A CWARM‘WE mm or CAL-swam MALE mm mm. mums mm WAY LIE mom. Am mow WHO mm out Tum L.» film mm M PL. 0.! ’ f * § MICHIGAN mm UNWERSIH; _; Kenneth L. Hardmg ‘ i ,’ ‘ 1966' ’" ”“’ fifiéia \|fl\\\\\|1\\\i||s\,\\fl\\\| Milli“ “W \ LIBRARY Michigan State This is to certifg that the thesis entitled _r . v -.'7 (:t H x. . o 57 A ComparanLJG Atuoy of CaucaSLah Aale 7v. _ n -\ a _‘ w \J_ o 1 diah echool btuoenbs iho beay 1n School, eno those Tho Dram Out presented by - Kenneth L. Hardin: has been accepted towards fulfillment of the requirements for -., Pl" . U. degree inwfihmnfi m1 Major professor Date Juno 13. 19;. 0-169 v— m—Awmrfi - .- (SIG-"n ‘ . --.._-oH-—-o~'y-o _ u .o—v...._..—. R Y Sm: MAGlC 2 Mi 33999 ;\ ABSTRACT A COMPARATIVE STUDY OF CAUCASIAN MALE HIGH SCHOOL STUDENTS WHO STAY IN SCHOOL AND THOSE WHO DROP OUT isi by Kenneth LR’Harding This study was undertaken to determine the ways in which Caucasian male high school dropouts differ from those students who remain in school in regard to selected characteristics. The major variables examined were a) general self—concept of academic ability, b) the student's perception of his parents' evaluation of his academic ability, c) the student's perception of how long his parents expected him to remain in school, and d) the level of education the student expected to attain. Also examined were three variables which had been identified in numerous 1 studies as contributing significantly to the decision of male students to leave high school prior to graduation: a) low academic ability; b) low socio—economic status of the family; and c) low grade—point average. Significant differences between dropouts and non—dropouts in regard to academic ability, socio—economic status, and grade— point average were partialled out by the method of matched groups prior to examining the major variables of the study. The dropout population of this study consisted of 95 drOpouts who had voluntarily withdrawn from school during Kenneth L. Harding the tenth, eleventh, or twelfth grade for reasons other than transferring to another school or poor health; the stay—in population consisted of 421 students. All 516 students were enrolled in the regular school program in one of three high schools in one Midwestern urban public school system. The major instrument used in this study was the Michigan State General Self—Concept of Ability Scale; only specific sections, however, were employed. Chi—square was used to determine the relative associations among the variables; and in all comparisons, a 2 x 2 contingency table design was used and the null hypotheses were rejected in those instances where a significance level of .05 or beyond was attained. Variables were dichotomized for chi—square analyses by deriving population mean scores for each vari— able. Scores falling below the mean were classified as low scores; those scores falling above the mean were 1 classified as high scores. Significant differences beyond the .001 level of con— fidence were found between dropouts and non—dropouts in regard to academic ability, socio—economic status, and » grade—point average. When matched groups were formed to control for these three variables, dropouts and non— dropouts differed significantly in regard to self—concept of academic ability (SCA), perceived parental expectations (PPEX), and educational expectations, or plans (Ed Pl). 4—_—_’—‘ Kenneth L. Harding No significant differences were found between dropouts and non-dropouts in regard to the student's perception of his parents‘ evaluation of his academic ability (PPEV). It was further determined that significant relationships existed between levels of high and low for SCA and PPEv, and between levels of high and low for PPEX and Ed Pl. The research project, of which this study was a part, was supported by funds granted by the U. S. Office of Education, under the direction of Wilbur B. Brookover. The project was entitled, Self—Concept of Ability and School Achievement, Research Project No. 1636. A COMPARATIVE STUDY OF CAUCASIAN MALE HIGH SCHOOL STUDENTS WHO STAY IN SCHOOL AND THOSE WHO DROP OUT By a Kenneth L& Harding A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Counseling, Personnel Services, and Educational Psychology 1966 ACKNOWLEDGMENTS The writer wishes to gratefully acknowledge the assistance of the following individuals who were instru— mental in the completion of this study: To Dr. James W. Costar, the writer's major advisor and chairman of the guidance committee, for his most generous and conscientious assistance. To the other members of the writer's guidance committee, Dr. Edward B. Blackman, Dr. Walter F. Johnson, and Dr. Bill L. Kell, for their critical examination of the manu— script. To Dr. Wilbur B. Brookover, director of the longi- tudinal study, of which the present study is a part, for providing the writer with data essential to the study, and for making available the services of his staff of research assistants. To Dr. Edsel L. Erickson, Dr. Lee M. Joiner, Corwin Krugh, and Natalie Sproull, for their assistance in the collecting of data, as well as their guidance concerning the design and statistical methodology of the study. And to his wife, Marilyn, for her constant encourage— ment and understanding. ii TABLE OF CONTENTS Chapter ACKNOWLEDGMENTS . . . . . . . . . . . . . LIST OF TABLES. . . . . . . . . . . . LIST OF APPENDICES. . . . . I. II. III. INTRODUCTION TO THE STUDY . . . . . . Introduction. . . . . . Statement of the Problem. . Justification for the Study . . . Relationships to be Explored. . . . . . Definition of Terms . . . . . . . . Limitations and Scope of the Study. Data Collection Procedures. . . . . Data Analysis Procedures. . . . . . Summary . . . . . . . . . . . . . . REVIEW OF THE LITERATURE. . . . . . . Introduction. . . . . . Reasons for Student Dropout . . . . Identifying Characteristics of Dropouts Efforts to Solve the Problem of Dropout Conclusion. . . . . . . . . . . . . . . RESEARCH PROCEDURES OF THE STUDY. . . . Identification of the Population. . The Stay-ins . . . . . . . . . . . . The Dropouts . . . . . . Hypotheses to be Tested Sampling Procedures . Instrumentation . . . . . . . . . . Reliability of the Instrument. Validity of Self—Concept Scales. Other Data Used. . . . . . . . . . Procedures for Treatment of the Data. Dichotomization of Variables Summary . . . . . . . . . . . . . . . Page ii 14 24 36 4O Chapter Page IV. ANALYSIS OF THE DATA. . . . . . . . . . . . . 59 Analysis of the Test Variables. . . . . . . 59 Analysis of Sampling Data . . . . . . . . . 64 Analysis of the Major Variables . . . . . . 66 Analysis of Relationships Between Major Variables. . . . . . . . 72 Addendum to the Study. . . . . . . . . . 79 Summary . . . . . . . . . . . . . . 80 V. SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS . . 82 Summary . . . . . . . . . . . . . 82 Variables Examined . . . . . . . . . . 83 Data Analysis Procedures . . . . . . 85 Summary of Significant Findings. . . . . 86 Conclusions . . . . . . . . . . . . . . . . 89 Recommendations . . . . . . . . . . . . . . 96 Closing Statement . . . . . . . . . . . . . lOO BIBLIOGRAPHY. . . . . . . . . . . . . . . . . . . 102 APPENDICES. . . . . . . . . . . . . . . . . . . . 107 iv Table No. 3.1 LIST OF TABLES Page No. Correlations of General Self—Concept with Specific Subject Self—Concept . . . . 53 Correlations of General Self-Concept of Ability with GPA. . . . . . . . . . . . 54 Mean Scores of Variables and Ranges of Scores for High and Low Categories . . . . 58 Chi—Square Analysis of Association Between Dropouts and Non-Dropouts and Level of Academic Ability. . . . . . . . . 61 Chi—Square Analysis of Association Between Dropouts and Non-Dropouts and Level of Socio—Economic Class . . . . . . . . . . . 62 Chi—Square Analysis of Association Between Dropouts and Non—Dropouts and Level of Grade-Point Average. . . . . . . . . . . . 63 Distribution of Dropouts and Non—Dropouts Scoring Either High (H) or Low (L) in . Regard to IQ, SES, and GPA . . . . . . . . 65 Chi-Square Analysis of Association Between Matched Groups of Dropouts and Non—Dropouts and Level of Self—Concept of Ability (SCA) 67 Chi—Square Analysis of Association Between Matched Groups of Dropouts and NoneDropouts and Level of Perceived Parental Evalua— tions (PPEV) . . . . . . . . . . . . . . . 68 Chi—Square Analysis of Association Between Matched Groups of Dropouts and Non-Dropouts and Level of Perceived Parental Expecta- tions (PPEX) . . . . . . . . . . . . . . . 69 Chi—Square Analysis of Association Between Matched Groups of Dropouts and Non—Dropouts and Level of Educational Plans (Ed Pl) . . 71 Chi— Square Analysis of Association Between Dichotomized Levels of SCA and PPEv When Matched Groups of Dropouts and Non— —Dropouts are Considered Jointly 74 Table No. Page No. 4.10 Chi—Square Analysis of Association Between Dichotomized Levels of SCA and PPEV When Non—Dropouts are Considered Separately . . . . . . . . . . . . . . . . 74 4.11 Chi—Square Analysis of Association Between Dichotomized Levels of SCA and PPEV When Dropouts are Considered Separately . . . . . . . . . . . . . . . . 75 4.12 Chi-Square Analysis of Association Between Dichotomized Levels of Ed P1 and PPEX When Matched Groups of Dropouts and Non—Dropouts are Considered Jointly. . 77 4.13 Chi-Square Analysis of Association Between Dichotomized Levels of Ed P1 and PPEX When Non—Dropouts are Considered Separately . . . . . . . . . . . . . . . . 77 4.14 Chi-Square Analysis of Association Between Dichotomized Levels of Ed P1 and PPEX When Dropouts are Considered Separately . . . . . . . . . . . . . . . . 78 vi Appendix LIST OF APPENDICES Self-Concept of Ability - General. . . . . Perceived Evaluations of Student's Academic Ability by Others (Parents) . . . Perceived Parental Expectations. . . . . Educational Expectations (Plans) . . . . . A Comparison of Eleventh Grade Transfer Students who Dropped Out and Eleventh Grade Non—Transfer Students Who Dropped Out on the Basis of Different Grade— Level Data . . . . . . . . . . . . . . . . A Comparison of Caucasian Male High School Students Who Dropped Out During the Tenth Grade and Those Who Dropped Out During Either the Eleventh or Twelfth Grade . . . Page 108 110 112 113 114 117 Chapter I INTRODUCTION TO THE STUDY Introduction The school dropout is by no means a new phenomenon. In fact, one source states that, "as an institution, it's probably just one day, or several hours, younger than the schools themselves.“1 The incidence of dropout has become so commonplace that, until recently, only minimal attention has been given to its occurrence or its magnitude. The primary reason for a general lack of concern in the past is the fact that students who dropped out of school were able to find employment as unskilled or semi—skilled laborers. Also, there was a considerable demand for per— sons having but minimal skills or formal education. This was especially true during World War II and the post-war period that followed. Prior to 1950, high schools were primarily concerned with preparing those students who had the necessary talents and abilities to continue their education in the nation's institutions of higher education. Consequently, those lDaniel Schreiber, editor, Guidance and the School Dropout, (Washington: National Education Association and American Personnel and Guidance Association, 1964), p. l. students who were unable (or unwilling) to compete with their more academically competent classmates were given but token assistance —— if that; and instead of being encouraged to remain in school, they were often "advised" to withdraw, or were discouraged to the point where they were unable to further tolerate the academic experience. Many educators might take issue with the preceding statement by citing the number of schools that have also been concerned with the student who is not planning on going to college. This could be substantiated by figures showing that many schools have been offering industrial or vocational education courses since the 1920's, as well as courses in home economics, clerical skills, and dis- tributive education. Despite the presence of many excellent courses spe— cifically designed for the non—college—bound student, he was required to compete with his academically—oriented Classmates in the basic high school program. For instance, he often had to complete three units of English, two units of social studies, and at least one unit in both math and science. Students having deficiencies in reading and language skills were unable to compete at a satisfactory level in these basic subjects; therefore, formal education became, for many students, a frustrating, degrading, unbear— able experience —— regardless of the "special programs" made available to them. In this sense, many schools are still basically geared to the academically—oriented college- bound student; other schools, however, have personalized curricular offerings in an attempt to meet the needs of all their students. Statement of the Problem It is the purpose of this study to determine the ways in which Caucasian male high school dropouts differ sig— nificantly from those students who remain in school in regard to selected characteristics. The main variables to be examined are: l. The Caucasian male high school student's general self-concept of academic ability (hereinafter referred to as self—concept of ability). 2. The Caucasian male high school student's perception of his parents' evaluation of his academic ability. 3. The Caucasian male high school student's perception of how long his parents expect him to remain in school. 4. The Caucasian male high school student's expressed educational plans; i.e., the level of education he expects to attain. In addition, the study will examine three other fac— tors which have been identified in numerous studies as contributing significantly to the decision of male high school students to leave school prior to graduation: low academic ability, low socio—economic status of the family, and low grade—point average. It is assumed that these factors are not operating in isolation from other factors and, consequently, do not give sufficient information to adequately identify the potential dropout. It is antici— pated that the present study will more clearly define the causes for students dropping out of school by comparing the relationships between male dropouts with non—dropouts in regard to the selected variables previously mentioned. Justification for the Study The dropout is no longer referred to as a phenomenon but as a national problem. Although the holding power of the high schools has risen from 30 percent to 65 percent,2 the problem has become increasingly more critical as a result of the drastic changes that have occurred in the labor market during the past ten years. Few jobs remain for persons with less than a high school education, and even more important is the outlook for the future. 2Ibid. , p. 2. Recent curricular innovations have been directed toward providing for individual differences regarding stu— dent interests and abilities, and thereby, have increased the holding power of the schools; but in many schools, attempts to identify potential dropouts and provide them with special guidance or academic assistance have been seriously lacking. With increasing demands for higher level technical and academic competencies there also arises the need to increase the percentage of students who suc- cessfully complete not only a high school education but additional training as well. It seems logical that the best place to start is by identifying those individuals who are most likely to drop out and then providing them with special counseling and assistance while they are still amenable to these services. Relationships to be Explored In a longitudinal study by Brookover and associates,3 of which the present study is a part, significant relation- ships were found between academic ability, grade—point average, and, to some extent, socio—economic level of the family. No attempt was made, however, to separate dropouts 3Wilbur B. Brookover, pp_§l;, “Self-Concept of Ability and School Achievement, II,“ U. 8. Office of Education, COOperative Research Project No. 1636, (East Lansing: Bureau of Educational Research Services, Michigan State University, 1965), p. 51. from non—dropouts in the analysis. A study by Fink4 did not examine academic ability and socio—economic factors on the assumption that such an examination would be repe— titious, it appears to be a necessary first step in the present study in determining the extent to which subsequent variables being tested are, in fact, significantly related. The following relationships are anticipated: 1. Dropouts and non—dropouts will differ sig— nificantly in regard to level of academic ability, as measured by a group intelligence test administered during the ninth grade. 2. There will be a significant difference between dropouts and non-dropouts in regard to the socioeeconomic level of the family. 