University of Southern California Dissertations and Theses

An Attitude Survey Of High School Dropouts By Means Of The Semantic Differential Process

(USC Thesis Other)

An Attitude Survey Of High School Dropouts By Means Of The Semantic Differential Process

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content This dissertation has been microfilmed exactly as received 6 6 -1 1 ,5 9 2 STREM, Bruce Elwood, 1924- AN ATTITUDE SURVEY OF HIGH SCHOOL DROPOUTS BY MEANS OF THE SEMANTIC DIFFERENTIAL PROCESS. U niversity of Southern California, Ph.D ., 1966 Education, psychology University Microfilms, Inc., Ann Arbor, M ichigan AN ATTITUDE SURVEY OF HIGH SCHOOL DROPOUTS BY MEANS OF THE SEMANTIC DIFFERENTIAL PROCESS by Bruce Elwood Strem A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (Education) June 1966 UNIVERSITY O F S O U T H E R N C A LIFO RNIA TH E GRADUATE SC H O O L UNIVERSITY PARK LOS A NGELES, C A LIFO R N IA 9 0 0 0 7 This dissertation, written by Bruce Elwood Strem under the direction of hXS—Dissertation Com mittee, and. approved by all its members, has been presented to and accepted by the Graduate School, in partial fulfillment of requirements for the degree of D O C T O R OF P H I L O S O P H Y Dean DISSERTATION COMMITTEE TABLE OF CONTENTS LIST OF TABLES Page Iv Chapter I. STATEMENT OF THE PROBLEM 1 Introduction Need for the Investigation Background of the Problem Statement of the Problem Statistical Hypotheses and Summary of Dropout Research Literature Hypotheses Anticipated Outcomes Presuppositions Delimitations Limitations Procedure Definitions of Terms Organization of the Remaining Chapters An Overview Extent of the Problem at Various Levels Family Correlates Physical Factors Absence from Class Intellectual Factors Emotional Concomitants Social Interaction Values, Attitudes and Motivation Attitude Measurement Semantic Differential Process Current Predictors of Dropouts Chapter Summary II. REVIEW OF THE LITERATURE 31 III. SOURCE OF DATA AND METHODOLOGY............. 176 Chapter Page Nature of the Sample Description of the Meaning of Words Test Selection of Concepts The Meaning of Words Test Research Design Statistical Treatment of Data Independent Variables Dependent Variable Chapter Summary IV. FINDINGS.................................... 237 V. SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS . . 303 Summary Summary of Major Findings Conclusions Recommendations Recommendations for Further Study BIBLIOGRAPHY........................................ 332 APPENDIX A. The Test Instrument: The Meaning of Words T e s t ......................... 352 APPENDIX B. Tabulation of Data for Seven Selected Descriptive Variables ................ 372 APPENDIX C. Tabulation of Mean Summed Factor Scores for Dropouts and Stay-ins, Tenth and Eleventh Grades, and Boys and Girls When Tested on the Evaluative, Potency and Activity Scales........ 380 APPENDIX D. Means and Standard Deviations of Summed Factor Scores for the Treatment Groups, Grade Level and Sex of the Student....................... 387 iii LIST OF TABLES Table Page 1. Number of High School Graduates for Every 1,000 Fifth Grade Pupils Enrolled Eight Years Earlier............................. 38 2. Number of Ninth Grade Pupils Not Enrolled, for Every 1,000 Eighth Grade Pupils Enrolled the Previous Year ..... ................... 51 3. Proportion of High School Graduates........ 52 4. Proportion of High School Age Youth Attending High School ............ 53 5. Class Status and Persistence in School .... 57 6. Proportion of Dropouts by Grade Level .... 59 7. Parent Education................ 71 8. Education of Parents . .......... 73 9. Years of Parent Education ........ ..... 73 10. Schools Attended and Percentage of Dropouts and Stay-ins................................ 83 11. IQ Scores for Dropouts and Non-Leavers .... 93 12. IQ Scores of Dropouts and High School Graduates.................................. 94 13. Nationwide IQ Scores for Dropouts.......... 95 14. National Statistics on Grade Retention .... 99 15. Dropout Grade Level ................ 101 16. Dropout Grade Achievement ................... 102 17. Dropout Grade Achievement ................... 103 iv Table Page 18. Distribution of Dropout Sample .............. 178 19. Summary of Averages of Selected Descriptive Variables ............................ 239 20. Median Test for Independence of Socioeconomic Status Rating......................... .. . 240 21. Mtn Ratios for Mean Differences Between Drop outs and Stay-ins for Selected Descriptive Variables.................................. 242 22. nt” Ratios for Mean Differences Between Boys and Girls for Selected Descriptive Variables 243 23. Percentage of Mean Number of Days Absent . . . 247 24. Analysis of Variance of Group Test Intelli gence Quotient Scores Between Treatment Groups, Grade Levels and Sex of the Student 250 25. Intelligence Quotient as a Correlate of Con cept Ratings on Selected Factor Scales for the Total Sample Population of Dropouts and Stay-ins................................ 252 26. Analysis of Dropout and Stay-in Mean Summed Factor Scores of Concepts Rated Against Selected Factor Scales................... . 256 27. Analysis of Tenth and Eleventh Grade Mean Factor Sum Scores of Concepts Rated Against Selected Factor Scales ..................... 258 28. Analysis of Boy and Girl Mean Factor Sum Scores of Concepts Rated Against Selected Factor Scales.......... 269 29. Treatment Groups and Grade Levels: Two-Way Interaction Effect, F Ratios for Evaluative, Potency and Activity Factor Sum Scores . . . 272 30. Treatment Group by Grade Level Two-Way Inter action Effect: Combined Mean Scores for Selected Concepts and Factor Scales for Dropouts and Stay-ins Compared with Tenth and Eleventh Grade Levels ............... 274 v Table Page 31. Treatment Groups and Sex of the Student: Two- Way Interaction Effect, F Ratios for Evalua tive, Potency and Activity Factor Sum Scores 278 32. Treatment Group by Sex of the Student Two-Way Interaction Effect: Combined Mean Scores for Selected Concepts and Factor Scales for Dropouts and Stay-ins Combined with Boys and Girls . ................................ 280 33. Grade Level and Sex of the Student: Two-Way Interaction Effect, F Ratios for Evaluative, Potency and Activity Factor Sum Scores . . . 282 34. Treatment Groups and Grade Level and Sex of the Student: Three-Way Interaction Effect, F Ratios for Evaluative, Potency and Activ ity Factor Sum Scores................. 284 35. Treatment Group by Grade Level by Sex of the Student: Three-Way Interaction Effect-- Mean Scores for Selected Concepts and Factor Scales for Dropouts and Stay-ins Compared with Tenth and Eleventh Grade Pupils and with Boys and Girls ................. 286 36. Frequency Distribution of Father Occupation Rating.................................. 372 37. Group Test Intelligence Quotient Scores . . . 373 38. Age in Months at Time of Testing........ 374 39. Mean Standard Score of Subtests 5, 6, 7 of the Iowa Tests of Educational Development . . . 375 40. Total Absences During Period of Enrollment . . 376 41. Total Number of Days Enrolled in School . . . 377 42. Dropout Leaving Age in Months.......... 378 43. Means and Standard Deviations of Summed Evalu ative Factor Scores for Tenth Grade Dropouts and Stay-ins............................ 380 44. Means and Standard Deviations of Summed Potency Factor Scores for Tenth Grade Drop outs and Stay-ins ........ ........ 381 vi CHAPTER I STATEMENT OF THE PROBLEM INTRODUCTION A disquieting truism faces the educators today as it did their predecessors. Each year a certain portion of the school population is diminished by those who leave before completing their prescribed courses of study. Frequently, these boys and girls, upon leaving school, are absorbed in the maelstrom of life about them and everything is lost. The fact that the schools have not made a concerted effort to prepare these early leavers adequately to face their occupational and social problems is a blot upon an enlightened era of education. . . . (Green, 1961:173) A Fundamental Concern Ristow (1964), in a discussion of the school1s responsibility for the mounting problem of school dropouts, remarks incisively that it is unfair and unreasonable to chastise the public schools for failing in a task that could not reasonably be considered rightfully theirs: Assuming such responsibility would imply that the school has a greater responsibility than that of the parents, the community, welfare agencies, youth assistance organizations, the church, busi ness and commercial interests and not the least of all, the individual concerned. The basic issue is whether the individual exists for the 1 good of society or whether society exists for the good of the individual. The entire burden of responsibility for the dropout cannot be borne by either school or community. Ideally, remediation must begin with the attitudes and values created by the home and community, because it is too late to apply corrective measures when the pupil reports that he is quitting school, or when a number of symptoms appear together and the student becomes a candidate for adminis trative dismissal. Certainly, as Ristow (1964) observed, institutions have a responsibility for evaluating their product, "but to assume that the school must be a panacea is unthinkable." Acceptance of the moral responsibilities of adulthood becomes increasingly important as youngsters come of age. Faris (1928:801) brought this concept to the attention of public school educators from the field of social psychology, suggesting that the school assume a greater role in attending to the needs of children. But, again, "to suggest that the school be responsible for the entire job is ridiculous." Herrold, discussing pupils who do not complete their course of study, observed that change in education is not achieved with one great call for intellectual and technical rearmament: It is brought about by the individual only when new needs are clearly and operationally defined, when there is readiness to discard concepts and behaviors no longer useful, and when there is total support for the required learning and