The actual retention for eighth-grade arithmetic and Freshman-Sophomore arithmetic is 69.02 per cent; for eighth-grade English and Freshman-Sophomore Enghsh is 77.87, and for eighth-grade English and Freshman-Sophomore Latin, 74.75 per cent. The higher retention between grammar school No. 6 and high school No. 6 may be partly due to the fact that the same system of mar ing withi i a narrow range or scale is used, but it is also no doubt due to a somewhat closer corre- lation of institutions on the basis of single subjects compared. " This result will be supplemented in sec. V. One additional grammar-school and high-school comparison will there be included. Sec. V will deal only with the pupils who went on to college. COMPARISON OF RELATIVE STANDING OF PUPILS 6 1 e) Comparison between grammar schools Nos. 2', j' , 4', and high schools Xos. 2, J, 4 in mathematics, English, and Latin {not in Kansas). — Since the practice in both elementary schools and high schools is somewhat different from the previous schools compared in this section, a few explanatory statements are appropriate here. The elementary schools of this city do not have any eighth grade. Pupils who satis- factorily complete the seventh grade are promoted to the high school.^ Since it was necessary to go back at least four years for the first grammar-school records of the high-school graduates, it was not easy to recover the marks of pupils who had completed the grammar school even eight or ten years ago. There was great difficulty in securing the grammar-school marks, partly because of the size of the city, partly because many records were either scattered among individual teachers and pupils, or were destroyed. The percentage system of marking is used in the various ward schools, but the letter sytem is used in the high schools, namely, e for excellent, g for good, m for medium, and p for poor. Those receiving the grade of poor are graduated from high school, but are not recom- mended for college. For purposes of comparison the letters were reduced to percentages by using e to represent 95 per cent; g, 85 per cent; m, 75 per cent, and p, 65 per cent. The high-school marks were in the first place charted separately and compared with the marks of the grammar-school pupils who came to the high schools, respectively.^ Composite charts were afterward made for the grammar schools and high schools. Grammar school No. 2' , for example, represents the total group of students coming from the different ward schools to high school No. 2. In the majority of instances charts 66, 68, 70, 73, 75, 77, 80, repre- senting the grammar schools, together with the composite charts 83, 85, and 87, show a tendency to a normal distribution of marks. In a few grammar schools the standings of pupils were recorded in terms of a, b, c. Where this was found to be the case, these letters were respec- tively transferrred to 95, 85, 75 per cent. This does not vitiate the ^ The seventh-grade records of pupils are placed upon the diplomas received and from these the high-school principal gets some idea of the previous career of the pupil. If these could be permanently preserved, together with the high-school record of the four years' work, they would furnish good data for a comparison with the same pupil's career in case he goes on to college. 2 It was possible to determine from the records kept in high school No. 3 in pre- cisely what order a pupil had pursued a certain branch. If, for example, a pupil had pursued a first-year subject during his Senior year, it was so recorded. 62 STANDARDIZATION OF SCHOOLS OF KANSAS results of plotting the curve, because these same marks, if they had originally been in terms of percentage, would have been grouped around these percentages. This fact explains the frequency of marks over the multiples of five. Together with this explanation it may be added that there is obviously frequent use of the multiples of five. Charts 83 and 87 are composed of the same pupils. The distribu- tions of the marks in the seventh-grade arithmetic and the seventh- grade English are much alike. This is an indication that somewhat similar standards have been used in the two subjects. The mode in chart 67, representing the distribution of the marks in Freshman-Sophomore mathematics in high school No. 2, occurs over 85 per cent, while in chart 74, representing the marks in high school No. 3, it occurs over 75 per cent. Both tend toward a normal distri- bution. Although the curves of distribution are somewhat more nor- mal than in some of the high schools already charted, yet in 67 there is an upward skew, in 74 a downward skew. Composite chart 84 includes an additional list of students who were not included in the original chartings of schools Nos. 2, 3, and 4 separately.^ The accompanying graph shows a distribution more normal than in the majority of the high schools studied, but there is a slight downward skew. Chart 69, representing Latin, skews toward the top more than the distribution in chart 76, representing school No. 3. The composite chart 86 also includes some additional marks of pupils. It tends toward a normal curve, with the group shifted a little toward the top of the scale. In high school No. 4 numerous marks appear toward the bottom of the scale in chart 81 which does not appear justifiable, either when compared with chart 80 in seventh-grade English or with chart 82, represent- ing the work of the same people during the Freshman and Sopho- more years.^ Again, this distribution in chart 81 may be compared ^ The arrangement of marks in chart 81 furnishes ground for the statement that school 4 tends to give many low marks during the first year's work, as is evidenced by the fact that 22 pupils out of 73 receive a standing of 65 per cent, or the rating as poor. Most pupils who remain in school after the first yeai get above this standing, at least a little, and so in chart 82 some who had an average of pm for the two years' work appear over the grade of 80, and the others have improved beyond this orignal standing. 2 In a few of the composite charts additional pupils have been included. This is due to the fact that some of the data could not be gotten until after some of the sepa- rate charts had already been completed. COMPARISON or RELATIVE STANDING OF PUPILS 63 with charts 78 and 79 of school No. 3, where there is apparently greater consistency, within the high-school marking at any rate, or with charts 71 and 72, representing school No. 2, or finally, with the com- posite of 299 pupils in chart S8, where there is a more normal distri- bution. rr o iiarts 66, 67 Ay. Fr.&JeSBcMatl: Tot. Ret. 46.05 H. S. Mo. 2 Charts 70, 71 2 Sem.Fy Bnfi. Ter. 3 Ret. Tot. Ret. 50.51 50.00 39.39 ,H. S. No. 2 Charts 70, 72 AT.ITr.&So.Eng Ter. Ret. 59.37 H. S. tip. 3 Charts 75, 74 AT.rr.& ao.Math. akaris 75, 76 Presh. Lat. Ter. 5 |Ret. Tot. Ret. 61-28 49. ??, 1.2^ 56. 9S, Tot .Ret. 56.12 Ter. Rei^ H. S. No. 5 Charts llTTa 2 Stm.gr H. S. No. 3 ^?.94 Tot .Ret. 59.13 _5Sfi. ter. 2 13 iRet. Charts 77, 1*9 Ay.Fx.A So.Eng Tot .Ret. 46.23 R, S. No, 4 sag. ?r 3 3 11 10 y9t. Re^- 81 H. S. No. 4 H.S. Ho3. 2.3.4 Ter. Ret. Charts 80, 82 fur St & So.'SaR 5S.35 44,00 3 11 10 Ter. 5 IRet. £LM Tot. Ret. Charts 83, 84 AvJgr.&So.Math 40.00 41.66 Ter. 2 13 IRet. 46.57 Charts 85, 86 Av gr & SO. Lat 33.33 45.00 Tot .Ret. 43.97 [. S. Noa. 2.3.4 Charts. 87 AT. Fr. & 13138149 Tyt. Ret. Ter. Ret. 60.00 26.26 Talsle VZI shows a nummary of the oompariaons of echools Nob. 2', 3*, 4* and E. &» Ko8. 2, 3, 4 •• not in Eansae. 49^ The percentages recorded in table VII are summaries of the com- parisons made between grammar schools Nos. 2', 3', 4' and high schools Nos. 2, 3, 4 in mathematics, English, and Latin. According to the tertile method high school No. 3 stands highest. This may be partly due to the fact that in one of the other high schools there has been a considerable shifting of students because of a new building in construc- tion, and also because, in the redistricting of pupils, a rather large pro- portion of weaker pupils came in. And again, in the remaining high 64 STANDARDIZATION OF SCHOOLS OF KANSAS school, probably the old-line subjects do not get the same emphasis as in high school No. 3. The total retention when these schools are charted separately com- pares favorably with the summary tables representing the schools of Kansas, yet the composite charts for schools Nos. 2, 3, 4 show a rather low retention. This is no doubt in part due to the fact that the majority of the added pupils were taken from school No. 4, where the retention is lower than in the other two high schools, which were charted sepa- rately. The correlation, as before pointed out, is lowest in schools Nos. 4' and 4, so that the addition of students in the composite charts from school No. 4 no doubt lowers the total retention. The total retention between the seventh-grade English and the Freshman-Sophomore Latin in the composite chart is higher than either that between seventh-grade English and Freshman-Sophomore EngUsh or than that between seventh-grade arithmetic and Freshman-Sopho- more mathematics, and the retention for the mathematics is the lowest. It is interesting to note that in the majority of the comparisons made in grammar schools and high schools the correlation between English and Latin has been higher than the correlation between grammar-school EngHsh and high-school EngHsh. While the retention in terms of the modified median method for the composite charts is a little below yo per cent^ yet when the chartings of the high schools are regarded separately, the general result is about the same as in the schools of Kansas, namely, yo per cent."- ^ The actual retention for the composite charts 'n mathematics is 65 . 5 per cent; for Latin, 68.27, and for English, 66. For the separate high-school comparisons it would be above this, as may be seen by comparing the tertile retention for the com- posite with the separate chartings in table VII. COMPARISON OF RELATIVE STANDING OF PUPILS 6s lOOOl Hi aif)t noc HHr ?3a; IHH 133 HHi !o ^lO 0>O CD HHi o to _ CD 4; ^i SS 00 — .1 ts S5 J>A <&.£ ?iS 4' nS loio J>l.J> <&i <& o>a> 00 61 a^ COMPARISON OF RELATIVE STANDING OF PUPILS 67 T ■4*9 15 hsi ^J: o!o O'l-i H •rHir-IH iU ■a o -_ .i> CD^Okii 01 tOa^^loSiO p I I ^3 it^ Vi 68 STANDARDIZATION OF SCHOOLS OF KANSAS IOB> :£S acoHK>|io t-ooo> Hi-tH C0O> t-it-{f-\ lOOO - NO c!.