3. There will be a significant difference between dropouts and non—dropouts in regard to grade—point average of four basic school subjects: English, social studies, math, and science. 4. When matched groups are formed to control for differences in levels of academic 4Donald D. Fink, "The Efficiency of Certain Criteria in Predicting School Dropout,“ (unpublished doctoral dissertation, Michigan State University, East Lansing, 1962). ability, socio—economic status, and grade— point average, dropouts and non—dropouts will differ significantly in regard to self- concept of academic ability. 5. Matched groups of dropouts and non—dropouts will differ significantly in regard to per— ceived parental evaluations of their ability to achieve in school. 6. Matched groups of dropouts and non-dropouts will differ significantly in regard to the amount of schooling they perceive their parents expect them to complete. 7. Matched groups of dropouts and non—dropouts will differ significantly in regard to level of educational expectations (or plans). 8. Self—concept of academic ability and per— ceived parental evaluations will be signifi— cantly related whether dropouts and non— dropouts are considered jointly or separately. 9. Educational expectations and perceived paren- tal expectations will be significantly related whether dropouts and non—dropouts are considered jointly or separately. Definition of Termg Although many of the terms to be described are gen— erally self—explanatory in nature, the following descrip— tions are considered essential to further clarify their specific application to the present study: 1. any student who interrupts his Dropout: high school education before graduation by voluntarily removing himself from school for reasons other than health or transfer to another school. Whether or not he con- tinues his education at a later date will not be considered a factor; he will still be classified as a dropout. scores on the California Test of I.Q.: Mental Maturity (CTMM), a group intelligence test, administered during the ninth grade. I.Q. and academic ability will hereinafter be considered as equivalent terms. Socio-Economic Class: a measure of the father's occupational level. Grade-Point Average: an average of the stu— dent's academic performance in four subjects: English, social studies, math, and science. 5. Self—Concept of Ability: the evaluation a person makes of himself in regard to the ability to achieve in academic tasks in general, especially as compared with others. 6. Perceived Parental Evaluations: a student's perception of his parents' evaluation of his ability to achieve in school in relation to other students. 7. Educational Expectations: the level in the educational system which the student expects to attain (not the level which he would like to attain). 8. Perceived Parental Expectations: a student's perception of his parents' expectations in regard to the amount of schooling he will acquire. Limitations and Scope of the Study As stated previously, the present study is part of a longitudinal study of a specific group of Caucasian male students enrolled in one Midwestern urban city school system. Although data is available for this group from seventh through twelfth grades, only tenth, eleventh, and twelfth grade dropouts will be compared with non-dropouts. -10- There are two main reasons for selecting dropouts in these three grades. First of all, the over—aged student will be eliminated from the study; i.e., if the over-aged student were predisposed to dropping out of school, he could have done so prior to the tenth grade. Secondly, most students become sixteen years of age —- the legal age for leaving school —— during either the tenth or the eleventh grade. Since this study is not concerned with the factor of over- ageness, all ninth grade dropouts have been excluded. The study will also be limited to Caucasian males in an attempt to further reduce the number of intervening variables that might contaminate the results of other relationships being measured. All special education students, or students not enrolled in the regular tenth, eleventh, or twelfth grade programs, will be excluded. Also, any student who has been selected for special treatment effects as part of the longitudinal study will not be included. Special treatment groups were quite small in number, however, and should not lower the total population to any appreciable degree. Only voluntary dropouts, as previously defined, will be used in this study. If a student has transferred to another city, has withdrawn because of illness, or has been suSpended or expelled, he will not be included. Stu— dents dropping out for financial reasons will not be excluded since recent surveys indicate that students seldom —11— leave school because they are financially unable to com— plete their high school education. It is realized that the population selected for this study may not be generalizable to other states, particularly those states with a markedly different socio—economic or sub-cultural milieu. For this study to be of maximum value, it would be necessary for replicative studies to be made in other states, as well as additional studies focussing on females and members of other races in another population or in the population presently being examined. Data Collection Procedures The following is a chronology of procedures used in collecting the data relevant to the present study: . . 5 . . . l. Questionnaires were administered in the fall of 1963 to all sophomores enrolled in one of three different high schools within (This was one phase 6) one Midwestern city. of a longitudinal study. 2. Questionnaires were administered to these 5The term "Questionnaire" refers to what is now called the Michigan State General Self—Concept of Ability Scale. 6Brookover, pp al., loc. cit. -12- same students during their junior year in the fall of 1964 and during their senior year in the fall of 1965. 3. The Bureau of Educational Research Services at Michigan State University tabulated and coded the responses on IBM cards along with data concerning race, sex, I.Q., grade— point average, and socio—economic class. 4. The writer recently examined the records of the Office of Child Accounting to determine whether the dropouts are, in fact, dropouts, and whether these dropouts meet all criteria as previously specified for this study. Data Analysis Procedures Dropouts and non—drOpouts will be compared in regard to the following variables: level of academic ability (I.Q.); grade—point average; and socio—economic level of the family. If a significant difference is determined between dropouts and non—drOpouts on any of the aforemen— tioned variables, matched groups will be formed to control for these differences. Matched groups will then be compared to determine whether dropouts and non—dropouts differ in regard to self—concept of academic ability, perceived _l3_ parental evaluations of the student's academic ability, educational expectations (or plans), and perceived parental expectations. A chi—square analysis with a significance level of .05 will be employed in all cases. The relatively large size of the population being studied justifies the use of this particular level of significance. Summary It has been the intent of Chapter I to delineate the problem being studied, the hypotheses it will test, and the limitations, scope, and procedures that will determine its direction and focus. Chapter II will present a review of the related research; the instrumentation and research procedures will be outlined in Chapter III. A presentation and analysis of the data will be conducted in Chapter IV, followed by a summary and recommendations for further study in Chapter V. Chapter II REVIEW OF RELATED RESEARCH Introduction Almost any recent issue of any educational journal will contain at least one article concerning what schools should do, or, are doing, to make their curricula more meaningful for all students. It may have an article on underachievers and overachievers describing the many fac— tors influencing these differences in academic motivation. Numerous articles describe what a particular state or city is doing to provide maximum learning experiences for its gifted; or, how certain funds provided by government or industry have made it possible to organize such programs as the work—study program for those students who would not otherwise be content to remain in school. There is often an article describing the increasing demand for the educated person or an article describing what a given community is doing to rehabilitate its former dropouts through the use of funds provided by federal legislation, such as the Manpower Development and Training Act of 1962. , All the aforementioned topics are related in varying degrees to the problem of dropout and add considerably to the professional educator's understanding of its enormity. -14- -15- There is also a considerable amount of information and data specifically concerned with the plight of the dropout at both the high school and college levels; and yet, alarming numbers of students continue to leave school prior to graduation. It is also possible for parents and students to be informed through the media of radio, television, newspapers, and nationally circulated magazines. In addition, many excellent books have been written which vividly portray the myriad of problems facing the schools and youth of this decade. Such nationally renown authors as Friedenbergl, Goodmanz, and Conant3 have caused considerable attention to be focussed upon these problems of youth, especially the disadvantaged youth, through their examinations of societal pressures, conflicts, and role expectations encountered by the adolescent. Empirical research on the dropout is also plentiful, but many studies reviewed seem to draw conclusions based on factors that have not been carefully defined or controlled. lIt is believed that despite the abundance of existing information, as previously mentioned, additional research lEdgar Z. Friedenberg, The Vanishing Adolescent, (New York: Dell Publishing Company, 1959). 2Paul Goodman, Growing Up Absurd, (New York: Random House, 1956). _ 3James B. Conant, Slums and Suburbs, (New York: McGraw— Hill Book Company, 1961). —l6— is still needed to more clearly define the problems and characteristics of the high school dropout. Reasons for Studgnt Dropout The student who voluntarily withdraws from school apparently does so for a number of reasons, and the com- binations of reasons vary with each case. Davis4 states that there seems to be a fairly general consensus regarding the common causes of students dropping out of school: a lack of self—esteem and a feeling of not being wanted. A study by Brookover and associates5 substantiates Davis“ statement by determining that one's self—concept is modi- fiable through the images and expectations he perceives others hold for him; and also, through increased "self~ esteem," his academic achievement level will be raised. This also implies that a loss of self-esteem, or feeling of not being wanted, would have a negative effect on achievement. Lambert6 states this concisely: "The problem 4Donald A. Davis, "An Experimental Study of Potential Dropouts," The Personnel and Guidance Journal, XL, (May, 1962), pp. 799—802. 5Wilbur B. Brookover, Ann Patterson, and Shailer Thomas, "Self-Concept of Ability and School Achievement,“ U. S. Office of Education, Cooperative Rgsearch Project No. 845 (East Lansing: Office of Research and Publications, Michigan State University, 1962), p. 55. 6Nadine Lambert, "The High School Dropout in Elementary School," Guidance and the School Dropout, op. cit., p. 62. —l7— I. presented by the school dropout is one of alienation -— alienation from peers, school and society." A study by Davis7 attempted to determine the effects of special attention given potential dropouts. He found that when this attention was employed systematically, students tend to drop out less often, get higher marks, and are referred less often to the office for disciplinary reasons. Sando8 also found that the basic need of students who had dropped out was to feel that someone in the school really cared for them and genuinely wanted to help them. Davis9 further states that recent studies indicate the major causes for dropouts to be school—centered, that dropouts stem from inadequate curriculums and unsatisfactory student—teacher relationships. Specifically listed are: a history of school failure; dislike for school subjects; unsatisfactory student-teacher relationships; a feeling of not belonging; non-participation in school activities; and parents with low socio—economic status (not directly a responsibility of the school). A study by Thompson and Nelson10 indicates that 70 percent of the mothers and ) 7Davis, loc. cit. 8Rudolph F. Sando, ”A Comparative Study of Early School Leavers," (Unpublished doctoral dissertation, University of California, Berkeley, 1952), cited by Davis, ibid., p. 799. 9Davis, loc. Cit. 10Michael J. Thompson and Robert H. Nelson, "Twelve Approaches to Remedy the Dropout Problem," The Clearing @132, XXXVIII, (December, 1963), pp. 200-204. —l8- ’, 80 percent of the fathers of dropouts have not completed high school. It appears that parents may have had unplea— sant educational experiences and dropped out before pre— paring themselves for an adequately—paying job. Therefore, the school may be indirectly reSponsible for the socio- economic status of the parent. Conantll indicates that the dropout rate tends to increase to 70 percent in slum areas, whereas the national 12 average is 35 percent. Schreiber further cites that 25 percent of the mothers and 30 percent of the fathers have not completed the sixth grade. He cites two recent studies, one in rural Louisiana and another in New York State, which found that "two—thirds of the parents of dropouts held negative or indifferent attitudes toward the value of education.“13 A study of dropouts in Maryland14 uncovered the fol— lowing data: . . . Only about one-tenth were following the academic course; the great majority were enrolled in the so—called "general course;" and the remainder were scattered in the commercial and vocational courses. One—tenth, at the time they left school, were reading below the third—grade level-— llConant, loc. cit., cited by Thompson and Nelson, loc. cit. lechreiber, op. cit., p. 5. 13Loc. cit. l4Loc. cit. -19- they might as well have been illiterate-— and a full 45 percent were reading at the sixth—grade level or lower. Half of them had been classified in the category of "below average" mental ability —— and most of them, in one way or another, prob— ably knew it . . . O'Neillls believes it to be "pretty well established" that the following are specific factors leading to school dropout: low socio—economic status of the pupil's family; low intelligence, but especially, his lack of motivation and consequent poor performance that results in constant failure; emotional instability and personality defects; an excessive number of transfers from school to school; and an inability to adjust to instructional procedures or to find personal fulfillment in the subjects being offered. He states further that it is possible, in individual cases, for none of these aforementioned factors to apply. In these cases one or more of the following reasons usually apply: lack of participation in school activities; lack of interest in subjects; retardation with concomitant ridicule and non— acceptance; unsympathetic attitudes of parents toward educa— tion, or broken homes.16 The fact that ability is not always the deciding factor 15John H. O'Neill, ”High School Dropouts," Education, LXXXIV, (November, 1963), pp. 156-159. 16There appears to be considerable overlap between these two groups, and the writer questions just how Ewell estab- lished" these factors are if some dropouts manifest none of the factors listed. -20- 17 is emphasized by Lee. She believes the theory of "sur- vival of the fittest“ to be inapplicable because some dropouts who are fit are eliminated while others less fit receive a high school diploma. Dipasqualel8 feels that the structure of the graded school is one of the main reasons for students dropping out of school. He states that grade repetition is one of the results of such a structure. Also, gradedness necessi— tates special corrective programs which are costly to operate, always stigmatic, and quite often traumatic to both the Child and the parents. He further states: Grade failure was conceived originally to give the child a second year to learn the essentials prerequisite to the next grade. However, it does not always work. Often the repeater does no better than he did the year before. Research indicates that he might have done as well, if not better, had he been permitted to go into the next grade. 19 Chamberlin and Catterall20 engaged in a study in l7Beatrice Crump Lee, "School Dropouts,“ NEA Research Division Bulletin, April, 1963, cited by Vincent C. Dipasquale, "The Relation Between Dropouts and the Graded School," Phi Delta Kappan, XLVI, (November, 1964), pp. 129—133. l8Dipasquale, ibid., p. 129. 