change. (1964:8) 3 Reed wrote that the school environment influences a pupil's self concept through his perception of what school meant to him. In this respect, it is important to know whether the pupil is accepting or rejecting, open or defensive toward the things the school represents and expects of him. Reed believed that we do not need to know for the moment why the pupil feels the way he does. If he feels holistically that the school is a threatening place, reading, sitting, and other activities will be rejected, not because he is opposed to them individually, but because they are symbols of something he dislikes and therefore he reacts in a manner which adults dislike, which in turn convinces him that he was right in the first place— that is, teachers and school are just no good. (1962:3) The Potential Dropout Lack of interest in school and especially lack of interest in cocurricular activities often characterizes the potential dropout. There is an implication in the literature that pupils lack interest because of some fault with the curriculum or the educational structure. Little is written about the individual's own lack of maturity and underdeveloped sense of responsibility. Studies of older youth and young adults returning to adult schools clearly indicate that the dropouts of five to ten years earlier are enrolling in the very same course work they failed previ ously. If the curriculum were as inadequate as we are led to believe, they why do these young adults return to retake the same course work (Los Angeles Unified School Districts, 1963:2)? Adequate functioning in this society requires certain bits of knowledge. The public schools are aware of -these needs, and have been alert to them for scores of years, because education, by definition, is in part the transmission of the needs of the society, whether academic or cultural. The statistics to be presented later in this report will verify that increasing progress has been made in conveying this information to the youth of the nation, indicated by the later leaving-age and higher grade-on- leaving. The symptoms of tardiness, unexcused absences, discouragement with studies, lack of initiative or self- confidence, nonparticipation in class, and academic failure, are but a few of the superficial, observable manifestations of much deeper feelings, attitudes and values. This study is an attempt to evaluate some of these attitude-value systems, to determine whether or not the semantic differential process can differentiate between pupils who remain in school and those who drop out. Ristow (1964) comments that it is doubtful that youth could be made to realize the value of the curriculum or school by simply changing the program. This realiza tion, he notes, requires not the alteration of the curric ulum but rather a recognition of its value. It may be unrealistic to require that we place emphasis upon waiting for youngsters to mature emotionally and personally, and create an atmosphere within the present school setting in which the individual will have an opportunity to develop new values and attitudes. It may well be that the only reasonable answer is to let the pupil drop out of school for a few years, gain needed maturity and insight, and establish for himself some measure of goal orientation. This leads logically to a consideration of develop ing better ways of preserving favorable attitudes toward school and facilitating the return of the dropout. This is evidenced by. the magnitude of the adult education program and the numbers of young adults returning for further education. Lichter (1962:2) stated that "*educational' dis ability, moreover, cannot be attributed simply to limited mental endowment." About half of all dropouts have the mental capacity to complete a high school curriculum successfully. Elliott (1963) believed that the dropout was running away from something, and that dropping out was a positive attempt at corrective action. Nachman (1963:35), on the other hand, felt that dropouts, in leaving school, were taking positive steps toward a successful future, not running away from anything. Glasser (1964) stressed the point that the youngster who dropped out was attempting to find an environment that to him was "reality." Finding himself unable to establish a real relationship within the school setting (probably assisted by a lack of reality -contact within the home) he attempts to find reality out side of the school. The gap between the potential dropout and his perception of the school environment is widened as the school and community display growing impatience with youth who have problems. "Impatience is especially strong when the youngster's behavior is of an aggressive, attack ing, or clowning nature" (Lichter, 1962:268). The delicate balance between the welfare of the majority and the welfare of the individual places teachers and school administrators in a difficult position (Faris, 1928:800). A Transition in Dropout Research Fenty gave a concise account of the changes in emphasis of dropout research: The early American Council on Education studies (which include The Inventory of Youth in Pennsyl vania, prepared by Harlan Updegraph in 1936; tne 'Maryland Study, Youth Tell Their Story by Howard Bell in 1938; and the Report of the New York Regents' Inquiry. When )fouth Leave School."made by Ruth Eckert and Thomas Marshall in 1938) served as models in many respects for subsequent studies of high school dropouts. (1956:2) These studies established a pattern of contacting the dropout individually in efforts to determine through personal contact the expressed reasons for having left school, as well as attempting to evaluate the underlying reasons for dropping out. The National Education Associa tion (1961:6) reported that youth who dropped out did not readily give out information about their problems. Many gave vague and meaningless reasops--£requently relating the final item in a long series of incidents as the basis for leaving. In many instances the dropout was probably unaware of the real reason for his decision to leave. Elliott (1963) commented that there was little point in asking a dropout where he was going; rather, one needed to know ■'where he had been emotionally." An evaluation of these factors might well tell us more clearly than his words where he is going. Dropouts most often will give two reasons for leaving: financial need (which may include anything from helping with family support to buying a car), and dissatis faction with or dislike for school. The latter is an all- inclusive statement reflecting an attitude about school. This may mean that the dropout is discouraged with his academic progress, that he has developed a decided dislike for certain teachers, or it may mean that he sees no real value in education, and almost certainly feels excluded from school social life. Student Attitudes The researches of Munoz (1957), Lee (1958), Shanley (1961), Charlap-Hyman (1964) and others signify a trend toward procedures that utilize the evaluation of percep tions and attitudes as they affect persistence in school. The attitude that causes a student to respond in a given manner Is more than the sum of his perceptions and motiva tions. It is an organization of many different aspects of the individual, not an itemized list of characteristics (Argyris, 1954:21). If change in attitude and consequent change in personality is to occur, then one must incor porate new ideas (about the self and about school) into a remodeled self concept. Lecky (1954) based his theory of personality on the individual's tendency to assimilate only those perceptions that are consistent with the existing personality. Argyris (1954:23) stated that this resistance to change guaranteed a degree of constancy, and that the individual's specific personality or attitude can be inferred from the constancy of his behavior. Schneider and Lysgaard (1953) and Rosen (1956) presented views regarding the influence of middle-class attitudes and values upon the desire to achieve and upon success. According to these authors, among the middle- class adolescents there exist characteristic patterns of attitudes. Schneider and Lysgaard (1953) referred to the "deferred gratification pattern," and Rosen (1956) identi fied a similar willingness to postpone present gratifica tions for later rewards as the "achievement syndrome." Schneider and Lysgaard (1953:142) felt that the effects of social class on the education achievement depended upon the factors that entered into the development of the attitude-pattern rather than upon the more direct effects 9 of teacher-pupil interpersonal relationships, as Rollings- head (1949) implied in the study of Elmtown. The presence of these attitude-patterns appears to he an essential characteristic that facilitates the differentiation of youth who will achieve in accordance with established middle-class standards. NEED FOR THE INVESTIGATION Stice wrote in his discussion of the homogeneity of the dropout group: If it [the dropout group] were found to be random it might have been possible to conclude that a kind of "residual” level had been reached and that much further improvement could not be expected. Instead, dropout is closely associated with all the logical and obvious environmental factors, includ ing family characteristics; school, community, and regional characteristics; peer group orientation; and the student's own self-picture. These, of course, are the same factors that numerous studies have shown must be taken into consideration in predicting whether or not a high school graduate will enter college. We may therefore expect that any early efforts to "conserve manpower resources," made with children in grade school or in junior high school, for example, will bear fruit all the way through high school and college. (1960:12) Identification of Attitudinal Concomitants During interviews with dropouts, Sando (1952) noted that certain questions elicited critical reactions or attitudes that seemed to distinguish between the dropout and the stay-in. Perfection of an instrument capable of quantifying the magnitude and the direction of these less 10 tangible signs associated