THE UNIVERSITY 
 
 OF ILLINOIS 
 
 LIBRARY 
 
 370 
 116 . 
 
 No. 26-34 
 
University of lUinoisUbra^ 
 
 UA1 
 
 Mi '• 
 
 OCT 3 1 M 4 
 JAM 2 3 W5 
 
 OCT 2 5 1985 
 OCT 2 3 198S 
 
 FES 4 f I 
 
 MR 
 
 
 L161 
 
 _-H4l 
 
Digitized by the Internet Archive 
 
 in 2011 with funding from 
 
 University of Illinois Urbana-Champaign 
 
 http://www.archive.org/details/arecollegestuden34odel 
 
9. Terms (continued) 
 
 (668) What arc Traditions? Theories: Laws? 
 
 (669) Contrast: Tribunal and Popular Sentiment: Cooperation and 
 Approbation. 
 
 (670) What is meant by the "Age of Science?" 
 
 (671) Explain "Balance of Power.'" 
 
 (672) Explain: Internationalism. Imperialism, Socialism. Nullification. 
 
 (673) What is meant by "The Totem Pole:"' 
 
 fKE 
 
 
 I '-8 1 
 
BULLETIN NO. 34 
 
 BUREAU OF EDUCATIONAL RESEARCH 
 COLLEGE OF EDUCATION 
 
 ARE COLLEGE STUDENTS A 
 SELECT GROUP? 
 
 Bv 
 
 Charles W. Odell 
 
 Assistant Director, Bureau of Educational Research 
 
 PUBLISHED BY THE UNIVERSITY OF ILLINOIS, URBANA 
 
 1927 
 
TABLE OF CONTENTS 
 
 PAGE 
 
 Preface 5 
 
 Chapter I. Introduction and Statement of the Problem 7 
 
 Chapter II. The Evidence as to the Selection Among Col- 
 lege Students in General 13 
 
 Chapter III. Differences Between the Groups of Freshmen 
 
 Attending Various Institutions of FIigher Learning. . 30 
 
 Chapter IV. Differences Between the Groups of Freshmen 
 
 Carrying Various Subjects 37 
 
 Chapter V. Summary and Conclusions 44 
 
PREFACE 
 
 Studies of the intelligence of high-school seniors and their intentions 
 of attending college, notably the Indiana study by Book, have created 
 the impression that practically no selection takes place between the high 
 school and the college. This hypothesis has been strengthened by ob- 
 servation of college freshmen and by reports showing that large num- 
 bers of them fail to do satisfactory work. In interpreting Book's 
 findings and the result of similar studies, most persons appear to have 
 failed to realize that expressing the "intention of attending college" was 
 not the same as "attending college," and that the group which enters 
 college may differ in significant respects from the group which indicated 
 the intention of attending college. 
 
 In this Bulletin Dr. Odell presents information relative to a group 
 of high-school graduates who actually entered college. In addition to 
 the intrinsic value of the facts reported, the study demonstrates the 
 need for more critical evaluation of data in conducting educational re- 
 search. In many investigations, the researcher works with substitute 
 data and frequently he changes the label attached to them without 
 keeping in mind that he has done so. For example, Book and many of 
 those who have used his data substituted "college entrance" for "expres- 
 sion of an intention of attending college" and failed to keep this fact in 
 mind. 
 
 Dr. Odell's study is significant in another respect. He calls atten- 
 tion to certain faults of his data and then shows that the nature of 
 these faults is such that they do not seriously weaken his conclusions. 
 The researcher in the field of education seldom works with perfect data; 
 frequently they involve errors, both constant and variable; and in many 
 cases they are lacking in validity. The researcher must know his 
 data if he is to be scientific; he must ascertain their faults. But he 
 should not stop here; he should inquire concerning the probable effect 
 of these faults when the data are applied to his problem. Sometimes, 
 as in the case of the present study, it may be shown that the existing 
 faults do not seriously limit the conclusions. 
 
 Walter S. Monroe, Director. 
 March 17, 1927. 
 
 [51 
 
ARE COLLEGE STUDENTS A SELECT GROUP? 
 
 CHAPTER I 
 INTRODUCTION AND STATEMENT OF THE PROBLEM 
 
 The recent increase in college enrollment. One of the most out- 
 standing and significant recent educational tendencies in this country 
 is the notable increase in school enrollment. It is true that this tendency 
 has existed and grown steadily during the whole history of our coun- 
 try, but it has become much more marked during the past generation, 
 especially since the World War. so that it has been a major factor in 
 creating what many consider our outstanding educational problem of 
 today. Although this increase has occurred in all the major divisions 
 of our educational system, it has been unusually rapid in high school 
 and college. 1 The figures in the most recent statistical summary of 
 education issued by the United States Bureau of Education- reveal this' 
 fact very clearly. At present there are in round numbers four million 
 students in our secondary schools and one million in our institutions of 
 higher learning. In both cases, the five years beginning in 1920 show 
 a greater increase, both absolute and relative, than any previous period 
 of similar length. The number of students in secondary schools in pro- 
 portion to our population is now about six times as great as in 1890 and 
 the number of those in colleges more than three times as great. Ex- 
 pressing the same facts otherwise, in 1890 approximate!}' one-fourth of 
 1 per cent of the whole population or 1 per cent of the total school en- 
 rollment consisted of college students, whereas in 1924 more than three- 
 fourths of 1 per cent of the whole population or about 3V-< per cent of 
 the total school enrollment was in college. 
 
 A problem resulting from the increase in school enrollment. Such 
 a rapid and largely unanticipated increase in enrollment cannot occur 
 without creating a number of problems. Undoubtedly the one of these 
 problems which, because of its very practical and pressing nature, has 
 received the most attention, is that of providing the necessary financial 
 support. In the first place, this great increase in enrollment, occurring 
 
 'Throughout this discussion the term "college'" will be used for the sake of brevity 
 to refer to colleges, normal schools, universities and all other institutions of higher 
 learning. 
 
 ^Phillips, Frank M. '"Statistical summary of education 192.1-1924." U. S. Bu- 
 reau of Education Bulletin. 1926. No. 19. Washington. 1926. 7 p. 
 
 [7] 
 
at a time when the purchasing power of the dollar has suffered a very 
 marked decline, necessitates greatly augmented expenditures merely to 
 preserve the previous scope and efficiency of our schools. In addition 
 to this, however, recent decades have witnessed a growing demand on 
 the part of those charged with the administration of our schools, and 
 also to a large extent upon the part of the general public, for broaden- 
 ing and enriching the offerings of our schools, securing teachers with 
 better training, providing better buildings and equipment, and other 
 advances, all of which require greater expenditures. Such demands 
 have served to render the problem still more urgent and it has been 
 practically impossible to secure from state legislatures, local taxing 
 bodies, and the general public the amounts of money which educators 
 have regarded as necessary. This reluctance or refusal on the part of 
 those who determine school revenues has been manifested in the case of 
 state universities, normal schools, and other publicly supported institu- 
 tions of higher learning as well as in that of elementary and high schools. 
 Institutions not supported by public taxation have likewise suffered be- 
 cause, although they have been the recipients of more or less steadily 
 increasing benefactions from private sources, their incomes have in 
 many instances not grown in proportion to the increases of sizes of stu- 
 dent bodies and in cost. 
 
 A suggested solution of the problem. A number of solutions for 
 this problem have been suggested, but the one which has probably re- 
 ceived the most attention is the proposal that our colleges exercise a 
 much greater degree of selection among those who desire to enter and 
 admit only those high-school graduates who can offer rather definite 
 evidence of their ability to derive and give back to society considerable 
 profit from a college education. In connection with this proposal it is 
 frequently assumed that the intellectual quality of the large number of 
 students now entering college is, on the average, much lower than was 
 that of the comparatively select few who entered a generation or more 
 ago. In some cases, this has not merely been assumed, but has been 
 expressed as a definite opinion or even stated as a fact. Many college 
 presidents and others 3 in such positions as to give their opinions weight 
 
 3 One of the strongest statements in support of this point of view has been made by 
 Thorndike who is probably as well qualified as anyone to render an opinion. Although 
 he refers to the secondary school rather than the college it would be decidedly surprising 
 if his statement did not also apply to the latter. Referring to the fraction of those of 
 the proper age now attending high school, he states that "we lack measures of the in- 
 born capacities of the one in ten or eleven of a generation ago and have only very 
 scanty measures of the capacities of the one in three of today. We have, however, 
 excellent reasons for believing that the one in ten had greater capacities for algebra and 
 
 [3] 
 
have supported this view and have advocated more selective entrance 
 requirements or, in some cases, more rigid selection during the first year 
 or two in college. On the other hand, many educators who likewise 
 are in close touch with the situation and should be able to form valid 
 opinions deny that there is any appreciable lowering of the average in- 
 tellectual quality of those who seek admission to college. They point 
 to the generally granted fact that a selective process is going on 
 throughout our elementary and secondary schools, more especially in 
 the latter, and maintain that this process is all or practically all that is 
 needed to control the situation. In other words, they assert that al- 
 most all those who are being allowed to graduate from our secondary 
 schools possess sufficient ability to profit by a college education and 
 therefore that our institutions of higher learning, more particularly those 
 supported by the state, should open their doors with no or almost no 
 reservations to all high-school graduates who wish to enter. The pro- 
 ponents of neither side in this controversy have offered enough objective 
 and relevant data to support their position. 
 
 A second important problem. In addition to the problem of secur- 
 ing the funds necessary to provide sufficient facilities for the rapidly 
 increasing number of students in institutions of higher learning, a sec- 
 ond problem is that of what facilities should be provided for the less 
 capable students. Those who believe that the intellectual level of college 
 students is below what it was in the past consider this a more import- 
 ant problem than do those on the other side of the controversy. The 
 latter, however, do not and can not deny that large numbers of students 
 who enter college are not successful in carrying the work offered. 
 Whether this failure is due to lack of mental ability, to poor habits of 
 study, to higher standards of work or to some other cause the question 
 of what to do with such students is generally regarded as of vital im- 
 portance. Many of those who contend that the intellectual level of col- 
 lege students is lower today than formerly advocate some plan of pro- 
 viding a differentiated type of education for students of inferior ability. 
 A few even go so far as to urge that such individuals should not attempt 
 to continue their education beyond the secondary level, but most edu- 
 cators believe that they should be given the opportunity to continue 
 their education or perhaps even required to do so. Industrial or other 
 vocational work with related subjects, in either full-time or part-time 
 schools, is most often suggested for this purpose. On the other hand, 
 
 for intellectual tasks generally than the one in three of today." (Thorndike, E. L. 
 "Changes in the quality of pupils entering the high school," School Review, 30:355-59, 
 May, 1922.) 
 
 [9] 
 
most of those who believe that the college students of today are the in- 
 tellectual equals of those of the past maintain that the colleges them- 
 selves are primarily to blame for the large per cents of failures and that 
 the remedy lies in better selection and organization of subject matter, 
 improved instruction, training in study habits, and, in general, in do- 
 ing those things which both induce and enable students to do approxi- 
 mately the maximum work of which they are capable. 
 
 This problem has been referred to in describing the present situ- 
 ation but it is not the intention of the writer to offer even a partial so- 
 lution. However, as will be shown in the following paragraphs, the 
 data to be presented do have a bearing upon the question of the intel- 
 lectual ability of college entrants and in so far as they offer an answer 
 to this they enter into the solution of the second problem. This state- 
 ment does not imply, however, that determination of the intellectual 
 ability of those who seek admission to college even though it be an ac- 
 curate one is a satisfactory basis for deciding the type of training which 
 each should receive. Not mere intellectual ability but other abilities, 
 aptitudes, and interests as well as the needs of society must be consid- 
 ered. 
 
 Possible methods and difficulties of determining differences in the 
 ability of college students of the past and present. Although it would 
 not completely 'solve the problems stated above it would be a definite 
 step in this direction if those on the one side or the other of the conten- 
 tion as to the intellectual quality of college students could prove their 
 point. To do this, however, is very difficult if not absolutely impossible. 
 It is true that mental tests have exhibited almost phenomenal develop- 
 ment within the past few years until the best of them are fairly satis- 
 factory instruments for measuring the ability or trait commonly called 
 ''general intelligence." especially when group averages and not individ- 
 ual scores are employed. We have, however, no comparable data from 
 the past with which to compare the results obtained from the present 
 use of such tests. 
 