19Loc. cit., citing Virgil E. Herrick, "Elementary Education Programs," EncyclOpedia of Educational Research, Chester W. Harris, editor (New York: The Macmillan Company, 1960), p. 438. 20Gordon L. Chamberlin and Calvin D. Catterall,‘ "Acceleration for the Overage Potential Dropout?” Phi Delta Kappan, XLV, (November, 1963), pp. 98—100. -21- California, called "Core 87," and their results seem to substantiate the findings in the aforementioned study. Thirty—four overaged youth were placed in a special class which covered the seventh and eighth grades in one year. Nine of these persons, however, did not remain in the 21 special class for the entire year. From this group of twenty—five who completed the one-year program, only two dropped out before their junior year in high school. Both of these students were girls. The reasons for dropout, as determined by the compos— ite results of the 1963 dropout campaign22 are as follows:23 1. Disinterest in school, compounded from such matters as discouragement because of failing grades; other adverse school experiences, including rejection by fellow students or the school staff; and dissatisfaction with the school pro— gram, particularly because of the spe— cific vocational training desired by the student. 21Five were returned to their regular classes for behavioral reasons, two moved away, and two asked to be returned to their regular classes —— a total of 26 percent. 22The dropout campaign in 1963 was undertaken, at the request of the late President Kennedy, as a national effort to combat school dropout. Sixty—three communities in twenty— three states and the District of Columbia participated. 23U. S. Department of Health, Education and Welfare, "A New Look at School Dropouts," Hpalth, Education and Welfare Indicators, (April, 1964), cited by Maurice Voland, editor, Conference Summary — Fact Book. 1964 Conference on Rural Youth (East Lansing: Cooperative Extension Service, Michigan State University, 1964), pp. 22-23. -22- 2. Home responsibilities, including a real need to work to help support the family or to stay at home and care for younger brothers and sisters, or ailing parents. 3. Unfavorable pargppal attitudp, including lack of feeling for the importance of school, general lack of concern for the child, inability to cope with or influence the child's behavior, desire for financial assistance from the child or to train child in father's business, and support of child's desire to leave school and be “independent.“ Parents of most dropouts were dropouts themselves; however, contrary to popular belief, many parents of poten— tial or actual dropouts are in favor of their children remaining in, or returning to school. 4. Peer influence, social pressure, some— times translated by the student into economic pressure —— the compulsion to earn money in order to buy cars, to dress as well as classmates, have equivalent Spending money and the like. In some cases, peer influence had resulted in general delinquent behavior and trouble with school authorities. 5. Unhappy home situations coupled with the desire to be independent of parental ties led many youngsters to take jobs, enter the military service, or get married rather than continue to live at home and attend school. The lack of finances as a reason for drOpping out was 2 one of many factors studied by Cervantes. Interviews with disadvantaged youths indicated that leaving school was rarely precipitated by financial conditions. He states 4Lucius F. Cervantes, The Dropout, Causes and Cures (Ann Arbor: The University of Michigan Press, 1965), p. 96. -23- that "poverty is the condition of dropout, but inability to pay for his education is not the reason why the dropout 25 withdraws." Hollingshead's study26 found that all of the youths of high school age from upperclass parentage were in school, whereas 75 percent from the lowest social class had dropped out prior to becoming sixteen years of age. Davie27 reported only 1.8 percent of sixteen and seventeen—year-olds from upperclass families not attending school, but 42.6 percent not attending from the lowest class. The attitudes of these lowest social class families toward education, and the conditions contributing to these attitudes, are described by Davie as follows:28 . . . The parents of the children had generally received less than a high school education. Both parents worked in semi- skilled and unskilled capacities to earn enough to pay the rent for their cold— water flat and to keep the family fed. Sickness and industrial accidents often necessitated the children leaving school to help out. When the children weren't ”needed” at home, they often left school 25Loc. cit. 26August B. Hollingshead, Elmtown's Youth (New York: John Wiley and Sons, 1949), pp. 329—359, cited by Cervantes, ibid., p. 97. 27James S. Davie, "Social Class Factors and School Attendance,” Harvard Educational Review, (Summer, 1953), cited by Charles M. Allen, Combating the Dropout Problem, (Chicago: Science Research Associates, Inc., 1956), pp. 10-12. 28 Ibid., p. 11. —_._..___ -24- anyhow in quest of a job and “some spend— ing money in the jeans" which the parents could not provide. The one positive influence keeping them in school was the knowledge that more and more employers were asking for high school diplomas. However, the attraction of earning one's own money was often too great. Matthews and Bowman29 found that only 5 percent of the dropouts had to leave school for financial reasons. They state, however, that “dropping out of school is essentially a function of the lower class."30 Identifying Characteristics of Dropouts According to Stebbens3l, many educators claim an ability to identify the potential dropout in kindergarten by means of assessing his background, environment and attitude; and with each successive year, he or she becomes easier to identify. He further states that although children react in different ways, the following characteristics (all or in part) describe the junior and senior high school 29C. V. Matthews and Peter Bowman, The Motivations of Youth for Leaving School, Quincy Youth Development Project, Quincy, Illinois (Washington: Office of Education, U. S. Department of Health, Education, and Welfare, 1960), p. 23, cited by Cervantes, pp; cit., p. 99. 30Loc. cit. 31Marion Stebbens, “Flint Offers the Potential Dropout a Personalized Curriculum," The Clearing House, XXXVIII, (December, 1963), pp. 205-209. -25.. potential dropout:32 1. Negative or hostile attitude toward teachers and school 2. Severe retardation in reading, i.e., below 5th grade level 3. Irregular attendance pattern 4. Poor work habits and study skills 5. Indifference and lack of ambition 6. Disruptive classroom behavior 7. Withdrawal tendencies 8. Preoccupation with non-school activities 9. Unacceptable personal habits lO. Deficiency in basic citizenship skills 33,34 In 1954, Herrman administered a questionnaire designed by Cottle35 to a group of 1,834 junior high school students in a large Kansas city. Three semesters later, responses between dropouts and stay—ins were compared. A randomized sample of the stay—ins was used for comparison purposes. Later, those items found common to the Herrman 321bid., p. 206. 33W. L. Herrman, "Partial Validation of a Dropout Scale,“ (unpublished Master's report, University of Kansas, 1957), pp. 1—16, cited by Margaret W. Epps and William C. Cottle, "Further Validation of a Dropout Scale,“ Vocational Guidance Quarterly, VII, (Winter, 1958-59), pp. 90—92. 34W. L. Herrman and William C. Cottle, "An Inventory to Identify High School Dropouts," Vocational Guidance Quarterly, VI, (Spring, 1958), pp. 122—123, cited by Epps and Cottle, loc. cit. 35William C. Cottle, Life Adjustment Scale, No. 1, University of Kansas, 1953, cited by Epps and Cottle, loc.cit. —26— study and to a study by Epps and Cottle36 were grouped in four sections: those concerned with the home, school, economic stress, and the students' personal feelings "toward things in general.“ These studies found highly significant differences between the early dropouts and the stay—ins. No attempt was made to study those dropping out during the later years of high school. It was sug— gested that a cut—off score could be established to determine the potential dropout. Lambert37 made a follow-up study on high school students who had been clinically diagnosed by Bower38 as being emo— tionally disturbed when they were enrolled in fourth, fifth, and sixth grades. At this time, children were grouped into three categories: Group I. Those diagnosed as being emotionally disturbed. Group II. The “unidentified" poorly adjusted pupil. These individuals were rated by their teachers as being “one of the two most maladjusted in my class." 36Epps and Cottle, loc. cit. 37Lambert, op. cit., pp. 40—65. 38Eli M. Bower, A Process for Early Identification of Emotionally Disturbed Children. (Sacramento: California SEate Department of Education, 1958), Cited by Lambert, loc. cit. _27_ Group III. The "average or better adjusted" pupil (control group). Lambert compared these groups five years later and found insignificant differences between high school students pre- viously categorized as Group I and Group II.39 Group I, however, showed less scholastic potential and achievement. Comparisons between Group I and Group III showed Group I to (a) require more coaxing (or force) to work with other pupils, (b) have more difficulty learning, (c) be more fre- quently depressed, (d) have lower intellectual potential, and (e) have lower than average emotional adjustment as observed by the classroom teacher. In comparing Groups II and III, the aforementioned differences were also found, except that no difference was observed regarding intellectual potential (d). Teachers classified persons in Group II, however, as making more frequent responses in class that were either unusual or inappropriate. On the basis of this study, Lambert concludes that the school dropout does not exhibit a single behavior pattern, but instead, manifests one or more of the follow— ing characteristics: low socio—economic status, poor scholastic aptitude, limited school achievement, emotional 39Only the results for boys are being reported; however, Lambert gives the results for both boys and girls. -28— problems in elementary school, difficulties with peers and parents, and resistance to help from existing guidance facilities. Some of the students classified in Group III did drop out. The reasons appeared to be the result of low achievement, low I.Q., or low socio—economic status.40 A study by Fink4l found I.Q., level of occupational aspiration, and age to be important factors distinguishing dropouts from non—dropouts, but did not find a significant difference in socio-economic class. He states that being overage, due to repeating grades, and/or being a member of a family having low socio—economic status causes a pupil to be "highly dropout prone;" also, that overage—ness . 2 “ascribes a less stable or predictable role for I.Q. . . ."4 L . . 43 A review of numerous writings was interpreted by Sando as indicating that only repetition of grades was consist— ently and positively related to early school leaving. In contrast with the aforementioned study, no consistency was revealed in regard to the factor of I.Q. 40These are the three "typical" (or test) variables to be examined in the present study. 41Fink, op. cit., p. 112. 421bid., p. 113. 43 19. Sando, Op. cit., cited by Fink, ibid., p. -29_ In 1960, Bertrand and Smith44 studied two rural Louisiana school districts and found that 26 percent of the dropouts had repeated two grades, came from families with low socio-economic status, but were not consistently lower than stay—ins in regard to level of I.Q. The capriciousness of I.Q. scores as a reliable pre— dictor of potential dropouts is reflected by Matthews and Bowman in the following statement:45 On the average the dropout has the intel— lectual potential to successfully finish high school, but the intellectual ability of dropouts as a group is distinctly below average when measured by group intelli- gence tests. Surely large numbers of children with no greater intellectual potential than the average dropout finish high school. The following list of dropout characteristics are based on several studies reviewed by Matthews and Bowman:46 Dropouts were more often boys. Dropouts were more often from the loWer class. Dropouts were more often from minority groups. Dropouts came more often from broken homes, but the percentage difference was not pronounced. Dropouts' parents had little education. 44Alvin I. Bertrand and Marion B. Smith, Environmental Factors and School Attendance. (Baton Rouge: Louisiana State University, Bulletin No. 533, May, 1960), cited by Fink, pp; cit., p. 25. 45Matthews and Bowman, op. cit., cited by Fink, op. cit., p. 25. 46Fink, ibid., p. 23, citing Matthews and Bowman. _30_ Dropouts had below—average intellectual potential as measured by I.Q. tests. Dropouts were usually retarded in reading and other skill areas. Dropouts were usually overage for their grade as a result of being retained in one or more grades. Dropouts (particularly boys) usually failed several courses in the year prior to leaving school. Dropouts had frequently moved from school to school. Dropouts showed a marked regression in attendance in junior and senior high. Dropouts usually were not active in the extra— curricular 1ife of the school. Dropouts usually had enough money to con- tinue in school. Although lack of money was not usually a major factor in early school leaving, indirectly it had an influence. Dropouts often had difficulty in getting and keeping a job. Dropouts came from families in which parents less strongly support the schools' and their children's interest in education. Dropouts placed little value on education as a help to themselves, and consequently had little further interest in school even though the adjustment to adult life had not been easy. Dropouts felt that their education should have been more practical; it should have prepared them for a vocation. Dropouts felt insecure and lacked a feeling of belonging in school. Dropouts felt poorly treated by their teachers, and were fearful of asking for help. Dropouts were often dissatisfied with their social relationships in school, and they lacked friends. Another source states that studies of dropouts have shown a fairly consistent pattern of identifiable charac— teristics. -31- These are paraphrased as follows:47 1. Most dropouts are under—achievers, have a record of failure in school, and are usually behind grade level in reading ability. Many have been retained in grade at least once in elementary school, and at the time of dropping out, are at least one grade behind their age level in class placement. 2. Irregular attendance and frequent tardiness are characteristic of the potential dropout. Many have changed schools frequently; and most show a marked disinterest in school and are failing in one or more subjects at the time they drop out. 3. Most potential dropouts do not participate in any extracurricular school functions. All activities are cen— tered outside the school. 4. A sizeable proportion of dropouts have been discipline problems at one time or another, although suSpension from school may not have resulted. Many have had difficulty with community agencies and the law. 5. Dropouts are usually from low—income groups and the education of their parents is usually below the eighth grade M 47 u U. s. Department of Health, Education' and Welfare' A New Look at School DrOpOUtS:" Health. Education, and Welfare Indicators, c1t. (April, 1964), cited by Voland, op. I p. 24, -32- level. 6. Many drOpouts have unhappy home situations. Their parents frequently have a negative attitude toward the school and attach little importance to high school graduation for their children. 7. Dropouts frequently have certain characteristics or situations which separate them from their successful classmates, such as: lack of funds for the normal expendi— tures of adolescent life; marked differences in physique, personality development, dress, social class, or interests. These physical or emotional handicaps often retard their school performance or acceptance by their peers. 8. Potential dropouts are often purposeless and have no clearly defined goals. They fail to see the long— range value of schooling and settle for short—range satis— factions. Cervantes, on the basis of his recent study, lists the following characteristics common to youths who are potential or actual dropouts:48 School 1. Two years behind in reading or arith— metic at seventh grade level. of grades are below average. 2. Failure of one or more school years (lSt, 2nd, 8th, 9th grades most commonly failed; 85% of dropouts behind one year; Majority .-________~___ Cervantes, op. cit., pp. 198—199. Family 9. 10. ll. 12. l3. l4. Peers 15. 16. 