with the decision to leave school early coqld, conceivably, be a significant contribution to the betterment of education. Lichter (1961:247) wrote that the early detection of emotional factors and an orientation to factors that contribute to learning difficulties and poor school adjust ment are essential to staying in school. The Maryland Study (1963) included in its comments about the student's attitude toward school and the learning process that many children who drop out later in high school develop patterns and habits which may be recognized in elementary school as symptoms which are peculiar to the dropout. Children not succeeding in school and who see themselves as poor students or that their work is too hard or of little interest, develop at an early age an attitude against school. (Reed, 1962:4) Lichter commented that a student who has always failed has a much greater tolerance for failure than students who have rarely failed. The first has learned to accept the fact, the second finds it a new experience. Prediction of dropout from data other than that available from school and community records is indicated. Case study and Interview techniques are perhaps the most valuable; however, the expense in time and personnel renders these approaches impractical for the study of large groups. Two currently available tests have been used in related dropout research. Reference to the Illinois Inventory of Pupil Opinion (World Book Company, 1948) has appeared in the doctoral studies of Sando (1952) and Lee (1958). Certain items have demonstrated a correlation with early school leaving, suggesting that this inventory may have value as a predictor of dropouts. The School Interest Inventory (Cottle, 1963) offers the first direct approach to the identification of the potential dropout. This inventory appears to be the first commercially available test specifically designed to predict dropouts. The valid ity of these instruments requires that the individual answer accurately and truthfully. Several disadvantages exist with question-and-answer type tests. A yes-no forced choice series of items tends to obscure information which may be quantitative in nature. Subjects may avoid a nnoM response because the situation may not be absolutely false. The sophisticated test taker can usually identify the socially acceptable response. Biased response patterns may well result from the drive for social acceptance expe rienced by middle-class youth, according to Schneider and Lysgaard (1953) and Rosen (1956). The semantic differential process involves no directly observable right or wrong answers. The subject responds according to the way he feels about the particular concept being rated. The oblique nature of rating concepts against bipolar adjectives requires that the subject respond in accordance with his attitudes and feelings or values. Osgood (1957:89) declared that it is possible (and practical) to measure attitude as one of the major dimen sions of meaning In general and thus to extend measurement of semantic differential to the area of social psychology. The semantic differential procedure Is a generalized measure of attitude; an estimate of the way a person per ceives himself in his environment. The adult dynamics of personality and attitude with respect to one's self are well established by ages sixteen or seventeen. The semantic differential procedure is based upon selected bipolar adjectives identified by a group of 100 college students as having meaning. We must assume that the dimensions of attitude expressed by the sample will be representative of the standardizing group. Studies by Suci (1960) and others have substantiated the stability or generalizahility of the semantic differential procedure. There is, however, no generally-accepted and con venient measure in wide use today that has significant value for the prediction of those pupils whose attitudes and temperament predispose them to leave school before graduating. 13 BACKGROUND OF THE PROBLEM While efforts at early identification of school maladjustment are being made in the elementary grade levels, there is a present need to help those pupils not enrolled in high school but who will not complete the school year because of an undetected chronic sense of frustration, failure and defeat (Liddle, 1962:280). Gragg (1950), Sando (1952), Lee (1958), and Lichter (1961) offer comments representative of previous literature attesting to the lack of correlation between manifest behavior and persistence or nonpersistence in school. Sando (1950), Cook (1956), Lee (1958), and Shanley (1961) discuss the inadequacies of questionnaires, opinion- naires and interview techniques in assessing pupils expressed explanations for early school leaving. These investigators emphasize the pupil's reluctance (undoubtedly the result of inability or unawareness) to assess criti cally their own ability, the school, the teachers, or peer and family relationships. These statistical studies have described but one aspect of the factors associated with the predisposition to drop out of school. Linton (1945:112) asserts that attitude-value systems operate automatically and, for the most part, subconsciously. Sando (1952:263) noted a need to assess attitudes and motivations if effective dropout prediction is to be accomplished. Livingston (1958:202) observes that 14 there is as yet no generally available device for the assessment of attitudes, values, and motivation. This lack of an effective attitude“measuring device persists even today. STATEMENT OF THE PROBLEM The purpose of this study is to identify critical attitudinal variables in the motivational patterns of dropouts and stay-ins among tenth- and eleventh-grade high school students in three urban high schools. STATISTICAL HYPOTHESES AND SUMMARY OF DROPOUT RESEARCH LITERATURE This section presents a concise summary of research related to the seven areas of interest in this investiga tion: school experience, family relations, social interac tion, self concept, authority relations, goal orientation, and moral and social values. Each area is further described in terms of the specific concepts selected for study. The null hypothesis is advanced for all dependent variables. Results of the review of related research are summarized in Figure 1, which indicates the favorableness or unfavorableness with which stay-ins (S/I) and dropouts (D/0) respond with respect to specific stimuli. Direc tional alternative hypotheses may then be advanced in FIGURE 1 15 STATISTICAL HYPOTHESES AND SUMMARY OF DROPOUT RESEARCH LITERATURE ^Direction IndicatecT Hypoth- Area of by Related Dropout esls Investigation ________. ____ Research* Date Number School Experience 7.0 Teachers S/I> D/0 Aseltine** 1941 7.1 Hollingshead 1949 Sando 1952 McCreary, Kitch 1953 Cook 1956 Penty 1956 Munoz 1957 Ullmann 1957 Lee 1958 Bowman, Matthews 1960 Shanley 1961 Liddle 1962 Lee 1963 Grades S/I>D/0 Pond 1953 7.2 Cook 1956 Lee 1958 Bowman, Matthews 1960 Liddle 1962 School S/I> D/0 Beckstead 1940 7.3 Aseltine 1941 Sando 1952 Cook 1956 Snepp 1956 Ullmann 1957 Livingston 1958 Van Dyke, Hoyt 1958 Bienstock 1961 Wolfbein 1959 Bowman, Matthews 1960 Shanley 1961 Lichter 1962 Maryland Study 1963 Plott 1964 *S/I = Stay-ins; D/0 = Dropouts. **Tit'ies listed in bibliography. FIGURE 1 (continued) 16 STATISTICAL HYPOTHESES AND SUMMARY OF DROPOUT RESEARCH LITERATURE Direction Indicated Hypoth- Area of by Related Dropout esis Investigation______________ Research Date Number School Experience (continued) 7.0 Reading S/l>D/0 Lanier 1949 7.4 Warren 1954 Snepp 1956 Pond 1953 S/I= D/O Thomas 1954 S/i> D/O Penty 1956 Lee 1958 Schreiber 1962 Ideal Teacher No related research 7.5 Family Relations 8.0 Home S/I> D /0 Dillon 1949 8.1 Cook 1956 Peters 1957 Ullmann 1957 Epps, Cottle 1958 Ideal Parent - S/I>D/0 Abbott 1939 8.2 S/I>D/0 Dresher 1954 S/I= D/0 Charlap-Hyman 1964 Parents S/I> D/0 Schneider, Lysgaard 1953 8.3 Snepp 1956 Munoz 1957 Ullmann 1957 Liddle 1962 Mannino 1962 Social Interaction 9.0 Most People No related research 9.1 My Best Friends S/I>D/0 Lee 1958 9.2 Classmates S/l>D/0 Hollingshead 1949 9.3 Sando 1952 McCreary, Kitch 1953 Munoz 1957 Epps, Cottle 1958 FIGURE 1 (continued) 17 STATISTICAL HYPOTHESES AND SUMMARY OF DROPOUT RESEARCH LITERATURE Direction Indicated Hypoth- Area of by Related Dropout esis Investigation______________ Research Date Number Social Interaction (continued) 9.0 Classmates Lee 1958 (continued) Livingston 1958 Steadman 1959 Bowman, Matthews 1960 Swan 1961 Liddle 1962 Plott 1964 Grownups S/I> D/0 Lee 1958 9.4 Shanley 1961 Self Concept 10.0 Me S/I> D/0 Abbott 1939 10.1 Allen 1956 Penty 1956 Ullmann 1957 Livingston 1958 Lichter 1962 Liddle 1962 Huber 1963 Maryland Study 1963 My School S/I> D/0 Aseltine 1941 10.2 Ability Penty 1956 Epps, Cottle 1958 Lefever 1961 How I Would S/I>D/0 Ullmann 1957 10.3 Like to be Charlap-Hyman 1964 How My Class S/I> D/0 Abbott 1939 10.4 Sees Me Sando 1952 Penty 1956 Livingston 1958 Gordon 1959 Swan 1961 FIGURE 1 (continued) 18 STATISTICAL HYPOTHESES AND SUMMARY OF DROPOUT RESEARCH LITERATURE Direction Indicated Hypoth- Area of by Related Dropout esis Investigation ______ Research Date Number Authority Relationships 11.0 Authority S/I>D/Q Beckstead 1940 11.1 Sando 1952 Schneider, Lysgaard 1953 Cook 19 56 Peters 1957 Bloch, Neiderhoffer 1958 Cassel, Coleman 1962 Greene 1962 Charlap-Hyman 1964 Rules S/I>D/0 Schneider, Lysgaard 1953 11.2 Lee 1958 Lichter 1962 Punishment S/l>D/0 Sando 1952 11.3 McCreary, Kitch 1953 Nachman 1963 Goal Orientation 12.0 Future S/I> D/0 Schneider, Lysgaard 1953 12.1 Rosen 1956 Argyrus 1957 Kvaraceus, Miller 1959 Bowman, Matthews 1960 Liddle 1962 College S/I>D/0 Schneider, Lysgaard 1953 12.2 Shanley 1961 A Job D/0> S/I Hollingshead 1949 12.3 Dillon 1949 Hecker 1953 Schneider, Lysgaard 1953 Cook 1956 Penty 1956 Rosen 1956 Bowman, Matthews 1960 Lombardi 1960 Savitsky 1961 Swan 1961 FIGURE 1 (concluded) 19 STATISTICAL HYPOTHESES AND SUMMARY OF DROPOUT RESEARCH LITERATURE Direction Indicated HypofH^ Area of by Related Dropout esis Investigation Research Date Number Goal Orientation (continued) 12.0 A Job (continued) Lichter 1962 Liddle 1962 Graduating S/l>D/0 Schneider, Lysgaard 1953 12.4 Bowman, Matthews 1960 Shanley 1961 Quitting School D/0> S/I Dresher 1954 12.5 Munoz 1957 Ullmann 1957 Lee 1958 Cohen 1961 Moral and Social Values 13.0 Trying Hard S/I>D/0 Munoz 1957 13.1 Kvaraceus, Miller 1959 Savitsky 1961 M&ryland Study 1963 Cheating D/0>S/I Schneider, Lysgaard 1953 13.2 Argyrus 1957 Something Easy D/O >S/I Schneider, Lysgaard 1953 13.3 Woollatt 1961 Something S/I>D/0 Allen 1949 13.4 Important Penty 1956 Shanley 1961 Money D/O>S/I Ellis 1903 13.5 Hollingshead 1949 Schneider, Lysgaard 1953 Allen 1956 Snepp 1956 Epps, Cottle 1958 Swan 1961 Liddle 1962 Success S/I>D/O Ullmann 1957 13.6 Livingston 1958 Bowman, Matthews 1960 Woollatt 1961 Liddle 1962 20 accord with the findings of prior research. In Figure 1, the notation (S/I>D/0) indicates that stay-ins respond more favorably than dropouts with respect to the area of investigation. The notation (S/I = D/O) indicates no significant difference between the responses of the stay-ins and dropouts with respect to the given area of investigation. HYPOTHESES Descriptive Variables The following research hypotheses (H-^) have been advanced in accordance with the reviewed literature for selected descriptive variables. 