 Another possible line of attack is through the measurement of the 
 intelligence of college students and the comparison of the results with 
 those obtained from high-school seniors. However, few if any studies 
 of this sort have yielded valid evidence. They have usually revealed 
 the anticipated fact that the intelligence of the college group is on the 
 whole above that of high-school seniors but they have not presented 
 evidence to show how much of this increase resulted from the added 
 vear or more of maturitv and how much from the factor of selection. 
 
 [10] 
 
Another possible method of answering the question would be to 
 secure the answers of students of a generation or more ago to college 
 entrance examinations and compare them with answers to the same 
 questions obtained from a present-day group of entering students. Even 
 if such a comparison were made and the measures assumed to be reliable 
 and valid, the results would not furnish conclusive evidence for the 
 reason that there are at least two chief causes which might explain any 
 difference. These are, of course, the two familiar factors of heredity 
 and environment. It may be that the supposedly superior training now 
 given in our elementary and secondary schools, perhaps supplemented 
 by more favorable environmental conditions outside of school, operates 
 so that the college entrants of today would make higher marks upon 
 examinations or. for that matter, higher scores upon intelligence tests, 
 than did those of the past, even though the latter possessed as high or 
 higher native ability. 
 
 A still further complication is caused by the fact that the work 
 offered in most of our colleges covers a much broader scope and oth- 
 erwise differs a great deal from that given twenty-five or fifty years 
 ago. Therefore we do not know whether the same, a higher, or a lower 
 degree of intellectual ability is possessed by students able to do suc- 
 cessful work in college at present as compared with those who did so in 
 the past. 
 
 The problem of this bulletin. It is the purpose of this bulletin to 
 present some data which bear upon the point at issue, though they in 
 no sense form the basis for a complete answer to the question. They 
 are the result of the writer's continuing a stud} 4 begun in 1923 and sim- 
 ilar in the general outline of its first part to several investigations 3 previ- 
 ously carried on by others. In these studies. Book and Williams in In- 
 diana. Trabue and Mann in Xorth Carolina. Colvin and McPhail in 
 
 ''Odell, C. W. "Conservation of intelligence in Illinois high schools." University 
 of Illinois Bulletin, Vol. 22, Xo. 25. Bureau of Educational Research Bulletin Xo. 22. 
 Urbana: University of Illinois. 1925. 55 p. 
 
 5 Book, W. F. The Intelligence of High School Seniors. New York: The Macmil- 
 lan Company, 1922. 371 p. 
 
 Book, W. F. "A preliminary report on the state-wide mental survey of high 
 school seniors." Indiana University Extension Division Bulletin, Vol. 6, Xo. 1. Bloom- 
 ington, Indiana: Indiana University, 1920, p. 31-67. (Contains a brief summary' of the 
 study. ) 
 
 Manx, G. W. "Selective influence of desire to attend college," The High School 
 Journal, 7:8-9, January, 1924. 
 
 Colvin, S. S., and McPhail, A. H. "Intelligence of seniors in the high schools of 
 Massachusetts." U. S. Bureau of Education Bulletin. 1924. Xo. 9. Washington, 1924. 
 39 p. 
 
 [11] 
 
Massachusetts, and the writer in Illinois secured intelligence test scores 
 and other data from rather large numbers of high-school seniors and 
 compared such of these data as were pertinent with the expressed in- 
 tentions of the seniors regarding the continuation of their education. In 
 so far as the writer knows, however, the studies in Indiana, North Car- 
 olina, and Massachusetts have not been continued to the point of follow- 
 ing up the seniors after they left high school to ascertain which ones en- 
 tered college and how well they were able to do their college work. In 
 the Illinois study, however, this has been done and it is the chief pur- 
 pose of this bulletin to present what has been found regarding the re- 
 lationship of ability as shown by intelligence test results and high-school 
 marks to actual college entrance. Some slight reference will also be 
 made to the group of students who remained in college through the 
 freshman year. As a minor part of the discussion the findings in all 
 four studies as to the relationship between intelligence and college in- 
 tentions will be briefly summarized. Furthermore, the scores and marks 
 of those who attended college will be grouped to show what differences 
 existed between the students attracted by ten different large institutions 
 or groups of institutions and also those between the students carrying 
 different freshman subjects. 
 
 [12] 
 
CHAPTER II 
 
 THE EVIDENCE AS TO SELECTION AMONG 
 COLLEGE STUDENTS IN GENERAL 
 
 The evidence offered by the studies in Indiana, North Carolina, 
 and Massachusetts. Since the studies made in these three states were 
 not continued to the point of following up the high-school seniors in- 
 cluded into college, only brief reference will be made to them to show 
 what differences were found between those seniors who stated that they 
 intended to continue their education and the total groups. In his re- 
 port of the Indiana study Book does not give data from which the exact 
 median or mean scores can be obtained, but it appears that the differ- 
 ence in the averages of the two groups did not exceed one point on the 
 test used. 6 He divided the seniors into three classes, high, average, and 
 low on the basis of intelligence, and found that of all the seniors, 22 per 
 cent possessed high and 26 per cent low intelligence, whereas for those 
 who planned to continue their education the corresponding per cents 
 were 23 and 25. Therefore it may be said that the investigation in In- 
 diana indicated that very little selection had taken place in the group 
 of high-school seniors who planned to attend college. 
 
 The North Carolina study showed a somewhat higher degree of 
 selection. Those who were planning to attend college had a median 
 point score 7 of 112, whereas the whole group of seniors had one of 109. 
 Dividing the seniors into high, average, and low groups with scores of 
 120 and 100 as the dividing points, 30 per cent of those planning to go 
 to college were in the high and 19 per cent in the low group, whereas 
 of all high-school seniors only 23 per cent were in the high and 25 per 
 cent in the low group. In other words the excess of those in the high 
 group over those in the low was 1 1 per cent for seniors who planned 
 to go to college, whereas for all seniors the low group was 2 per cent 
 larger. 
 
 In the Massachusetts study the seniors were divided into three 
 groups and designated as good, questionable, or bad college risks ac- 
 cording to the records of students at Brown University who had taken 
 the same test 8 in previous years. On the basis just mentioned 26 per 
 
 This was the Indiana University Intelligence Scale, Schedule D. 
 
 'The point scores mentioned were made upon Trabue's Mentimeters. 
 
 8 In this case the Brown University Psychological Examination was employed. 
 
 [13] 
 
cent of all seniors were classed as good and 54 per cent as bad risks, 
 whereas of those planning to go to college 34 per cent were good and 
 46 per cent bad risks. This difference of 8 per cent in each case is 
 great enough to indicate that some definite selection took place. 
 
 The collection and tabulation of data in the Illinois investigation. 
 The Illinois data which the writer will utilize were obtained in a study 
 which was begun with high-school seniors of 1923-24 and was contin- 
 ued to follow a number of them through their freshman year in college. 
 In the autumn of 1923, all of the four-year public high schools in Illi- 
 nois were asked to have their seniors take the Otis Self-Administering 
 Test of Mental Ability, Higher Examination, Form A, and fill out an 
 "Information Blank for High School Seniors," 9 which called for the fol- 
 lowing information: 
 
 Name 
 
 Sex 
 
 Date of Birth 
 
 Age on September 1 , 1923 
 
 Name of school 
 
 Town or city 
 
 Intentions concerning further education 
 
 Intention of continuing 
 
 Institution 
 
 Course 
 
 Major subject 
 
 Vocational choice 
 
 Father's occupation 
 
 Information as to previous intelligence tests taken 
 
 Units of high-school credit 
 
 High-school subjects liked most 
 
 High-school subjects liked least 
 
 Number of failures in high school 
 
 Average school mark in high school 
 
 Three hundred and sixty-eight schools, or more than half of those in 
 the state, cooperated at that time and contributed data for more than 
 twelve thousand seniors. 
 
 The second step in the investigation was taken a year later when 
 most of the seniors concerning whom data had been obtained had been 
 graduated from high school. At this time requests were sent to the par- 
 ticipating schools for the complete scholastic records of these individu- 
 als. A small number had not been graduated and a few high schools 
 
 "The tests were administered by high-school principals or teachers designated by 
 them and the blanks filled out by the seniors themselves. All scoring of test papers and 
 all tabulating of test and other results was done at the Bureau of Educational Re- 
 search under the supervision of the writer. 
 
 [14] 
 
no longer continued their cooperation, but the complete records of over 
 eleven thousand graduates were secured. 
 
 In the fall of 1925 all of the more than three hundred institutions 
 of higher learning named by the seniors were asked to supply the com- 
 plete 1924-25 records of all of their students who had been included in 
 this study. A few of these institutions which rather large numbers of 
 students had indicated their intention of attending and a large number 
 of those named by only one or a few students each did not furnish the 
 information requested, but the majority of them did so. In spite of this 
 generous cooperation and also despite the fact that approximately seven 
 thousand seven hundred seniors had stated definitely that they expected 
 to attend, the records of less than two thousand who actually entered 
 college were secured. 10 
 
 After the mental test papers had been scored and the point scores 
 computed these were turned into intelligence quotients according to the 
 directions given by Otis. Since the general reader is undoubtedly more 
 familiar with the use and meaning of intelligence quotients than with 
 point scores upon the Otis or any other particular test the former will 
 be the chief basis of making the comparisons of intelligence presented 
 in this report. A second means of comparing different groups of stu- 
 dents will be the average marks received by them in high school. Since 
 the marks secured from the several hundred high schools which partici- 
 pated were given according to a large number of different marking sys- 
 tems they were all reduced to a comparable basis. This was done by 
 securing from each school as complete an explanation of its system as 
 possible and, using this information as a guide, transmuting all marks 
 to a percentile system with the passing mark at 70 and no conditions. 11 
 
 The reliability of the data to be presented. It must be admitted 
 that in both the intelligence test results and the school marks large 
 variable errors are present. Xo group intelligence test yields highly re- 
 liable individual scores and the one used in this case is probably less re- 
 liable than a test which requires two or three hours to give instead of 
 the half hour which the Otis test consumes. Furthermore, the fact that 
 
 10 In addition to the refusal of a number of colleges to cooperate the two other 
 most important factors causing a decrease in the number of records secured were that 
 the reports from some institutions were not complete and that many of the high-school 
 graduates found it necessary to be out of school a year or more before entering the col- 
 leges of their choice. 
 
 "This system was adopted because after the first forty or fifty high schools had 
 sent in their data it was found to be used by more of them than any other one. Later 
 when all schools had reported it was found that a percentile system with a passing mark 
 of 75 instead of 70 was in more common use. 
 
 [15 J 
 
the tests were not administered by trained examiners, but by several 
 hundred different persons, many of whom had probably never given a 
 standardized test before, undoubtedly increased the errors present in 
 the scores. On the other hand, because of the fact that the directions to 
 be given by examiners are reduced to a minimum, scores on the Otis 
 Self-Administerin? Test are probably less liable to errors due to sub- 
 jective qualities of the persons giving the tests than are these yielded by 
 most group tests. 
 
 At this point, attention should be called to a constant error intro- 
 duced into the intelligence quotients employed in this study by the 
 method which Otis provides for determining I.Q.'s. This method is such 
 that no person fourteen years of age can be credited with a higher I.Q. 
 than 143, no person of fifteen with one higher than 139, of sixteen 
 higher than 136, of seventeen higher than 134, and of eighteen higher 
 than 133. On the other hand, it is possible for those of any age to re- 
 ceive I.Q.'s as low as 58. Inasmuch as very few r of those included in 
 the study were under fifteen years of age when tested and the majority 
 seventeen or older it can be seen that his method sets an absolute and 
 arbitrary upper limit to intelligence quotients not so far above normal 
 or 100 as the lower limit is below that point. Furthermore, Otis' method 
 not merely sets this definite upper limit, but also more or less scales 
 down I.Q.'s above average, this effect being greater the nearer they ap- 
 proach the upper limit. We know from other evidence that the highest 
 I.Q.'s found in a large group of high-school seniors will run well above 
 130 or 140. Terman 12 states that the highest 1 per cent of children 
 have I.Q.'s above 130 and most other investigators agree approximately 
 w r ith this statement. For high-school seniors who have undergone some 
 selection as indicated by a mean I.Q. of 103 instead of 100 the propor- 
 tion above 130 is undoubtedly somewhat greater than 1 per cent. There- 
 fore, in considering the intelligence quotients to be presented later in 
 this discussion the reader should bear in mind that those above average 
 or at least certainly those above 110 should be somewhat higher than 
 they are. This of course means that the averages computed from the 
 distributions should also be slightly higher than those actually obtained. 
 Attention will be called to this fact later in interpreting the facts pre- 
 sented. 
 