^ COMPARISON OF RELATIVE STANDING OF PUPILS 69 1; w ss J)At- Ji j\\0 r^J,4 to H 4.U0 O 9> to lO 70 STANDARDIZATION OF SCHOOLS OF KANSAS lO ^OlO as i-iHH sss .i.i H OIL. as: t-trii-ti-ir1H f-IHH H ^2^ 16 H »t 03 O WlO> fH|HJHr lislo CO 1 U5 CM n t-kn OHjg Jo> 00 Joo o Sj«o CO ssg 72 STANDARDIZATION OF SCHOOLS OF KANSAS Al U^^ SrtMe ld> ^§ ^^ 3- COMPARISON OF RELATIVE STANDING OF PUPILS 73 lO to 25 » ojo 5^a 16 ^6 O»k0 hi '9S J3?1SS3 SSi SSSi ss 33 PI3 3a- sot Ok OIA HO ^ Si aSi" MN04 -sasssse^^^^^ s:SSgg§S:sgSgS|SS SSpS5g|g^§g|i2S '"Hskg^siSssggpsi 1 -r^ssgdiigss «58S5SSgSgSSSrSS^gysU t- o> e- (M «o £jtWT> >o ssgsgs^Sgsssssg •— — siii^ s^|s d5S|g ^sl3§SS?di§ 1- NN esgas^sF -gpSi^ "SS^SgS S£Sg^3|§; 38- ?s lal^ SS I' %st i 2 1 i Si "^ 00 a iHHf- o q M CM W^' HH(m «o^■ ►■ r- SS ?399S3 ^3( s Ob o oo ta ^ Hrt »>H» 83* si 5 SssH= a =s§spsSI3sS I ia>o(D sSSSSSSsas- CQO s w OMCO sSSsPjsifs* 00)0 0)^ iou>r-0Dnt~!o>5f Hio a S- r -^ «> CM HtQ^MNCM<0 ^Sf^SPSEiHis^ H^NNCOOOO. CO la o>ia cO c- otjoi h s^ia'j^jspps" gs ii gSgds:5bsE:?^g «] oo ot N eo H r-iioit-t) en t- M « «D*- H ^ » Ojfe "^ H|HHHH...H 1 M a> ^ CO lo lo la <0kn so »iioSa>^ioocvi|&ao i! -«5isbsS§S5 rtPJN » e- '»«!"s3aS3s° at §S2S||sggaS3 sssdsa§S||S 10 N sStsSSH- S52g?S ssssasssssg e-aoB>'«MO>M»- r« ^«oa>o>no HO) Hioa>a>^m o «o^-lOlO^-co o »85ssa|5ss 53S8IF «o«oio»fr- s M^OiHHn > HHM !35SSS^^aSS dsE^ggg riot oa> # MM S HM g ^- M !- «L H> c t- 3 J CM !- (l f, Jt g lO 04 ^ 8 i coca ?- 1 1 t s MM 4i -1 ' ^^ ^t sg, iJ. J>^A J^ 52 lofor HMO Uj OHr^M IS too -OH Hhh ^51 249 1?.P 5 216 159 80 1 293 fiR 10 262 _12& 16 218 160 85 2 R95 77 11 127 4 253 133 20 219 161 86 12 296 87 22 134 R 258 136 25 221 162 RR 4 297 9S 52 154 19 264 337 28 222 163 89 .6 299 96 59 156 26 271 29 223 164 93 18 300 114 70 157 36 274 _141^ 32 '24 165 97 21 302 116 102 184 40 276 169 166 101 23 303 121 113 186 45 279 .70 34 226 ;.67 103 117 188 46 283 .72 41 229 .04 124 1R9 47 288 76 67 230 07 31 310 30 135 196 49 292 178 68 235 OR 143 197 50 294 179 60 236 ■ 74 110 37 312 42 147 199 56 298 182 61 247 176 112 39 313 146 ! 149 211 64 301 185 62 256 177 119 42 315 148 1 187 239 73 304 192 in 260 180 120 43 317 191 206 _ 209 242 81 305 201 74 262 181 128 44 320 207 245 82 306 202 76 263 .83j 132 48 321 217 215 248 90 203 78 267 :90 138 61 322 220 251 255 95 314 79 269 :.93 53 323 2P,P 280 257 98 316 208 83 270 .94 :.44 54 326 2B7 261 99 318 212 84 272 196 .45 55 327 233 290 266 105 324 226 91 273 198 150 63 329 234 319 268 106 325 231 94 277 200 161 65 330 238 335 276 109 3?.fl 237 110 2R1 205 152 67 332 240 337 285 118 331 241 111 282 210 163 69 333 244 |338 |286 123 |334 243 116 1 1284 213 155 72li336 260 1 r Vb 80 Sb 1 L____ - ~ ■=^^=^ pCbar^-Ho, 61. Avers >««-© 25&- -n ^KatV^f- 47^^ 224 150 13* 11- ; 227 16 264- 144 ifl* 229 156- 16» 270 145 19- 262 163 72* 230 167- 37* 242- 118- 2* 27&. 151 23,* 263 164 75' 1" 232' 159 24' 7- 247 122 3 278* 154. 26 267 166 78 4- 234* 160 27' 124- 8- 248- 133 ff 283- 168 40- P71- 172 79 6 237- 161 SR* 126 22- 255- 136 10- 51* 274- 174 84 9- 845- 162 48* 127- 23* 257- 137 20 fiB* 277 1R3 86* 12* 244' 166 54' 135- 26- 266- 148* 27* 286- 187- 60 279- 185 88T 14' 246- 167 67* 143- 32 268- 149- 29 293* 192 66* 280- IRR- 89* 28 256 171 69* 147- 33 276- 153 36- 298- 194 76 282 189- 90- 30' 258- 173 71 45- 281 169 39^ R6* 299* IQl* 95- 31* 259* 175 74 170 47- 287- 176 41 301-202 91 303* 197- 96* 34 260 179 77* 199- 50- 290- 178 44* 304- ?,03 98- 308* 200 97* 35' 265' IBl 80* 209- 52- 291* 184- 54 305- 206- 207* 108* 42' 269 182 87* 216- 68 295* IRfr. 59- 309* 219 103* 818- 212- 114' 43* 272 190 92' 239^ 61 296'' 195- 64- 311* 220* 106- 321' IK 241- 82- 297* 196 70- 312* 226- 10&. 323* 128* 49— 288- 198 lOl' 261- 204 73- 313' 112*" 326* 130*" 53' 289* 210 107* 294- 109- 306- 206 1- 3 .4- 233* i,:6 328- 226* 131' 66- 292- 2 .3 MO' 307- 111 322' 208 83 3,5* 236 19* 329* 240* 132* 62 310' 2 6. 16, 319- 324- 113- 334- 117- 335- 211- 231- 94 ^vr 99- 327* 136 121' 249- 134- .520i 331- 246- 260* 140* 142* 63* 320* 66* 333* 2 7* .20* 221 29* 13*5- 123rl|338- 23?; 100 1 1336' 251- 138'i 1332' 252- 152 68-1 1337- 222 146 1 75 ^MI 85 9 l_ _:_:.. J> COMPARISON OF RELATIVE STANDING OF PUPILS 77 BlBtrlbutlon of HarkB tn Chart So. 62. ' RO I fiK ' on ' DlBtrl'bxition of Marks In Chart Bo. 63. 75 80 65 90 95 -mr JJlgl *^ ?AS^M^f 5 ° ■"-geg^ pupils. — lOl pupil! 246 ElO 160 146 115 84 101 pupils 888 248 211 181 147 116 85 E89 250 213 182 148 117 86 66 28 1 29 1 2BZ. 214 183 149 118 87 57 29 29 253 215 184 151 120 88 58 30 3 292 254 216 185 152 121 89 59 31 4 293 255 217 186 153 122 90 60 32 6 294 256 218 187 164 123 91 61 3K fi 295 257 219 188 165 125 92 62 34 7 296 26R 220 189 157 126 93 63 35 8 297 259 221 190 94 64 36 9 319 298 260 191 159 95 65 3R 10 320 299 261 192 160 129 96 66 29 11 321 300 262 225 193 161 130 97 67 40 12 263 227 194 131 98 99 68 41 13 323 302 264 228 195 132 69 42 14 324 265 229 196 100 70 43 15 325 305 266 230 197 ,65 101 71 44 16 326 306 267 232 198 ,66 102 72 45 17 327 307 269 234 199 168 136 104 73 46 18 329 30R 270 235 169 137 105 74 47 19 330 309 271 201 170 138 106 75 48 20 831 310 273 237 202 171 :39 107 76 49 21 332 3:.i 276 238 203 172 ,40 108 77 50 22 S3S 312 278 239 204 174 141 109 79 51 23 334 313 280 2.41 205 175 142 110 80 52 24 335 314 284 242 206 176 143 111 81 53 25 337 316 285 243 207 177 14/ 112 82 54 26 1338 ^17 1 1286 244 208 |179 145 113 Bl_|^ 27 1 75 u S5 8B 266- ^^ ' ^erne^g^ 64:^ 2fi9- 26* 192 24* 273- 73* 197 33* 278- 200 46' 280- 93* 201 47' 121 4' 110 205 60* 214 m 14'- 122 11* 286^ 113 208 69* 215- 118 26* 125 19' 112 288— 117 22^- 83* 217- 126 28* 139 20' 116 35* 1' 289- 123 232- 88* 219- 132 42* 143 22* 116 38* 2* J "Hfc" 151 "242^ iT _2ao^i3i_ 221- 145 IM S3* 128 39' 3' j 147 36* 130 40* 5* 1 299- 159 244- 224- 148 55* 149 44* 134 41' 6' 301- 165 £52- 131 234- 162 56* 153 49* 138 43* 7* 30&- fe53- m- 241- 24&£ 163 57* 1B5 6^: 141 45' 8' 306- 191 m- 59* 161 -64* 170 _53*_142 51* 9* 307- 196 258- 62^ 152 54* 10* 310- 199 276- ■ 40 259- 171 65* 180 70* 154 58* 12" 290- 146 270- 372 66* 193 76* 169 K 13* 316- 204 294- 157 295- 176 68* 198 82' 184 63* 15* 211 71* 207 86* 210 67* 16* 325- 225- 30&- " .60 298- 81 75* 222- 87* 21fr. 72* 17' 1 326- 230- 302- ' 66 306- '. 83 79* 228- 89* 2l8- ■'tt 18* 32$- 308- 74 311- 87 96* 2S9- 77* 21* 309- 175 317- 90 84* 238- 80* 27' '179 194 92* 248- 99* 250- 85* 29* 256- 313- 182 324- 195 94* 254- 104* 261- 91* 30* 333- 260- 320- 185 327- 202 100* 262- 106 264- 97' 31* 335- ?,63- 322- 188 334- 206 101* 267- 108 271- 102* 32* (338- 265-11323- 189 1 P37- 213 107 1 1285- 120 11291- 105 34*1 75 8C 85 ■ 90 ■■ 95 102 pttpilSi 78 STANDARDIZATION OF SCHOOLS OF KANSAS Dlstrfbntlon of HaiicB In Chart Chart So. 64. Eighth Grade Eng. of 338 pttplle. 90 • 95 t ns p«i,4if/o _._.. J 288 160 118 86 289 161 119 87 290 152 120 88 56 SR 291 260 233 20 163 121 89 57 P.9 1 292 261 234 20 180 354 122 90 68 30 2 293 262 235 20 ! IBl 165 123 91 69 31 294 263 236 20 1R2 156 124 92 60 3S 295 264 237 PIO 183 167 125 93 61 33 896 265 23fl 211 184 168 125 94 6P, 34 297 239 212 185 169 127 95 63 36 S67 213 186 160 1E8 96 64 36 299 26B 241 214 187 161 129 97 66 37 300 P.fi^ ?,4?, _£lfi_ 188 162 130 98 66 38 10 322 243 216 189 163 131 99 67 39 11 :23 802 271 244 217 190 164 132 100 68 40 12 303 272 245 191 166 133 101 69 41 13 304 273 P46 219 19P 166 134 lOP 70 4? 14 305 274 247 820 193 136 71 43 15 306 276 248 221 ,94 168 :.36 ,04 72 44 16 307 276 249 222 96 169 37 05 73 46 •7 308 P77 Pr<>0 ?,r,? 96 170 138 106 74 46 ft 309 278 251 224 197 107 76 310 279 262 225 198 ; ,72 108 76 48 20 280 253 226 199 .73 109 77 49 21 312 281 254 227 200 174 .10 78 50 22 313 282 255- 228 201 175 43 79 51 23 314 283 256 229 202 176 ;.44 80 62 24 315 284 267 230 177 45 81 63 25 316 285 204 178 46 82 64 26 (338 1 1317 ( 1286 869 232 205 |179 147 116 II ,83 55 27 1 ■■50 ^86 90 9B 1 Dlstrlhution of Ke.rks in Chart I7o. 65. hrhr^ Chert No . 65. Arerage of Fresh.S: Soph . -Ens. Of 338 pupils. 95* 58^ 43f. 66^ 1 118 1 119 160 54* 151 161 58* 1* 178 168 63* 2* 170 64* 4* 270- 171 67* 5* 275- 190 25* 177 86* 176 70* 6* 209 277- 196 33* 180 86* 3* 184 71* 7* 215 ?,7B- 1P7 45* 234- ^ 187 87* 1ft* 191 74* R* 26a- 284- 200 59* 239- 125 22* 189 9P* iS* 193 11* 9* E6&- 285- 203 60* 241- 127 24* 192 q4* 14* 195 82* 31* 266- 287- 210 78* 246- 129 37* 201 96* 20* 199 jMT'ia*": 288- 211 79* 251- 131 46' 202 1)4* P6* 204 89* TB* 289- 214 113* 256- 33 2 52* 205 105* ?7* 212 lo* 15* EBl- 292- 219 114 257- 134 56* 207 106* ?8* 215 ll* 17* 295- 221 123 258- 135 57* 213 108* ?,18 97* 18* 297- 223 137 267- 136 61* 227- 116 34* 304- 225 140 271- 149 52* 230- 120 35* 224 99* 21* 311- 232- 146 272- 155 69" 231- 124 4P* 226- 101* 23* 3 .3- 237- .150 273- 158 72* 233- 126 44* 229- 107* "29* 3.6- 240- .56 280- :.6' 73* 236- 128 48* 235- 109* 30* 3 .8- 243- :.i>7 293- .6" 80* 238- 133 49* P4^- 111* 31* 321- 244- ;.i;9 294- .7: 81* 242- 138 53* 248- 112* 36* 322- 247- 305- 174 93* 253- 143 «>'^* 249- 115 38* 320- 329- 250- 165 307- 175 100^*" 264- 144 65* P55- 130 39* 323- 330- 252- ' 66 309- 188 102* 276- 148 6fi* PfiO- 139 40* 324— 331^ 254- 179 314- 198 103* 285. 152 68* 261- 141 41* 326- 333- 259- 181 317- 208 110* 290- 154 75* 291- 142 43* 328- 835- 263- 182 319- 217 117 302- 162 7-7* 296- 145 47* 332— 336- 268- 183 325- 220 121 308- 169 83* 298- 147 50* |338-| 1337- 269- 185 1 1334- 228- 122 1 1310- 172 -^4^1327- 153 51*1 75 80 85 ■ 90 9B ■■' COMPARISON OF RELATIVE STANDING OF PUPILS 79 8o STAANDARDIZTION OF SCHOOLS OF KANSAS *hU\ * c on ^- NCMN P^-^-fct cfxi men c4p * » aj3 S;s 4il^<&cS.A>i! iio «* CO 00 a ^ DO ^ CO rHbO loS COMPARISON OF RELATIVE STANDING OF PUPILS 8l t(- ^ >0 '^U set w- Aciui e^^ii « Jo a » Mi CJlig ss ^■•o 3? ^^J^ i AiL'hoiJ) ri CJ 3> IK IC u tc € 9> JIT. t" o o lit 82 STANDARDIZATION OF SCHOOLS OF KANSAS f ^ » t ^ a ' it &S AJ, H,icii4i cirtco u> lo to ^- 00 o> i iioS ><0 COMPARISON OF RELATIVE STANDING OF PUPILS 83 A-* P< • o o ll MM CMMOI ^^ J>6 oolco n .i ioinn«S « e u o 9.j^ r OH 6^ t » i>ll iSio as: W DO O 10 84 STANDARDIZATION OF SCHOOLS OF KANSAS ss ^;s 0AJ, S (ic!. A% i^ 1AJ> iAA J^ MM «;, S lO evT s s 1 - s lO i^^ s lO <0 COMPARISON OF RELATIVE STANDING OF PUPILS 8s I- r^; I ? s§^ a?l Hi-«H ^n g§ 00r^»-IH e- CO to M ( lO OIOI MNOIC (DOkO>^»t-C- gS8 una la lo r M M gCiS I-4H S 0> M COMPARISON OF RELATIVE STANDING OF PUPILS 87 SEC. IV. SOME COMPARISONS OF HIGH SCHOOLS AND COLLEGES This section includes a comparison of the relative standing of pupils between high schools and colleges, together with some comparisons on the basis of absolute marks.