1.0 Dropouts, in general, come from families of lower socioeconomic status rating than pupils who stay in school. 2.0 Group test IQ scores will not differ signifi cantly between groups. 3.0 Pupils who later drop out of school will be older at the time of testing than those who remain in school. 4.0 Dropouts will receive lower reading index scores than will stay-ins. 5.0 Dropouts will be absent from school a greater number of times than will stay-ins. 6.0 Dropouts will be enrolled in school for fewer days during the school year than will stay- ins. Dependent Variables The Null Hypothesis (Ho) has been advanced for the mean summed factor scores (evaluative, potency and activ ity) of the Meaning of Words Test for each of the following areas of investigation. 7.0 School experiences 7.1 Teachers 7.2 Grades 7.3 School 7.4 Reading 7.5 Ideal teacher 8.0 Family relationships 8.1 Home 8.2 Ideal parent 8.3 Parents 9.0 Social interaction 9.1 Most people 9.2 My best friends 9.3 Classmates 9.4 Grownups 10.0 Self concept 10.1 Me 10.2 My school ability 10.3 How I would like to be 10.4 How my class sees me 11.0 Authority relationships 11.1 Authority 11.2 Rules 11.3 Punishment 22 12.0 Goal orientation 12.1 Future 12.2 College 12.3 A job 12.4 Graduating 12.5 Quitting school 13.0 Moral and social values 13.1 Trying hard 13.2 Cheating 13.3 Something easy 13.4 Something important 13.5 Money 13.6 Success ANTICIPATED OUTCOMES The experimental design of this study was struc tured in terms of certain expectations. It was anticipated that: 1. Attitudinal variables expressed in terms of the mean summed factor scores will differentiate those pupils who drop out and those who stay in regular high school classes. 2. The semantic differential process will be validated as an effective means of identifying attitudinal variables that are associated with early school leaving and with persistence in school. 3. The association of factor score response paradigms with ex post facto dropout will provide a contribution to the prediction of 23 persistence in school. PRESUPPOSITIONS Fundamental to the research design were certain assumptions or presuppositions: 1. The dropouts represent a random sample of all factors affecting early school leaving. 2. Attitudes and values are developed in the drop out and stay-in groups to a degree commensurate with that of the standardizing population from whose responses the bipolar adjectives were derived. 3. The smaller the mean summed factor score, the more negative the attitude toward the concept being rated. 4. Home, community and school experiences as per ceived by the pupil are the motivating factors which influence the decision to drop out or to remain in school. 5. Responses to the Meaning of Words Test repre sent the generalized attitudes of the subjects toward selected concepts in such a way that the attitudes associated with dropping out and staying in school can be identified. 24 DELIMITATIONS 1. This study dealt with the dropouts from grades ten and eleven from three high schools in a southern California community of 114,000 population. 2. The stay-in group consisted of a random sample matched for IQ score (89 and below, 90-110, and 111 and above), grade level, and sex drawn from each grade level at the close of the 1963 school year. 3. This was a cross-sectional study; the test was given and supplemental information obtained upon one occa sion in the Fall semester of 1963. 4. The investigation concerned those attitudes, expressed as factor scores, that pertained to school, family, peers, self, authority, goals and moral and social values which are associated with, but do not necessarily determine, the fact that some pupils drop out of regular high school classes during the academic year. LIMITATIONS 1. Osgood (1957:31) stated that a problem approach in this way concerned the representativeness of the sample. This type of study differs from other factor studies in that it must deal with three sources of variability: sub jects, scales and concepts judged. 2. The validity of the semantic differential 25 procedure has not been well established according to the f authors consulted. 3. The selected stimulus items may represent personality dimensions other than those factor loadings upon which the standardizing population were established. 4. Any single measure administered only once evaluates the way a person perceives himself in his environment at the particular moment. 5. There was no control over the day of the week, nor the time of day at which the Meaning of Words Test (MWT) was administered. The MWT was administered as a part of the required state-wide testing program, for which there is no established sequence. The MWT was administered when ever the testing supervisor felt a break in activity was needed or when there was sufficient time to include it at the end of another test period. The biases introduced by this procedure could not be evaluated. PROCEDURE Details of test construction and administration are described in Chapter Three. The Meaning of Words Test was administered to all tenth and eleventh grade pupils in three high schools of one district in southern California, as a part of the regular state-wide testing program. The counselors and teachers responsible for the supervision of the group testing program at each school met with the 26 examiner a day or two ahead of the testing period to become acquainted with the test and to review its administration. The overall testing extended over a three-week period, each school being assigned one week in which to administer the required group tests. All tests for a given grade level were pooled. The names of the dropouts were secured by individual reports from each school throughout the year, concluding in June. There was no control over the day of the week, nor the time of day when the MWT was administered. Teachers and counselors reported that it provided a convenient break in the tedium of longer tests. Some test supervisors administered it as the first, some as the last test. The test cards were scored by converting the electrographic marks to holes in the same IBM card. Raw scores for each of the three evaluative scales were summed and entered as whole numbers on new cards. Data from the new cards were stored on magnetic tape from which computa tions were made by means of computer programs. DEFINITIONS OF TERMS Attitude. The term attitude refers to a readiness state that exerts a directive influence upon an individ ual* s behavior (after English and English [1953:50] and Gerberich [1962:132]). 27 Authority relations. This term is applied to the attitudes which an individual expresses regarding the rules and regulations that govern individual and group behavior. Bipolar adjectives. Two adjectives, opposite in meaning and constituting the extremes of a semantic con tinuum (Osgood, 1957:20), are referred to as bipolar adjectives. Concept. In the survey phase of this study, "con cept” refers specifically to a stimulus word or phrase to be rated against one or more bipolar adjective scales. Congruent meaning. Congruent meaning refers to the meaning of a concept in the subgroup culture defined as the set of averaged rating scores representing that concept (Osgood, 1957:88). Dropout. A pupil who leaves regular full-time high school classes and for whom there is no evidence of further formal education is referred to as a dropout. Family relations. The term family relations refers here to attitudes associated with parents and home; more specifically, it refers to the feelings which the individ ual experiences in his relationships with others within the family and the home setting. Goal orientation. Goal orientation is the 28 evaluation placed upon the nature of future activities, or the maintenance of motive or drive toward indeterminate educational rewards. Meaning. For the purposes of this investigation, "meaning" has been defined operationally as the set of rating scores for a given concept on the Meaning of Words Test. Moral and social values. Moral and social values refer in this study to those concepts, held to be of value by the middle-class culture, associated with success and ascendancy (Schneider and Lysgaard, 1953; Rosen, 1956). School experience. Those attitudes regarding the school environment (reading, grades and teachers) are referred to here as school experiences. Self concept. The individual's awareness of his own existence, that is, his conception of who and what he is (Jersild, 1961:123), is referred to as his self concept. Semantic differential. The semantic differential is essentially a com bination of controlled associations and scaling procedures. The subject is provided with a concept (stimulus word or phrase) to be differentiated against a set of bipolar adjectives. The task is to indicate the direction and intensity of associa tion along a seven-point scale. (Osgood, 1957:20) 29 Social interaction. Social interaction refers to "the mutual stimulation of one person by another and the responses that result" (English and English, 1953:508). Stay-in. A pupil who remains in regular full-time high school classes for the entire school year is referred to here as a stay-in. ORGANIZATION OF THE REMAINING CHAPTERS Chapter Two presents an overview of the research literature specifically related to the problem of school dropouts, attitude measurement, and the semantic differen tial process. The statistical studies are organized according to national, regional, state and local investiga tions. A comprehensive review of all available literature would be impractical. Representative studies were selected to illustrate the findings and trends in research. Rele vant research on high school dropouts is presented accord ing to the principal aspects of personality development: family correlates, physical factors, absence from class at school, intellectual features, emotional concomitants, social interaction, and value and motivational issues. The chapter concludes with discussions of attitude measure ment, an explanation of the semantic differential process, and a description of currently available predictors of dropouts. 