17. -33- 53% two or more years). Irregular attendance and frequent tar— diness. Ill—defined sickness given as reason. Performance consistently below potential. No participation in extracurricular activities. Frequent change of schools. Behavior problems requiring disciplinary measures. Feeling of "not belonging“ (because of size, speech, personality development, nationality, social class, family dis- grace, retardation in school, dress, lack of friends among schoolmates or staff, etc.). More children than parents can readily control (e.g., only child for divorced and working mother; five or more for non-divorced and working mother of blue and lower white—collar class). Parents inconsistent in affection and discipline. Unhappy family situation ance, communication, experiences lacking; minimal). Father figure weak or absent. Education of parents at eighth grade level. Few family friends; among these few many problem units (divorced, deserted, delin- quents, dropouts). (common accept— and pleasurable family solidarity Friends not approved by parents. Friends not school oriented. Friends much older or much younger. TAT (Psychological orientation) school, 18. Resentful of all authority (home, police, job, church). 19. Deferred gratification pattern weak. 20. Weak self—image. -34- A study by Larson49 indicated that the most pronounced differences between high school dropouts and graduates were in their family relationships. This was determined by responses to items on the Minnesota Counseling Inven- tory (MCI) in categories dealing with the family, and is reported by Larson as follows: ... Male and female dropouts more often described their homes as unpleasant, and lacking in understanding, love and affection. Friction frequently existed between the dropouts and their parents and other family members. Lines of communication had often broken down. The dropout appeared to resent his parents. From the study one gets the feeling that a strong under-current of dissatisfaction existed relative to home and family relationships. 50 Larson also found that categories dealing with social life and social affairs appeared to be particularly descrip— tive of girl dropouts. He states, however, that both boys and girls “experienced considerable tension in group situations."51 Responses to items referring specifically to school and classroom situations conveyed a feeling of uneasiness and "not fitting in.“52 He further states: M..— 49 . Roland 3- Larson, “A Study of Selected Personality FaCtOrS Associated with High School Dropout,“ (unpublished dOCthal dissertation, Michigan State University, East Lansxng. 1964), p. 108. 50Loc. cit. 51Loc. cit. 52 Loc. cit. -35- Boy and girl dropouts appeared to lack self—confidence in many life situations. They seemed concerned about their lack of ability to concentrate. Feelings of guilt were more frequently present in dropouts than in graduates. There were strong feelings of hostility and mistrust, with more frequent feelings of wanting to violate society's codes through aggressive acts such as breaking things or hurting someone. 53 Although a general consensus obviously exists regarding the identifying characteristics of the dropout, or potential dropout, there have also been several misconceptions or “long-held myths" concerning his identity. These are pre— sented as follows:54 1. That dropouts are usually delinquents. Actually, the large majority of drop— outs, 79% of the group studied, had not been considered serious behavior problems by their counselors or prin- cipals and 76% had never been suspended from school. That dropouts are usually homeless, or the product of broken homes. Actually, 91% of the pupils studied lived with their parents (or with parent and step-parent) and 72% were with both parents living together. That lack of intelligence is a dominant cause of school dropout. Actually, approximately half (49.8%) of the drop— outs studied had average or above average intelligence. ~—-——-.________ 31bid., p. 109. 54 U. S. Department of Health, Education, and Welfare, ADC. cit. ~36— Warner55 further diSproves the myth of I.Q. as a factor determining the incidence of dropout by citing the combined results of numerous studies involving a total of 21,497 dropouts. Approximately 19 percent of these drop— outs had I.Q. scores less than 80; 20 percent had scores betWeen 80 and 89. Dropouts scoring in the average range of 90 to 109 totaled 50 percent; 11 percent scored 110 or above. Warner's data Would indicate that 61 percent of the aforementioned group of dropouts had average or above intelligence and had the ability to complete additional training, either post—high school vocational—technical programs or a college degree. Dropouts scoring above 110, according to Warner, are potentially capable of earning a baccalaureate degree. Efforts to Solve the Problgm of Dropout In July, 1963, President Kennedy sent letters to school officials throughout the nation in an attempt to reduce the number of students leaving high school prior to graduation. Also, he wished to encourage, through local 550. Ray Warner, "The Scholastic Ability of School Dropouts," Selected Rpports and Statistics on School Drop— outs, U. S. Department of Health, Education, and Welfare, Office of Education OE20063 (Washington: United States Government Printing Office, 1964), pp. 11—13. -37- school officials, those students who had previously dropped out to return to school in September. Progress reports of community efforts, or projects, were to be submitted to the U. S. Commissioner of Education by the end of September. An article by Bayley56 describes many of these community projects.57 New York City's "STEP" (School to Employment Program) provides special guidance, a Special curriculum, and a part—time job to sophomores who are considered potential dropouts. They receive supervision and school credit in addition to being paid for their work. "The Co—op Program" provides work experience with municipal agencies for youths from ”difficult environments.“ “Operation Return“ is a special program designed Specifically for dropouts who would be unable to finish a regular high school program before their twenty-first birthday. "Higher Horizons Programs" provide special guidance services, cultural enrichment for children, and education for parents in low socio- economic neighborhoods. It also provides remedial classes and after—school study programs. As a result of the Higher Horizons Program, 40 percent more pupils than previously 56Monica Bayley, "A Renewed Effort to Solve the Problem of Dropout," Selected Reports and Statistics on School Dropouts, ibid., pp. 2—7. . 57Several projects reported to the commissioner were initiated prior to receiving President Kennedy's letter. \ ’. \ ~38— are finishing high school in New York. Columbus, Ohio has attempted to combat the dropout problem through broad curriculum improvements. They have already made gratifying progress; the dropout rate in 1950 ‘5” was 50 percent, but now is 33 percent. Reasons cited for this reduction in the dropout rate are: a more diversified instructional program; smaller classes; and an expanded staff of counselors, visiting teachers, and psychologists. Chicago held special orientation days for pupils at the elementary and first—year high school levels. They also employed attendance officers for an additional week prior to the opening of school to encourage potential dropouts to return. Washington, D.C., has attempted to reduce the number of dropouts for many years through programs designed to meet the needs of individual students. During the summer of 1963, a campaign was conducted in which all high school officers visited the homes of students, Worked with families and community agencies in an attempt to persuade the drop— out to return to school in the fall. An American Legion Post in Atlanta provided $3,000 for scholarships for the stay—in—school campaign. In Boston, 1,600 letters were sent to the previous year‘s high school dropouts, inviting them to attend Special guidance clinics. The dropouts were offered a new oppor— tunity to return to school, to receive training for a better -39- job, or to obtain part—time or full—time employment. In Dunkirk, New York, teachers have been Specially assigned to help Puerto Rican students overcome their educational problems. All teachers selected were able to speak Spanish. In Miami, educators, local government officials, and other community leaders met to organize a county-wide campaign against the occurrence of dropout. Several mea- sures were prOposed: changing the curriculum to meet the needs of the potential dropout; expanding the evening school program; improving reading instruction in the lower grades; establishing free nursery schools and kindergartens in key areas; acquiring the services of good students to tutor the less capable students; and assigning more visiting teachers. Many other communities, too numerous to mention, reported the establishment of work-study programs; also, many communities set up, or greatly expanded, vocational— technical facilities and course offerings. 58 Bayley summarizes the effectiveness of these community efforts as follows: It is clear from the replies to President Kennedy's letter that the methods which school and college officials consider most effective in combating the dropout problem are the counseling of students and parents; 58Bayley, op. cit., p. 7. -4o_ special studies leading to curriculum changes, more flexible student schedules, improved teaching methods, and the devel— opment of new technical and vocational education programs and expansion of those in existence. It is further indicated by Bayley that when all com- munity resources are united in concert, the problem of dropout is most effectively approached. Conclusion During the past ten years, many communities have become deeply concerned about their high school dropouts. In 1961, Myers59 reported that 128 of Michigan's 318 high schools had recently conducted systematic dropout studies. Although this figure of 128 schools amounts to only 40 percent, Michigan would undoubtedly rank high when compared with the other states. Allen6O aptly sums up the importance of studying the potential dropout in the following statement: For the large proportion who have dropped out, the school can only offer out—of—school service with the expectation that some of the former students who once rejected its help will later welcome it. Schools can, however, learn to spot potential dropouts 59George R. Myers, “CUrriculum Improvement Through Holding Power Follow-Up Studies,“ Michigan Journal of Secondary Education, II (Winter, 1961), p. 110. 60Allen, op. cit., p. 5. -41- and do everything in their power to help them stay in school. This can be done by studying dropouts and the combination of characteristics and circumstances that has led them to leave school. Chapter III RESEARCH PROCEDURES OF THE STUDY The research procedures of this study will be dis— cussed under five general headings: 1) Identification of the population; 2) Hypotheses to be tested; 3) Sampling procedures; 4) Instrumentation; and 5) Procedures for treatment of the data. Identification of the Population The total population used for this study consisted of 516 male Caucasian high school students who were clas— sified as sophomores during the fall semester of 1963, as juniors during the fall semester of 1964, as seniors during the fall semester of 1965, and who were in attendance at one of three high schools in one Midwestern urban pub- lic school system. The population included only those students who were enrolled in the regular school program; students enrolled in any type of special education pro— gram were excluded. Since the present study is part of a larger longitudinal study, all students who participated in experimental groups, or other special treatment phases -42- -43- of the investigation by Brookover and associatesl, were also excluded. The Stay—ins. The stay-in population of this study has therefore been limited to a more homogeneous classification by having been selected on the basis of the following criteria: 1. Male 2. Caucasian 3. Sophomore, as of September, 1963; junior, as of September, 1964; senior, as of September, 1965. 4. Enrolled in the regular school program; i.e., not enrolled in a special education program of any type. 5. Participated in the regular phases of the longitudinal study by answering the questionnaires administered annually. 6. Did not participate in other phases of the longitudinal study (e.g., experiments to enhance self—concept of ability). lWilbur B. Brookover, Ann Patterson, and Shailer Thomas, "Self-Concept of Ability and School Achievement," U. S. Office of Education, Cooperative Rpsearch Project No. 845 (East Lansing: Office of Research and Publica- tions, Michigan State University, 1962). -44_ 7. Did not withdraw from school at any time for any purpose. This stay—in population numbered 421 students and included all those persons meeting the seven aforementioned criteria. The Dropouts The dropout population of this study consisted of all those persons who met all the requirements previously Specified for the stay—in population, except for the fact that they had voluntarily withdrawn from school during the tenth, eleventh, or twelfth grade. Special attention, however, was given to the following criteria: 1. Neither suspended nor expelled. 2. No transcript sent to another school 'p system. 3. Had not withdrawn because of illness or poor health. The above criteria eliminated all involuntary with— drawals and also accounted for all students who withdrew for the purpose of transferring to another school. It is reasonable to assume that some students may have dropped out of school, even though transcripts had been forwarded, but this was difficult to verify. The Size of the dropout -45- population numbered 95 persons and was large enough to minimize the effects of such cases on the results of the study. Dropouts were initially identified by listing all students who were not present when questionnaires were administered; those students who were absent, but had not withdrawn, were removed from the list and re—scheduled to complete the questionnaire at a later date. Whether a student had withdrawn or had merely been absent was determined by examining the records of the three high schools involved in this study. A revised list consisting of only male Caucasian dropouts was then compiled and a thorough check was made with the central Child Accounting Office to determine the exact cause or nature of their withdrawal. Those students moving to another city were identified and excluded from the list of dropouts only in ‘ 1 those cases where transcripts had been sent to the appro- priate school system. Also excluded from the dropout population were those individuals who had been involun— tarily withdrawn. In addition, two persons who had been initially classified as dropouts were excluded from the study when it was determined that a chronic health problem had necessitated their withdrawing from school. Forty—two of the ninety—five dropouts being studied withdrew from school during either the eleventh grade or the first three months of the twelfth grade. From this -46— group of 42 dropouts, 21 were students who had transferred as eleventh graders from other cities to one of the three high schools under study. For the purposes of this study, dropouts and non—dropouts were compared on the basis of tenth grade questionnaire data and grade—point averages; but since no tenth grade data was available for the 21 eleventh grade transfer students who dropped out, they were compared with non—dropouts on the basis of eleventh grade questionnaire data and grade—point averages. It should be noted, however, that only tenth grade question- naire data and grade—point averages were used for the non- drOpout population. A statistical comparison of the 21 eleventh grade transfer students (using eleventh grade data) and the 21 eleventh grade non—transfer students (using tenth grade data) was made to determine whether significant differences existed between these two dropout groups in regard to the grade level of data being used. The results of this com— parison did not Show any significant differences existing between these two groups on any of the variables examined in this study2 (see Appendix E). 