 In addition to the fact just mentioned, another point should also be 
 taken into consideration in interpreting the intelligence quotients ob- 
 tained in this and also in other similar studies. It was stated in the 
 
 "Termax, L. M. The Intelligence of School Children. Boston: Houghton Mif- 
 flin Company, 1919, p. 8. 
 
 [15 J 
 
preceding paragraph that it is possible for I.Q.'s on the test used to be 
 as low as 58 and, as will be seen from the data presented later, a few 
 below 65 were actually secured and many others in the upper sixties, 
 the seventies and eighties. Both the evidence which we have on the 
 matter and a common sense interpretation of the meaning of the I.Q. 
 indicate that it is unlikely that individuals with I.Q.'s below 90 can 
 successfully complete a high-school course and still more improbable 
 that they can do satisfactory work in college. The explanation of this 
 discrepancy between the intelligence quotients found in this and other 
 studies and the evident facts in the case undoubtedly results from nega- 
 tive errors, some of which are decidedly large, in the test scores. Al- 
 most half of the seniors tested stated that they had never before taken 
 an intelligence test and undoubtedly a number of these wasted a consid- 
 erable amount of time in getting started because of failure to under- 
 stand instructions or to realize that it was important to start promptly. 
 Others probably began at once but did not understand exactly what 
 they were to do or spent an undue amount of time trying to make sure 
 that each item was correctly answered. Still others, whether they had 
 been tested before or not, probably did not put forth their best efforts 
 either because they were below par physically or mentally or because 
 they were not particularly interested in the test and cared little whether 
 they made high scores or not. Because of these and other reasons few 
 if any of the low I.Q's obtained should be regarded as reliable measures 
 of the intelligence of the seniors tested. 
 
 The subjectivity and unreliability of ordinary school marks has 
 been too often shown and discussed to need extended comment here. In 
 this study the errors of this sort present were undoubtedly increased 
 somewhat by the fact that marks from several hundred schools with 
 different standards and ideals were thrown together and also by the fact 
 that it was necessary to convert them to a common basis. Although 
 this was done as carefully and by as sound statistical procedure as pos- 
 sible the information at hand was not in all cases sufficient to render the 
 transmutation highly accurate. 
 
 The factor of unreliability or inaccuracy in the test results and 
 school marks should not, however, be overemphasized. When scores 
 for comparatively large numbers of individuals are dealt with, the prob- 
 ability is very strong that, even though the variable errors are both nu- 
 merous and large, they will balance one another so that the averages 
 secured will not be seriously in error. It should also be remembered 
 that although there may be and undoubtedly are constant errors pres- 
 
 [171 
 
ent in particular groups of scores, 13 yet when these groups are thrown 
 together into one large group they become variable errors with respect 
 to the total group or distribution and therefore are unlikely to affect its 
 average any more seriously than the variable errors present in individ- 
 ual scores or marks. In view of these facts it is extremely unlikely 
 that averages will be in error by as much as one point. For the whole 
 group of those entering college there is only about one chance in sixty 
 thousand that the mean I.Q. is one point in error and one in almost in- 
 finity in the case of the mean high-school average. 14 
 
 The four groups compared in this study. The data to be presented 
 have been tabulated to differentiate the four following groups: (1) all 
 high-school graduates; (2) graduates who stated that they were defi- 
 nitely planning to attend college; (3) those known to have entered col- 
 lege and for whom college records were secured; 15 and (4) those 
 still in college at the end of the freshman year and receiving marks on 
 the last semester's or term's work of that year. This last group is in- 
 cluded because it may be a matter of interest to some readers to see 
 how it compares with the others, though such a comparison is not an 
 integral part of the purpose of this bulletin. It will be seen that each 
 group is more inclusive than the next one, including all of the following 
 group and also many others not in it. 
 
 The intelligence quotients of the different groups in respect to 
 college careers. Table I presents the distributions of intelligence quo- 
 tients of the four groups both by actual numbers and by per cents. 
 Reading down the first column, for example, one sees that 32 of the 
 whole number of high-school graduates had intelligence quotients from 
 131 to 135, inclusive, that 186 had I.Q.'s from 126 to 130, and so on, 
 the total number of graduates being 11,321. The first column in the 
 
 "For example, the person giving the test may have allowed too much or too little 
 time or given too much help and thus made all the scores of the group too high or too 
 low. Similarly in the case of school marks a particular school may tend to mark much 
 higher or much lower than most schools and thus introduce a constant error into the 
 records of all its graduates. 
 
 "The standard error of the mean I.Q. is about .23 of a point and that of the 
 mean high-school average about .12. Therefore a distance of approximately 4.3 sigma 
 in the first case and 8.3 sigma in the second is required to include the chance of an 
 error of one point. These distances correspond to the chances stated above. 
 
 "Except in the case of business colleges, very few of which supplied any records, 
 there was no marked difference in the degree of cooperation received from institutions of 
 various types. Therefore the writer believes that the group of individuals whose college 
 records were secured may be considered as fairly representative of the total group at- 
 tending college in 1924-25, a group which he estimates was from SO to 100 per cent 
 larger or, in other words, included from three to four thousand persons. 
 
 [18] 
 

 
 OOu- . 
 
 
 
 
 
 
 
 
 ose 
 inin 
 id o 
 yeai 
 
 *ONOVOONNO>OOhij<MOh 
 
 o 
 
 
 
 
 
 
 
 
 heSs 
 
 n ^o — r-- c> r~- c*j -o n — i 
 
 o 
 
 
 
 
 
 
 — — — 1 
 
 o 
 
 
 
 
 
 
 Z - ° 
 
 
 
 
 
 
 
 c 
 
 
 
 
 
 
 
 
 ■s 
 
 
 
 
 
 
 
 
 o u -a w 
 
 
 
 
 
 
 
 
 c > CU 00 
 
 ^ K «■ u 
 
 vO\SOW ( OvC'tON«n'tNO'-< 
 
 o 
 
 
 
 
 
 
 -c £!== 
 
 Mvoo\ooM^i-t^n-' 
 
 o 
 
 
 
 
 
 
 u ^ i- o 
 en O c X 
 
 
 o 
 
 
 
 
 en 
 
 O « « ° 
 
 
 
 
 
 
 1 a 
 
 _c 
 
 
 
 
 
 
 
 o 
 
 H 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 CO 
 Oh 
 
 duates 
 nning 
 ttend 
 liege 
 
 TfC\rNu-)coTfTf— i r^ n t-^ — w — r »— < 
 
 o 
 
 
 
 
 
 
 "•ti^cir^h-vc-vOri« 
 
 d 
 
 
 
 
 
 
 a a a o 
 
 
 O 
 
 
 
 
 
 
 l«-S <-> 
 
 
 
 
 
 
 
 
 O a 2 
 
 
 
 
 
 
 
 
 O en 
 
 
 
 
 
 
 
 
 O eo 
 
 
 
 
 
 
 
 
 -C ■" 
 
 
 
 
 
 
 
 
 — u rt 
 
 fO\o^on>ox>oooix — 'fuitN'-' 
 
 o 
 
 
 
 
 
 
 
 •^nr- - 1 ui c» t^ n \o r ; - i 
 
 o 
 
 
 
 
 
 
 -C td 
 
 
 o 
 
 
 
 
 
 
 60 u 
 
 
 
 
 
 
 
 
 £^ 
 
 
 
 
 
 
 
 
 00 t- 
 
 
 
 
 
 
 
 
 en i- W 
 
 o>^ i oNM"in\co''Ooovonc- 
 
 ■* 
 
 t~~ 
 
 oo 
 
 ■* 
 
 
 
 0.-T3 >> 
 
 H^00\»f j 0i'- | , 0'-i 
 
 lO 
 
 o 
 
 CO 
 
 
 
 
 JC « C u 
 
 H E « c 
 
 — — « CN CO CN CN — 
 
 SO 
 
 
 
 
 
 
 
 rt 
 
 
 
 
 
 
 U u O 
 
 
 
 
 
 
 
 
 1- 03 
 
 
 
 
 
 
 
 £ 
 
 
 
 
 
 
 
 
 i 
 
 
 
 
 
 
 
 
 o u-n jj 
 
 
 
 
 
 
 
 
 5 - ti M 
 
 i-^\aMfOF<xtN?.-vor~>nO'-i 
 
 , — 1 
 
 r^ 
 
 r~- 
 
 ■* 
 
 
 
 ■* 2 S u 
 
 rt«i-.-Nxnt^n\0(S 
 
 ■* 
 
 O 
 
 co 
 
 
 
 
 - 1 — — ; 
 
 -h CN CO r~0 CO CN -— 
 
 ON 
 
 
 
 
 
 
 10 ^ i- O 
 en O c y 
 O »-> co u 
 
 
 — 
 
 
 
 
 
 
 
 
 
 
 
 
 CD 
 
 -e 
 
 
 
 
 
 
 
 -a 
 £ 
 
 r - 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 3 
 
 
 
 
 
 
 
 
 £ 
 
 eu MT) 
 
 
 
 
 
 
 
 ■w c c u 
 
 
 
 
 
 
 
 
 a .— eu 60 
 
 oooooccNco-*-r-^r^oocoow-,so — ' ^t 1 
 
 CO 
 
 •* 
 
 r» 
 
 ON 
 
 
 
 3 C *-> co 
 
 oi-t — t^xrtnf, O'XOxtNrt 
 
 r~~ 
 
 o 
 
 cs 
 
 
 
 
 -a c t; = 
 
 i-h n w o>. t-i « ri x * cn 
 
 vO 
 
 
 
 
 
 
 « c« rt O 
 
 
 r^ 
 
 
 
 
 
 
 rh"E° ° 
 
 
 
 
 
 
 
 
 *0 en 
 
 
 
 
 
 
 
 
 O eu 
 
 
 
 
 
 
 
 
 -c *i 
 
 
 
 
 
 
 
 
 r i OS 
 
 NvOr- — r ; vCO — 'NM^ro-H'tr- 
 
 
 CO 
 
 **« 
 
 o 
 
 
 
 3^3 
 
 nx — M — x — nxvOTnuin 
 
 CN 
 
 o 
 
 CN 
 
 
 
 
 < . T3 
 
 p-Tj-xror^ — 3'Mr^n" 
 
 CO 
 
 
 
 
 
 
 J= nl 
 
 
 
 
 
 
 
 
 00 l- 
 
 
 
 
 
 
 
 
 I 
 
 
 
 
 
 
 
 
 cx 
 
 
 
 
 O 
 
 eu 
 > 
 
 o 
 
 d 
 
 ON 
 
 _o 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ,_; 
 
 
 C3 
 
 
 
 CD 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 O 
 
 e 
 
 a! 
 
 c 
 
 CD 
 
 O 
 
 C 
 
 CD 
 O 
 
 
 
 
 1 1 1 l 1 l 1 l 1 1 l l l l l 
 
 
 
 
 — . so — . so — • sO — ' vO-h\0 - .sO— ■ \0 — 
 
 
 U 
 
 u 
 
 L> 
 
 
 
 
 r", n r) — — 'COO-oxxr^r-^OvO 
 
 
 s 
 
 eu 
 
 CD 
 
 Oh 
 
 [19] 
 
second or per cent half of the table shows that the 32 graduates who 
 had I.Q.'s from 131 to 135 constituted .3 per cent of the whole group. 
 with I.Q.'s from 126 to 130 were 1.6 per cent, and so on. The 
 second column in each part of the table contains the I.Q. distribution 
 of those who planned to attend college, the third that of those known to 
 have entered college, and the last column the distribution of those who 
 remained in college until the end of the freshman year. It can easily 
 be seen, especially by looking at the per cents, that there is a marked 
 tendency for each group to have more students with high intelligence 
 quotients and fewer with low quotients than do the more inclusive 
 groups which are shown in the columns to the left. 
 