30 Detailed justification is provided in Chapter Three for the selection of the clusters of concepts to be rated. Chapter Three concludes with a description of the sample and the population from which it was drawn; the Meaning of Words Test, its development and administration and the statistical treatment of the data. Chapter Four presents the findings of the study. The discussion of the findings, conclusions, recom mendations and summary comprise Chapter Five. CHAPTER II REVIEW OF THE LITERATURE AN OVERVIEW This chapter presents a summary of the dropout research applicable to the immediate problem. An opening statement describes pertinent dropout research at national, regional, state-wide, city-wide, and local levels. Inas much as this study is concerned with attitudes, feelings and perceptions of pupils who drop out of high school, factors that are associated with the development and expression of attitudes were organized according to the following selected concomitants of personality development: (1) family correlates, (2) physical factors, (3) absence from class at school, (4) intellectual, (5) emotional, (6) social, and (7) value and motivational issues. A critique of attitude measurement has been included to describe the background and to present some of the available measurement techniques. The semantic differential process as described by Osgood (1957) and others is presented in detail so that the reader may have a capsule review of the development, functioning and evaluation of this rather new procedure. 31 32 Finally, brief consideration is given to the only published inventory specifically directed to the identification of the potential high school dropout. Background of the Problem "For all its current urgency in the public concern, dropout is no rare phenomenon. ... It is the problem, not the fact, of dropout which is new and contemporary" (Schreiber, 1964:1-2). Published concern for the early school leavers dates back to Ellis (1903) who became interested in boys who left Boston high schools without graduating. Research reports appear at sporadic intervals in the literature until about 1950 at which time an intense acceleration in the study of the dropout problem occurred. At the turn of the century Ellis (1903:796) reported that 3 per cent of entering first graders were graduated from high school. This was not inconsistent with reports issued during that decade that from 6 to 8 per cent of ninth grade pupils were graduated from high school. Retention in high school reached the half-way mark in 1930 when the United States Office of Education (1963:13) estimated that 495 of every 1,000 pupils enrolled in the ninth grade four years earlier actually graduated from high school. Currently the nation's schools are graduating nearly seven of every ten ninth grade pupils, according to the United States Office of Education (1963:3). Wolfbein in 1959 perceived the magnitude and the national implications of the dropout problem when, based upon information from the Bureau of Labor Statistics, he estimated that as a direct consequence of the postwar baby boom (which began in July 1946) and at the then current rate of dropout, by 1970 at least 7.5 million youth would leave school before graduation; moreover, 2.5 million would have less than an eighth grade education. Nationally, this would mean that during the 1960-1970 decade, about 26 mil lion youth would attempt to enter the labor market poorly prepared to meet the demands of a society that can effec tively utilize only 5 per cent of the unskilled labor work force (National Education Association, 1962:52). These estimates were predicated upon the existing dropout rate information without mitigation for the future. Compounding of the present problem was inferred in a news release from the Census Bureau in which it was stated that there would be a tremendous increase in the number of youngsters reaching the age of 18 in the twelve month period beginning July 1964. "The estimate of 3,728,000 is about one million, or 35 per cent, greater than in the comparable 1963-1964 period" (Census Bureau, News Release, May 20, 1964). Who are these dropouts? The dropout per se is not an odd fellow; rather he is more like than unlike his peers. Collectively, dropouts are as heterogeneous a group as stay-ins (Shanley, 1961; Wolfbein, 1959; Stice, 1961). Certain traits appear to distinguish the dropout as an entity from the stay-in, but ultimately such characteriza tions falter when the coexistence of identical traits are found among the stay-ins. The dropout displays no pathog nomic symptoms. Diagnosis of the potential dropout is dependent upon the identification of various symptoms or clusters of indicators. Researchers have reported that the dropout is generally lower in intelligence than the stay-in. This can be argued, in that the very tests by which he is evaluated demand unlearned skills. Correspondingly, the dropout does not usually achieve academically at the aver age level for his class; may well be retarded one or more grades; and is usually a social misfit. Thus, through lack of success and rejection by peers, the dropout is forced into an unremitting isolation. This discernment provides important clues for further investigation. The studies which follow have been identified according to scope of population and are presented for identification only. Relevant findings are reported in later discussions. Trends in Dropout Research College and university dropout research. Studies of the college dropout have had as their focal concern a loss of potential talent analogous to that of the high 35 school. Recent reports of college dropouts not only corroborate the findings of elementary and high school investigations, but portend a concern for the individual in a setting of higher education. Slater (1960:3-8) studied 1,323 male dropouts at the University of Illinois. The findings supported the assumption that the incorporation of institutional, curric ular and professional objectives into self was a central task which occupied those who persisted in higher educa tion. Lecky (1945:150) suggested that the process of incorporation included only that which was perceived as having a relationship to self. Slater (1960:7) demon strated that home and family experiences, health, peer relations, and relations with adults all had a bearing on the facility with which students had to develop if they were to achieve persistence in college. Lecky (1945) and Slater (1960) identified as singularly important the fact that the; incorporation of new values and ideas was a process. There was a remarkable similarity between the characteristics identified by Slater (1960:8) and the characteristics of high school pupils who remained to graduate. Slater (1960:8) concluded that students who were confident, experienced a minimum of threat and fear in new settings; who found an easy identification with others and with new ideas, goals and values needed only an opportunity to pursue an academic career. Conversely, he found that 36 the student with feelings of inadequacy regarding the acceptance of new relationships with ideas and with people would fail in the process of incorporation which facili tated persistence in college. Gekoske and Schwartz (1961:194) found essentially the same factors among a group of freshman dropouts. Adverse home life, ineffective social interaction and poorly defined goals were identified as significant corre lates in the student's decision to drop out. Parent interviews. The technique of holding inter views with parents represents an innovation in the identi fication of the potential dropout. Sando (1952) inter viewed 400 parents of both dropouts and stay-ins. During the interview he asked them to complete the Illinois Inventory of Parent Opinion (World Book Company, 1949). Medinnus (1963:100) developed his own instrument, "Parental Q-Sort." McDonald (1963) relied upon an informally- structured interview with the parent. Each of these studies has made a significant contribution to the identi fication of those parental attitudes and values to predis pose dropout. Essentially, these studies demonstrate that those parental attitudes that convey a positive evaluation of education are most significant in influencing a pupil to remain in school. 