2If significant differences had been found between these two groups, it would have been necessary to statistically compare tenth and eleventh grade data for only the non— transfer drOpout group; otherwise, any significant differences found between the two eleventh grade dropout groups might have been the result of differences existing between transfer and non—transfer dropouts and not the effects of using different grade level data for the tw0 groups. -47- Hypothespp to be Tested Many previous studies reviewed have indicated that dropouts differ from students who stay in school on at least three criteria: 1) level of academic ability; 2) socio—economic level of the family; and 3) grade—point average in basic high school subjects —— English, math, social studies, and science. Although these criteria may indicate measurable differences between dropouts and non— dropouts, it is believed that they are not necessarily causal factors but resultant factors; i.e., intervening psychological variables may very well be the major causes for these measured differences. The first three hypotheses, stated in null form, were actually a retest of the findings in earlier studies to determine the extent of the relationships between dropouts and non—dropouts regarding academic ability, socio—economic class, and grade—point average. Hypothesis I: There are no significant differences between male Caucasian high school dropouts and non-dropouts in regard to academic ability (IQ).3 3Academic ability was determined by using scores derived from the California Test of Mental Maturity (CTMM) admin— istered during the ninth grade. In some cases, it was necessary to use CTMM scores from tests administered during the eleventh grade. Academic ability and IQ have been considered synonomous terms in this study. -48— Hypothesis II: There are no significant differences between male Caucasian high school dropouts and non—dropouts in regard to socio—economic class (SES).4 Hypothesis III: There are no significant differences between male Caucasian high school dropouts and non—dropouts in regard to grade-point average (GPA). The following are the major hypotheses of this study. The treatment of these hypotheses, also stated in null form, was dependent upon the results of the tested hypotheses I, II, and III. Hypothesis IV: There are no significant differences between male Caucasian high school dropouts and non—dropouts in regard to level of self—concept of ability (SCA). Hypothesis V: There are no significant differences between male Caucasian high school dropouts and non—dropouts in regard to level of perceived parental eval— uations (PPEV). 4Socio-economic status was assessed by a scale developed by Otis Dudley Duncan. See Reiss, Occupation apd Social Status (Glencoe, Illinois: Free Press, 1961). -49- Hypothesis VI: There are no significant differences between male Caucasian high school dropouts and non—dropouts in regard to level of perceived parental expecta- tions (PPEX). Hypothesis VII: There are no significant differences between male Caucasian high school dropouts and non—dropouts in regard to level of educational expectations, or plans (Ed P1). Previous studies by Brookover and his associatess’6 found that SCA is influenced by PPEV, and Ed P1 is influenced by PPEX. These findings involved the same population being used in the present study but did not consider dropouts and non-dropouts separately. The final two hypotheses were ' 1‘ tested for this purpose. I Hypothesis VIII: There are no significant differences between level of SCA and level of PPEV when dropouts and non-dropouts are considered jointly or separately. 5Brookover, pp al., 1962, op. cit., p. 74. 6Wilbur B. Brookover, et al., "Self—Concept of Ability and School Achievement, II,7_U. S. Office of Education, Cooperative Research Project No. 1636 (East Lansing: Bureau of Educational Research Services, Michigan State University, 1965), p. 167. -50- Hypothesis IX: There are no significant differences between level of Ed P1 and level of PPEX when dropouts and non—dropouts are considered jointly or separately. Sampling Procedures The total population, as previously defined and described, was used to test the first three hypotheses in this study. Since it has been established in prior studies that dropouts and non-dropouts differ significantly regard- ing academic ability, socio-economic class, and grade— point average, it was anticipated that these factors would have to be partialled out before examining the data for further relationships. It this were not done, it was believed that considerable contamination of the measured differences of the main variables being examined would result. If Significant relationships were determined in HO: I, II, or III, or any combination thereof, the effects were to be partialled out by the method of matched groups. Since the number of non—dropouts exceeded the number of dropouts in any given cell, it was necessary to randomly select the non-dropouts (on the appropriate criteria) to equal the number of dropouts in each cell. A table of -51- random numbers was employed for this purpose.7 Instrumentation The major instrument used in this study was the Michigan State General Self-Concept of Ability Scale; however, only specific sections of this instrument were employed. They are: 1. General Self—Concept of Academic Ability (Appendix A) 2. Perceived Parental Evaluations (Appendix B) 3. Perceived Parental Expectations (Appendix C) 4. Educational Expectations, or Plans (Appendix D) Reliability of the Instrument The General Self—Concept of Academic Ability Scale W consists of eight five—choice items (see Appendix A) and was developed from a pre—test analysis. Items were coded from 5 to l with the higher self—concept alternatives receiving the higher values. These eight items were origi— nally tested by the Guttman scale and received coefficients of reproducibility of .95 for males and .96 for females. 7Wilfrid J. Dixon and Frank J. Massey, Jr., Introduction to Statistical Analysis (New York: McGraw—Hill Book Company, 1957), pp. 366—370. -52- These coefficients are for 1050 seventh grade students8 (the same population presently being studied, but limited to male Caucasians). Brookover and his associates further state:9 In the eighth and ninth grades, ran— dom samples of 35 males and 35 females indicated that these items retained a scale form with reproducibilities of .96 and .97 for males in the two years and .92 and .93 for females in the same two years. In addition, the scale has reliabilities cal— culated by Hoyt's Analysis of Variance of .82, .91, .92, and .86 for males, and .77, .84, .84, and .84 for females for the 7th, 8th, 9th, and 10th grades, respectively. These correlations are very high, especially when it is considered that self—concept is not constant but fluc— tuates through changes individuals will undoubtedly experience in interpersonal relationships. Validity of Self—Concept Scales Correlations of self-concept of ability in each of four subject matter areas (English, math, social studies, and science) with general self-concept of ability proves significantly the validity of this instrument. These correlations are presented in Table 3.1. 8Brookover, pp al., 1965, Op. cit., p. 51. 9Loc. cit. 10 11219-- p. 55. -53_ Table 3.1 Correlations of General Self-Concept with Specific Subject Self—Concept Male Female Factors 7th 8th 9th 10th 7th 8th 9th 10th Grd. Grd. Grd. Grd. Grd. Grd. Grd. Grd. Math Self-Concept X GSC .65 .70 .69 .68 .64 .67 .63 .58 English X GSC .63 .64 .62 .64 .61 .65 .66 .63 Social Studies X GSC .64 .69 .67 .69 .62 .72 .71 .70 Science X GSC .66 .71 .67 .73 .54 :67 .68 .64 N=513 N=731 N=446 N=425 N=537 N=751 N=508 N=517 Also in evidence is a high degree of predictive validity. This is indicated by the following correlations between self— . 11 concept of ability and grade-pOint average (GPA): llIbid., p. 57. _54- Table 3.2 Correlations of General Self—Concept of Ability with GPA Grade Males Females Jan. June No. Jan June No. 7th Grade .57 513 .57 537 8th Grade .63 .60 731 .65 .63 751 9th Grade .61 .57 446 .64 .60 508 10th Grade .56 425 .59 517 The following statement summarizes the degrees of 12 reliability and validity of this instrument: It is concluded that the general self- 1) has a high concept of ability scale: internal consistency as measured by Hoyt's analysis of variance; 2) has a consistently high coefficient of reproducibility making it an excellent Guttman scale; high test-retest correlation; and 4) has concurrent, construct and predictive The consistency of the relia— validity. 3) has a bility measures and indices of validity over a three-year period indicates that this scale is a consistent and valid measure. 12 IPA-1;: pp- 57-58- -55- W Additional data were made accessible by the school officials of the three senior high schools presently under study. Specifically, these data are: l. I.Q. scores, derived from scores on the California Test of Mental Maturity (CTMM) which was administered by the schools to all ninth and eleventh grade pupils. 2. Semester grades for both fall and spring semesters. 3. Enrollment records, when needed. 4. Information from the students' cumulative records when basic information for the study was needed. A considerable amount of information was also derived from the records of the Office of Child Accounting -~ especially enrollment information on those persons who had withdrawn from school. The nature and cause of with- drawal was, in many instances, determined from this source of information. Procedures for Treatment of ppp Data The first three hypotheses were tested by using the _55- total population (as previously Specified). Chi—square was used to determine the relative associations among the variables. For all nine hypotheses, a 2 x 2 contingency table design was used and the null hypotheses Were rejected in those instances where a significance level of .05 or beyond was attained. The level of significance was deter- mined by the following formula:13 2 N lAD-BC' —‘p 2 d.f. = 1 (A+B)(C+D)(A+C)(B+D) According to Siegel,l4 this is the best formula to use when comparing two independent groups on two discrete categories. Also, it has the advantage of incorporating a correction for continuity "which markedly improves the approximation of the distribution of the computed X2 by the chi—square distribution."15 The design of the present study has met the requirements for using this particular formula since comparisons were made between dropouts and non—dropouts on variables that had been dichotomously categorized as being either high or low. l3Sidney Siegel, Nonparametric Statistics for the Behavioral Sciences (New York: McGraw—Hill Book Company, 1956), p. 107. l4Loc. cit. 15ibid., p. 108. -57- The study was designed in such a way that if signifi— cant relationships were determined between dropouts and non-dropouts on any of the first three hypotheses, or any combination thereof, the remaining six hypotheses would be tested by using matched groups to control for these significant differences. Hypotheses VIII and IX were also analyzed by using chi—square tests of significance to determine whether or not two dichotomized variables differ significantly when matched groups of dropouts and non—drOpouts were considered either jointly or separately. Dichotomization of Variables In order to classify the data into two discrete cate— gories for chi—square analysis, a mean score was computed for each of the seven variables to be examined in this study. In each case the mean was determined by utilizing the total population of the study.16 Any score exceeding the mean was classified as a high score; those scores falling below the mean were classified as low scores. The mean scores of the variables and the subsequent ranges of Scores for the high and low categories are presented in Table 3.3. l6Ten dropouts did not respond to PPEX and Ed P1 questionnaire items. —58— Table 3.3 Mean Scores of Variables and Ranges of Scores for High and Low Categories Variable Mean Low High N I.Q. 105.6 105 or 106 or 516 Below Above GPA 1.96 O — 1.75 2.00 ~ 4.00 516 SES 39.19 1 - 39 40 — 100 516 SCA 27.76 0 — 27 28 — 40. 516 PPEV 18.77 0 — 18 19 - 25 516 PPEX 5.15 l — 5 6 - 7 506 Ed Pl 5.02 l — 5 6 - 7 506 Summary In Chapter III an account has been given of the research procedures involved in the present study. Specifically discussed were the methods utilized in identifying the population under study, the hypotheses to be tested, and the sampling procedures anticipated to be necessary to partial out significant differences between dropouts and non—dropouts prior to testing the major variables of the study. Also, a.description was given of the instrument that was employed, as well as a description of the statistical procedures to be used in the treatment of the data_in Chapter IV. Chapter IV ANALYSIS OF THE DATA The purpose of this chapter is to present an analysis of the data which was theoretically supported in Chapter I and further delineated and described in Chapter III. The first section will consist of an examination of the relation— ships between dropouts and non—dropouts in regard to three variables that have been examined in many previous drop- out studies. For the purposes of this study, these variables have been classified as “test variables." The second section will contain a description of the population dis— tribution in regard to the aforementioned test variables, as well as the sampling procedures that were necessary to control for significant differences found between drop— outs and non—dropouts. The third section will be a presenta- tion and analysis of the data related to the major variables of the study. A brief summary will conclude the chapter. Analysis of the Test Variables Academic ability, socio—economic status of the family, and grade-point average were the three test variables examined in this study. A comparison of dropouts and non-dropouts was made to test the hypotheses, stated in -59- -60— null form, that no significant differences existed between the two sub—populations on these three criteria. Hypothesis I: There are no significant differences between male Caucasian high school dropouts and non—dropouts in regard to academic ability. The population mean score for academic ability (IQ), which was presented in Table 3.3, was 105.6. Of the 421 non-dropouts being studied, 244 (58 percent)1 were classi— fied as having high scores by scoring 106 or higher; only 31 of the 95 dropouts were classified as having high scores (33 percent). Chi—square analysis was used for the test of signifi— cance to measure Hypothesis I; a rejection limit of .05 or beyond was established. Results of this analysis are presented in Table 4.1. , Examination of Table 4.1 indicates that dropouts and _ ' non-dropouts differed significantly in regard to level of academic ability. This difference was significant beyond the .001 level of confidence; therefore, the null hypothesis that no significant differences exist between dropouts and 1Since the dropout and non-dropout groups are of unequal size, percentages have been parenthetically included to further illustrate the differences between these two groups in regard to the number of individuals receiving a high classification. This same procedure will be used in presenting the analysis of data for H0 II and Ho III. ~61- non~dropouts regarding level of academic ability was rejected. Table 4.1 Chi—Square Analysis of Association Between Dropouts and Non-Dropouts and Level of Academic Ability Level of Academic Ability Groups Totals High Low Non-Dropouts 244 177 421 Dropouts 31 64 95 Totals 275 241 516 X2 = 18.90 1 d.f. Significant Beyond .001 Level Hypothesis II: There are no significant dif— H ferences between male Caucasian high school dropouts and non— dropouts in regard to socio- economic class. The population mean score for socio-economic status (SES) was 39.19, as assessed by a lOO—point scale developed by Otis Dudley Duncan. Of the 421 non—dropouts, 224 (53 percent) were classified as having high scores; and 28 of —62- the 95 dropouts (29 percent) were classified as having high scores. Chi—square analysis was also used for the test of Significance to measure Hypothesis II; also, a rejection limit of .05 or beyond was established. Results of this analysis are presented in Table 4.2. Table 4.2 Chi—Square Analysis of Association Between Dropouts and Non-Dropouts and Level of Socio—Economic Class Groups Level of Soc10—Economic Class Totals High Low Non—Dropouts 224 197 421 Dropouts 28 67 95 '1 Totals 252 264 516 i . X2 = 16.54 1 d.f. Significant Beyond .001 Level It is indicated in Table 4.2 that dropouts and non— dropouts differed significantly in regard to level of socio- economic class (or status). The difference between the two sub—populations of this study was significant beyond the .001 level; therefore, the null hypothesis that no sig— nificant differences exist between dropouts and non—dropouts —63— regarding socio—economic class was rejected. Hypothesis III: There are no significant dif— ferences between male Caucasian high school dropouts and non— dropouts in regard to grade—point average. The test of significance for Hypothesis III was a chi— square analysis with a rejection limit of .05 or beyond. In Table 4.3 the results of this analysis are presented. Table 4.3 Chi—Square Analysis of Association Between Dropouts and Non—Dropouts and Level of Grade—Point Average Level of Grade—Point Average Groups High Low Totals Non-Dropouts 265 156 421 Dropouts 8 87 95 Totals 273 243 516 X2 = 90.31 1 d.f. Significant Beyond .001 Level -64- Inspection of Table 4.3 indicates that dropouts and non—dropouts differed significantly in regard to level of grade—point average. Nonrdropouts