^ The object of this section is to discover as nearly as possible what the actual existing relation is between high schools and colleges, and then farther along, on the basis of these results and those found in sec. V, attempt to determine about what should really be expected to be the extent of correlation between the secondary and higher institutions of learning. The variety of the systems of grading used in the following schools concerned, here again, as before, so complicates the process of comparing schools that it is not possible to draw conclusions without allowing for some modification of statements relative to the results.^ A general tendency, previously noted, is obvious when we compare such graphs as 89, 91', loi, 103, 105, 107, 109, or some of the smaller graphs representing the 23 different high schools in chart 1 09 A or 117 in the advance section; namely, that the great majority of the distributions of high-school marks tend to skew toward the top of the scale. This may be partly accounted for by the fact that none of the records of eliminated students are included, but in spite of this explanation it may be in part due also to the use of a too-narrow range of estimates. Taking up more in detail some of the graphs representing the dis- tributions of marks, it may be noted that the skew in chart 89 is much more exaggerated in case of the Freshman year than it is in the average of three years of English of precisely these same pupils as shown in chart 91. But do not charts 90 or 92 indicate that the rating in chart 91 is probably more justifiable than that in 89, since the high-school Freshman ^ Since a three-estimate basis of marking practically amounts to ranking students, a few schools were charted and compared on the basis of the original grouping rather than by dividing them into equal tertile groups. The width of the broken base lines in charts loi, 102, 103, 104, 105, 106 indicates as well as the graphs the upward- skewing tendency in the high school and college already pointed out in the discussion. 2 College No. i uses marks, i, 2, 3, to represent students' standings from high to low, and these stand respectively for 90-100, 80-90, 70-80 per cent. College No. 2 uses the percentage system, ranging from 70-100; college No. 3 uses the letters A, B, C. High school No. i uses the number system, i, 2, 3, indicating respectively 95-100, 85-95, 75~^5 P^r cent. Other high schools, as, for example, No. 7, No. 6, No. 5, use numbers, letters, and the ordinary percentage system, respectively. . 55 STANDARDIZATION OF SCHOOLS OF KANSAS class as a group does not hold its position in the Freshman year of college ? And furthermore, the 86 pupils out of these 266 represented in chart 91' who go on to college and graduate, as a group, hold their place pretty well, as shown in the graph for chart 92'. The 81 pupils who go on to college No. 2, represented in charts 93, 95, and 97, are taken from the previous group of 212 pupils in school No. 5. Numerous marks toward the lower end of the scale here occur, as was previously the case, with the whole group. When chart 99, representing Freshman-Sophomore mathematics, is compared with the above charts, it indicates that the standards are somewhat different in the two depart- ments. Again, the shifting of the whole group of pupils in college mathematics toward the lower end of the scale, as shown in chart 100, indicates that the two institutions are not using similar standards. For in chart 99 the pupils are grouped about the upper end of the scale. Charts 84, 96, or 98 indicate a sort of bimodal distribution, with a somewhat larger number of marks toward the top of the scale, while in chart 100 marks are more numerous toward the lower end. Consequently the departments within college No. 2 are using different standards, although these are more alike than those used by the high school and college. Graphs 101-9 indicate on the whole that either the standards of the two institutions are not similar or that the students who go from the high schools are not strong enough to maintain, as a group, their posi- tions. Whenever there has been any considerable number of pupils involved in these comparisons, in very few instances do the graphs show a normal distribution of high-school pupils, examples of which, not before used, are charts loi, 103, 105, 107, indicating absolute marks; while on the other hand college No. i, as evidenced by graphs 102, 104, 106, 108, has in the majority of cases distributed its marks somewhat according to the normal curve. A very brief discussion of some of the charts representing the 23 dif- ferent high schools, together with composite charts of these same pupils, will furnish some notion of the relation of these schools to college No. i. See charts 107-9 A. After finding out the standing of these pupils in terms of percentage, they were translated into terms of i, 2, 3, and then charted and graphed, separately, in the first instance, as well as charted and graphed in com- posite form later.^ ^ The percentage system is used in practically all of these high schools. Since college No. i uses the marks i, 2, 3, it was thought that it would be interesting to find COMPARISON OF RELATIVE STANDING OF PUPILS 89 Schools Nos. 2 2, 25, and 5 are exceptions to the skew upward.^ While schools Nos. 5 and 22 hold their positions in the college or prob- ably improve as a group, school No. 25 as a group does not do so well in maintaining its relative position. School No. 21 has a peculiar rectangular distribution which is hardly possible with any large num- ber of pupils, but this group, too, improves as a whole in the college. The different relations between the standings of the high-school pupils in schools Nos. 35 and 18, and in college No. i, either show a differ- ence in the use of standards by the college, or it shows that high school No. 25 is the weaker of the two. It might be concluded from the graphs in chart 109 A that in such schools as Nos. 12, 23, 14, 15, 17, 20, 8, 24, 27 only the stronger pupils enter college, if it were not for the distribution of marks which occurs during the Freshman year of college work. It may be noted that the groups as a whole shift toward the lower end of the scale in college No. i. The actual percentage ratings were charted in chart 109 to indicate that the translation of the percentages to i, 2, 3 did not distort in any way the grouping of the marks. For chart 107 shows the same tend- ency through its graph to skew toward the top as is found in chart 109. And while there are exceptions to this tendency, found in the sepa- rate graphs of the 23 schools, yet the composite charts 107 and 108 warrant the statement that there is a more normal distribution of grades in college No. i than in the 23 high schools considered as a whole. ^ As has been said relative to previous charts, so here it may be reiterated that it is possible to determine what the relative standing of individuals is, as well as of the group, by following out the numbers accompanied by the characters plus and minus. For illustration, in high school No. 11 out of the 15 pupils who had a standing of i in the high school, 6 retained this standing in the college, 7 of them out from all the principals concerned precisely what is the range of the scale used in the various high schools, and exactly what percentages which they do use are equal to the I, 2, 3 marks of the college. From this investigation it was learned that the large majority of the high schools are using a range of 70-100 per cent, in which i equals 90-100; 2, 80-90; 3, 70-80. In the other several schools i equals 90-100; 2, 80-90; 3, 75-80; or i equals 90-95; 2, 85-90; 3, 80-85; or I equals 95-100; 2, 85-95; 3, 75-85; or A+ equals 97-100; A, 90-97; B+, 85-90; B, 80-85; C, 70-80. * The 23 high schools do not appear in any logical order because it was necessary to rearrange the charts for the purpose of printing them. 2 Charts 107 and 108 have been used in finding the retention between the composite 23 high schools and college No. i. 90 STANDARDIZATION OF SCHOOLS OF KANSAS fell back to the standing represented by 2, and 2 of them fell back to a standing of 3. The fact just pointed out is indicated by the accom- panying stars. This indicates that the standards of the two institutions are not the same, and probably, too, that not all of the high-school pupils are able to do the work according to the standard set up. It may mean that the standard of the college ought to be modified, together with the standards of the high schools. Since pupils need to readjust themselves whenever they enter dif- ferent institutions, it was thought that it would be of some significance to compare the first year of the high-school English with the first year of college English, as well as to make the comparison between the aver- age of the three years' high-school English and the Freshman college ft i w i * i ■A Col. No. 1 i i CO 1 D Col. NO. 1 Presh. Eng. Charts 89. 90 Presh. Eng. Charts 91. 92 Fresh. Eng. Charts 93. 94 Aver, of 4 yre Charts 91'. 92 1 2 3 Ter. Ret. 1 2 3 Ter. Ret. 1 2 3 Ter. Ret. 1 2 3 Ter. Ret. 1 50 34 5 56.18 59 27 3 66.28 10 13 4 37.07 18 7 4 62.06 ?. ?.9 .„ 31 31.81 24 6 38 23 2L 60 42i4S 67,41 11 f 10 22,22 11 13 4 44.26 i 11 55 61.79 6 e 13 48, H _^ 8 21 69,?6 Tot. Ret. 50.00 Tot .Ret. 59.02 Tot .Ret. 35.80 TotTRet. 60.46 10 i to 1 <* in Ofll. No. 2 i o •a i in CO Col. NO. 2 Table ?II1 showing sum- mary of the relative standing of pugils in' high schoal and college Frea'n. Eng. Charts. 95^ 96 Fresh. Eng. Charts. 97. 98 Fresh. Math. Charts 99.100 I 14 ?, 3 Ter, Ret. 14 2_ 11 2 Ter. Ret._ 51.85 1 2 3 Ter. Ret. 10 3 51.85 8 5 7 40.00 2 8 12 7 44.44 9 8 10 29.62 9 7 4 35.00 3 5 17 68.00 4 8 15 55.55 3 8 9 45.00 Tot. Het. 53.08 Tot. Ret. 45.67 T:6t.Reti 40.00 English. Table VIII indicates that the total retention is 59.02 per cent in the latter comparison and 50 per cent in the former, which probably signifies, in harmony with statements already made, that it takes the high-school student some time to get adjusted in his first year's work. A further comparison in charts 91' and 92' of the three years' average of high-school English with the four years of EngHsh taken in college corroborates this statement. For the 86 pupils out of these 266 show a somewhat similar retention to that in charts 91 and 92, namely, 60.46 per cent, as is shown in table VIII. The results of the comparisons in charts 93 and 94, together with the results in charts 97 and 98, as shown in table VIII, also justify the former statement. The total retention for high school No. 5 and college No. 2 COMPARISON OF RELATIVE STANDING OF PUPILS 91 is 53.08 per cent between the Sophomore high-school EngHsh, and Freshman college English; the total retention between Freshman high- school and Freshman college English is low, namely, 35.80 per cent; while that between the four years' average and the Freshman college English is 45 . 