37 EXTENT OF THE PROBLEM AT VARIOUS LEVELS National Statistics Reports published by the Office of Education regarding dropout rates are estimations based on the number of children enrolled in the fifth grade throughout the country. This base grade was selected because many chil dren repeat earlier grades and compulsory attendance laws have kept them in school. The dropout rate was estimated by comparing the number of high school graduates in a given year with the number of ninth graders eight years earlier. There is no generally-accepted method for computing the number of dropouts or the dropout rate. An excellent reference regarding definitions of the dropout and methods for computing dropout rates is found in Putnam (1963:24). Comparability of statistics is particularly difficult in the absence of uniform criteria concerning base grade, definition of dropout, and method of reporting. Statistics cited here are consistent with the practice of reporting the number of high school graduates for every 1,000 fifth grade pupils enrolled eight years earlier. Nation-wide figures in Table 1 are representative of the continuous increase in number of high school gradu ates extending over a period of fifty years. This table illustrates the steady increase in number of high school graduates over the past seventy-four years. Variations in 38 TABLE 1 NUMBER OF HIGH SCHOOL GRADUATES FOR EVERY 1,000 FIFTH GRADE PUPILS ENROLLED EIGHT YEARS EARLIER Year of Number of Graduation Graduates Source 1890 60 Maryland Study (1963:n.p.) 1910 100 Maryland Study (1963:n.p.) 1914 140 Tompkins (1951:310) 1918 139 Phillips (1930:16) 1931 270 Tompkins (1951:310) 1935 350 Tompkins (1951:310) 1940 450 Tompkins (1951:310) 1945 400* Tompkins (1951:310) 1948 480 Tompkins (1951:310) 1952 522 Office of Education (1959:13) 1956 581 Office of Education (1959:13) 1962 636 Lee (1963:2) 1964 698 Cohen (1964) *World War II 39 reporting practices, definitions, and computational methods make direct comparisons uncertain. However, since these figures represent nation-wide summaries, perhaps limited generalizations are warranted. With the single exception of the years of World War II, there is evident a firm growth in the holding power of the high school. Nearly twelve students are being graduated today for every high school graduate in 1890. More impressive figures would obtain if ninth grade enrollment had been selected as a base grade; however, the dropouts which occurred between the fifth and ninth grades would have been obscured. Within this group also would have been the mentally and physically retarded who tend to drop out at earlier grade levels than non-handicapped pupils. Lee (1963:12) assumed "that the factors not taken into consideration in compiling the data are fairly con stant, T l and used these figures for general comparisons. She presented an excellent summary of national figures on the school dropout. The reader is referred to the article for extensive detail that could not be included in this discussion. Since 1950 more than one half of the entering fifth graders have been graduated from high school. At . the present time (1964) we can expect to graduate nearly 70 per cent of all children entering the fifth grade. These figures will, in all probability, begin to plateau 40 within the next few years. Retention of all children until graduation is not a realistic hope, but certainly there is considerable margin for improvement of the holding power of the high school. Regional Studies Instances in which more than one community or more than one state are represented are less amenable to gener alized comparison than are nation-wide statistics. Each study developed its own definitions of dropouts, base grade, methods of reporting, and dates to be included. Thorndike (1908) conducted the first large-scale investigation of the dropout problem. Twenty-three cities with populations over 25,000 were selected for study. An average of 8 per cent of those enrolled in the first grade were graduated from high school. Ayer (1909) conducted a study of Boston high schools one year later and found that 10 per cent of all first grade pupils graduated from high school. Ayer's findings are not inconsistent with Thorn dike's larger study. Dillon (1949) directed a study involving a total of five communities in Indiana, Michigan and Ohio. School records of 1,300 dropouts were examined to determine what identifiable signs might appear that would be significant in determining the pupil vulnerable to dropout. The more significant findings include: 54 per cent of the dropouts left at age 16, the legal minimum age; an additional 26 per 41 cent left at age 17; about 40 per cent of the dropouts had satisfactory attendance in high school; and 60 per cent of the dropouts had IQ scores below 95. Wolfbein (1959) reported a study in which seven communities, with populations from 30,000 to 350,000, participated in an intensive investigation of dropouts. The cities participating were: Providence, Rhode Island; Utica, New York; Harrison County, West Virginia; Port Huron and Saginaw, Michigan; Phoenix, Arizona; and Evansville, Indiana. A summary of the results described the dropout as being: below average in group intelligence test score; a dropout at age 16 to 17; retarded one or more grade levels; more probably a boy than a girl; disliked school; and having left to go to work or to earn money. State-wide Dropout Studies Bowden (1917) published an early comprehensive survey of all high schools in South Dakota. He found that for every 100 ninth grade students enrolled, 28 dropped out during the freshman year, 13 dropped out as sophomores, 14 as juniors, and 45 had dropped out during the senior year. Undoubtedly these figures reflect the influence of World War I. No statement was made regarding the propor tion of students who were graduated. Four factors were identified as significant correlates with leaving school: lack of a suitable place to study, large number of children in the family (five or more), lack of effective study 42 habits and economic factors. Regarding the lack of effec tive study habits, Bowden commented that these should have been acquired in junior high or in elementary school; high school was no place to learn how to study. Borden (1917: 448) concluded that ". . . the most important duty of high school teachers is the creation of suitable attitudes in students.” He extended this comment by stating that the home environment was a very important factor in creating a favorable attitude for persistence in school. Bowden's conceptions of the dropout problem nearly sixty years ago are consistent with the findings of current research. Lehman (1926) reported a study of Ohio high school graduates in which he found that about 50 per cent of the ninth grade students were graduated. Eckert and Marshall (1938) in a massive study of the schools of the State of New York found that about 30 per cent of all high school students dropped out before graduation. The figures in these studies are somewhat higher than the national rate of retention, as might be expected when the sample is restricted. Now Hear Youth (McCreary and Kitch, 1953) represents the only attempt made to date by the State of California to summarize information concerning the dropout. The writers compiled information from twenty-eight study projects throughout the state with no indication that the areas reporting were pre-selected in any way. The chapter on dropouts was based primarily upon the 1950 census data which, at best, was of limited value. According to these figures, most California youth remain in school until they reach the age of 16, and that there was an increase in school enrollment during the years 1910 to 1950. This would not be inconsistent with the national trend for these same years. There were 301 high school dropouts in 1951 for every 1,000 eighth grade pupils enrolled four years earlier; in 1952, 296 students failed to graduate for every 1.000 eighth grade pupils enrolled four years earlier; and in 1953, 295 students dropped out of school for every 1,000 eighth grade pupils enrolled in 1949 (1953:28, 29). The authors recognized that these figures did not take into consideration pupils retained in grade, nor the in-migra- tion of youth to the state. Shifts in private to public school attendance and enrollment in continuation high schools tend to affect the number of reported dropouts and influence the California data to an unknown extent. Two studies comprise the major portion of the chap ter. The first, an investigation conducted in the San Diego City schools, dated back to the seventh grade enrollees of 1940. This group was followed up in 1946 after they had graduated from high school. The intervening "war years" undoubtedly biased the results. For every 1.000 seventh grade pupils enrolled in 1940, 380 had dropped out by 1946, and 620 were graduated. Dissatisfac 44 tion with school was given as the reason for leaving by 57 per cent of the dropouts; 13 per cent listed financial need as a reason for leaving. Marriage was the most frequently- given single response (25 per cent) by the girls, M. . . though in some cases it appeared to be a result of rather than a cause of withdrawal" (1953:53). Modal age for all dropouts was between 16 and 17 years. Forty per cent of the dropouts obtained a group test IQ score between 90 and 110 (average); 34 per cent were below an IQ score of 90; and 12 per cent were above 110 In IQ scores. These find ings were consistent with the concept that dropouts were a heterogeneous group with respect to group test intelligence scores. Further evidence was found in that two thirds (62 per cent) of the parents of the dropouts were unskilled or semiskilled workers. About one fourth (24 per cent) of the fathers of dropouts were classified as skilled workers, and only 5 per cent were classified as professional. Fewer than one fourth of the parents of dropouts had completed high school. In general, the dropouts did not discuss the problem of leaving school with their parents--undoubtedly indicative of a considerable lack of communication between parent and child. This situation might imply that the parents did not have a positive evaluation of education per se. "The Contra Costa Study," presented by McCreary and Kitch (1953:37), was abstracted from the doctoral study of Rudolph Sando (University of California, Berkeley, 1952). The inclusion of this investigation implied the generaliza- bility of Sando's findings to the state at large, an exten sion of inference not offered by Sando. The comments which follow are taken from Sando (1952). This investigation is examined in detail in the latter part of this chapter, and in Chapter Three of this study. Sando (1952) interviewed 100 dropouts and their parents within one week following termination of school. A matched control group of stay-ins and their