earning grade—point averages above the population mean of 1.96 numbered 265 (63 percent); only 8 of the 95 dropouts (8.4 percent) earned grade-point averages above this mean. The chi—square analysis of Hypothesis III indicated significant differences between the two sub—populations of this study beyond the .001 level of confidence; therefore, the null hypothesis that no significant differences exist between dropouts and non—dropouts regarding grade—point averages was rejected. Analysis of Sampling Data It was stated in Chapter III that if significant differences were found between dropouts and non—dropouts in regard to Hypotheses I, II, or III, or any combinations thereof, these differences would be partialled out by the method of selecting matched groups on the appropriate criteria. It has been determined that significant dif- ferences did, in fact, exist between dropouts and non— dropouts on all three criteria; i.e., academic ability (IQ), socio—economic status (SES), and grade—point average (GPA). Matched groups were thus selected on the basis of the data presented in Table 4.4. Ninety—five of the 421 non—dropouts were randomly —65- selected to match the number of dropouts on each of the eight combinations of variables presented in Table 4.4. A table of random numbers was used for this purpose. Distribution of Dropouts and Non-Dropouts Table 4.4 Scoring Either High (H) or Low (L) in Regard to IQ, SES, and GPA Variables Groups Matched Totals IQ SES GPA Non-Dropouts Dropouts H H H 129 4 8 H H L 24 6 12 H L H 69 1 2 H L L 22 20 40 L H H 32 0 O L H L 39 18 36 L L H 35 3 6 L L L 71 43 86 Totals 421 95 190 -66— Analysis of the Major Variables The four major variables that have been examined in this study are: 1) general self—concept of academic abil— ity (SCA); 2) the student's perception of his parents' evaluations of his academic ability (PPEV); 3) the student's perception of the level of education his parents expect him to achieve (PPEX); and 4) the student's expressed educational expectations, or plans (Ed Pl). AS stated pre— viously, matched groups of dropouts and non-dropouts have been used to test the null hypotheses to be examined in this section of Chapter IV. Hypothesis IV: There are no significant dif— ferences between male Caucasian high school dropouts and non-dropouts in regard to level of self—concept of ability (SCA). Chi—square analysis was used to test Hypothesis IV; and a rejection limit of .05 or beyond was established. The results of this analysis are presented in Table 4.5. Examination of Table 4.5 indicates that dropouts and non—dropouts differed significantly in regard to level of SCA, even when IQ, SES, and GPA differences had been par- tialled out. The population mean score for SCA was 27.76 —67- Table 405 Chi-Square Analysis of Association Between Matched Groups of Dropouts and Non-Dropouts and Level of Self—Concept of Ability (SCA) Level of SCA Groups Totals High Low Non—DrOpouts 34 61 95 DrOpouts 18 77 95 Totals 52 138 l9O 2 X = 5.96 l d.f. Significant at .02 Level on a 40—point scale; 34 non-dropouts and 18 dropouts scored 28 or above. This difference was significant at the 002 level of confidence; therefore, the null hypothesis that no significant differences exist between dropouts and non» dropouts regarding level of SCA was rejectedo Hypothesis V: There are no significant dif— ferences between male Caucasian high school dropouts and non-dropouts in regard to level of perceived parental evaluations (PPEv)° —68~ To test Hypothesis V, chi—square analysis was employed and a rejection limit of .05 or beyond was established. In Table 4.6 the results of this analysis are presented. Table 4.6 Chi—Square Analysis of Association Between Matched Groups of Dropouts and Non-Dropouts and Level of Perceived Parental Evaluations (PPEV) Level of PPEV Groups Totals High Low Non—Dropouts 42 53 95 Dropouts 29 66 95 Totals 71 119 190 2 r}! ‘ X = 3.24 l d.f. Not Significant at .05 Level " Inspection of Table 4.6 indicates that matched groups of dropouts and non—dropouts did not differ significantly at the .05 level or beyond in regard to level of PPEV. The population mean score for PPEV was 18.77 on a 25-point scale. Forty—two non—dropouts and 29 dropouts scored above the pOpulation mean. A chi—square coefficient of 3.84 was required to reject the null hypothesis that no sig- nificant differences exist between dropouts and non—dropouts —69- regarding the level of PPEV; therefore, Hypothesis V is accepted. Hypothesis VI: There are no significant dif— ferences between male Caucasian high school dropouts and non— dropouts in regard to level of perceived parental expectations (PPEX). Chi-square analysis was used for the test of signifi- cance to measure Hypothesis VI; a rejection limit of .05 or beyond was established. Results of this analysis are presented in Table 4.7. Table 4.7 Chi-Square Analysis of Association Between Matched Groups of Dropouts and Non-Dropouts and L' Level of Perceived Parental Expectations (PPEx) Level of PPEX Groups Totals High Low Non—DrOpouts 43 52 95 Dropouts 21 64 85 Totals 64 116 180 X = 7.40 l d.f. Significant at .01 Level -70.. It is indicated in Table 4.7 that matched groups of dropouts and non-dropouts differed significantly at the .01 level of confidence in regard to level of PPEX. The p0pulation mean score was 5.15 on a 7-point scale. Forty- three of the ninety—five non—dropouts were classified in the high category by scoring either 6 or 7; 21 of the 85 dropouts2 were classified in the high category. Since a rejection limit of .05 or beyond was established, the null hypothesis that no significant differences exist between dropouts and non—dropouts regarding level of PPEX was rejected. Hypothesis VII: There are no significant difa ferences between male Caucasian high school dropouts and non~ dropouts in regard to level of educational expectations, or plans (Ed P1). The test of significance for Hypothesis VII was a chi-square analysis with a rejection limit established at .05 or beyond. In Table 4.8 the results of this analysis are presented. 2It was indicated in Chapter III that 10 dropouts did not respond to PPEX questionnaire items. -71- Table 4.8 Chi—Square Analysis of Association Between Matched Groups of Dropouts and Non-Dropouts and Level of Educational Plans (Ed Pl) Level of Ed Pl Groups . Totals High Low Non—Dropouts 35 6O 95 Dropouts 18 67 85 Totals 53 127 180 X2 = 4.57 l d.f. Significant at .05 Level Examination of Table 4.8 indicates that matched groups of dropouts and non—drOpouts differed significantly in regard to level of Ed P1. The population mean score for Ed P1 was 5.02 on a 7-point scale. Of the 95 non-dropouts, 35 were classified in the high category; 18 of the 85 drop— outs responding to this questionnaire item3 were classified as having given responses categorized as high responses (6 or 7). The difference betWeen these two groups in regard to Ed P1 was significant at the .05 level of con- fidence; therefore, the null hypothesis that no significant _ 3Ten dropouts did not respond to Ed P1 questionnaire items (Cf. ante). -72- differences exist between dropouts and non—dropouts regarding level of Ed P1 was rejected. Analysis of Relationships Between Major Variables Previous studies by Brookover and his associates4’5 found that SCA is influenced by PPEV, and Ed P1 is influenced by PPEX. These findings involved the same population that has been used in the present study but did not consider dropouts and non—dropouts separately. By using matched groups of dropouts and non—dropouts, additional controls have been established by partialling out the effects of academic ability (IQ), grade-point average (GPA), and socio~economic status of the student°s family (SES). Hypotheses VIII and IX were tested to determine whether or not two dichotomized variables differ significantly in regard to levels when matched groups of dropouts and I; nonudropouts are considered either jointly or separately. Hypothesis VIII; There are no significant difa ferences betWeen level of SCA 4Wilbur B. Brookover, Ann Patterson, and Shailer Thomas, "Self—Concept of Ability and School Achievement,“ U. S. Office of Education, Cooperative Research Project No. 845 (East Lansing: Office of Research and Publications, Michigan State University, 1962), p. 74. 5 Wilbur B. Brookover, 2: al., “Self-Concept of Ability and School Achievement, II," U. S. Office of Education, Cooperative Research Project No. 1636 (East Lansing: Bureau of Educational Research Services, Michigan State University, 1965), p. 167. -73... and level of PPEV when dropouts and non-dropouts are considered jointly or separately. To test Hypothesis VIII, three chi-square analyses were employed, and in each case, a rejection limit of .05 or beyond was established. The results of the test for significance between levels of SCA and PPEV when matched groups of dropouts and non—dropouts are considered jointly" is presented in Table 4.9. The results of tests for sig- nificance between levels of SCA and PPEV when non—dropouts and dropouts are considered separately are presented in Tables 4.10 and 4.11, respectively. Examination of Table 4.9 indicates that a significant difference existed between levels of high and low when SCA and PPEV were compared by using matched groups of drops outs and non—drOpouts. The chi—square analysis coefficient ‘ was 49.47, which was significant beyond the .001 level of confidence. It was further indicated in Table 4.10 that a significant relationship existed between the dichotomized levels of high and low when SCA and PPEV were compared by using only the non—dropout group. The chi~square analysis coefficient was 38.18, which was also significant beyond the .001 level of confidence. In Table 4.11 a comparison was made of the same dichotomized levels of the same two variables as those examined in Tables 4.9 and 4.10; however, -74- Table 4.9 Chi-Square Analysis of Association Between Dichotomized Levels of SCA and PPEV When Matched Groups of Dropouts and Non—Dropouts are Considered Jointly Level of SCA I Level of PPEV Totals High I Low 1 High PPEV 41 29 70 Low PPEV 12 108 120 Totals 53 137 190 X2 = 49.47 1 d.f. Significant Beyond .001 Level Table 4.10 Chi—Square Analysis of Association Between Dichotomized Levels of SCA and PPEV When Non—Dropouts are Considered Separately Level of SCA Level of PPEV Totals High Low High PPEV 30 ll 41 Low PPEV 5 49 54 Totals 35 6O 95 X = 38.18 1 d.f. Significant Beyond .001 Level -75- Table 4.11 Chi—Square Analysis of Association Between Dichotomized Levels of SCA and PPEV When Dropouts are Considered Separately Level of SCA Level of PPEV Totals High Low High PPEV ll 18 29 Low PPEV 7 59 66 Totals 18 77 95 X2 = 8.08 l d.f. Significant at .005 Level only dropouts were used. A chi—square analysis coefficient of 8.08 indicated that a significant difference existed between the two dichotomized levels of SCA and PPEV at the .005 level of confidence. The null hypothesis that no significant differences exist between level of SCA and level of PPEV when dropouts and non-dropouts are considered jointly or separately was rejected. In all three instances, the level of significance was greater than the predetermined rejection level of .05. This indicated that students who scored high on SCA also tended to score high on PPEV, or tended to score low on PPEV if they had scored low on SCA. ~76— Hypothesis IX: There are no significant dif— ferences between level of Ed P1 and level of PPEX when dropouts and non—dropouts are considered jointly or separately. Hypothesis IX was tested by using three chi—square analyses; a rejection limit of .05 or beyond was estab- lished for each of the three analyses. The results of the test for significance between levels of Ed P1 and PPEX when matched groups of dropouts6 and non—dropouts are considered jointly is presented in Table 4.12. The results of tests for significance between levels of Ed P1 and PPEX when non-dropouts and dropouts are considered separately are presented in Tables 4.13 and 4.14, respectively. Inspection of Table 4.12 indicates that a significant difference existed between levels of high and low when Ed P1 and PPEX were compared by using matched groups of dropouts and non—dropouts. The chi-square analysis coeffi- cient was 80.55, which was significant beyond the .001 level of confidence. Significant relationships Were also indicated in Tables 4.13 and 4.14. In Table 4.13, levels of high and low for the variables Ed P1 and PPEX were examined by using only the non—dropout group. A chi~square 6Ten dropouts did not respond to Ed P1 and PPEX questionnaire items. -77_ Table 4.12 Chi—Square Analysis of Association Between Dichotomized Levels of Ed P1 and PPEX When Matched Groups of Dropouts and Non—Dropouts are Considered Jointly Level of Ed Pl Level of PPEX Totals High Low High PPEX 46 19 65 Low PPEX 7 108 115 Totals 53 127 180 X2 = 80.55 1 d.f. Significant Beyond .001 Level Table 4.13 Chi—Square Analysis of Association Between Dichotomized Levels of Ed P1 and PPEX When Non—Dropouts are Considered Separately Level of Ed Pl Level of PPEX Totals High Low High PPEX 32 12 44 Low PPEX 3 48 51 Totals 35 6O 95 X2 = 42.50 1 d.f. Significant Beyond .001 Level -78- Table 4.14 Chi-Square Analysis of Association Between Dichotomized Levels of Ed P1 and PPEX When Dropouts are Considered Separately Level of Ed Pl I Level of PPEX Totals High Low I High PPEX 14 7 21 Low PPEX 4 6O 64 Totals 18 67 85 X2 = 31.58 1 d.f. Significant Beyond .001 Level analysis coefficient of 42.50 was measured, which was sig- nificant beyond the .001 level of confidence. Results of Table 4.14 likewise indicated that significant relation— ships were found when levels of high and low for the variables Ed P1 and PPEX were compared by using only the dropout group. For this chi—square analysis the coefficient was 31.58, which was significant beyond the .001 level of confidence. The null hypothesis that no significant differences exist between level of Ed P1 and level of PPEX when dropouts and non—dropouts are considered jointly or separately was rejected. The level of significance, in all three cases, was greater than the predetermined rejection level of .05. _79_ Results of the three previous tables indicated that students who scored high on Ed Pl also tended to score high on PPEX, or tended to score low on PPEX if they had scored low on Ed Pl. Addendum to the Study The present study has examined relationships between male Caucasian high school dropouts and non—dropouts in regard to selected characteristics. The assumption was made that the dropouts used in this study were a homogene— ous group, regardless of the differences in grade—levels they had attained at the time of dropping out. After analyz- ing the data, however, as presented above, it was felt that a test of this assumption of homogeneity in regard to grade—level attained was a necessary addition to the study. Comparisons were therefore made between tenth grade drop— outs and those students dropping out during either the eleventh or twelfth grade to determine whether significant differences existed between these two dropout groups in regard to the three test variables and the four major var— iables that were examined in this study. Analysis of the data indicated that no significant differences existed between these two dropout groups on any of the seven var- iables of the study (see Appendix F). —80— H Mir/f) Summary The first three hypotheses Were tested by using the total population of the study. Differences between drop— Outs and non-dropouts in regard to academic ability (IQ), socio-economic status (SES), and grade—point average (GPA) were partialled out by the method of matched groups prior to the exploration of other relationships. Chi—square was used to determine the relative associations among the variables. For all nine hypotheses, a 2 x 2 contingency table design was used and the null hypotheses were rejected in those instances where a significance level of .05 or beyond was attained. The following relationships were found: 1. Dropouts and non—dropouts differed signifi— cantly in regard to all three test variables: a) academic ability (IQ); b) socio—economic status (SES); and c) gradeapoint average (GPA). 