 The condition just described is perhaps shown more clearly by 
 Figures 1 and 2 which represent the same data graphically. The first 
 of these shows the actual numbers of cases, the second 16 the per cents. 
 It will be seen that the broken line representing those who planned to 
 attend college is further to the rieht than the solid line representing all 
 high-school graduates and that the dotted line representing those known 
 to have entered college is markedly further to the right than is the 
 broken line. Likewise, in Figure 1. the line of mixed dots and dashes 
 representing those in college at the end of one year is slightly further 
 to the right than the dotted line. Considering the means, which are re- 
 spectively 105. 104. 107. and 107. 17 we may summarize the differences 
 by saying that those who planned to continue their education averaged 
 one point I.Q. above all high-school graduates, that those who were 
 known to have entered college averaged three points more than all 
 those who planned to do so. and that there was very little difference be- 
 tween those who remained in college for a year and the whole group of 
 entrants. Were it not for the effect of Otis' method of computing I.Q.'s 
 in reducing those at the higher levels the means of the last two groups 
 would undoubtedly be at least one or two points higher and the differ- 
 ences more marked. 1 ' 
 
 Probably the best means of describing the situation, however, is to 
 state the per cents of individuals in each group above and below certain 
 
 'This figure contains no curve corresponding to the last column of Table I. The 
 
 is that such a curve would come so close to the one representing all students 
 
 known to have entered college that very little difference could be detected between the 
 
 i it would therefore serve to confuse the figure. 
 
 1T The mean I.Q. of those remaining in college at the end of one year is slightly 
 
 'han that of all students known to have entered college, but the difference is so 
 
 small that it is not shown in units' place. 
 
 >ere not for the other fact stated above, that undoubtedly most of the low 
 represent the true degi ."elligence of the individuals for whom they 
 
 were computed, the true means would be increased another point or two. 
 
 [20] 
 
Number 
 
 Cases 
 
 r 2lQQ 
 
 61 7/ II 91 101 III 121 131 
 
 Intelligence Quotient 
 
 All high-school graduates 
 
 Graduates planning to attend college 
 
 Those Known to have entered college 
 
 Those remaining at end of one year 
 
 Figure 1. Numbers of Tuosr i\ Each Group With 
 Various Intelligence Quotients 
 
 I 21 ] 
 
4/ 
 
 // 
 
 81 91 tot m ill 
 
 Intelligence Quotient 
 
 •All high-school graduates 
 
 ■Graduates planning to attend college 
 
 •Those known to have entered college 
 
 Figure 2. Per Cents of Those in Each Group With 
 Various Intelligence Quotients 
 
 points. Since the most common plan of threefold classification of mental 
 ability is to call all those with I.Q.'s above 110 superior, those with 
 I.Q.'s from 90 to 110 average or normal, and those with I.Q.'s below 
 90 inferior, this division has been employed here. Twenty-four per 
 cent of all graduates were found to possess superior intelligence accord- 
 ing to this criterion whereas 27 per cent of all those who planned, to 
 attend college, 37 per cent of those known to have entered college, and 
 38 per cent of those still in college at the end of one year belonged to 
 this group. On the other hand, just as there is a marked increase in 
 these per cents so there is a decrease in the per cents of those possessing 
 inferior intelligence, these being respectively 10, 9, 4, and 4. The excess 
 
 [ 22 ] 
 
TABLE II.— PER CENTS OF INTELLIGENCE QUOTIENT GROUPS OF 
 
 ALL HIGH-SCHOOL GRADUATES FALLING IN VARIOUS 
 
 CLASSES WITH REGARD TO COLLEGE CAREERS 
 
 
 Graduates 
 
 Those 
 
 Those 
 
 Intelligence 
 
 planning 
 
 known to 
 
 remaining 
 
 Quotients 
 
 to attend 
 
 have entered 
 
 at end of 
 
 
 college 
 
 college 
 
 one year 
 
 131-35 
 
 88 
 
 34 
 
 31 
 
 126- 
 
 80 
 
 77 
 70 
 75 
 75 
 63 
 64 
 65 
 63 
 
 27 
 28 
 26 
 25 
 21 
 16 
 14 
 10 
 8 
 
 24 
 
 121- 
 
 26 
 
 116- 
 
 23 
 
 Ill- 
 
 21 
 
 106- 
 
 18 
 
 101- 
 
 14 
 
 96- 
 
 11 
 
 91- 
 
 8 
 
 86- 
 
 7 
 
 81- 
 
 59 
 
 7 
 
 5 
 
 76- 
 
 56 
 
 5 
 
 4 
 
 71- 
 
 51 
 
 6 
 
 6 
 
 66- 
 
 46 
 
 
 
 
 
 61- 
 
 57 
 
 14* 
 
 14* 
 
 
 
 Total 
 
 68 
 
 17 
 
 15 
 
 *These rather large per cents are caused by a single individual as 
 may be seen by referring to Table I. 
 
 of individuals with superior intelligence over those with inferior intelli- 
 gence is thus seen to be 14, 18, 33, and 34 per cent, in the same order 
 previously used. If the true I.Q.'s were available all these per cents 
 would undoubtedly be increased, each slightly more than the one pre- 
 ceding it. The figures as well as the means indicate that the most 
 marked selection occurs in the group of those who actually attend col- 
 lege. Those who plan to do so are only slightly more intelligent than 
 the whole group of high-school graduates and those who remain in col- 
 lege at the end of one year differ still less from the whole group of those 
 entering college. 19 The difference between those who actually entered 
 college and all high-school graduates or even all who planned to go to 
 college is so marked, however, as seemingly to leave no doubt that there 
 was very definite selection. 
 
 Another method of bringing out the same facts is shown in Table 
 II. This gives by intelligence quotient groups the per cents of all grad- 
 uates who planned to attend college, who were known to have entered 
 
 "It is probable that if those who returned to college at the beginning of their sec- 
 ond year were compared with all those who entered a greater difference would be shown. 
 Many of those still in college at the end of the year had such low marks that it seemed 
 evident they would be dropped. On the whole these were students of lower I.Q.'s than 
 the average. 
 
 [23] 
 
college, and who remained at the end of one year. Taking the first row 
 of the table, for example, 88 per cent of all seniors with intelligence quo- 
 tients from 131 to 135, inclusive, planned to attend college, 34 per cent 
 are known to have done so and 3 1 per cent were still in college at the 
 end of one year. In each column there is a marked though not regular 
 tendency for the entries to decrease toward the bottom, that is, for the 
 per cents of individuals of high intelligence to be greater than the per 
 cents of those of low intelligence. Furthermore, the decrease is rela- 
 tively greater in the case of those known to have entered college than 
 for those who planned to do so and still slightly greater for those who 
 remained throughout the year. Although only 17 per cent of all grad- 
 uates are known to have entered college, 34 per cent of those in the 
 highest I.Q. group, 27 per cent of those in the next, 28 per cent in the 
 next, and so on, did so as compared with per cents of from zero to 14 
 near the bottom of the column. Again in this case the differences would 
 be somewhat greater if the higher I.Q.'s were not too low. 
 
 The high-school averages of the different groups in respect to 
 college careers. Since the evidence offered by the high-school averages 
 is so nearly the same as that of the intelligence quotients the discussion 
 thereof will be much briefer. Tables III and IV and also Figures 3 and 
 4 present the high-school averages for the same groups and in the same 
 manner as did Tables I and II and Figures 1 and 2 the intelligence quo- 
 tients. From Table III or Figure 3 it can be seen that the mean high- 
 school average for all graduates was 82.9, the mean for those who in- 
 tended to go to college 83.3, for those who actually entered college 84.6, 
 and for those remaining at the end of the year 84.8. Likewise, the sec- 
 ond half of Table III and Figure 4 show a definite tendency for the per 
 cents of those with high averages to increase and of those with low av- 
 erages to decrease from each group to the next more limited group. The 
 points dividing averages into high, medium, and low were taken as 90 
 and 80 per cent. On this basis there were in the whole group of grad- 
 uates 16 per cent more with low than with high averages, whereas 
 among those who planned to go to college the difference was 12. among 
 those who actually entered college it was only 1. and among those who 
 remained until the end of the freshman year the per cent of high aver- 
 ages was 4 greater than that of low ones. Table IV 7 offers similar evi- 
 dence by showing in each column a greater downward decrease in the 
 per cents than is found in the column preceding it. In all these cases 
 it will be noticed that the differences between all graduates and those 
 who planned to attend college were not very great, those between the 
 
 [24] 
 

 t>Ou- 
 
 
 
 
 
 
 
 
 U.S ° 
 
 :« 
 
 
 
 
 
 
 
 S.S"S 
 
 u 
 
 -HMM l 'lMW , '13NO\noO'} , at^tS 
 
 o 
 
 
 
 
 
 HaS 
 
 u 
 
 3 
 
 ^Ohx-r,r,riOoo , t-< 
 
 o 
 
 
 
 
 
 _ ,— i .-h _ i — . 
 
 o 
 
 
 
 
 
 S * 
 
 o 
 
 
 
 
 
 
 c 
 
 
 
 
 
 
 
 
 =s 
 
 
 
 
 
 
 
 
 O 1J-T3 
 
 u 
 
 
 
 
 
 
 
 C > u 
 
 60 
 
 «MOOON\0^n-0\N«a;N 
 
 o 
 
 
 
 
 
 
 u 
 
 
 o 
 
 
 
 
 -HcoM^-ioicT+- 1 (j ( ion 
 
 
 u -. 1-1 
 
 C/5 O G 
 
 o 
 
 
 o 
 
 
 
 
 *2 
 
 O 4-1 W 
 
 u 
 
 
 ~^ 
 
 
 
 
 c 
 
 JC 
 
 
 
 
 
 
 
 1) 
 o 
 
 H 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 V 
 
 c/> 
 
 
 
 
 
 
 
 On 
 
 U 60_- 
 
 ** c — 
 
 1) 
 
 1 
 
 
 
 
 
 
 ««.- c 
 
 60 
 
 — vO»0'<t , 0\M^X-HN'<)'!^n'<J< 
 
 o 
 
 
 
 
 
 3 C « 
 
 W 
 
 
 
 
 
 
 
 t3 s t; 
 
 
 — Tfcococo-t'trir^t-)-! 
 
 o 
 
 
 
 
 
 
 "3 
 
 
 o 
 
 
 
 
 
 ■- "3 
 
 (J 
 
 
 
 
 
 
 
 O a 2 
 
 
 
 
 
 
 
 _ 
 
 
 
 
 
 
 
 
 O en 
 
 
 
 
 
 
 
 
 O U 
 
 
 | 
 
 
 
 
 
 
 _e -^ 
 
 
 
 
 
 
 
 
 All 
 -Sc 
 dua 
 
 
 i-iviuiuio\MOI^OOS0^1 l 'l , 'i 
 
 o 
 
 
 
 
 
 
 -Hroi^MOfO' / i , flMoofl-" 
 
 o 
 
 
 
 
 
 J3 cs 
 
 
 
 o 
 
 
 
 
 
 60 t- 
 
 
 
 
 
 
 
 
 s o 
 
 
 
 
 
 
 
 
 60 
 
 
 
 
 
 
 
 
 <U .- O 
 
 V 
 
 rtrt\o^oa\i^nnfOi — -f cn oi co 
 
 <N 
 
 oo 
 
 O 
 
 NO 
 
 
 C- "3 
 
 >> 
 
 N«ioino\(Nconr^'tr^n-H 
 
 r^ 
 
 
 <N 
 
 
 
 s n ^ 
 
 3 
 
 — . — I rH (S N (S - 1 1— 1 
 
 NO 
 
 ^ 
 
 
 
 
 H £ u 
 
 
 rt 
 
 oo 
 
 
 
 
 1) 4-> 
 
 O 
 
 
 
 
 
 
 
 >-i rt 
 
 
 
 
 
 
 
 C 
 
 
 
 
 
 
 
 
 S 
 
 
 
 
 
 
 
 
 O u -73 
 
 u 
 
 
 
 
 
 
 
 C > i» 
 
 60 
 
 rHt^ocr^»oor^\oxr-i''i(^r-i<^^ 1 
 
 (N 
 
 NO 
 
 ON 
 
 oo 
 
 
 — £ 
 
 D 
 
 (suinuin-ticr^-jic - *™ 
 
 no 
 
 
 
 
 
 
 — — n n N n ts -i « 
 
 ON 
 
 Tf 
 
 
 
 
 u ^ r> 
 M O C 
 
 o 
 
 
 <-l 
 
 oo 
 
 
 
 c/5 
 
 o 4-. u 
 
 
 
 
 
 
 U 
 
 J3 
 
 
 
 
 
 
 
 M 
 
 H 
 
 
 
 
 
 
 
 s 
 
 
 