parents were also interviewed. Dropouts expressed greater dissatisfac tion with social interactions than did stay-ins. The dropout often felt left out, that he didn't belong to the group. In general, dropouts were more critical of "unrea sonable discipline" and felt teachers were inconsistent in the administration of punishment. The dropout tended to feel as though the teachers were not interested in him as an individual; consequently, he avoided needed help with course work. The parents of dropouts appeared less aware than did the parents of stay-ins that their children were having trouble in school. This finding was consistent with the San Diego study (1946), reported above, in that there was a critical lack of parent-child communication. Marked differences were evident between dropouts and stay- ins; the dropout did not (or rarely) participated in extracurricular activities. This might be expected since 46 the dropout also expressed dissatisfaction with his social interactions. A sample of 1,381 early school leavers in Kentucky was examined by Hecker (1953), who reported that 14 per cent of the first graders, 40 per cent of the seventh, and 54 per cent of the ninth graders were graduated from high school in 1952. In this study, 59 per cent of the students in school were rural youth, so that it was not surprising that a large (64 per cent) proportion of dropouts were from rural homes. Grade retention appeared to be a significant factor in identifying the potential dropout, as did the pupil's desire to seek farm work. Hecker found, as did Dillon (1949), that about three fourths of the dropouts were living with both parents, tending to dispel the con cept that the dropout comes from a broken home. Teachers reported that the dropouts lacked sound study habits and skills in basic subjects. When Hecker queried regarding the attitudes of dropouts, teachers indicated that the dropout was not particularly worried, immature, withdrawn, resentful or aggressive. These results are inconsistent with findings concerning dropout attitude reported in the bulk of the literature and might indicate a lack of aware ness of pupil needs. Van Dyke and Hoyt (1958) carefully planned a drop out research project in Iowa so as to provide a stratified sample, representing about 5 per cent of all pupils in the 47 public high schools of the state. Holding power for the 73 selected high schools was determined to be 84.2 graduates for every 100 ninth grade pupils enrolled four years earlier. This was well above the national average of 69.2 for the same year (Office of Education, 1959:13). The mean group test IQ score was 96.6 for the dropouts and 106.4 for the stay-ins. This finding was consistent with Aseltine (1941) and others who have established a significant IQ score differential between dropouts and stay-ins. Twenty- nine of the dropouts had group test IQ scores of 120 or above. This comprised 17.6 per cent of the 165 students with IQ scores at or above 120, and 3.9 per cent of all dropouts, signifying, again, that as a group the dropout is indeed heterogeneous. The noteworthy Pilot Study of Ohio High School Drop Outs. 1961-1962 (Nachman, 1963) is another example of a carefully executed state-wide research project. Informa tion was obtained from the school counselors in each secondary system on each of three occasions--October, 1961, February, 1962, and June, 1962. All reports were converted to coded information suitable for electronic data proces sing. The annual dropout rate of 2.75.per cent of all enrollees in grades nine through twelve appeared unusually low when compared with other studies; however, the defini tion of a dropout excluded "illness; death; transferring to another school in or out of the state; commitment to a 48 correctional institution; expulsion.; or inability to meet state standards for slow learning programs." Practice among prior investigators appears to have been to define the dropout in such general terms that it would yield a larger dropout sample. Contrary to Cannaday (1962), the dropout rate in Ohio was greater in the larger city school districts than in smaller city districts. The eleventh grade had the highest proportion of dropouts and the ninth grade the lowest. Most dropouts occurred in the months of September and January for all grades examined, including both boys and girls. Over-age for grade appeared to be an indicator of potential dropout, especially in grades nine and ten; here, 90 per cent of the ninth grade and 66 per cent of the tenth grade dropouts were over-age for grade. Eighty-five per cent of the dropouts scored below the national average (group test IQ score of 104) on the Cali fornia Test of Mental Maturity (California Test Bureau, 1964:11). Chronic absence and truancy characterized both boys and girls who dropped out in grades nine through twelve. Emotional and social maturity, as rated by the counselors, indicated that only 4 per cent of the dropouts were rated "average or above," whereas 40 per cent were rated "below average." Nearly 55 per cent of the fathers of dropouts were employed in semiskilled and unskilled occupations, whereas the Ohio census for 1960 reported only 30 per cent of the general population so employed 49 (Nachman, 1963:25). Correspondingly, only 1.7 per cent of the dropouts were from homes where the father was classi fied as "professional." The education of the parents was described as "slightly below" the average of the 1960 Ohio census. Persistence in school appeared to be positively correlated with parent education. Families of dropouts were rated as belonging to an average socioeconomic class in some 53 per cent of cases; 44 per cent of the families were rated as below average; and fewer than 3 per cent of the dropouts came from above-average socioeconomic class homes. The reasons given by dropouts for leaving school tended to minimize the factors that counselors believed were most important. Students gave such reasons as going to work, marriage, and joining the armed service; whereas counselors felt that poor academic performance, problems of school adjustment, and lack of interest were more realistic causes for leaving. Marriage was the most frequent single reason (and cause) for girl dropouts. Woollatt (1961) reported previously unpublished data from the New York State Holding Power Project. This investigation was unique in that individual students were followed from the seventh through twelfth grades, beginning in 1954 and continuing through June 1960. Ninety school districts were selected from throughout the state for study. Modal dropout grade was the ninth at age sixteen, consistent with the time when compulsory attendance laws 50 permitted leaving. Median group test IQ score was 96 for the dropouts and 111 for the total group. These data were consistent with those of the large-scale study of Van Dyke and Hoyt (1958). There were no significant differences in IQ scores between boys and girls among the dropouts in either Van Dyke and Hoyt (1958), or Woollatt (1961). The reasons given by the dropouts for leaving concerned a general dislike for school, failure in classes, and finan cial need. Girl dropouts added marriage to their list of reasons for leaving. The most significant reason for leaving reported by school counselors concerned parental attitude toward school, and the family situation. Lack of success was second on the counselors' list of reasons for leaving. Cannaday (1962) reported a more recent study of enrollment and graduation from elementary schools in Arkansas. This rather thorough tabulation of data revealed that, in Arkansas, size of school district was a factor in the number of dropouts. There were 40 per cent dropouts from larger districts (over 8,000) compared with 52 per cent in smaller districts. For every 1,000 first grade children enrolled in 1940, 221 graduated from high school in 1952, representing a gross dropout of 78 per cent before graduation. Three hundred sixty first grade children out of 1,000 enrolled in 1949 graduated from high school in 1961— a dropout rate of about 64 per cent over twelve 51 years. For every 1,000 fifth graders enrolled in 1944, 451 graduated from high school in 1952, indicating a dropout rate of about 55 per cent. Fifth grade children enrolled in 1953 showed a dropout rate of only 40 per cent before graduation (601 pupils graduated for every 1,000 enrolled eight years earlier). These figures are comparable to national statistics (Lee, 1963:2). Cannaday (1962:8) illustrated the increased holding power of the elementary schools by noting the decrease in Msummer loss” between eighth and ninth grades at five-year intervals beginning in 1940, as is shown in Table 2. TABLE 2 NUMBER OF NINTH GRADE PUPILS NOT ENROLLED, FOR EVERY 1,000 EIGHTH GRADE PUPILS ENRMXED THE PREVIOUS YEAR Year of Attendance Proportion Not Enrolled 1940-41:1941-42 285 1946-47:1947-48 184 1950-51:1951-52 152 1955-56:1956-57 98 1959-60:1960-61 68 Summarizing high school graduate data for the year 1945, Cannaday (1962:8) showed that an increasing proportion of students were completing the twelfth grade. Table 3 illustrates this increase. These figures are comparable with national reports for the year 1961 (Lee, 1963:3). TABLE 3 PROPORTION OF HIGH SCHOOL GRADUATES Base Grade Percentage Graduating in 1945 Percentage Graduating in 1961 8 9 10 30.5 58.0 69.0 56.0 62.6 73.0 The Maryland State Department of Education (1963) conducted a state-wide survey of all school dropouts. A thirty-item questionnaire was completed on each early school leaver. The report was an excellent tabulation of pertinent information, which can only be summarized here (Table 4). Maryland data on enrollment and graduation were reported for the past seventy years. There is a remarkable similarity between the proportion of graduates each year and national statistics (Lee, 1963:2-3). The lack of information concerning the base year from which these data were determined limits generalization. Eleven factors were identified as the most important findings from among the 53 thirty bits of information obtained on each dropout. TABLE 4 PROPORTION OF HIGH SCHOOL AGE YOUTH ATTENDING HIGH SCHOOL* Year Percentage Attending School Percentage Who Graduate 1890 8 6 1910 16 10 1930 50 29 1950 74 58 1960 85 69 ^Maryland State Department of Education (1963:n.p.) 1. Most dropouts left school at age 16, the mini mum legal age. 2. Over half of the dropouts came from homes where the principal occupation was either unskilled or unemployed; emphasizing socioeconomic status as a factor or indicant of potential dropout. 