2. Matched groups were formed to partial out differences between dropouts and non-dropouts in regard to IQ, SES, and GPA. The four major variables were then analyzed and sig- nificant differences were found between drop- outs and non-dropouts in regard to: a) self- concept of academic ability (SCA); b) perceived -81— parental expectations (PPEX); and c) edu- cational expectations, or plans (Ed Pl). Differences between dropouts and non- dropouts in regard to perceived parental evaluations (PPEV) were not significant at the .05 level. Significant relationships were found between levels of SCA and levels of PPEV, whether dropouts and non-dropouts were considered either jointly or separately. Significant relationships were also found between levels of Ed P1 and levels of PPEX, whether dropouts and non—dropouts were considered either jointly or separately. Chapter V SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS The basic purpose of this study was to determine the ways in which Caucasian male high school dropouts differ from those students who remain in school in regard to selected characteristics. The total population selected for this study con- sisted of 516 Caucasian male high school students who were classified as SOphomores during the fall semester of 1963, as juniors during the fall semester of 1964, as seniors during the fall semester of 1965, and who were in attendance at one of three high schools in one Mid— western urban public school system. Only those students who were enrolled in the regular school program were included in this study; those students enrolled in any type of special education program were excluded. Since this study was part of a larger longitudinal study, only those students who had participated in the regular phases of the longitudinal study were included; but any student who had taken part in Special experimental phases of the longitudinal study were excluded. The stay-in population numbered 421 students and included all those persons who met the aforementioned criteria. The dropout population consisted of 95 persons who met all the requirements -82- -83— previously specified for the stay—in population, except for the fact that they had voluntarily withdrawn from school during the tenth, eleventh, or twelfth grade. Dropouts were excluded from the study if they had been involuntarily withdrawn, had transferred to another school system, or had withdrawn for reasons of illness or poor health. 5 Variables Examined The first three variables examined were academic ability, socio—economic status of the family, and grade— point average. These variables were classified as “test variables," since many earlier studies reviewed had found significant differences between dropouts and non—dropouts in regard to these three variables. Academic ability was determined by obtaining the i! score for each student in the study pOpulation on the ( California Test of Mental Maturity (CTMM) which had been mass-administered to all ninth grade students. In some instances, ninth grade CTMM scores were not available and eleventh grade CTMM scores had to be used; both ninth and eleventh grade CTMM scores were obtained from the three high schools involved in the longitudinal study. Grade— point averages for the tenth grade —- and, in some instances, the eleventh grade —— were also obtained from this same -84.- source and included only those grades earned in the basic high school subjects. Specifically, these subjects were English, social studies, math and science. The socio- economic status of the student was actually a measure of the father's occupational level and was assessed by a scale developed by Otis Dudley Duncan. The four major variables examined in this study were: 1) general self—concept of academic ability; 2) perceived parental evaluations; 3) perceived parental expectations; and 4) educational expectations, or plans. Data for these four major variables were attained through the mass— administration of a questionnaire which was developed as a part of the larger longitudinal study. The term “ques— tionnaire" refers to what is noUVclassified as The Michigan State Ggperal Self—Concept of Ability Scale. This ques— tionnaire (or scale) was administered to all students in -. one Midwestern urban public school system who were seventh graders during the 1960—61 school year. This questionnaire was re-administered annually to the same group of students for the next five consecutive years; therefore, data were collected on one specific group of students from the seventh grade through the twelfth grade. For the purposes of this ! study, dropouts and non—dropouts were compared on the basis of tenth grade questionnaire data, except for twenty- one drOpouts who transferred to the school system under study at the beginning of the eleventh grade. For this —85— Specific group of dropouts, eleventh grade questionnaire data and grade-point averages were used. Two additional relationships were examined to deter- mine whether students who scored in either the high or low range on one major variable had also scored in the same range on one other major variable. These two relationships were further examined by using dropouts and non—dropouts combined, then by examining the data with dropouts and non—dropouts being considered separately. The first rela— tionship examined was a comparison of the levels of self— concept of academic ability and levels of the student“s perception of his parents' evaluation of his academic ability; the second relationship examined was a comparison of the levels of the student's educational plans and the levels of the student's perception of his parents” expecta- tions in regard to the amount of education he would complete. Data Analysis Procedures The three test variables and four major variables examined in this study were first dichotomized into high and low categories by computing mean scores for each var— iable. All scores falling below the mean for any given variable were classified as low scores; those scores exceed— ing the mean were classified as high scores. Chi~square was then used to determine the relative associations —86- between dropouts and non—dropouts in regard to these dichotomized variables. For each analysis a 2 x 2 contin— gency table design was used and the null hypothesis was rejected if a significance level of .05 or beyond was computed. The total population of the study was used for compar~ ing dropouts and non-dropouts in regard to the three test variables. Prior to comparing dropouts and non—dropouts in regard to the four major variables of the study, matched groups were randomly selected in order to partial out significant differences between dropouts and non-dropouts that had been determined from the analyses of the three test variables. The major variables of the study were then examined by using 95 dropouts and 95 non—dropouts -— a total matched population of 190 subjects. Summary of Significant Findings The findings of this study are listed here according to results obtained by testing the nine null hypotheses found in Chapter IV. 1. There was a significant difference between Caucasian male high school dropouts and non» dropouts in regard to academic ability. This difference was significant beyond the .001 level of confidence. Ll -87— There was a significant difference between Caucasian male high school dropouts and non-dropouts in regard to socio-economic status of the family. This difference was significant beyond the .001 level of con- fidence. There was a significant difference between Caucasian male high school dropouts and non—dropouts in regard to grade-point average. This difference was significant beyond the .001 level of confidence. There was a significant difference between Caucasian male high school dropouts and non—dropouts in regard to self-concept of academic ability. This difference was significant at the .02 level of confidence. There was a significant difference between Caucasian male high school dropouts and non~dropouts in regard to perceived parental expectations. This difference was signifi- cant at the .01 level of confidence. There was a significant difference between Caucasian male high school dropouts and non—dropouts in regard to educational —88— expectations, or plans. This difference was significant at the .05 level of con— fidence. There were significant differences between levels of self-concept of academic ability and levels of perceived parental evalua— tions when dropouts and non—dropouts were considered either jointly or separately. When dropouts and non-dropouts were con~ sidered jointly, the differences between levels of self—concept of academic ability and levels of perceived parental evalua— tions were significant beyond the .001 level of confidence; when dropouts and non—dropouts were considered separately, these differ— ences between levels of the aforementioned .( variables were significant at the .005 level ‘ of confidence, and significant at the .001 level of confidence, respectively. There were significant differences between levels of educational plans and levels of perceived parental expectations when drop— outs and non—dropouts were considered either jointly or separately. When dropouts and non-dropouts were considered jointly, the -89— differences between levels of educational plans and levels of perceived parental expectations were significant beyond the .001 level of confidence; when dropouts and non—dropouts were considered separately, these differences between levels of the aforementioned variables were -— in both instances -— significant beyond the .001 level of confidence. W2 On the basis of the statistically significant findings summarized above, the following conclusions have been formulated. Each conclusion should be viewed as being most applicable to Caucasian male high school students who reside in Midwestern urban communities. 1. It is concluded that dropouts are different from those students who remain in school on the basis of academic ability (or IQ) scores. This conclusion is supported by the findings of numerous studies reviewed in Chapter II of this study. It is believed, however, that differences in academic abil— ity scores do not provide sufficient informa— tion, when considered in isolation from -90- other factors, to adequately identify the potential dropout. Thirty—one of the 95 dropouts studied had IQ scores of 106 or above (8 had scores of 120 or above); of the 421 non—dropouts studied, 177 scored below 105.6 -- the mean score for the total study population. 2. It is concluded that dropouts are different from those students who remain in school on the basis of the socio—econgmic status of the student's family (SES). This conclusion is supported by the findings of numerous studies reviewed in Chapter II of this study. It is believed, however, that examination of socio-economic factors alone do not pro» vide sufficient information to adequately identify the potential dropout. In the 1 present study, 29 percent of the dropouts were classified as having a high socio— economic status level; and 47 percent of the non-dropouts were classified as having a low socio-economic status level. 3. It is concluded that dropouts are different from those students who remain in school on the basis of grade—point average (GPA) for -91- the basic school subjects —- English, social studies, math, and science. This conclusion is supported by the findings of numerous studies reviewed in Chapter II of this study. Of the three test variables examined (IQ, SES, and GPA), it appears to the writer that GPA may be the best single predictor in regard to identifying which students are most likely to drop out of school. Only 8 of the 95 dropouts in the present study had high grade-point averages, whereas 265 of the 421 non-dropouts had high grade—point averages. It is concluded, from examining the distribu~ tion of dropouts and non-dropouts in regard to high and low levels of the three afore— mentioned test variables, that potential I drOpouts cannot be adequately defined solely on the basis of IQ, SES, and GPA. Although significant differences were found between dropouts and non—dropouts on all three test variables beyond the .001 level of confidence, the distribution of non—dropouts did not indicate that a predictable distribution had occurred (see Table 4.4). The largest -92- number of non-dropouts (129) did fall into the highest of the eight possible categories (H—H—H), but the second largest number of non-drOpouts (71) was recorded in the lowest of these eight categories (L—L—L). It is believed that a distribution such as this indicates that other variables were also responsible for the differences between drop— outs and non—dropouts being measured in this study. It is concluded that dropouts are different from non-dropouts on the basis of a student“s self—concept of his academic ability (SCA), even when IQ, SES, and GPA differences have been partialled out. Although a student"s self-concept may be affected by his academic ability (or IQ) and his previous academic performance (GPA), it is believed that the assessment of the student‘s attitudes about his ability to achieve in academic endeavors is an essential factor in predicting whether or not a student will drop out of school prior to graduation. It is concluded that dropouts are not different from non—dropputs on the basis of perceived _93- parental evaluations (PPEV). Although a significant relationship was determined between dichotomized levels of SCA and PPEV when dropouts and non-dropouts were considered either jointly or separately, the examina— tion of differences between dropouts and non—dropouts in regard to PPEV did not prove to be significant at the .05 level of con- fidence. Further examination of the data indicated that 17 of the 95 dropouts scored in the high category on PPEV and in the low category on SCA, whereas only 6 of the 95 dropouts scored in the high category on SCA and in the low category on PPEV. It appears that the dropout may tend to perceive his parents as having a higher evaluation of n his academic ability than he has. 7. It is concluded that dropouts are different from non—dropouts on the basis of perceived parental expectations (PPEX), even when IQ, SES, and GPA have been partialled out. It is believed, however, that potential dropu outs cannot be identified by asking students 7 whether or not they think their parents expect them to graduate from high school. -94- The study population mean score for PPEX was 5.15 on a 7—point scale; high school completion was the third lowest item on the scale (see Appendix C). Further exam— ination of the data revealed that only four dropouts and one non—dropout perceived that their parents did not expect them to graduate from high school. It is therefore concluded that a graduated scale, such as the one employed in the present study, can discriminate between dropouts and non—dropouts, whereas a single—item response cannot. It is concluded that dropouts are differepp from non—dropouts on the basis of educational expectations, oryplans (Ed Pl), even when IQ, SES, and GPA have been partialled out. It is believed, however, that potential dropouts cannot be identified by asking students whether or not they expect to gradm uate from high school. The study population mean score for Ed P1 was 5.02 on a 7—point scale; high school completion was the third lowest item on the scale (see Appendix D). Further examination of the data revealed that only four dropouts and one non-dropout 10. -95_ did not expect to graduate from high school. It is therefore concluded that a graduated scale, such as the one employed in the pre— sent study, can discriminate between drops outs and non—dropouts, whereas a single-item response cannot. Since a significant relationship was deter— mined between dichotomized levels of Ed P1 and PPEX when dropouts and non—dropouts were considered either jointly or separately, it is concluded that potential dropouts are different from the student who will remain in school when compared by using either one of the two aforementioned variableg. It is not believed, however, that the use of both Ed P1 and PPEX measures would increase ,- to any appreciable extent ~- the degree of reliability in identifying the potential dropout. It is concluded that there are many variables which help describe potential dropouts from those students who will remain in high school until graduation. Although this and previous studies have demonstrated that academic abil— ity, socio—economic status, and grade-point ~96- ‘ average appear to be essential variables in identifying potential dropouts, it is believed, on the basis of this study, that a student“s self—concept of his academic ability and his expressed educational plans are two additional variables that would assist school personnel in their attempts to identify potential dropouts. Recommendations Based on the data obtained in this study and the information acquired from reviewing the literature, the following recommendations for further research are made: 1. Replicative studies should be made by com“ paring Caucasian male high school dropouts and non-dropouts on the same criteria as that examined in the present study but using populations from areas other than Midwestern urban cities. 