67 per cent. On the basis of the single subjects compared, the results warrant the conclusion that the correlation between the high school and college is better for high school No. i and college No. i than it is for high school No. 5 and college No. 2. The amount of retention for the schools compared on the basis of absolute marks is somewhat similar to that of the comparisons on the basis of the relative standing, as shown in table IX. The total reten- tion for English between high school No. 7 and college No. i is 53.57 W X 1 Col. NO. 1 1 Col. No. 3 ft_181 56 _182" J7 _i88 62 184~ 701186 76_ie6: 9 p n p ll B . 168 90 46 gg 164 91 X65 94 166 yy BI £^ 79 187 167 108 62 93 237 ~ggOT~95~2 23 6 4 189 nT^9~ T.l "2g9~20ri ~96~~224 87~~190 1 70 | " . 125 I 161 266 113 2 42 207 I 100 2 26 101 192 l7g~pL26 228 102 1 93 1 73 128 231 105 19^^ 174 "I5ff~ 262 2 61 157 248 2 16 127 26 6 263 138 2 49 2 17 129 £66 I 1264 139 | jESO 218 131 ) BO 5B 90 117 232 106 196 176 li37:eS3Z107__196_17i 236_10e 197 177 114 238 178 241 1161 1243 179 53^— ' ' "119" 351 "84 11 TT 166 69 40 17 168 75 43 18 169 82 9 7 .6 JU Graph Bhowlng dlBtrlhutlon of mexka In BngllBh of seae 266 etnd^Bts— Collego So. 1. T6 ' 80 ' 85 ' 90 ' Chart Ko. 90. Freeh. English of the saroe 266 BtndentB, College Ho. 1. 61> 230- 162* 112 80- 24 128* 3* 187 143- 282 166* 115- 88 25^ 135- 18* 188 14^ 69* 22* J» 233 167' 114 84^ 27* US- 56' 189 162* 62^ 86* 2*1 200- 284- 172 115 86* SB* 1B4- '21* 190 1E7- 63* 28* 4* 203 66- 235 173 116* 87' 37*" 180 23* 192 1"CT* 77 E9*- 6' 1 208 67* £36 175 120- 88_^ 38"m86 33* 193 160' 79 30' 6' 136- 41- 211 68- 241 176 121 89* 39' 194 34* 196 163- 62- 31 7* 139- 71- 218- 70 246- 181 122- 91* 42 205 43* 196 164' 90' 32' < 4^ 76 220- 78- 248- 186 123 92 46* 212 i?t 197 165* 101 36* .?„ : 56- 96 22B- 100 260- 191 126 94 Be* £2B 49* 201- 168* 102 40* 16* 198- 99 223 108 251- 199- 134- 95 6T)* 226 60* 202- 169* 106 44* 11* 206- 117 229- 137- 263- 207- 136- 97* 61- 237- 54 264 170 110* 46* 12* 217- 24- 238 142- 264- 210- 146- 98 66* 23^ ^^ 209- 174 116- 47* 13* 219- .27- 242- 144- 266- 213- 150' 104 69* 240- 64* 216- 176 119* il* "! 227- .29- 243- 147- 263- 214- 161- 106 72 243 61. 228 179 125* i2* '^^, 262- .30 266- 168' 264- 216 163- 107 73 247- 86* 231 182 132* 53* 16* 268- .31- 269- 171 265- 221- 166» 109- 74- 249- _93- 103*^ S44- 183 140- 6B 17* E61- 13^1 1260- 177 1 1266- 224 161^111- 75^267- 1262- 184 141 66' 19*1 ^ 7B 60 86 90 95 Graph showing dlstrlhation of marka In Boellah of 266. p^p l l B , In High Sohool lo. 1. 80 85 f-f-4 90 95 97.6 Chart Ho. 91. Areraga of 3 years In EngllBh for each of the 266 '7js,-ssA" 001 DO. 1. 88 pnpllB 89 pnpllB 162 26 1 153 61 166 87 S 154 68 210 148 159 38 8 156 71 Ell 145 10J_ 26 161 89 4 E06 74 212 42 164 40 f, 62 20 165 48 6 214 78 220 148 107 57 184 66 21 166 44 7 41 218 80 221 157 108 22 167 45 8 72 219 83 227 172 "116 66 205" 79 24__168__jl7_ 26 169 48 9 10 111 222 .88 230. 113 223 96 234 173 117 70 209 ar 28 170 60 11 122 224 99 236 24" 229 100 236 177 118 73 216 102 29 183 52 12 180 121 76 216 115 80 165 53 18 1 133 245 0O4 237 181 123 84 225 160 31 187 63 14 .39 260 109 £39 186 125 86 226 160 33 lee 69 16 142 £61 HE 240 198 ,126 87 £28 162 34 189 62 16 1 161 263 lEO 241 199 129 69 231 163 46 190 „_65__ 4F 256 £55 136 244 201 130 131 91 £32 171 92 £88 174 64 192 iO_ 51 193 97 18 19 £69 266 137 246 20g 132 93 238 175 . 49 194 103 28 261 267 138 248 203 134 94 E43 176 66 195 110 £7 £65 260 140 262 20? 136 96 247 178 68 J.96 119 32 IE66 1 ^64 144 1 |E63 £08 141 98 11249 179 69 11197 1E8 85 1 ^n BP M \»^97.E 94 STANDARDIZATION OF SCHOOLS OF KANSAS H o O lO 1^" g ip El 1 . o CO l&^&&(S Alt *A 5 I. in O C n 3 § lOCQ « 3 n in 14 *? ^ lO §5 — : 1 a COMPARISON OF RELATIVE STANDING OF PUPILS COOt> SEJ? CO «>o H rl to to D I r » 1 1 ■ H ^- rlfoie Q J^ili. gssg ^ 160 149 114 116 71 46 20 i 151 150 117 116 72 47 21 ■ 162 151 IIB 121 73 4R 22 1 163 1S2 l?.o 122 74 49 M 153 154 122 123 143 '146 ~ 76 60 -II- l^ G E Dlstrlhntlon of McrltB in Chart Ho. 106, C ' B ■ A »«H»-* 15* 9 46* iO 20' )1* 21* >3* 22* ^\ 1- 56 23' 4" 3* 58 25' 6' 5' 69 27' 8* 7* 24' 60 28' 35' 9' Pfi* 61 P»' 36' 10* 52* 66 30* ?fli 11* 67 68 31' 39' .2' 62-. 69 32* 48* 3* 66 71 37* 54 14' 67 74 ^K 70 V' 63 17* 73 ZS 42* 64* 18' '^Jt 43' 72 19* (i3- 70- 44' 78- 33* 64- 81- 45* 80- 34' lit « ' 1 Dletrlbntion of Marks in Chart Ho. 104. Chart Ho. 104. Presh. Math, of the same 184 pupils, College Ho, 8, 7* 4" 10* 13* 14^ 72* 70* 125 16* fll* 31* 77* 73' 130 25* 86* 32* 83' 82' 132 37* 87* 33* 85* 89* 36* 93' 88' 134 39* 90* 36* 102 94* 135 44* 91' 40' 104 . 103 140 45*: 92* 42' 109 105 141 46* 95* 43' 153 "164 ^ 158 110 142 48' 111 145 53* 112 147 54'*' 96* 49' 98' 50* 1* 99*^—51*- a* E* 159 113 148 67*^ 101 62* 5* v^^~ 114„160_„_62' 115 162 65* 108 68' fi* 123 69*- 16* 11' 70- 118 155 69' 131 60' 17* 12' ,71- 120 160 78' 136 61' Ifl* 20' 174- 122 161 97* LS7 63* 19* 24* 175- 126 164 100 L43 64* PI* 34' 176- 127 165 106 144 66* ?,?,* 41 177- 128 166 107 146 71' 23' 47' 179- 56 180- 129 167 116 138— 169^r-Tr7— 139 172- 119 149 173-121 161 74* 26* L56 75* 27' 156* 182- 67* 183- 157 76*- L68- 79* 28* 29* " SS' 184- 153 181- 124 178- 80* 30- u S A Chart Ho. 106. Fresh. Eng. of the s pupixu, * pp** 118 70* 23* 30* 71' 24* 31* 1* 121* 72* 25* 32* 2* 1P.3 73* 28* 33* 3' 126 74* 35* 34* 4* W^^ 128 76* 37* 36* 6* 129 81 38' 39* 6* 57* 133 82 41' 40' 7' 26 62* 134 84 42' 44' 8' 127 63* 138 87* 43* 47' 9* 180 68* 146- 88* 46* 49* 10' 131 76* 141- 91* 46* 6&* 11* 136 77' 142 94^ 48* 68' 12* 136 78* 149* 96^ 6T^ 83 13' 79* 146 97 63' 06 14' 139 80* 14fl 86* 15* 145' 89 149 100' 65' 93' le- 151 90 iRn ge 17* J 63 92* 152 102* 60* 116* 18' 156- 154 156- 159- 160- 162- 103' „61' 104* 64* 105^ 651 110, __60* 113' 67* 116' 19' "168^-107 Lisi- 108* J,53- 109 119- 20' 1 122._21*1 144- 27 164- 111 165- 117 69* 147 29* J-^ B ' A 98 STANDARDIZATION OF SCHOOLS OF KANSAS ^1X c c cv a o> ( [ I fl ri ^-|o^ 3s ss SIS ^ M <4 o gs "S 3P u*c Sis: MM Jci? IM Mil s H H IHHi SS COMPARISON OF RELATIVE STANDING OF PUPILS 99 ' — ^ ^1 —fl IT qcw ^ » s f",: il •^ 2 MM 5:5 :S n b i-fH 1 1 ^ o *• a lat"! r n • »- SSCJ Y M 5 ~ S •> r) 1 . 52 sgs ra i-H «> t i i-t 1 i.U n ;t Liui r>m.A ,«| HiHH s <^ === LSJ N^iO i-HH s ^ , «o»-a3 (Q 4 t o»o« a rirfi H 3S^ H M y SP r 5 _la 5 ^2 aSS H 1 ea 1 >» ? rt • IS s^i H iii. ssk khs M 3 s [^ bsa lOtCt- H l— Kl fl n? S S^q! A m rt 1 •a O M s m * h — -1 « 1 » » 1 N OJ CJ i » » 5 £5o II H' sss 1 10 WW w" li •S.O w n ! , ; T n ■"1 ^1 £ CMM 01 " • ► » y » » t * fc-S c cJgiSS: ^ ,r ^a WCMcScj 1* I2U L± _.£> H SS ^m .1 ■; SSSisL e w« 1 -i Xi • ■ J . __ . 5S sg^^,.gS5 1 ^ M IT f iitf » jL -rr l-l i2 "*$ ss ■^ »* . Issi fr ^ 1 1 • t' 5 " ki S) sss'ss U n lii M 5 4[ isi3s ^ •go* tot- oa a hSS M |rt <4 JSS 14 1 •&0* "--I 4o a" "* of" 00 y i-t «.kJ.« Jj ^^ « sds i » ^Uu,iJ. s? p] ri 5 s a S __ '{** _ 1 ssS H sii!}± J8- rd i! WD £S i^ SiS »1 a" q" Lei&lJ . r — l:?3;^R 1 =- Si H ■- K W kl S li ' aa r -^ — 1 'ff*' H 8. 2iS H Sa- SS'flO OM »r s O 10 t> CD onrt s^::5 •fto "" a' . en (0 > c lOO STANDARDIZATION OF SCHOOLS OF KANSAS SEC. V. COMPARISON BETWEEN THE RELATIVE STANDING OF THE SAME PUPILS IN THE GRAMMAR SCHOOL, HIGH SCHOOL, AND COLLEGE The purpose of the previous comparisons has been to try to deter- mine the existing relation between the grammar schools and high schools, and also between the high schools and colleges, but not necessarily deal- ing with the same pupils throughout the three institutions. The pur- pose of this last section of chap, iii is to try to determine the relation of the grammar school to the high school and the relation of the high school to the college on the basis of the marks made by the same pupils who have attended all three of these institutions of learning. In the former sections naturally there were more different pupils involved, because it is very difficult to find reliable records of large numbers of pupils who have attended the three institutions.^ In order to make such a comparison as this in sec. V it is necessary to have available records covering at least the eighth grade, the four years of high school, and the first year of college work. But it is easy to see what a tedious process it is to gather much of this sort of reliable data when it is remembered that one-half of the grammar-school pupils drop out somewhere near the completion of the fifth grade, and that not more than 5 per cent, approximately, go on to high school, and not more than I per cent enter college. So far as the writer is aware, the collec- tion of the marks of pupils attending the three institutions has not pre- viously been done in an extensive way. And it will need to be carried much farther in order to justify wider conclusions.