 
 
 
 
 
 3 
 
 i> 60^3 
 
 
 
 
 
 
 
 
 <u 
 
 
 
 
 
 
 
 rt ." u 
 
 60 
 
 rj^vOMMM^OO^O^OOt^ — N 
 
 o 
 
 CO 
 
 CO 
 
 "-> 
 
 
 3 3" 
 
 •~> 
 
 ■t'tnosoocivOt^X'i'r^wOM^ 
 
 
 
 
 CN 
 
 
 T) C ^ 
 
 ~ 
 
 -h to i" \0 x o - O O*. >o N 
 
 r-» 
 
 CO 
 
 
 
 
 O 
 
 
 r~- 
 
 oo 
 
 
 
 
 fS o 
 
 u 
 
 
 
 
 
 
 
 o "•*- 
 
 
 
 
 
 
 
 ( 
 
 
 
 
 
 
 
 
 O v> 
 
 
 
 
 
 
 
 
 O «J 
 
 
 
 
 
 
 
 
 
 
 •t-trTJ"*\DOrOOOxuTtn^ 
 
 r~- 
 
 ON 
 
 ,_, 
 
 r- 
 
 
 3°?-3 
 
 
 ^MJih-'tni^ — rHvoo — r^ io 
 
 NO 
 
 
 
 (N 
 
 
 
 -nnvOO^. --^r^t^^a-t — 
 
 CO 
 
 CN 
 
 
 
 
 J= <3 
 
 
 
 
 OO 
 
 
 
 
 60 t, 
 
 
 
 
 
 
 
 
 £ o 
 
 
 
 
 
 
 
 
 u 
 
 60 
 
 
 
 
 
 V 
 
 > 
 
 
 
 
 
 
 
 
 
 n 
 
 
 
 
 
 o 
 
 
 
 
 
 
 
 
 XI 
 
 © 
 
 oo 
 
 
 
 
 
 > 
 < 
 
 
 
 
 
 u 
 
 
 
 
 
 _ 
 
 
 
 
 
 
 
 _o 
 
 
 
 
 
 
 
 
 
 u 
 
 
 o 
 
 _3 
 
 
 
 "cs 
 
 
 cd 
 
 M 
 
 
 
 
 V. 
 
 
 
 O 
 
 3 
 
 i> 
 
 c 
 
 ONI — wi CO — CT\t^>nf5-- ifJM^uin-" 
 
 
 
 
 i ooav^aNxxxxxM^r^M^ 
 
 
 ni 
 
 u C 
 
 U 
 
 
 60 
 
 
 i i i i i i i I i I i i i i i i 
 
 
 U 
 
 h 0* 
 
 t. 
 
 
 
 
 xO'T'rlOx>o-tr'lox\0'tMO 
 
 
 s 
 
 u 
 
 U 
 
 
 — 
 
 
 r-^^3^a\xxxxxr^r^r^r^t^ 
 
 
 p^ 
 
 G* 
 
 [25] 
 
TABLE IV.— PER CENTS OF HIGH-SCHOOL AVERAGES OF ALL HIGH- 
 SCHOOL GRADUATES FALLING IN VARIOUS CLASSES WITH 
 REGARD TO COLLEGE CAREERS 
 
 
 Graduates 
 
 Those 
 
 Those 
 
 High-school 
 
 planning 
 
 known to 
 
 remaining 
 
 average 
 
 to attend 
 
 have entered 
 
 at end of 
 
 
 college 
 
 college 
 
 one year 
 
 98-99 
 
 75 
 
 25 
 
 25 
 
 96-97 
 
 83 
 
 43 
 
 39 
 
 94-95 
 
 84 
 
 34 
 
 32 
 
 92-93 
 
 84 
 
 35 
 
 32 
 
 90-91 
 
 73 
 
 23 
 
 21 
 
 88-89 
 
 73 
 
 23 
 
 21 
 
 86-87 
 
 73 
 
 22 
 
 20 
 
 84-85 
 
 68 
 
 17 
 
 15 
 
 82-83 
 
 67 
 
 16 
 
 14 
 
 80-81 
 
 64 
 
 13 
 
 10 
 
 78-79 
 
 64 
 
 13 
 
 10 
 
 76-77 
 
 63 
 
 11 
 
 8 
 
 74-75 
 
 62 
 
 10 
 
 8 
 
 72-73 
 
 53 
 
 9 
 
 7 
 
 70-71 
 
 54 
 
 7 
 
 5 
 
 Total 
 
 68 
 
 17 
 
 15 
 
 latter group and those who actually entered college several times as 
 large, and finally those between the college entrants and those who re- 
 mained a year rather slight. This parallels very closely what was shown 
 by the intelligence quotients. 
 
 Summary and conclusion. The data presented in this chapter may 
 be summarized briefly as follows. The study made in Indiana shows 
 that very slight if any selection was taking place among those who 
 planned to attend college as compared with all high-school seniors, 
 whereas the North Carolina and Massachusetts investigations show 
 greater differences, large enough that there is no doubt of their being 
 reliable. The results obtained in Illinois indicate that the group already 
 mentioned was only slightly selected, that there was considerably more 
 marked selection of those who actually entered college and a still further 
 though relatively small selection of those who remained until the end of 
 the freshman year. On the other hand, even with the amount of se- 
 lection indicated it appears that a fairly large fraction of the ablest 
 high-school seniors are not attending college and also that some individ- 
 uals of inferior ability are doing so. 
 
 In view of these facts, the proponents of both sides of the contro- 
 versy may base a portion of their arguments upon the results obtained. 
 There appears to be no doubt that students who enter college are defi- 
 
 [26] 
 
Number 
 
 Cases 
 
 t/?00 
 
 jo so go 
 
 High-School Average 
 All high-school graduates 
 
 100 
 
 Graduates planning to attend college 
 
 Those known to have entered college 
 
 Those remaining at end of one year 
 
 Figure 3. Numbers of Those in Each Group With 
 Various High-School Averages 
 
 [27] 
 
so go 
 
 High-School Average 
 ■ All high-school graduates 
 
 100 
 
 Graduates planning to attend, college 
 
 Those known, to have entered college 
 
 Figure 4. Per Cents of Those in Each Group With 
 Various High-School Averages 
 
 [28] 
 
nitely above the average of high-school seniors in mental ability or, in 
 other words, that some selection is taking place. The degree of selec- 
 tion may be summarized by saying that the difference in mean I.Q.'s 
 is probably five or six points, and that in mean high-school averages 
 about 2 per cent, or that the excess of superior over inferior students is 
 at least 25 "per cent (of the whole) greater among college entrants than 
 among all high-school graduates. On the other hand, the amount of se- 
 lection occurring is not great enough that it can be said that inferior 
 students do not enter college in relatively large numbers. As was 
 pointed out in the beginning, no conclusion can be reached from this 
 evidence as to how college students of today compare in intellectual 
 ability with those of a generation or more ago, except perhaps that they 
 certainly are not as inferior as some persons have claimed. 
 
 [29] 
 
CHAPTER III 
 
 DIFFERENCES BETWEEN THE GROUPS OF FRESHMEN 
 
 ATTENDING VARIOUS INSTITUTIONS OF 
 
 HIGHER LEARNING 
 
 The grouping of the institutions of higher learning. For the pur- 
 poses of this chapter all except three of the 120 institutions of higher 
 learning which contributed the records of their freshmen were classed 
 in several groups. The data for each of the three outstanding univer- 
 sities in the state of Illinois — Illinois, Chicago, and Northwestern — 
 were tabulated separately. All the others were thrown into seven 
 groups according to the types of institutions and their location. The 
 first of these groups consisted of large universities outside of the state, 
 the ones included being Columbia, Cornell, Harvard, Iowa, Pennsyl- 
 vania. Michigan, and Wisconsin. In the second were placed all colleges 
 and small universities located within the state, and also a few, such as 
 Beloit at Beloit, Wisconsin, and Washington at St. Louis, so near that 
 students graduated from certain high schools in Illinois live nearer to 
 them than to any college in the state. All other liberal arts colleges and 
 small universities were placed in the third group. The five Illinois 
 state teachers' colleges composed another group and all other teachers' 
 colleges and normal schools, whether within or without the state, an- 
 other. Institutions of the type of Armour, Bradley, Carnegie, Purdue, 
 and so on, were grouped together as technical schools. Such schools or 
 colleges within universities were not included here, but were included 
 with the universities of which they are .parts. Finally, the last group 
 contains all the others, that is, professional, business, art, and music 
 schools. Likewise here, departments of institutions not primarily exist- 
 ing for these purposes were not included. 
 
 For other purposes the number of groups was still further reduced, 
 only four being used and the last or miscellaneous group neglected. 
 These four were: (1) the University of Illinois, (2) Chicago, North- 
 western, large universities outside of the state, and technical schools, 
 (3) liberal arts colleges and small universities regardless of location, 
 and (4) all teachers' colleges and normal schools. 
 
 Numbers of students attending each group of institutions. It will 
 be seen from Table V that slightly over one-fourth of the freshmen in- 
 cluded in this study attended the University of Illinois and about one- 
 
 [30] 
 
TABLE V.— NUMBERS, PER CENTS, AND MEANS OF FRESHMEN 
 
 ENTERING DIFFERENT INSTITUTIONS OF 
 
 HIGHER LEARNING 
 
 Institution or Group 
 
 Number 
 
 Per 
 cent 
 
 Mean* 
 
 Age 
 
 Mean 
 Point 
 Score 
 
 Mean 
 I.Q. 
 
 Mean 
 
 H. S. 
 
 Average 
 
 
 520 
 79 
 56 
 50 
 
 747 
 98 
 
 251 
 
 35 
 78 
 
 57 
 
 26.4 
 4.0 
 
 2.8 
 2.5 
 
 37.9 
 5.0 
 
 12.7 
 
 1.8 
 4.0 
 
 2.9 
 
 18.4 
 17.9 
 18.3 
 18.1 
 18.3 
 18.3 
 18.4 
 
 18.6 
 18.7 
 
 18.6 
 
 49.5 
 50.8 
 46.7 
 52.1 
 48.6 
 49.6 
 44.4 
 
 45.9 
 50.9 
 
 44.9 
 
 108 
 110 
 105 
 111 
 107 
 108 
 103 
 
 104 
 109 
 
 103 
 
 84 6 
 
 
 84 9 
 
 Northwestern University 
 
 Large Universities not in Illinois. 
 Colleges in and very near Illinois . 
 
 Illinois State Teachers' Colleges . 
 Other Teachers' Colleges and 
 
 Normal Schools 
 
 Technical Schools 
 
 Professional, Business, Art, and 
 
 Music Schools 
 
 83.4 
 83.4 
 85.2 
 85.0 
 84.3 
 
 81.3 
 83.6 
 
 84.0 
 
 
 
 Total 
 
 1971 
 
 100.0 
 
 18.4 
 
 48.5 
 
 107 
 
 84.6 
 
 *The mean ages given in this table are those at the approximate time of entering college. 
 
 eighth the Illinois state teachers' colleges. Combining these gives al- 
 most two-fifths who attended institutions supported by the state in 
 which they lived. Chicago and Northwestern drew only a few, 3 or 4 
 per cent in each case, as was likewise true of the whole group of uni- 
 versities outside of the state. Well over one-third of the whole num- 
 ber attended colleges or small universities within or very near the state 
 and only about one in twenty colleges not so located. The other three 
 groups were all small. Making comparisons according to types of 
 schools we see that more than one-third attended some large university, 
 more than two-fifths colleges or small universities, and more than one- 
 seventh teachers' colleges. From the standpoint of location approxi- 
 mately seven-eighths of the freshmen attended institutions within the 
 state. 
 
 The mean ages of the different groups. The mean ages found in 
 the third column of Table V show that on the average the freshmen at- 
 tending the University of Chicago were the youngest, their mean being 
 17.9 years. The next youngest group, which had a mean of 18.1, was 
 that attending large universities outside of the state. The three groups 
 with ages definitely above the average are the last three in the table, 
 those attending teachers' colleges and normal schools not supported by 
 the state of Illinois, technical schools, and the miscellaneous group. For 
 the University of Illinois, Northwestern University, both groups of col- 
 leges, and the Illinois state teachers' colleges mean ages were very close 
 to the general mean. 
 