2. Replicative studies should be made by using the same criteria as that examined in the present study but comparing high school dropouts and non—dropouts whose race is other than Caucasian. _97_ Replicative studies should be made by using the same criteria as that examined in the present study but comparing female high school dropouts and non—dropouts. Similar longitudinal studies should be made, using upper elementary or junior high school students, to determine whether the variables examined in the present study have predict— ability by identifying students in other populations who will drop out of school prior to high school graduation. Many previous dropout studies have found significant differences between dropouts and non-dropouts on the basis of variables other than those examined in the present study. Only a few of these previous Studies, however, controlled for variables such as academic ability, socio—economic status, and grade—point average. It is recommended that replicative studies be made of these previous studies but first controlling for academic ability, socio—economic status, and grade—point average. —98— Based on the data obtained in this study and the information acquired from reviewing the literature, the following recommendations are made to public school teachers, counselors, and administrators: 1. Every possible effort should be made to identify —— as early as possible —— those individuals who are not performing at a satisfactory academic level, as well as those individuals who are not making satis~ factory social adjustments. Results from the present study and other recent studies indicate that dropouts are persons who experience numerous failures in both aca~ demic endeavors and social contacts. 2. Elementary school administrators should schedule regular periods of time when w teachers will be free from teaching responsibilities to meet individually with the parents of each student in their classes. Most of the students who drop out have a long history of school failure and are often two or more years behind their peers in reading level. Parents and teachers working in concert seems to ‘ be the optimum method of overcoming some of these problems. -99- High school counselors should make use of questionnaires or personality inventories as a means of assessing a student"s atti- tude toward himself, others, and his envi- ronment. Research indicates that many students who withdraw from school have low self—concepts and feel they are incapable of competing with their peers; they also feel that no one is concerned about them or their problems. Contacts should be made with those students who have dropped out of school to deter— mine the reasons for their leaving school. This contact would also serve as a means of encouraging these dropouts to return to school to complete their high school educa» tion. This contact might also indicate to the dropout that someone really does care. Counselors should spend more time doing personal-social counseling with students. Many articles reviewed indicated that stu— dents who had dropped out of school had not been encouraged by their counselors to remain in school. The major complaint of high school counselors seems to be that they have ~100— insufficient time to counsel with those students having the greatest need for their services. Closing Statement "Every dropout is a unique individual with unique characteristics . . ."1 This statement illustrates the difficulty involved in attempting to identify the poten= tial dropout. The basic problem, however, is not the identification of the potential dropout; this is merely a first step in defining the problem and its causes. The basic problem is how to make formal education a more meaningful and rewarding experience for all students. Most students need but minimal guidance and assistance in order for them to attain goals commensurate with their respective abilities; a few students, however, need con~ siderable guidance and direction before such results can be attained. The problem of helping those students who are in need of assistance is compounded by the fact that many of these students do not overtly seek the help that is available. challenge to school administrators and counselors; and lLucius F. Cervantes, The DrOpout: Causes and Cures (Ann Arbor: This is a situation which poses a tremendous The University of Michigan Press, 1965), pjml98. -lOl- unfortunately, too many counselors have not been trained to provide this type of student with the kind of help that is needed. In too many instances, the counselor is a former classroom teacher who has taken additional course— work in basic guidance procedures in order to qualify for a state guidance certificate; consequently, he may have neither the training nor the temperament to help those students who have the greatest need for a counselor"s assistance and understanding. It is believed that the task of making formal education a more meaningful and rewarding experience for all students can best be accomplished when competent administrators, dedicated teachers, adequately-trained counselors, and well—informed parents are all working in concert toward this goal. m P m G O I L B I B —102— BIBLIOGRAPHY Allen, Charles M., Combating the Dropout Problem (Chicago: Science Research Associates, Inc., 1956). Bayley, Monica, “A Renewed Effort to Solve the Problem of Dropout," Selected Reports and Statistics on School Dropouts, U. S. Department of Health, Education, and Welfare, Office of Education OE20063 (Washington: United States Government Printing Office, 1964), PP. 2—7. Bertrand, Alvin I., and Smith, Marion B., Environmental Factors and School Attendance (Baton Rouge: Louisiana State University, Bulletin No. 533, May, 1960). Bower, Eli M., A Process for Eerly Identification of Emo~ tionally Disturbed Childrep (Sacramento: California State Department of Education, 1958). “Self-Concept of Ability and School Achievement," U. S. Office of Education, Cooperative Research Project No. 845 (East Lansing: Office of Research and Publications, Michigan State University, 1962). Brookover, Wilbur B., LePere, Jean, Hamachek, Don E., W Thomas, Shailer, and Erickson, Edsel, "Self=Concept H of Ability and School Achievement, II," U. S. Office ‘ of Education, Cooperative Research Project No. 1636 j (East Lansing: Bureau of Educational Research Services, Michigan State University, 1965). Brookover, Wilbur B., Patterson, Ann, and Thomas, Shailer, Cervantes, Lucius F., The Dropout, Causes and Curee (Ann Arbor: The University of Michigan Press, 1965). Chamberlin, Gordon L., and Catterall, Calvin D., l"Acceleram tion for the Overage Potential DrOpout?“ Phi Delta Kappan, XLV (November, 1963), pp. 98—100. Conant, James B., Slums and Suburbs (New York: McGraw_ Hill Book Company, 1961). Cottle, William C., Life Adjustment Scale, No. l (Univer— sity of Kansas, 1953). ~103~ ~104= Davie, James 5., ”Social Class Factors and School.Attend— ance," Harvard Educational Review, (Summer, 1953). Davis, Donald A., “An Experimental Study of Potential Dropouts," The Personnel and Guidance Journal, XL (May, 1962), PP. 139—145. Dipasquale, Vincent C., "The Relation Between Dropouts and the Graded School,“ Phi Delta Kappan, XLVI (November, 1964); pp. 129-133. Dixon, Wilfrid J., and Massey, Frank J., Jr., Introduction 1 to Statistiqel Analysis (New York: McGrameill Book Company, 1957). Epps, Margaret W., and Cottle, William C., “Further Valida~ tion of a Dropout Scale,"I Vocational Guidance Quarterly, VII (Winter, 1958-59), pp. 90—92. Fink, Donald D., “The Efficiency of Certain Criteria in Predicting School Dropout,“ (Unpublished Doctoral Dissertation, Michigan State University, East Lansing, 1962). French, Joseph L., "High School Dropouts of High Ability,“ Vocational Guidance Quarterly, XIV (Winter, l965~66), ' pp. 123-136. Friedenberg, Edgar Z., The Vanishing Adolescent (New York: Dell Publishing Company, 1959). Goodman, Paul, Growing Up Absurd (New York: Vintage Books, 1956). Herrick, Virgil E., “Elementary Education Programs,“ gpcyclopedia of Educational Research, Chester W. Harris, editor (New York: The Macmillan Company, 1960). Herrman, W. L., “Partial Validation of a Dropout Scale,“ (Unpublished Master's Report, University of Kansas, Lawrence, 1957). Herrman, W. L., and Cottle, William C., "An Inventory to Identify High School Dropouts,” Vocational Guidance Quarterly, VI (Spring, 1958), pp. 122-123. Hollingshead, August B., Elmtown“s Youth (New York: John Wiley and Sons, 1949). ~105— Lambert, Nadine, "The High School Dropout in Elementary School,” Guidance and the School Dropout, Daniel Schreiber, editor (Washington: National Education Association and American Personnel and Guidance Association, 1964), pp.40—65. Larson, Roland S., "A Study of Selected Personality Factors Associated with High School Dropout," (Unpublished Doctoral Dissertation, Michigan State University, East Lansing, 1964). Lee, Beatrice Crump, "School Dropouts,"l NEA Research Division Bulletin, April, 1963. Matthews, C. V., and Bowman, Peter, The Motivations of Youth for Leaving School, Quincy Youth Development Project, Quincy, Illinois (Washington: Office of Education, U. S. Department of Health, Education, and Welfare, 1960). Myers, George R., "Curriculum Improvement Through Holding Power Follow—up Studies," Michigan Journal of Secondaryygducation, II (Winter, 1961), p. 110. O'Neill, John H., “High School Dropouts,“ Education, LXXXIV (November, 1963), pp. 156—159. Sando, Rudolph F., "A Comparative Study of Early School Leavers," (Unpublished Doctoral Dissertation, University of California, Berkeley, 1952). Schreiber, Daniel, editor, Guidance and the School Dropout, (Washington: National Education Association and American Personnel and Guidance Association, 1964). Siegel, Sidney, Nonparametric Statistics for the Behavioral Sciences (New York: McGraw-Hill Book Company, 1956). Stebbens, Marion, "Flint Offers the Potential Dropout a Personalized Curriculum," The Clearing House, XXXVIII (December, 1963), pp. 205-209. Thompson, Michael J., and Nelson, Robert H., ”Twelve Approaches to Remedy the Dropout Problem,“ The Clearing House, XXXVIII (December, 1963), pp. 200-204. U. S. Department of Health, Education, and Welfare, ”A New Look at School Dropouts," Health, Education, and Welfare Indicators, (April, 1964). Voland, Warner, -106= Maurice, editor, Conference Summary -= Feet Book, 1964 Conference on Rural Youth (East Lansing: Cooperative Extension Service, Michigan State University, 1964). 0. Ray, "The Scholastic Ability of School Drop— outs," Selected Reports and Statistics on School U. S. Department of Health, Education, Dropouts, and Welfare, Office of Education, OE20063, (Washington: United States Government Printing Office, 1964), pp- 11-13. APPENDICES ~107- APPENDIX A SELF-CONCEPT OF ABILITY - GENERAL Circle the letter in front of the statement which best answers each question. 1. How do you rate yourself in school ability compared with your close friends? a. I am the best b. I am above average c. I am average d. I am below average e. I am the poorest 2. How do you rate yourself in school ability compared with those in your class at school? am among the best am above average average am below average am among the poorest o m aratfw H+4F4H|4 3. Where do you think you would rank in your class in high school? a. among the best b. above average c. average d. below average e. among the poorest 4. Do you think you have the ability to complete college? yes, definitely yes, probably not sure either way probably not no (DQJOO‘SD 5. Where do you think you would rank in your class in college? a. among the best b. above average c. average d. below average e. among the poorest -108w —lO9- In order to become a doctor, lawyer, or university professor, necessary. work beyond four years of college is How likely do you think it is that you could complete such advanced work? a. very likely b. somewhat likely c. not sure either way d. unlikely e. most unlikely Forget for a moment how others grade your work. In your own opinion how good do you think your work is? o (DQJOO‘W What kind getting? my work is excellent my work is good my work is average my work is below average my work is much below average of grades do you think you are capable of mostly A's mostly B's mostly C's mostly D's mostly E's APPENDIX B PERCEIVED EVALUATIONS OF STUDENT“S ACADEMIC ABILITY BY OTHERS (PARENTS) Please answer the following questions as you think your PARENTS Would answer them. If you are not living with your parents answer for the family with Whom you are living. Circle the letter in front of the statement that best answers each queetion. 1. How do you think your PARENTS would rate your school ability compared with other students your age? a. among the best b. above average c. average d. below average e. among the poorest 2. Where do you think your PARENTS would say you would rank in your high school graduating class? a. among the best b. above average V 3‘ c. average “ d. below average e. among the poorest 3. Do you think that your PARENTS would say you have the? ability to complete college? yes, definitely yes, probably not sure either way probably not , definitely not (DQ-IOU’DJ —llO- -111~ In order to become a doctor, lawyer, or university professor, work beyond four years of college is neces- sary. How likely do you think your PARENTS Would say it is that you could complete such advanced work? very likely somewhat likely not sure either way somewhat unlikely very unlikely a m mvuo‘m What kind of grades do you think your PARENTS would say you are capable of getting in general? a. mostly A's b. mostly B's c. mostly C's d. mostly D's e. mostly E's APPENDIX C PERCEIVED PARENTAL EXPECTATIONS Circle the letter in front of the statement which best answers the question. How far do you think your PARENTS expect you to go in school? a. They expect me to quit as soon as I can. They expect me to continue in high school for a while. c. They expect me to graduate from high school. d. They expect me to go to secretarial or trade school. e. They expect me to go to college for a while. f. They expect me to graduate from college. g. They expect me to do graduate work beyond college. ~112~ APPENDIX D EDUCATIONAL EXPECTATIQN§ (PLANS) Circle the letter in front of the statement which best answers the question. Sometimes what We would like to do isn't the same as what we expect to do. How far in school do you expect you really Will go? a. I think I really will quit school as soon as I can. b. I think I really will continue in high school for a While. ; , c. I think I really will graduate from high ; school. d. I think I really will go to secretarial or trade school. e. I think I really will go to college for a While. } f. I think I really will graduate from college. g. I think I really will do graduate work beyond I 1 college. ~113~ WHO WHO I Non- Tral 7“” Non- Tral Gle‘ APPENDIX E A COMPARISON OF ELEVENTH GRADE TRANSFER STUDENTS WHO DROPPED OUT AND ELEVENTH GRADE NON—TRANSFER STUDENTS WHO DROPPED OUT ON THE BASIS OF DIFFERENT GRADE-LEVEL DATAl Grade—Point Average Groups Totals High Low Non-Transfers 4 17 21 Transfers 2 19 21 Totals 6 36 42 X2 = .19 d.f. Not Significant at .05 Level Groups Socio-Economic Status Totals, i High Low , Noanransfers 6 15 21 Transfers 7 14 21 Totals 13 29 42 X2 = .00 l d.f. Not significant at .05 Level lTenth grade data was used for non—transfer students; eleventh grade data was used for transfer students. —ll4— ~115~ Groups Self—Concept of Ability Totals High Low Non~Transfers 5 16 21 Transfers 6 15 21 Totals 11 31 42 X2 = .00 l d.f. Not Significant at .05 Level Groups Perceived Parental Evaluations Totals High Low Noanransfers 10 11 21 Transfers 4 17 21 Totals 14 28 42 2 X = 2.68 l d.f. Not Significant at .05 Level Perceived Parental Expectations Grou s . P ngh Low Totals Non—Transfers 7 14 21 Transfers 7 14 21 Totals 14 28 42 2 , . , . X = .17 l d.f. Not Significant at .05 Level Educational Plans Groups Totals High Low Noanransfers 6 15 21 Transfers 5 16 21 Totals 11 31 42 X2 = 00 l d.f. Not Significant at .05 Level APPENDIX F A COMPARISON OF CAUCASIAN MALE HIGH SCHOOL STUDENTS WHO DROPPED OUT DURING THE TENTH GRADE AND THOSE WHO DROPPED OUT DURING EITHER THE ELEVENTH OR TWELFTH GRADE Academic Ability -117~ Dropout Groups Totals High Low 10th Grade 14 39 53 11th or 12th Grade 17 25 42 Totals 31 64 95 X2 = 1.51 l d. Not Significant at .05 Level SociomEconomiC Status Dropout Groups Totals High Low 10th Grade 15 38 53 11th or 12th Grade 13 29 42 Totals 28 67 95 X2 = 26 l d f Not Significant at .05 Level -118— Grade-Point Average Dropout Groups Totals High Low 10th Grade 2 51 53 11th or 12th Grade 6 36 42 Totals 8 87 95 X2 = 2.12 1 d.f. Not Significant at .05 Level Self-Concept of Ability Dropout Groups Totals High Low 10th Grade 7 46 53 l l 11th or 12th Grade 10 32 42 H 1 l Totals 17 78 95 X2 - 2.60 l d.f. Not Significant at .05 Level -1l9- Perceived Parental Evaluations Dropout Groups Totals High Low ' 10th Grade 15 38 53 11th or 12th Grade 14 28 42 Totals 29 66 95 2 . . . X = .44 l d f. Not Significant at .05 Level Perceived Parental Dropout Groups Expectations Totalsl High Low 10th Grade 7 36 43 11th or 12th Grade 14 28 42 Totals 21 64 85 X2 = 3.32 l d.f. Not Significant at .05 Level 1Ten of the 53 tenth grade dropouts did not respond to Perceived Parental Expectations and Educational Plans questionnaire items. —120— Educational Plans Dropout Groups Totals High Low 10th Grade 7 36 43 11th or 12th Grade 11 31 42 Totals 18 67 85 X2 = 2.01 l d. Not Significant at .05 Level 300%! (BE DNLY,__. "I11111111111“