^ While this section also involves a separate comparison between the grammar school and high school, and between the high school and college, yet it is not a mere duplication of the former sections. It will be of some interest to see whether the percentage of retention for the pupils who attend all three of the institutions is similar to the results in the former sections, although the largest group of pupils used in sec. ^ The difficulties in securing reliable data for such a comparison as this are obvious. Many pupils who graduate from high school have completed their eighth-grade work in cities other than the one in which the high-school work has been done. Some pupils have come in from the rural sections where records often are poorly kept. Some pupils when partly through the high school either move to another city or drop out temporarily and consequently records are not continuous. Many high-school pupils who are reported by principals as entering college leave before any record worth noting is made. These are only a few of the veritable difficulties met with in the actual collection of marks. * High schools Nos. 5,1, and 6 have furnished the majority of the records for this triple-institution comparison. COMPARISON OF RELATIVE STANDING OF PUPILS lOI V in comparing the grammar and high school has not been previously used in this thesis. Many of the pupils already included in the grammar- school discussions never went on to college. The charts, tables, graphs, and diagrams used in this section are similar to those used in earlier sections, and so will need no further explanation at this juncture. The diagram used is probably the main original supplement to the graphic representation by the use of charts. Since it is not possible to compare more than two subjects or two institutions at any one time by the use of the charted marks of pupils, it was necessary first to compare the marks of the grammar-school pupils with those of the high-school pupils, and then to compare the marks of these same pupils in the college with their marks received in the high school. In order to read charts no, in, 112, then, first it is necessary to compare chart no with chart in. This will show how the pupils have maintained their relative positions in the groups, or how they have shifted their relative positions in the high school. The starred num- bers in chart in indicate that originally the pupils represented by these starred numbers were located within the high group of the grammar- school work in eighth-grade English. The numbers accompanied by the minus characters indicate that these pupils originally began their grammar-school work with a position in the low group. The plain numbers in chart in represent pupils who have come from the middle group of the grammar school. Ignoring, then, in the second instance the stars and minuses In chart in, it may be read again in a similar fashion in relation to chart 112, which shows the positions of pupils in the college work who have come from the high school. The tendencies in distribution of the groups which have appeared in other sections relative to grammar school No. 5', high school No. 5, and college No. 2 may be seen in these later comparisons in charts iio- 15. In no or 113 the marks of the grammar-school pupils are most numerous between 86 and 95 per cent, while the marks of the same high-school pupils are most numerous between 75 and 90, and 78 and 89, respectively, in charts in and 114, while the marks of these same pupils are bimodally divided in college No. 2, as is indicated in either chart 112, or II 5 . Is this amount of zigzag shifting within precisely the same group of pupils as they pass from one institution to another justifiable ? Or is it probable that if the institutions should agree more definitely upon I02 STANDARDIZATION OF SCHOOLS OF KANSAS the rating of students among themselves and between each other the shifting of groups of this kind would be quite materially reduced ? When the sixth and seventh grades of school No. i are included in charts ii6, 119, 123, 126 the tendency of the upward skew occurs in a similar fashion to that pointed out in previous sections where only the eighth-grade work was considered, in other schools.^ On the other hand, as we have seen previously with respect to gram- mar school No. 6 in charts 60 and 64, so here again we may observe a tendency toward a normal distribution curve, as is evidenced by such charts as 129, 132, 135, 141, 148. Charts 120 and 124 are both skewed more toward the top than is 122. This likely shows that there is some irregularity in the standards used by teachers. Otherwise chart 124 would be more like chart 122 than like chart 120. Another explanation may be offered, namely, that a select body of pupils is represented in 124, since chart 125 showed that as a group they have held their place quite well in college. Graphs 116, 117, 118 indicate as well as graphs 119, 121, and 122 that the grouping is more similar between the high school and college than the grouping between the grammar school and high school, and a similar conclusion may be made regarding the charts 123-28. However, the graphs representing the sixth-, seventh-, and eighth- grade English of 50 pupils, and an average of their four years' English in high school together with an average of their three years' work in English in college, as shown respectively in charts 123, 124, 125, are enough alike to justify the assertion that the standards used in grammar school No. i', high school No. i, and college No. i are in the main some- what similar. In a modified form this statement will hold also with respect to charts 126, 127, 128. The 74 pupils in mathematics represented by charts 132, 133, 134 are part of the group of 90 students in English represented by charts 129, 130, 131. There is a general corresponding similarity in the charts for the respective institutions. But, is the almost equal distribution of marks in chart 134 justifiable ? Charts 129-40 indicate that the grouping of pupils is more alike between grammar school No. 6' and college No. 3 than between high ^ The system of grading in grammar school No. i in the records used was whole numbers and fractions, as, for example, i, i|, ij, i|, 2, 2j, 3, etc. i is used to repre- sent 95 per cent; i|, 90; 2, 85; 2I, 80; and 3, 75 per cent for purposes of charting. As before noted this is not an absolutely accurate method, for i, or 95 per cent, really stands for something in the range of 95 per cent. This translation was made into percentages because some of the later marks had been recorded only in integral numbers. COMPARISON OF RELATIVE STANDING OF PUPILS 103 school No. 6 and the college.^ The grammar school has a more normal distribution of marks than the college, particularly in the subjects of mathematics and history. Judging from charts 143 or 145 or 147 the distribution either in chart 142 or 146 is more justifiable than that of 144, for the pupils in 145 do not maintain their position so well as in 142 or 146 when they go on to college. Furthermore, the distribution in 142 and 146 corre- sponds more to 141 than does that in 144. Chart 141 represents 35 pupils in eighth-grade English from grammar school No. 6' who go on to college.^ Charts 141, 142, 143 indicate a general similarity of groupings used in the respective institutions just In ■ H.S.K0.5 Cb.110.111 u P4 Col.Ko.2 Ch. 113. 112 ' H.S.N0.5 Ch. 113 .114 1 Col .110.2 Ch.114.115 i-r.EnK. Fr.Eng. 60 c 1 A -u CO Fr.and So.Eng. Pr.EnK. in ^. I. A 00 1 2 3 Ter. Ret. 1 2 3 Ter. Ret. . 2 3 Ter. Ret. 1 2 3 ret. Ret. 1 7 4 7i 50.00 7 5 2 50.00 7 5 2 50.00 7 5 2 50.00 ?. 5 4 30.76 5 4 4 30.76 4 5 4 38.46 5 5 3 38.46 ^ 2 S 7 50.00 2 4 8 57.14 3_ 3 8 57.14 2 3 9 64.28 Tot. Ret. 43.90 Tot .Ret . 46.34 Tot .Ret. 4^.76 Tot. Ret. 51.21 TABLE X Showing the retention in grammar school No. SS high school No. 5, and coUege No. 2. as 141, 144, 145 and 141, 146, i47 and 148, 149. 150 do when compared. But in the majority of cases there is a closer likeness in the grouping between grammar school No. 6' and high school No. 6 than between high school No. 6 and college No. i, and this may be due in part to the fact that the college is not located in the same city and consequently is not so likely to dominate over the high school in setting standards. The summaries of the retentions are presented in the tables that follow. Table X indicates the results of the comparisons made between the marks of 41 pupils who attended the three institutions, namely, grammar school No. 5', high school No. 5', and college No. 5. The retention for charts iii and 112 is higher than that for charts no and III, and the retention for charts 114 and 115 is higher than that for charts 113 and 114 in the subject of EngUsh. According to this result the relation between high school No. 5 and college No. 2 is closer than the relation between grammar school No. 5' and high school No. 5. ' This college is located in the same city with the grammar school and high school. ^ These pupils are numbered in order of their standing determined from the exponents accompanying the marks. I04 STANDARDIZATION OF SCHOOLS OF KANSAS The retention for the middle third and for the upper third is, however, the same for each institution. The retentions here correspond rather closely to the comparisons made between high school No. 5 and college No. 2 in the larger group of the earlier section. The general result of the comparison made between grammar school No. 1' , high school No. i, and college No. i may be seen in table XL The retentions both between the grammar school and high school and between the high school and college are considerably higher than was i to H. 3. Ho. 1 S w i Col. Ho. 1 i 00 ii. S. Ho. 1 a w 4 2 Ch . HO. . J Begin Ger. Ch. 116. 117 tiod. I Ch 1] jang. 7. 118 _ Freeh. Eng. Ch* 119. 120 120!"!21 1 \ ?. 3 Yex. Ret. 1 2 . Ter. Ret.. 1 2 3 Ter, pet. 1 2 3 Ter. 1 7>0 6 ■■. 81.08 28 8 1 1 75.67 40 9 4 76.4^ 33 20 62.2^ 2 7 t 11 51^33 8 19 10 61»33 13 26 14 48,07 16 21 }-^ 40.38 3 12 26 67,66 li 10 26 70^27 18 36 66.03 4 11 38 71.69 Tot. Eet. 66.66 Tot.Ret. 66.76 Tot.Ret. 63.29 Tot.Ret. 67.69 i CO H. S, Ho. 1 i i Col. Ho. 1 i s" to H. S. Ho. 1 1 to 1 col. HO. 1 1 CI rer .i,iii^' freeh. £ng. £hl22^ 121 Aver. 3 yrs.ilng Ch, 123. 124 AV« Ch r.4 yre.Engl ,124.. 126 1 1 2 3 Rer. Ret. 1 2 3 Ter. Ret. 1 2 3 Ter. Ret. 1 2 3 Ter. Ret, 1 33 15 6 62.26 32 19 2 60,37 10 6 1 68,82 11 B 1 64,70 2 16 25 12 14 23 15 44.23 e 6 6 37,50 6 7 4 43.76 3 5 12 36 s 9 36, 67.92 2 4 11 64.70 2 4 11 64.70 Tot. Ret. 59.49 Tot.Ret. 56.96 Tot.Ret. 54.00 Tot.Ret. 68.00 1 < i 00 H. S. Ho. 1 i >> 5 Col. Ho. 1 Table XI tlon t) school K school H , ehowing reten- etveen Gresmaz 0. l.High 0. 1. and CH. 126. 127 trreeh. llath. Cb, 127. 128 1 ?. 3 T.P. 3 T.R, r 16 2 6 saVee 61.11 2 2 13 4 68.10 6 8 6 42.10 3 4 14 mil 78.18 1 4 13 72,22 T< t. Rel (. Tot.Ret, 58.18 Ci] Liege ii:o. 1 found to be the case in grammar school No. 5', high school No. 5, and college No. 2. Part of this may be due to the necessity of charting on the three-estimate basis rather than on a wider per-cent basis, but it is no doubt safe to assume upon the basis of the results as they appear in table XI that the correlation between the primary, secondary, and higher institutions is the closer in the latter comparison.^ ^ The high-school marks used were i, 2, 3. In this comparison i equals 97.5; 2 equals 90; and 3 equals 80; an average of 2, i equals 93!; an average of i, 3 equals 88. 75; an average of 2, 3 equals 85. For convenience, 93! is charted as 95, and 88 . 