 [31] 
 
The mean point scores of the different groups. The point scores 
 of the different groups indicate that the freshmen in the large univer- 
 sities outside of Illinois were of the highest intellectual ability, those in 
 the University of Chicago and in technical schools next, and those in 
 the University of Illinois and in colleges outside of the state slightly 
 above the general mean. The lowest mean point score belonged to 
 those attending Illinois state teachers' colleges and the next lowest to the 
 miscellaneous group. Others definitely below the general average were' 
 those for Northwestern University and for other teachers' colleges. 
 The only group practically at the average for all was that attending col- 
 leges in or very near the state. 
 
 The mean intelligence quotients of the different groups. As is 
 practically always the case when dealing with numbers of individuals 
 the mean I.Q.'s show about the same conditions relative to intelligence 
 as do the mean point scores. The freshmen attending large universities 
 outside of the state, the University of Chicago, and the technical schools 
 were again distinctly above the average and those in both groups of 
 teachers' colleges, professional, business, art and music schools, and 
 Xorthwestern University below it. 
 
 The mean high-school averages of the different groups. The evi- 
 dence of the high-school averages as to the intellectual ability of the 
 different groups does not agree closely with that of the point scores and 
 I. Q.'s. Xone of the mean averages are very markedly above the gen- 
 eral mean, but that of one small group, those attending other teachers' 
 colleges and normal schools, is decidedly below it and those of fresh- 
 men at Xorthwestern University and the large universities outside of 
 the state also are more than one point below. It will be seen that this 
 evidence directly conflicts with that of the Otis Test for large universi- 
 ties outside of the state and that in the case of about half of the other 
 groups does not agree at all closely with it. This fact is in accord with 
 rather general findings that school marks do not correlate highly with 
 intelligence tests results. 
 
 Distributions of intelligence quotients for the four large groups. 
 Tables VI and VII present the detailed distributions of intelligence 
 quotients, the former by actual numbers and the latter by per cents, for 
 the four groups of institutions used in the second plan of grouping de- 
 scribed near the beginning of this chapter. For each institution and 
 also for all institutions the figures are given for all those included in this 
 study who entered college and for those who remained at the end of 
 their freshman vear. The first three columns of each table merelv re- 
 
 [52] 
 
"g en 
 
 8 "3 
 
 H^ 
 
 Those 
 Remain- 
 ing 
 
 CNxrl + r^-r', wit, 3nvOC-CC 
 cs cn co w-, Tf cn — ■> 
 
 CN 
 
 CN 
 
 co 
 
 O 
 
 CO 
 CM 
 
 o\ 
 
 Those 
 Enter- 
 ing 
 
 CN CN "tf" sC ■ "~. co CN 
 
 co 
 oo 
 CN 
 
 CO 
 
 O 
 
 CN 
 CN 
 
 oo 
 
 Colleges 
 
 Those 
 Remain- 
 ing 
 
 — C>COMN- isO^O — O O ^f — C: O 
 
 o 
 
 CO 
 
 r^ 
 
 o 
 
 ON 
 CO 
 
 ■* 
 
 Those 
 Enter- 
 ing 
 
 
 co 
 
 oo 
 
 o 
 
 oo 
 
 CO 
 
 ■* 
 
 Other Large 
 
 Universities 
 
 and Technical 
 
 Schools 
 
 Those 
 Remain- 
 ing 
 
 - n 't ui fi tN 
 
 -r 
 cn 
 
 o 
 
 CO 
 
 ~ 
 
 
 t^OVOOWPTttNOOcotOr- < O © © 
 
 — tN * \0 ■* n -h 
 
 00 
 
 in 
 
 o 
 
 5- 
 
 CN 
 
 University 
 of" Illinois 
 
 Those 
 Remain- 
 ing 
 
 
 CN 
 
 o 
 
 ■* 
 
 ■* 
 
 CN 
 
 Those 
 Enter- 
 ing 
 
 
 o 
 
 oo 
 O 
 
 CN 
 
 oo 
 CO 
 
 CN 
 
 u c 
 
 js 2 
 
 Those 
 Remain- 
 ing 
 
 -r-tNrotNri- 
 
 SO 
 
 o 
 
 ■* 
 
 Those 
 Enter- 
 ing 
 
 --enr, -xn^-ct^ryO« 
 ^-uo — — rN x r ; h- r, va M 
 
 0\ 
 
 O 
 
 CO 
 
 -* 1 
 
 All 
 High- 
 School 
 Gradu- 
 ates 
 
 nx — tN — x — tNx<!iriinui^ 
 
 CO 
 
 co 
 
 o 
 
 CN 
 
 O 
 
 cy 
 
 
 
 
 O 
 
 u 
 
 > 
 o 
 
 ca 
 
 c 
 u 
 u 
 u 
 
 1) 
 
 0- 
 
 
 
 
 
 o 
 ON 
 
 
 
 
 
 O 
 
 H 
 
 c 
 u 
 
 
 * 
 
 
 
 
 
 u 
 
 
 -O 
 
 
 
 
 
 
 
 i I i : i I I I I 1 1 1 1 I i 
 
 ~- \C •— ' sC^^O'— • ^O — \C ~— ^C^- vO r-H 
 
 n n n - — oo^o\x«r^r^vo>o 
 
 a, 
 
 [33] 
 
o 
 
 z 
 
 2 
 w 
 
 H 
 Z 
 
 w 
 ^Z 
 
 oE 
 go 
 
 J< 
 ►J 
 
 w 
 
 as 
 ^< 
 
 5;* 
 
 °z 
 
 2g 
 
 HE 
 
 PZ 
 
 He/) 
 ZO 
 WI 
 OH 
 
 *Q 
 
 £* 
 i < 
 
 HO 
 
 Those 
 Remain- 
 ing 
 
 ©CCO-^CsrOCTWOCiO^O-'J-OO 
 
 o 
 
 d 
 
 o 
 
 C5 
 
 
 
 Sum 
 
 U 4J — 
 
 -C c.S 
 
 Hw 
 
 ONN^MOi'^'HO'-iooO'tOO 
 
 o 
 
 d 
 
 o 
 
 
 oo 
 
 
 jj 
 
 60 
 
 ~o 
 O 
 
 I'll 
 
 H « 
 OS 
 
 -H\000O0\' H OO0\^'t u i-iOO 
 n ^o n t-~ OM^ (O \o M -h 
 
 o 
 
 o 
 o 
 
 
 "* 
 
 u J. 
 
 g U 60 
 
 2 *■' e 
 
 riovont^-MrHnn-i't^-'OO 
 
 o 
 
 o 
 © 
 
 oo 
 
 ■* 
 
 Other Large 
 
 Universities 
 
 and Technical 
 
 Schools 
 
 1) C 
 
 g'« M 
 
 -<= £.£ 
 H <u 
 
 nooi<tot^r--xNO\o\ooocc 
 
 o 
 
 o 
 o 
 
 
 - 
 
 rt ,-H <N — 1 — 1 
 
 g <U 60 
 
 -c c.S 
 
 l^w-iOOt^-oocjsOOr-^-^OOO 
 
 M n N O \0 * \0 -n ro-i ^ 
 
 _ _ ri — i — > 
 
 o 
 
 o 
 o 
 
 5 
 
 cs 
 
 11 
 
 _> — 
 
 P o 
 
 oj G 
 
 -g £.2 
 
 H «■> 
 
 PS 
 
 
 o 
 
 o 
 
 o 
 
 
 CN 
 
 g <U 60 
 
 
 o 
 
 o 
 
 o 
 
 
 CN 
 
 -c.2 
 
 60 *^ 
 ffi.t! 
 
 u c 
 
 g'S 60 
 
 H OJ 
 
 PS 
 
 sor^c^vDOI^f-O^OO — i-ttSO- 1 
 
 o 
 
 o 
 o 
 
 OO 
 
 to 
 
 ■* 
 
 (S\o^r^a>Nn\on" 
 
 g <U 60 
 
 Hw 
 
 \o\ooav*\o*ooi-Hn'trJO'H 
 
 o 
 
 o 
 
 o 
 
 1^ 
 
 CO 
 
 "* 
 
 nvoo^o^rs-jifso-n 
 
 — ^1 
 
 3 „ 
 13 S 
 
 i- as 
 
 
 
 r)vo\on\ooo\ooNco^-tui(NiH 
 
 o 
 
 d 
 
 o 
 
 
 O 
 
 a 
 
 
 
 o 
 H 
 
 o 
 
 > 
 
 o 
 .n 
 
 e 
 
 OJ 
 CJ 
 tH 
 
 aj 
 
 Cm 
 
 d 
 
 On 
 •S 
 
 _o 
 <u 
 -O 
 
 C 
 u 
 u 
 
 ID 
 
 Ch 
 
 
 
 
 
 
 
 
 
 
 
 
 1 1 1 1 1 1 l l 1 l 1 l 1 1 1 
 
 i — i^O' — i so i— i \o ' — isO' — i so ^* so « — i SO' — ! 
 (^riN — — OOCT\0\»ooM^sC>0 
 
 [34] 
 
peat some of the data given in Tables I and II so that they will be con- 
 venient for purposes of comparison with the four sets of double col- 
 umns for the different groups. Comparing the first and second of each 
 of these pairs of columns we see that in the cases of the University of 
 Illinois and of other large universities and technical schools more selec- 
 tion took place during the freshman year than was true of the other 
 two groups. In the case of each of the former the mean I.Q. of those 
 remaining at the end of the year was one greater than of those who 
 entered and the excess of those of superior over those of inferior intel- 
 ligence 2 and 3 per cent greater respectively, whereas in the cases of the 
 groups of liberal arts colleges and of teachers' colleges the means were 
 no greater and the excess of superior over inferior students one and 
 zero. This fact indicates that, on the whole, those institutions, such as 
 the University of Illinois and the other large universities and technical 
 schools, which exercise a greater degree of selection in choosing their 
 freshmen from high-school graduates continue the same policy within 
 the institutions themselves, whereas those who exercise a smaller degree 
 of selection likewise continue this policy. Undoubtedly individual insti- 
 tutions constitute exceptions to this rule. Some probably believe that 
 they have selected candidates for admission rigorously enough that they 
 can well afford to grade rather leniently and strive to have little elim- 
 ination, whereas others are undoubtedly conscious that they have done 
 little if anything to select those who enter and plan to weed out many 
 of those of inferior ability early in their college careers. 
 
 Reasons for the differences in freshmen attending the different 
 groups of higher institutions. With possibly one exception it is rather 
 easy to suggest explanations for the superiority of the freshmen attend- 
 ing certain groups of institutions and the inferiority of those at others. 
 In general large universities and technical schools have somewhat more 
 severe entrance requirements and enjoy the reputation of requiring 
 higher standards of work from their students than do liberal arts and 
 teachers' colleges and therefore it is to be expected that freshmen enter- 
 ing such institutions would, as a group, have more than average ability. 
 The one exception referred to above is the case of Northwestern Uni- 
 versity, one of the group just mentioned. Two possible explanations of 
 the low means of the group entering that institution may be suggested. 
 One is that the number of cases concerned is too small to yield a reli- 
 able measure and the second that the Northwestern freshmen included 
 may represent a group of students who live rather close to that insti- 
 tution and who have entered it merely because it offers the most con- 
 
 [35] 
 
venient opportunity for securing further education and not because they 
 have definite vocational or other purposes. The apparently inferior in- 
 telligence of those entering teachers' colleges is in accord with several 
 reported results of the use of intelligence tests in different institutions. 
 Most of those who enter teachers' colleges are probably planning to take 
 only one or two years of work and it is likely that on the whole students 
 with such intentions are less able than those who plan to carry four- 
 year courses. Also the fact that the group entering business, art, and 
 music schools ranks somewhat below the average agrees rather well with 
 general observation and common opinion, and the number of those 
 entering professional schools included is so small as not to produce 
 much effect upon the means for this whole group. Since more fresh- 
 men were found in liberal arts schools and smaller universities than in 
 any other of the groups it is not surprising that their means should be 
 very close to those of the whole number of freshmen. 
 