75 as 90. In the college marks of i, 2, 3, number i is used to indicate 95; number 2, 85; number 3, 75; an average of i, 2, 90; an average of i, 3, 85; an average of 2, 3, 80. COMPARISON OF RELATIVE STANDING OF PUPILS 105 The triple-arranged columns opposite the numbers of the individuals show whether a pupil has shifted or maintained his or her position in passing from one institution to another. For example, 27, 17, and 16 did not shift out of the high group; 4, 14, 19, and 28 maintained their positions in passing to the high school, but 4 and 14 dropped down to the second group in the college, while 19 and 28 dropped to the third group. By a glance at the table as a whole it may be seen that there are comparatively few third-group pupils who have risen to the high tertile; comparatively few first-group pupils fallen to the low tertile; in the middle tertile there is more of a mixture and shifting of positions indicated. PapU Low Tortus Pupil Hid . Tertile Pupil Hl«}i Tertil* 1 6 3 3 3 10 2 2 2 27 1 1 30 3 3 3 41 2 2 2 17 1 1 8 22 3 3 3 3 3 3 12 23 13 2 2 2 16 4 14 1 1 2 2 3 1 1 1 2 2 3 3 2 2 1 40 3 3 2 34 2 1 1 19 1 9 9 24 9 S 1 3 20 25 2 2 1 1 2 2 28 31 1 3 1 3 2 1 2 26 3 2 3 35 2 1 2 18 1 2 1 32 3 2 2 2 2 3 3 38 1 2 1 37 3 2 1 6 2 3 3 29 1 2 3 11 3 2 1 21 2 3 3 36 1 3 1 1 3 1 1 7 2 3 3 33 1 3 2 39 3 1 ' 15 1 3 2 TABLE B* Showing the relative standing of each individual in grammar school No. 5', high school No. 5, and college No. i. * The numbers used for pupils in table B and table C are the same numbers given miscellaneously to pupils in the larger previous groups. This explains the fact that number 2 above and number 3 in table C, for example, are recorded in the second group of the grammar school, etc. This triple-table arrangement has been here suggested because it is applicable to small as well as to large groups. In the above instance the number is somewhat small. The second column of the table in the high tertile indicates that seven pupils held the same group position in the high school which they held in the grammar school. This same fact is indicated in diagram III by the number 7 in the first tertile of the high-school group. Four pupils in diagram III, as indicated by the second column of the middle tertile of table B, held the same relative position in the high school which they held in the grammar school. In this manner the shifting or retention of each individual pupil may be traced out by following the lines in the diagram. From this table it is easy to construct the diagram which traces the groups as a whole. Both the triple table and the diagram show that many pupils tend to keep within the same groups, respectively, as they pass through the different institutions. Out of the 14 pupils in diagram III who appear in the high third of the high school 7 have come from io8 STANDARDIZATION OF SCHOOLS OF KANSAS column of the high tertile shows, 25 in the middle, as the second column of the middle tertile shows, and 35 in the lower, as the second column of the low tertile shows. By reading the three columns simultaneously, under any one tertile, it shows how many pupils were retained through- out the three institutions, which in this case is 28 in the high, 10 in the middle, and 25 in the lower groups, respectively. Pupils low Tertile j Pupile Hid. Tertile Pup^lB High Tert4e 107 3 3 3 81 2 2 2 1 156 3 3 3 84 2 2 2 2 41 3 3 3 85 2 2 2 4 127 3 3 3 86 2 2 2 5 131 3 3 3 87 2 2 2 6 139 3 3 3 24 2 2 2 7 142 3 3 3 69 2 2 2 8 144 3 3 3 75 2 2 2 9 145 3 3 3 94 2 2 2 10 147 149 3 3 3 3 3 3 54 82 2 2 2 11 12 2 151 3 3 3 31 2 2 13 153 3 3 3 55 2 2 14 1 114 3 3 3 62 2 2 15 1 117 3 3 3 77 2 2 16 120 ,3 3 3 79 2 2 17 129 3 3 3 90 2 2 19 133 3 3 3 102 2 2 22 134 3 3 3 105 2 2 26 136 3 3 3 119 2 2 29 137 3 3 3 97 2 2 32 138 3 3 3 106 2 2 40 1 95 3 3 3 115 2 2 44 96 3 3 3 116 2 2 45 100 154 3 3 3 121 28 2 2 52 56 3 T- 2 2 157 3 3 2 30 2 59 148 74 3 3 3 3 2 2 36 47 2 2 1 1 63 18 _k_ 2 88 3 3 2 81 2 20 2 135 3 3 2 3 2 21 2 143 3 3 1 3^ 2 2 23 2 118 3 3 1 34 2 ? 27 2 140 3 3 1 37 2 2 33 2 146 126 3 3 1 3 43 46 2 2 2 2 38 39 2 2 3 2 155 3 2 3 48 2 2 49 2 158 3 2 3 53 2 2 50 2 150 3 2 3 83 2 2 58 2 130 98 3 3 2 2 3 3 92 93 2 2 3 3 2 2 60 42 2 2 2 108 3 2 3 66 2 3 3 57 2 2 25 3 2 2 68 2 3 3 64 2 2 89 3 2 2 99 2 3 3 65 2 2 91 3 2 2 104 2 3 3' 72 2 2 1>03 3 2 2 109 2 3 3 73 2 2 128 3 2 2 111 2 3 3 67 2 3 152 3 2 112 2 3 3 70 2 3 izi. 3 2 113 2 3 3 76 2 3 141 3 2 122 2 3 3 61 3 2 101 3 2 123 2 3 3 80 3 2 110 3 2 • 124 2 3 3 71 3 3. 125 3 2 78 3 iJ TABLE C Showing relative standing of each individual in grammar school No. i', high school No. i, and college No. i, of 158 pupils. From a glance at table C it may be seen that a great many of the pupils tend to remain within the group in which they started out in the grammar school. There are some extremes, such as numbers 71 and 78, or 118 and 140, but these are comparatively few. The relative decline or progress of a student's work throughout the three institutions may readily be COMPARISON OF RELATIVE STANDING OF PUPILS IO9 seen in such cases as 67, 70, and 76, or in such cases as 152, 132, and 141, respectively. The proportionate retention is higher between grammar school No. i', high school No. i, and college No. i, as shown by diagram IV, than is the retention for grammar school No.. 5', high school No. 5, and college No. 2, as shown in diagram III. In diagram IV there are no pupils who pass from the lower third in the grammar school to the higher third of the high school, but there are four pupils who pass from the higher group of the grammar school to the lower group of the high school. These latter four pupils never get back to the high group in the college; two of them remain in the lower group, and two of them get up to the second group in the college. On the other hand, in diagram III three pupils fall to the low group as they pass from the grammar school to the high school. But two of these get back to the middle group and one of them up to the high group in college. Diagram IV indicates that while many of the pupils do pass straight through the three institutions within the same group, yet some of those who appear in the respective groups in the college have arrived there by devious pathways. The retention for the higher and lower tertiles is clearly higher than that for the middle. The advantage of such a diagram as this is that it shows what sort of pupils, in the way of schol- arship, constitute the groups at the several stages of progress in the different institutions. Table XII is a summary of the comparisons made between gram.mar school No. 6\ high school No. 6, and college No. 3. These schools all use the letter system of grading but it was not possible to secure any large number of marks from the college records.^ In table D, column 2 in the high tertile shows a retention of 15 pupils in the high school; column 2 in the middle tertile, a retention of 12, and column 2 in the lower tertile, a retention of 20 pupils in the high school. Columns 1,2, and 3 in the high tertile show a retention of 11 pupils throughout the three institutions; columns i, 2, and 3 in the middle ^ The college uses the letters a, b, c, the high school and grammar school each e, g, and /. These were reduced to percentages in the same manner as has been indi- cated with the previous marks. While the cases are not numerous here, the attempt has been made to get as accurate a collection as possible. In chart 129 the pupils are numbered in order of their standing, e is used to represent 95 per cent; g, 85; c, 75, etc. no STANDARDIZATION OF SCHOOLS OF KANSAS tertile, a retention of 5, and columns 1,2, and 3 in the lower tertile, a retention of 16 pupils throughout the three institutions. i- • fi 00 H.S.K0.6 Ch. 129, 130 > 0) Col. No. 3 Ch. 130, 131 1 CO H.S.1J0.6 Ch.132.133 Col. He. 3 Ch. 133, 134 Aver. 4 yrs.Eng Fr.Eng. Fr.Math. Fr.Math,___ 1 2 3 Ter. pet. 1 2 3 Ter. Ret. 1 2 3 Ter. Ret. 1 2 3 Ter. Ret. 1, JL5 8 15 50.00 14 11 _5 4 46.66 15 8 2 15 , 3 60.00 2 12 10 40.00 14 12 40.00 8 6 10 25..0P 52.00 8 7 9 29.16 3 7 3 20 66.66 2 7 21 70.00 2 10 13 10 .3 52.00 Tot .Ret. 52.22 Tot, Ret,. 52.20 rot. Ret.. 45.94 rot. Ret. 47.29 i H.S.N0.6 Ch.135.136 CO J. Col. Kg. 3 Ch.136,137 1 CO H.S.N0.6 Ch. 138. 139 CO Col.lJro.3 Ch. 139,140 Aver.cf Fresh. &. Soph.Math. Fr.Math. Soph.Hist ._ rr_.Jii3t. 1 2 3 .Ter. Ret. 1 2 3 Ter. Ret. 1 2 .3. 3 Ter. Ret_. 1 2 3 Ter. Ret. 1 B 10 1 42.10 12 6 1 63.15 13 5 61.90 12 5 4 57,14 ?. B 4 B 20.00 7 4 9 20.00 5 11_ 4 5 52,38 4 61.90 3 3 6 10 52.63 10 T ' 9 147.36 4 13 61.90 5 3j ;3 61.90 Tot. Ret. 37.24 Tot .Ret. 143.10 Tot .Ret... 58^ Tot. Bet. 60.31 TABLE XII Showing retention in grammar school No. 6', high school No. 6, and college No. 2. PUpllB Low terlj lie Pupils Kid. Tertile | PupllB High Tertile 1 63 3 3 34 2 2 .2 2 1 68 7, 3 37 2 2 2 3 72 3 3 89 2 2 2 * 74 3 3 44 2 2 2 6 75 76 3 3 3 4B 33 2 2 2 6 7 1 2 2 1 78 3 36 2 2 14 80 3 41 2 2 21 82 3 43 2 2 22 83 9 51 2 2 23 1 84 85 87 3 3 3 2 ?; 24 A i 1 47 31 2 2 10 13 1 1 1 2 2 2 1 88 3 46 2 1 15 89 90 3 56 2 1 20 i 3 48 50 2 2 1 1 2 2 1 8 i 2 2 2 53 2 1 2 12 2 59 2 1 2 16 86 3 1 38 2 3 2 19 2 71 3 49 2 3 2 26 77 61 3 2 2 52 60 35 2 2 2 3 3 3 2 2 3 29 30 9 2 2 2 2 2 2 69 62 70 73 79 81 2 1 3 3 3 3 40 42 55 57 54 58 2 2 2 2 2 2 J 3 3 3 3 1 3 3 3 3 3 3 1 11 17 18 25 27 2 2 2 2 2 TABLE D Showing relative standing of each pupU in grammar school No. 6', high school No. 6. and college No. 6, of 90 pupils. Diagram V and table D show that no pupils in passing from the high third of the grammar school fall to the lower third in the high school and then pass back to the high third in college; but number 62, for COMPARISON OF RELATIVE STANDING OF PUPILS III _L j ' r-|-. I 1 I Ui.J » '^ IF --tI ^^- '?* CK OJ -vj Co R v: •i< s; ^^ ^ ^ -CJ c^ ^. ^ >^ I =t^ ^ ^ ■S-S to \) > » u 112 STANDARDIZATION OF SCHOOLS OF KANSAS example, passes from the lower group of the grammar school to the middle group in the high school and to the high group in the college, and numbers 70, 73, 79, 81 go from the low group in the grammar school to the high group in high school and back to the low group in college. As previously stated, it is important not only to know in what respective groups pupils appear in college, but it is equally important to know over what path, circuitous or straight, they have come. Table D and diagram V show that there is a larger proportionate retention than was the case in table B and diagram III, but that there is a smaller proportionate retention than was the case in table C and diagram IV. Consequently the result is that according to this dia- grammatic scheme the retention of pupils throughout the three institu- . Col. Ho. 1 OiartB 1427142 Tot. Ret. 65.71 Charts 146,147 Tot. Ret. Ter. Retj. 58.55 4?t45 66,6^ H. 3. Ko. 6 Charts 1417144 Tot. Ret. 50^00 66t66 lOt^Ret. Ter. Ret. Tliii 55. 55 55.00 Cal. Ko. 1 Charte 144,146 Bng. Col. I?o. 1 Charts 149,150 Tot. Ret. 71^ 70.00i TABLE XIII Showing retention in grammar school No. 6', high school No. 6, and college No. i. tions is highest in the case of grammar school No. i', high school No. i, land college No. i. On the whole, it is obvious that the percentages of retention are ower than in table XI. With the exception of the relation between charts 129 and 130 the retention is higher between the high school and college than it is between the grammar school and high school, as indi- cated by table XII, which is exactly the reverse of the results shown in table XI. The retention for both the grammar school and high school and for the high school and cohege is not far from 50 per cent. The number of pupils involved in the comparison summarized in tables XII and XIII are too few to make anything but tentative con- clusions. But on the whole, the percentages of retention between COMPARISON OF RELATIVE STANDING OF PUPILS 113 N --- — --•-— — - — — -.—. » r r- ■n 7 I < I < I • •k \ I 1 J ^ 5 1^^ ^ ..i .till *f -^j lii 1-4 r - -tz-:--: ._--L .:T—^ o a) "St <3 -xr 15 114 STANDARDIZATION OF SCHOOLS OF KANSAS high school No. 6 and college No. i are higher than the retentions between high school No. 6 and college No. 3. This is some indication that the standards in college No. 3 and college No. i are different. The results of the summary tables for the whole sec. V, which com- pares the grammar school, high school, and college marks of the same pupils, are as follows: The retention for grammar school No. 5', high school No. 5, and college No. 2 is about 50 per cent; that for grammar school No. i', high school No. i, and college No. i is about 60 per cent,^ and that for grammar school No. &, high school No. 6, and college No. 3 is somewhat above 50 per cent, according to the tertile method. When comparisons are made in terms of the average of the percentages of those pupils in the upper and lower tertiles who remain above or below the median there is a retention between grammar school No. 5' and high school No. 5 of about yo per cent;^ between grammar school No. i' and high school No. i of about 85 per cent;^ between grammar school No. 6' and high school No. 6 of about 75 per cent.^ The retention between high school No. 5 and college No. 2 is about jo per cent; between high school ^ It may be that a part of the higher retention in schools Nos. i', i, i, is accounted for by the influence of college No. i, as before stated. 2 The actual retention is 71.92 between eighth-grade and high-school Freshman English; 67.90 between high-school Freshman English and Freshman English, college No. 2; 74.99 between eighth-grade Enghsh and Freshman-Sophomore high-school English; 71 .47 between Freshman-Sophomore Enghsh and English, coUege No. 2. 3 The actual retention is 85 . 13 between sixth-, seventh-, and eighth-grade English and high-school German; 83 . 78 between high-school German and Modern Languages in college No. i; 80. 18 between sixth-, seventh-, and eighth-grade English and high- school Freshman English; 83 . 95 between high-school Freshman English and Fresh- man college English; 83.91 between sixth-, seventh-, and eighth-grade English and average of 3 years' high-school English; 83 .01 between average of 3 years' high-school English and Freshman college; 82.35 between sixth-, seventh-, and eighth-grade English and average of 3 years' high-school English; 88 . 23 between average of 3 years' high- school English and average of 4 years' college Enghsh; 91 . 66 between sixth-, seventh-, and eighth-grade arithmetic and average of 2 years' high-school mathematics; 88 . 88 between average of 2 years' high-school mathematics and Freshman college mathe- matics. 4 The actual retention is 79 . 99 between eighth-grade English and 4 years' aver- age of high-school Enghsh; 75 between 4 years' average of high-school English and Freshman college; 62 between eighth-grade mathematics and high-school Freshman mathematics; 80 between high-school Freshman mathematics and coUege Freshman mathematics; 73 .68 between sixth-, seventh-, and eighth-grade arithmetic and high- school Freshman-Sophomore mathematics; 81.57 between Freshman-Sophomore high-school mathematics and Freshman college mathematics; 83.33 between eighth- COMPARISON OF RELATIVE STANDING OF PUPILS II5 No. I and college N^o. i about 8j per cent; between high school No. 6 and college No. 3 about y^ per cent. The general result of the comparisons made between the grammar schools, high schools, and colleges in sec. V is that there are many pupils, as shown by the diagrams, who go through the three institutions without shifting their positions outside of the groups in which they began their grammar-school work. And although there is some shifting in the high and low groups, there are relatively few pupils who make extreme shifts in either the way of decline or progress in passing through the three institutions. Naturally there are fewer pupils who maintain their positions throughout the three institutions than between any two, respectively. While some of the retentions between grammar schools and high schools and between high schools and colleges are below 75 per cent, and one school is considerably above 75 per cent, yet it is safe to assume that for the schools as a whole there is a retention in terms of the modi- fied median method used of about 75 per cent between the three institutions of learning — namely, the primary, secondary, and higher institutions. grade history and high-school Sophomore history; 69.04 between high-school Sopho- more history and college Freshman history. There are some irregularities in the retention between high school No. 6 and college No. i, but the cases are not numerous enough to modify the above results materially. ii6 STANDARDIZATION OF SCHOOLS OF KANSAS HHH(W[ to ^ AA^i J, COMPARISON OF RELATIVE STANDING OF PUPILS 117 3 at n D a* > B» 10 at »» I i a «o n r »» » » ss^ ^ » » » » • ?:s;5?sgs A^fy, K n D 'rd&» 1 _ " . C ^ f r i-(B rdsss ■J,s» n^<« oD » ssss t ■„ a J 3 r 3 ^ .i. J: • 3 i, 1 4- • hI« ^ ; « * a NO V *" «f < ^ *^^s? 2 2 f iSiis c » H t 4 .. g ■• a> DO B» ^ J* to 01 n »» 1- s « « »v M ». * <# H H A 3 »» 1* r 0» » H D »«uJL» Ho (0 B L ^SS Vg » »^^g ooff VI HMej CI k tii^it i ^^^l u e n oaa> j <« a ^«i« K '4iI0 Ji4n || MD 1 *as %4 NJiH» * N c 1 .Us ^ i 1 t- > •8 •g 60 « L m e> Z t ic '^ 4* I a M Wi a • M 1; 1 8 1 1 1 ?5S ii8 STANDAEDIZATION OF SCHOOLS OF KANSAS i- e> ti^ .A^ K a 10 C « H te « i> »> Ji £ A I20 STANDARDIZATION OF SCHOOLS OF KANSAS ^^ tig' 4J - Oil cbOkftM^ 10 4.i^J>(^ ^i 4< OH Pi g| i^ • .& to H hiilTlH ■p a> • CO « to •H < o| H EiSi 55 H too OHO> wo 5^ SI I f-)« K ^l£ « i'c ^i ^iS ^g S^ J^i as ■ HW , .A.& SS3 i^^i AT] 10 _ J!! lO at s ;i II S ^: COMPARISON OF RELATIVE STANDING OF PUPILS 121 122 STANDARDIZATION OF SCHOOLS OF KANSAS iyA^^6J>J>l tU J.6A iJ, s 1 u HriririHriH \ot-\aa s 0>OMN(<} oiAioaai 4tA F-OBI COMPARISON OF RELATIVE STANDING OF PUPILS 123 ^ 4J> cici 4.C IN OD H s a, 3 'I' t- CD H s ;^ in •* E 10 w 2 H s H Cvj ki S •* CM H 4 ot >-o ^ I0 ^ A e- 0» » W( •* T)i «i> » ^ IF t i Tl • i JL i. V?, ^a X> cc a a. M «;^ < ^ u ^ u COMPARISON OF RELATIVE STANDING OF PUPILS 12 J>J, ni I I I i HiAooJo>h1n oQOt^'^lia h w Mi-«ciW!H Nto^io t-'c^nJ' |HMMCl|td §- ~ f — — s g oo 3 CO o 1 HtO ^ lO fc 3 ^; 3S h^ s i "^ N H to ;i5 «• 00 0» ^ s to H ^1 3k 4 .A I ^■^ ci.A Hhc $* SI UI CHAPTER IV General Conclusions A few conclusions incidental to the main issue of this thesis, and yet not irrelevant to it, will first be set forth. The completer standardi- zation of schools in any state will need to be based upon a series of investigations. Such a study as the present one, the writer believes, has value in establishing a more scientific attitude in the analysis of practical school problems. This thesis has dealt with marks or grades relative to the standardi- zation of schools. There can be no doubt but that scholarship of pupils is one important factor to be taken into account in attempting to get a measure of the efiiciency of school systems. Nevertheless, in order to get a balanced estimate of the working standards of school organizations, it will be necessary to carry out other studies supple- mentary to any study based upon marks. It would be very profitable if some such problems as the nature of school support, the value of material equipments, the measurements of teachers, the significance of the physical, moral, and social life of the pupils, could be worked over in detail in the light of our more scientific attitude in educational doctrine and practice. As is obvious from the previous sections of this discussion, there is considerable difference of opinion as to the manner of rating pupils within any one state. In the light of this, one cannot avoid the conclu- sion that there is a great need of clarification, together with more common agreement as to what is the best system of marking; furthermore, as to what is the most expedient form for the preservation of intelligible, accurate, complete, but simple records of the marks made by pupils. The variation in the distribution of marks referred to frequently in the earlier discussion is evidence that there is lack of agreement in estimating the abilities of pupils. The most general tendency through- out the schools as a whole was the skew toward the top of the scale. Furthermore, in some instances it was found that the grouping of pre- cisely the same pupils was quite different from year to year within the same subjects. We have seen reasons for emphasizing the fact that there is need for the agreement of teachers among themselves upon the rating of pupils in the subjects within any one school. 