 135 J 
 
CHAPTER IV 
 
 DIFFERENCES BETWEEN THE GROUPS OF FRESHMEN 
 CARRYING VARIOUS SUBJECTS 
 
 The data to be presented in this chapter. Although the topic of 
 this chapter is not an integral part of the main problem treated in this 
 bulletin the data and discussion which follow are presented because 
 they may be of interest to some readers. The forty-nine subjects 20 
 carried by ten or more freshmen are listed in Table VIII. Following 
 each subject may be found the number of students carrying that sub- 
 ject, the per cent that this number is of the total number whose fresh- 
 man marks are included, the mean age, point score, I.Q., high-school 
 average, freshman mark in the given subject, and freshman average of 
 the group. According to the first line of the table, which gives these 
 data for the whole group, 1892 freshmen remained in school long 
 enough to receive at least some marks, so this number is taken as the 
 base or 100 per cent. Their mean age at entrance was 18.3, mean point 
 score 48.5, mean I.Q. 107, mean high-school average 84.6, mean fresh- 
 man subject mark 82.5, and mean freshman average, of course, the 
 same, 82.5. 
 
 Numbers of freshmen carrying different subjects. Glancing at the 
 first two columns one may discover the not unexpected fact that more 
 freshmen carried rhetoric than any other subject and that the second 
 largest number took physical education, the per cents being about 85 
 and 70, respectively. No other subject was carried by more than half 
 of the whole group. Chemistry and history were carried by more than 
 one-third of the group, and algebra and hygiene by more than one- 
 fourth. At the other extreme are to be found dentistry and Italian 
 with less than 1 per cent in each, athletic coaching, Greek, and horticul- 
 ture with approximately 1 per cent apiece, and agriculture, clothing, 
 engineering, general home economics, pharmacy, philosophy, physics, 
 sociology, and stenography and typing with less than fifty freshmen, or 
 2y 2 per cent, in each. 
 
 The mean ages of the different groups. In general the differences 
 in the mean ages of the freshmen carrying the different subjects were 
 
 20 In some cases what is treated as a subject really embraces a group of closely re- 
 lated subjects. For example, all agricultural subjects other than horticulture are grouped 
 as agriculture, and all courses in freehand drawing, painting, designing, and so forth, are 
 grouped as art. 
 
 [37] 
 
w 
 
 to 
 to 
 
 to 
 O 
 
 CO 
 
 to 
 
 D 
 
 O 
 
 OS 
 
 o H 
 
 toU 
 
 ow 
 
 ^to 
 cd to 
 
 £3 
 
 to 
 o 
 
 OS 
 
 w 
 
 to 
 
 to 
 
 to 
 
 s 
 
 « t 5 
 to < 
 
 U fl u 
 
 S3*"! 
 
 to^^ 
 
 < 
 
 
 C 4-1 U 
 
 ns C i- 
 2 tocn 
 
 '^nnM^Mxo-nr-'tO't^ 1 ''!- oc -+■ <n — — or^-ooN-f 
 
 uiI^OvC*^' 
 
 OMvOM-* , -t"XMt^w1rf(SM*^0'*'H'th. 
 
 vocnNoriMi^'-'i^^r'iM'flri 
 
 ifloorovOOOrlxuiOMK 
 
 "^^•lu^voc^ l ^o^^o^■c^ l ^co^' H C^fO\0"^^^xvD»^'^o-t , 
 
 ooooooooooooooooooooooooooooooooooooooooooooooooooo 
 
 0-frir'r\Oai-i\0(Srf | 'i^i0^woo«r^oo\onoooo 
 
 O oo r) (J\ oo vo >h O O I s t^ N >C 
 
 0\un>w\ONMOO\nO') , nHTtioi , '-'0-Hr~Orl'H 1 iit*) 
 
 £%£ 
 
 be." 
 
 U 
 
 
 5>a g-c -^ n 
 
 s g 3.5 g S3 fco E g 
 
 to 
 
 —^ O oAJ— — to J-. *-<.— .« u -J-i ^z, <j> vs u \J i-> t-, v v <u <u <u u, • ~ u 
 
 <<<«<<«MPQUUUQKWWfcOUOOOOXI 
 [38] 
 

 
 •t ^ vO rl 3> M - w ; 
 
 X "•( n X ?T) O f. ^i >fl "1 N + M 
 
 Orixor^-"HU\ l 'iri-fvO'tON^MC>-'voi^c i 'i-'vD 
 
 2 = ^ 
 < 
 
 n x m o\ vo r^ "■/ — o^. n o rt o r^ n "■> ?> n « r~ o\ o c ••■ 
 
 
 ; o s 
 
 O C= — norioivoxnwnovB- <rooo(Nor^t^nc>c)(N 
 
 oooooor~-r^ooooocoooo3vooooooooocooooooooocoooooo 
 
 h[ — rfir, 0\5\-t^^O^\0vCt\C^it^W>^nO't0" 
 
 on u iOOto>^n | ^rH''Hvc i 'i'ti s -X'txrTt>o\o-t 
 m — i ri c- ; — n — cs r-4 <-o t^j -t- — I 
 
 2 s 
 
 5 « 
 
 ■S xi 
 
 »i 2 ni 
 
 _o — _2c/3. 
 
 o<JOcj2on-2« b ' oc O< 
 
 = >,>,>,- ^ « ■- = cm- 
 
 ^ m' c £ re. -■_.«_! — !— .-^-e-e-C-c o « 3* o o-i:, 1 - o 
 
 J3 U" 3 
 
 
 [39] 
 
not great. Athletic coaching and pharmacy alone had means above 19 
 years and Italian and Latin are the only two whose means fell below 
 18. Of these four Latin alone had enough cases that the mean can be 
 considered reliable to the nearest month or two. 
 
 The mean point scores and I.Q.'s of the different groups. The 
 point score and I.Q. columns show somewhat greater variability than 
 the one for age. Students carrying freshman geometry and philosophy 
 tied for the highest point score, 52.6, and likewise had mean I.Q.'s of 
 111, as high as any. Although the mean point score for those carrying 
 Latin was somewhat lower, their low age resulted in a mean I.Q. of 
 111 also. Other subjects which had groups of students distinctly above 
 the average are algebra, chemistry, engineering, French, general mathe- 
 matics, Italian, mechanical drawing, military, and trigonometry. The 
 one distinctly below all others, as measured by both point scores and. 
 I.Q.'s, was dentistry, but as there were only twelve students therein, the 
 figures probably have little significance. Other relatively low groups 
 were those in arithmetic, art, athletic coaching, biology, clothing, com- 
 mercial work, education, geography, home economics, horticulture, mu- 
 sic theory, pharmacy, physiology, and psychology. 
 
 The mean high-school averages of the different groups. The sixth 
 column contains the mean high-school averages, which vary from 80.2 
 for the small dentistry group and 81.3 for the only slightly larger group 
 in pharmacy up to 87.6 for those carrying freshman Latin. Other 
 groups than the Latin one with relatively high averages were those in 
 French, Italian, stenography and typing, and zoology, whereas home 
 economics and physics were markedly low. Comparing the high-school 
 averages with the point scores and I.Q.'s, it is apparent that in some cases 
 the different evidence as to the intellectual ability of the group of stu- 
 dents carrying a particular subject agrees, whereas in other cases it 
 does not. Thus the Latin group, which had as high an I.Q. as any and 
 also a rather high point score had the highest average, but the fresh- 
 men carrying geometry and philosophy, who had the highest point 
 scores and as high I.Q.'s as any, had high-school averages only some- 
 what above the general mean. Of the fifteen groups referred to in the 
 last paragraph as being low according to test results all except two had 
 mean high-school averages below the general mean, but several of the 
 groups with low high-school averages were not below the average in 
 point scores and I.Q.'s. 
 
 The mean freshman marks in the different subjects. The next to 
 the last column shows decidedly greater variations in the mean marks 
 
 [40] 
 
in the different freshman subjects than were found in the high-school 
 averages. The lowest mean freshman mark, 77.4, was given in phar- 
 macy, whereas the highest, 88.9, was in military work. Other subjects 
 having means below 80 were accountancy, geography, and physiology, 
 whereas those of engineering, Italian, music performance, and physical 
 education were above 85. It can readily be seen by glancing down the 
 column of mean freshmen subject marks and comparing the entries 
 therein with those in the previous columns that there is positive, but not 
 nearly perfect, correlation. For example, the military group received 
 the highest freshman mark, but had a high-school average below the 
 general mean. On the other hand it was well above the means in both 
 point scores and I.Q.'s. Of the other four subject groups in which the 
 freshmen subject means were above 85, three had high-school averages 
 above and one below the general mean. Two of the four were decid- 
 edly high according to the test results and two very close to the mean. 
 Of the four named as having freshman subject marks below 80, all had 
 high-school averages lower than the general mean and three were below 
 the mean in their point scores and I.Q.'s. 
 
 The mean freshmen averages of the different groups. The last 
 column of Table VIII contains the mean freshmen averages of the stu- 
 dents carrying each subject. The groups which made averages above 
 85 were those carrying dentistry, French, Italian, and Latin, whereas 
 the only one with an average below 80 was pharmacy. Others below 
 81 were accountancy, arithmetic, athletic coaching, economics, engineer- 
 ing, physiology, and zoology. As was the case with the mean freshmen 
 subject mark so in this case there is not a high degree of agreement with 
 the other measures given in the table. Three of the four groups earn- 
 ing high averages were also high on the test scores and high-school 
 averages. All of them were above the average in their freshman sub- 
 ject marks, but only one very much so. The one group which had a 
 mean freshman average below 80 was likewise low in all the other 
 measures except age, in which it was very high, but of those having 
 freshman averages between 80 and 81 some were high in each of the 
 preceding measures though most were below the average. 
 
 The agreement between the various data concerning groups of 
 freshmen. In the last few paragraphs various comments have been 
 made to the general effect that the correlation between mean ages, point 
 scores or I.Q.'s, high-school averages, freshmen subject marks, and 
 freshmen averages was not very high although there tended to be a 
 definite relationship. The amount of relationship is perhaps best shown 
 
 [41] 
 
TABLE IX.— COEFFICIENTS OF CORRELATION BETWEEN MEAN AGES, 
 POINT SCORES, I. Q.'S, HIGH-SCHOOL AVERAGES, FRESHMAN- 
 SUBJECT MARKS, AND AVERAGES 
 
 Ages and Point Scores 
 
 Ages and I. Q.'s 
 
 Ages and High-School Averages 
 
 Ages and Freshman Subject Marks 
 
 Ages and Freshman Averages 
 
 Point Scores and I. Q.'s 
 
 Point Scores and High-School Averages 
 
 Point Scores and Freshman-Subject Marks 
 
 Point Scores and Freshman Averages , 
 
 I. Q.'s and High-School Averages 
 
 I. Q.'s and Freshman-Subject Marks 
 
 I. Q.'s and Freshman Averages 
 
 High-School Averages and Freshman-Subject Marks. 
 
 High-School Averages and Freshman Averages 
 
 Freshman-Subject Marks and Freshman Averages... 
 
 -.53 
 -.53 
 -.67 
 -.21 
 -.46 
 + .97 
 + .48 
 + .24 
 + .21 
 + .54 
 + .21 
 + .19 
 + .30 
 + .36 
 + .42 
 
 by the coefficients of correlation, which will be found in Table IX. From 
 this table one can see at a glance that the only really high correlation 
 was that found between point scores and I. Q.'s. which would be ex- 
 pected to approach unit}-, but that all of the coefficients were large 
 enough, either positively or negatively, to indicate that some relation- 
 ship existed, and several of them that it was rather marked. As would 
 be expected the correlation of the other items with age were all nega- 
 tive. All except one were approximately — .50 or more. The writer 
 would have expected this one, that between ages and freshman marks, 
 to be larger than — .21 and more like the others, and has no reason to 
 suggest why this was not the case. It appears that point scores and 
 I. Q.'s agreed more closely with high-school averages than with fresh- 
 man marks or averages. There was little difference in the degree of 
 relationship between high-school averages and marks in particular 
 freshman subjects and freshman averages in general. 
 
 The relationship of intellectual ability to college success. Prob- 
 ably the most significant conclusion suggested by the table is that the 
 quality of freshman work in college apparently depends much less upon 
 intelligence as shown by an intelligence test than does the quality of 
 high-school work. This conclusion is borne out, although the differ- 
 ences are not so marked, by a comparison of the coefficients of corre- 
 lation between actual individual test scores and high-school marks with 
 those between test scores and freshman marks. The chief explanation 
 appears to be that in college conditions are such that other factors than 
 mere intellect or mentality in the narrow sense play a much more im- 
 
 [42] 
 
portant role in determining success or failure than is the case in high 
 school. The college freshman frequently enters a very different atmos- 
 phere and environment than that to which he has been accustomed. He 
 must work much more independently and with comparatively little 
 guidance. It is frequently more difficult to locate and secure the ma- 
 terials to be studied. Distractions and opportunities for activity other 
 than study are much more numerous. In view of these and other sim- 
 ilar facts such qualities as fixedness of purpose, determination, perse- 
 verance, and so on. would seem to determine the quality of work done 
 to a much greater degree than in high school. This fact or apparent 
 fact emphasizes the need for minimizing the break between high school 
 and college and providing by means of "freshman week" or other sim- 
 ilar procedures and an efficient advisory system for the better orienta- 
 tion and guidance of those entering- college. 
 