126 GENERAL CONCLUSIONS 1 27 There is no little difference of opinion as to the range of marks to be used. It is common to find one of two extremes — either a too-wide range, which frequently grows out of a percentage system, or a too- narrow range, which grows out of a letter or number system. When the theoretical scale is too wide, some points or marks are very likely not to be taken into account at all, or sometimes there is a theoretical differ- entiation too fine to be practicable. When the range or scale is too narrow, a lumping off frequently occurs with a consequent lack of differentiation. Why would it not be possible to adopt a theoretical scale which will be likely to be followed in practice, and which will, at the same time, overcome in some measure both of the above-mentioned extremes ? It will make comparatively little difference whether numbers, letters, qualifying words, or percentage systems are used, providing whatever marks we do decide to use are stated in translatable and comparable forms. Probably a six-, seven-, or eight-estimate system would serve very well for a compromise. And would it not be very profitable to keep a record of the degrees of failure as well as to keep a record of the points of difference between those who pass ? We have failed in a large measure to recognize that there is no absolute demarkation or abrupt dividing point between the eliminated pupils themselves as a group and also between the eliminated pupils and those who go on. It is valuable to know not only who has failed and who has passed, but also to know how much more some pupils have failed to pass than others. The writer is convinced through his experience in attempting to collect the data for this thesis that we are very deficient in the keeping of significant continuous records covering at least the period of years included by the eighth grade of the grammar school, the four years of the high school, and the first year of the college. Such records, of course, are indispensable if we care to make a study of pupils' progress through- out the three institutions. If typical school systems over our country would keep a careful record of the marks of several thousand pupils covering this period of six years of school life, such data in the form of certificates kept on file in the college vaults would furnish material for a check experiment to such an investigation as the present one, and consequently would furnish a means of testing the validity and value of present conclusions. The permanent records can be conveniently kept in the loose-leaf form within bound volumes in both the grammar school and the high school. The transfer of these to the college-entrance certificates would be a simple matter. 128 STANDARDIZATION OP SCHOOLS OF KANSAS From the few comparisons made in this study as well as in some other studies the results seem to indicate that the capacities of children reflected through the pursuance of one school subject are characteristic in a great many instances of the capacities for work in other subjects. In agreement with this Miles says: "These coefiicients would seem to show that if a pupil makes a good mark in one subject he will be quite apt to make good marks in all subjects. Similarly, the pupil who is poor in one subject will tend to be poor in all."^ After this brief statement of some of the incidental conclusions, it is in place to bring together in a summary way the general results found in the separate sections of the previous discussions with reference to the relative standing of pupils from year to year and from institution to institution, together with the consequent amount of retention. Percentages of retention have been determined throughout this thesis by means of the tertile method. These percentages have been stated in connection with the charts and in the tables used in the differ- ent sections. It is only necessary here to summarize by stating about what the average retention for all the schools compared is. It may be noted that in some cases a few schools show either a higher or lower retention than the following statement of the average. But when the schools are regarded as a whole the average retention within the grammar school is somewhere between 50 and 60 per cent for the upper and lower tertiles, respectively, and between 35 and 45 per cent for the middle tertile; within the high school it is between 55 and 60 per cent for the upper and lower thirds, respectively, and between 40 and 50 per cent for the middle third; from the grammar school to high school it is between 50 and 60 per cent for the high and low groups, respectively, while for the middle it is between 35 and 45 per cent; from high school to college it is between 55 and 65 per cent for the upper and lower groups, and for the middle between 40 and 50 per cent. If an average of only the upper and lower tertile retentions had been used without consideration of the middle third, the retention would, of course, be considerably higher, and this would be really a more representative statement; for two- thirds of the pupils are included in these upper and lower tertiles, and since much of the interchange in the middle third is of little importance, for reasons previously seen in the body of the thesis, the results as shown by the tertile retention which has been used are conservatively stated and easily warrant the conclu- sions which follow. ^ University of Iowa Studies, p. 10. GENERAL CONCLUSIONS 1 29 As previously noted, a modified median method has also been employed in ascertaining the retentions and correlations, and the general results of the comparisons are as follows: within the grammar school there is a retention of at least 75 per cent; within the high school, about 80 per cent; from grammar school to high school, between jo and 80 per cent, and from high school to college, between yj and 80 per cent. These results are in general agreement with previous studies in so far as the former studies have made these comparisons. Miles has pointed out that the Pearson coefficient of correlation between the average elementary-school grade and the average high-school grade is + . 71, and that the correlation between specific subjects is a little higher than this.^ Dearborn's results for the high school-university comparisons were as follows: Considering what percentage of those who were in the highest and lowest quarter of the group in high school remain in the upper and lower halves respectively of the class in the university, a little over 80 per cent of those who were in the lowest or highest quarter of the group in the high school are found in their respective halves of the group throughout the university. With the results of these two methods in mind, we are safe in concluding that three- fourths of the students who enter the university from the high schools will maintain throughout the university approximately the same rank which they held in high school. ^ Expressed in terms of the two methods used, namely, the tertile method, and the average of the percentages of those pupils in the upper and lower groups who remain above or below the median, the results justify the conclusion that the majority of the pupils who are classified within a certain original group on the basis of marks retain this same grouping, whether we consider their progress within the grammar school or within the high school, respec- tively, or whether we consider their progress from grammar school to high school or from high school to college. Perhaps the most striking illus- tration of this conclusion is to be found in the groups of pupils that have been followed from the grammar school through the high school and into college, represented by such charts as 116, 117, 118, including III pupils; by such charts as 119, 120, 121, including 158 pupils, and by such charts as 123, 124, 125, including 50 pupils. In the introductory chapter it was stated that the object first would be to see what the actual existing relation between institutions is and ' Studies in Education at the University of Iowa, Vol. I, No. i, pp. 8, 10. ^ Bulletin 312, High School Series No. 6, University of Wisconsin, p. 41. 130 STAND AEDIZATION OF SCHOOLS OF KANSAS then, on the basis of such results, make a statement as to what we have a right to expect in the way of retention between institutions. While the majority of the comparisons have been made between single subjects, yet it has been seen that pupils do about equally well or medium or poor work in all subjects respectively. Consequently the results obtained from the comparisons between single subjects do furnish a safe basis for measuring the efficiency of institutions. As already indicated, the comparisons made within the grammar school and high school were made as a sort of check experiments. In the light of the author's results found within the grammar school and high school, respectively, and in the light of the results of the other studies made of the relation between grammar school and high school and between high school and college, together with the author's results with reference to the grammar school, high school, and college, it may safely be assumed that we have a right to expect a retention between the grammar school and high school and between the high school and college of at least 75 per cent, or of three-fourths of the pupils. It is conceivable that this percentage of retention may be justifiable in some schools and not so in others. For it is admitted that it v/ill sometimes be necessary to take account of the exceptional and varying social factors that come in and affect the efiiciency of institutions. There may be instances where the correlation between high school and college is markedly higher than 75 per cent. When this is true it may be appropriate to inquire as to how far this is the result of the dominating influence of the college. Any standard of measurement which we attempt to set up ought to assume that institutions will be willing and free to modify practices whenever such modification is conducive to the best progress of the pupils concerned. If such a standard as this can be accepted as one means of measuring the efficiency of institutions — until we find a different standard superior to the one here suggested there will be some advantage in having a tentative standard of measurement — then those institutions which show a retention of at least three-fourths of their pupils may be pro- nounced as working efficiently, so far as scholarship is concerned. When the relation or retention between primary, secondary, and higher insti- tutions is markedly lower than 75 per cent, it may be rightly questioned whether such institutions are working in an efficient manner. m 13 1912 One copy del. to Cat. Div. MN ic LIBRARY OF CONGRESS - -.- .Ill 021 289 175 7