 [43] 
 
CHAPTER V 
 SUMMARY AND CONCLUSIONS 
 
 The problem of this bulletin. The recent marked increase in en- 
 rollment in institutions of higher learning coupled with the decline in 
 the purchasing power of the dollar and the demand for more varied ed- 
 ucational offerings and activities has resulted in a serious problem as to 
 what is to be done to meet the situation adequately. One of the im- 
 portant questions arising in the consideration of this problem is that 
 of the intellectual ability of the large numbers who now seek admittance 
 to college. It has been the purpose of this bulletin to present data 
 which throw some light upon the question of the extent to which college 
 freshmen constitute a select group as compared with all high-school 
 graduates. 
 
 The methods used in this study. In addition to reviewing briefly 
 the evidence offered by investigations in Indiana, North Carolina, and 
 Massachusetts as to the degree of selection taking place in the group of 
 high-school seniors who expected to attend college a more complete and 
 extended investigation was conducted with some thousands of Illinois 
 high-school seniors. They were followed through the freshman year of 
 college and data showing their ages, mental test scores, high-school av- 
 erages, and freshman marks tabulated. Although it is true that cer- 
 tain errors are present in these data, it is shown that some of these er- 
 rors have practically no effect upon the average measures used and 
 that others only strengthen whatever conclusions may be reached as to 
 the amount of selection taking place. 
 
 The amount of selection found. It appears that there is a small 
 but reliable difference in ability between high-school seniors who plan 
 to attend college and all seniors, the former of course ranking slightly 
 higher. A decidedly greater difference exists between those who actu- 
 ally become college freshmen and all high-school graduates. The mean 
 I.Q. of the former is probably at least five points higher than that of the 
 latter and the mean average high-school mark about two points higher. 
 The excess of superior over inferior college freshmen is probably about 
 25 per cent greater than the corresponding figures for all high-school 
 graduates. It is further shown that a comparatively small amount of 
 selection occurs during the freshman year and that those still remaining 
 
 [44] 
 
in college at the end of the year are slightly superior to the whole group 
 of those entering. 
 
 Differences between freshmen attending various institutions of 
 higher learning. Grouping the freshmen according to the institutions 
 or types of institutions attended it appears that those attending the 
 University of Illinois and other large universities and technical schools 
 possess on the whole mental ability well above the average, whereas 
 those entering teachers colleges and professional, art, and music schools 
 are decidedly below the average, and those entering liberal arts colleges 
 and small universities have approximately the same average ability as 
 the whole group. 
 
 Differences between freshmen carrying various subjects. Tabu- 
 lating the data according to the forty-nine subjects carried by the fresh- 
 men, it appears that there are marked differences in the groups carry- 
 ing different subjects. The evidence of the test scores, high-school aver- 
 ages, subject marks, and freshmen averages is conflicting in many cases 
 so that it is difficult to say with a high degree of certainty that the 
 groups of freshmen in certain subjects are superior and those of others 
 inferior. On the whole the conclusion seems justified, however, that the 
 freshmen carrying foreign languages, except perhaps Spanish, and most 
 mathematics courses, also philosophy, constitute somewhat superior 
 groups, whereas those in arithmetic, athletic coaching, geography, home 
 economics, pharmacy, and physiology are inferior. In the case of a few 
 other subjects there are marked contradictions between the evidence, 
 whereas in the case of many of them it appears that on the whole the 
 student groups therein do not vary greatly from the general average of 
 all freshmen. 
 
 [45] 
 

-u 
 
 UNIVERSITY OF ILLINOIS; BULLETIN 
 
 Issued Weekly 
 Vol. XXIV May 10, 1927 No. 36 
 
 [Entered as second-class matter December 11, 1912, at the post office at Urbana, Illinois, under the 
 Act of August 24, 1912. Acceptance for mailing at the special rate of postage provided for in 
 section 1103, Act of October 3, 1917, authorized July 31, 1918.] 
 
 BULLETIN NO. 34 
 
 BUREAU OF EDUCATIONAL RESEARCH 
 COLLEGE OF EDUCATION 
 
 ARE COLLEGE STUDENTS A 
 SELECT GROUP 
 
 By 
 
 Charles W. Odell 
 
 Assistant Director, Bureau of Educational Research 
 
 PRICE 25 CENTS 
 
 PUBLISHED BY THE UNIVERSITY OF ILLINOIS, URBANA 
 
 1927 
 
The Bureau of Educational Research was established by act 
 of the Board of Trustees June 1, 1918. It is the purpose of the 
 Bureau to conduct original investigations in the field of education, 
 to summarize and bring to the attention of school people the results 
 of research elsewhere, and to be of service to the schools of the 
 state in other ways. 
 
 The results of original investigations carried on by the Bureau 
 of Educational Research are published in the form of bulletins. A 
 list of available publications is given on the back cover of 
 this bulletin. At the present time five or six original investigations 
 are reported each year. The accounts of research conducted else- 
 where and other communications to the school men of the state 
 are published in the form of educational research circulars. From 
 ten to fifteen of these are issued each year. 
 
 The Bureau is a department of the College of Education. Its 
 immediate direction is vested in a Director, who is also an instructor 
 in the College of Education. Under his supervision research is 
 carried on by other members of the Bureau staff and also by grad- 
 uates who are working on theses. From this point of view the 
 Bureau of Educational Research is a research laboratory for the 
 College of Education. 
 
 Bureau of Educational Research 
 
 College of Education 
 
 University of Illinois, Urbana 
 
THE UNIVERSITY OF ILLINOIS 
 
 THE STATE UNIVERSITY 
 URBANA 
 
 DAVID KINLEY, Ph.D., LL.D., President 
 
 The University Includes the Following Departments 
 
 The Graduate School 
 
 The College of Liberal Arts and Sciences (Curricula: General with majors, in 
 the Humanities and the Sciences; Chemistry and Chemical Engineering; Pre- 
 legal, Pre-medical and Pre-dental; Journalism, Home Economics, Economic 
 Entomology and Applied Optics) 
 
 The College of Commerce and Business Administration (Curricula: General 
 Business, Banking and Finance, Insurance, Accountancy, Railway Administra- 
 tion, Railway Transportation, Industrial Administration, Foreign Commerce, 
 Commercial Teachers, Trade and Civic Secretarial Service, Public Utilities, 
 Commerce and Law) 
 
 The College of Engineering (Curricula: Architecture, Ceramics; Architectural, 
 Ceramic, Civil, Electrical, Gas ; General, Mechanical, Mining, Municipal and 
 Sanitary, and Railway Engineering; Engineering Physics) 
 
 The College of Agriculture (Curricula: General Agriculture; Floriculture; Home 
 Economics; Landscape Architecture; Smith-Hughes — in conjunction with the 
 College of Education) 
 
 The College of Education (Curricula: Two year, prescribing junior standing for 
 admission — General Education, Smith-Hughes Agriculture, Smith-Hughes Home 
 Economics, Public School Music; Four year, admitting from the high school — 
 Industrial Education, Athletic Coaching, Physical Education 
 
 The University High School is the practice school of the College of 
 Education) 
 
 The School of Music (four-year curriculum) 
 
 The College of Law (Three-year and four-year curricula based on two years of 
 college work) 
 
 The Library School (two-year curriculum for college graduates) 
 
 The College of Medicine (in Chicago) 
 
 The College of Dentistry (in Chicago) 
 
 The School of Pharmacy (in Chicago) 
 
 The Summer Session (eight weeks) 
 
 Experiment Stations and Scientific Bureaus: U. S. Agricultural Experiment 
 Station; Engineering Experiment Station; State Natural History Survey; State 
 Water Survey; State Geological Survey; Bureau of Educational Research. 
 
 The Library collections contain (June 1, 1926) 711.753 volumes and 155,331 pam- 
 phlets. For catalogs and information address 
 
 THE REGISTRAR 
 
 Urbana, Illinois 
 
BULLETINS OF THE BUREAU OF EDUCATIONAL RESEARCH, 
 
 COLLEGE OF EDUCATION, UNIVERSITY OF ILLINOIS, 
 
 URBANA, ILLINOIS 
 
 Price 
 No. 6. Monroe, Walter S. The Illinois Examination SO 
 
 No. 7. Monroe, Walter S. Types of Learning Required of Pupils in the Sev- 
 enth and Eighth Grades and in the High School IS 
 
 No. 8. Monroe, Walter S. A Critical Study of Certain Silent Reading Tests. . .50 
 
 No. 10. Bureau of Educational Research. Relation of Size of Class to School 
 
 Efficiency JO 
 
 No. 11. Monroe, Walter S. Relation of Sectioning a Class to the Effectiveness 
 
 of Instruction IS 
 
 No. 12. Odell, Charles W. The Use of Intelligence Tests as a Basis of School 
 
 Organization and Instruction SO 
 
 No. 13. Monroe, Walter S., and Foster, I. 0. The Status of the Social Sciences 
 
 in the High Schools of the North Central Association SO 
 
 No. 14. Monroe, Walter S., and Carter, Ralph E. The Use of Different Types 
 of Thought Questions in Secondary Schools and Their Relative Dif- 
 ficulty for Students 30 
 
 No. IS. Monroe, Walter S. The Constant and Variable Errors of Educational 
 
 Measurements 2S 
 
 No. 16. Odell, Charles W. An Annotated Bibliography Dealing With the 
 Classification and Instruction of Pupils to Provide for Individual 
 Differences SO 
 
 No. 17. Monroe, Walter S., and Souders, Lloyd B. Present Status of Written 
 
 Examinations and Suggestions for Their Improvement SO 
 
 No. 18. Streitz, Ruth. Teachers' Difficulties in Arithmetic and Their 
 
 Correctives 30 
 
 No. 19. Odell, Charles W. The Progress and Elimination of School Children 
 
 in Illinois SO 
 
 No. 20. Monroe, Walter S., and Mohlman, Dora Keen. Training in the 
 
 Technique of Study 50 
 
 No. 21. Monroe, Walter S. A Survey of the City Schools of Marion, Illinois. . .50 
 
 No. 22. Odell, Charles W. Conservation of Intelligence in Illinois High Schools JO 
 
 No. 23. Streitz, Ruth. Teachers' Difficulties in Reading and Their Correctives .30 
 
 No. 24. Seybolt, Robert Francis. The Evening School in Colonial America 50 
 
 No. 25. Monroe, Walter S., and Johnston, Nell Bomar. Reporting Educational 
 
 Research 50 
 
 No. 26. Brownell, William Arthur. A Study of Supervised Study JO 
 
 No. 27. Glick, H. N. Effect of Practice on Intelligence Tests JO 
 
 No. 28. Seybolt, Robert Francis. Source Studies in American Colonial Educa- 
 tion — The Private School JO 
 
 No. 29. Odell, Charles W., assisted by Blough, John H. An Annotated Bib- 
 liography Dealing with Extra-Curricular Activities in Elementary 
 and High Schools JO 
 
 I No. 30. Monroe, Walter S. The Duties of Men Engaged as Physical Direc- 
 tors or Athletic Coaches in High Schools 21 
 
 No. 31. Monroe, Walter S., assisted by Clark, John A. The Teacher's Respon- 
 sibility for Devising Learning Exercises in Arithmetic 50 
 
 No. 32. Odell, Charles W. The Interpretation of the Probable Error and the 
 
 Coefficient of Correlation » 50 
 
 No. 33. Monroe, Walter S., and Herriott, M. E. Objectives of United States 
 
 History in Grades Seven and Eight JO 
 
 No. 34. Odell, Charles W. Are College Students a Select Group? 25 
 
r 
 
 awiinmiifln wTWj.nrmmiiirwi iiMii M M iiii i i iiM w ii i ii i nBBWWi^ 
 
 Maaamsgaauaumasam 
 

 UNIVERSITY OF ILLINOIS-URBANA 
 
 3 0112 070071516