3. More than two thirds of the dropouts never participated in athletic programs; similarly more than two thirds failed to participate in extra-curricular activities. 4. Almost 66 per cent of the dropouts left school 54 during or before the tenth grade. 5. Most of the dropouts were not considered to have been behavior or disciplinary problems by teachers or administrators. 6. Four out of five of the dropouts were living with their parents. Again, this attenuates the concept that dropouts tend to come from one-parent homes. 7. Dropouts occurred in all mental age groups. Nearly half (49.8 per cent) had group test IQ scores of 100 or above; again indicating the heterogeneity of dropouts as a group. 8. Reading retardation was a crucial factor. Four out of ten dropouts read at or below the sixth reader level. 9. More than half (56.6 per cent) of the dropouts were not achieving at grade level. 10. A small proportion (1.3 per cent) of the drop outs left for reasons of social maladjustment. This appears contrary to much of the literature which indicates that social maladjustment is an important factor in the identification of potential dropouts. 11. More than 50 per cent of the dropouts expressed a lack of interest in school, and is consistent with previous literature. 55 This comprehensive study has provided information from which reasonable generalizations can be made. Perhaps the most significant recommendation, in addition to com ments about counseling and curriculae, concerns a recogni tion that the problem of the potential dropout can best be met through the study of motivations and aspirations. City-wide Dropout Research: Los Angeles Dropout research has been conducted by many school districts as a medium for self-evaluation and improvement. Ayer (1924:96) reported in a study of Seattle (Washington) Public Schools that in October 1923 there was a differen tial in enrollment of 4,749 in first grade and 1,912 in the twelfth grade, a gross dropout of 40.2 per cent. He con cluded that this illustrates a greater holding power than that of the national average. The national average for the year 1924 was about 200 graduates for every 1,000 enrolled in the fifth grade, or an 80 per cent dropout rate. Direct comparison is difficult because of the difference in base grade. In any event, had the base grade been taken at the first grade, the number of graduates would probably have been fewer than 200 on the national average. Inasmuch as Ayer's figures represent enrollment for the same year, little can be inferred about the proportion that would graduate twelve years later. Woody and Cashman (1936:183-187) wrote a concise report on dropouts in the Denver (Colorado) schools. From 56 1928 to 1934 the modal period for dropouts was between the tenth grade of one school year and the eleventh of the next. The second highest number of dropouts occurred after entering the eighth grade, but before entering the ninth. Nonpromotion was found to have been positively related to the student's decision to leave school early. Abbott (1939:137) conducted an extensive survey of six high schools in the Los Angeles (California) area. Dropouts were reported for grades nine through twelve, the total of ninth-grade dropouts being 136 for every 1,000 students. The median age for boy dropouts was 18.2 and 17.7 for girls. Numbers of boys and girls were about equal at each grade level. Abbott noted that the median age of the dropout was the same as the age of the graduate, again pointing to the importance of over-ageness, or grade retardation as a significant accompaniment in early school leaving. Throughout the study the variation between schools within the same district was emphasized. Aseltihe (1941) differed from previous researchers in that he examined in detail the general characteristics, school progress, curricular and extra-curricular experi ences and attitudes of the senior high school dropout. The difference in median age for boys and girls was significant at the .01 level of confidence. His findings differed with respect to median dropout age from those of Abbott (1939:137) who found the median age for boy dropouts to be 57 nearly one year higher. Aseltine (1941:53) found the group test IQ scores for boy dropouts to have a mean of 97.32 and a median of 95.6; for girls the mean was 96.5 and the median 95.3 Hollingshead (1949:329-359) discussed in some detail the characteristics of youth who did not complete high school. Table 5, adapted from his page 330, illus trates the relation between class status and persistence in school. TABLE 5 CLASS STATUS AND PERSISTENCE IN SCHOOL Class In School Out of School Number Per Cent dumber ter Cent 1 (upper) 4 100.0 0 0.0 2 31 100.0 0 0.0 3 146 92.4 12 7.6 4 183 58.7 129 41.3 5 (lower) 26 11.3 204 88.7 Hollingshead concluded that class position was strongly associated with persistence in school. Group intelligence test scores tended to reflect the same general pattern: the class IV stay-ins had significantly higher IQ test scores than did the class IV dropouts, but no difference 58 was noted between class V youth matched in the same manner. Hollingshead commented, "However, we do not believe that the lower intelligence scores of the withdrawees in class IV are an adequate explanation for these youngsters' with drawal from school before their class equal" (1949:335). Delaney (1950:22) examined Chicago public school health records of dropouts during the 1946-1950 period and concluded that the health records, indicating faulty vision and defective hearing in dropouts, were important dropout correlates. He neglected to report the extent to which stay-ins were also afflicted with faulty vision and hearing defects. Holbeck (1950:35) reported research in Passaic (New Jersey) in which a dropout rate of 45.4 per cent was found. Smith (1950:26) showed an average dropout rate of 46 per cent for the years 1922-1946 in Syracuse (New York). Bowman and Matthews (1960) conducted an eight-year study of all pupils in the Quincy (Illinois) public schools. Following the sixth-grade pupils of the 1951-52 school year for the next eight years, they found an overall dropout rate of 31.4 per cent. First (1950:252) reviewed a research project in Sault Saint Marie (Michigan), in which 18 per cent of the 1956 enrollees had dropped out by 1960, 23 per cent of those enrolled in 1958 had left by 1960, and 26 per cent of all pupils enrolled in 1959 had left in 1960. 59 Published Local Dropout Statistics: Los Angeles The bulk of data regarding dropouts in the Los Angeles area were obtained from the annual reports prepared by individual school districts. A recent report of the Los Angeles Unified School Districts (1963) summarized the dropout rate based on registrars' reports which identified nine reasons for leaving. The fact that 24 per cent of the dropouts were recorded as "unable to locate" diminished the meaningfulness of the figures because of the possibility that these pupils might still be in school. Table 6 sum marizes the estimated dropout rates for the years 1959- 1962. PROPORTION OF TABLE 6 DROPOUTS BY GRADE LEVEL* Estimated Percentage of Dropouts, Year Grades 7 through 12 1959-1960 31.8 1960-1961 31.7 1961-1962 29.6 *Adapted from Los Angeles Unified School Districts, Research Report No. 252, 1963, p. 5. These figures were based upon a hypothetical group of 1,000 seventh grade pupils from which the proportion of 60 dropouts for each grade was subtracted cumulatively. Median grade level for leaving was 11-2, almost one year higher than was reported in other studies reviewed. The median age-on-leaving was 16-11, again nearly one year higher than the majority .of other studies reported. Aver age over-ageness in grade was 9 months. This report identified a rate of mobility termed "transiency” which varied from 50.3 to 55.4 per cent for regular secondary schools and ballooned to 185.6 per cent for the special schools, representing extreme mobility. It would appear that the holding power of the Los Angeles schools was greater than would be expected in view of the excessive mobility of the population--which in itself contributes substantially to dropping out. General Characteristics of the Dropout The numerous local investigations reviewed and the few reported here corroborate, for the most part, the find ings of national and regional surveys. Variations in figures are due in part to variations in: (1) methods of reporting, (2) base grade level, (3) grade levels included, (4) losses due to special school assignment, community immigration and emigration, and (5) not the least, to the definition of the dropout. Several pertinent factors are absent from most research reports. For the most part, data collected for these investigations pertain to enrollment (head count) 61 figures reported from one period of time or one grade level to another. No attenuation is provided for immigration or emigration in the community. Some areas are considerably more mobile than others. Adult work-force demands can have an influence upon the direction and magnitude of population movement. Grade acceleration and retardation play an important role in the pupil*s decision to remain in school and seldom are considered in research studies. The impli cation concerns the recognition of the individual student, following each through the established period of research. Dropout literature tends to identify certain rather consistent characteristics or factors associated with early school leaving. Broad generalizations may reduce the list of factors to as few as four traits, while more detailed studies expand to as many as thirteen (Lee, 1958), or twenty-five (Cassel and Coleman, 1962). Examination of . these findings appears to resolve into five spheres of personality development. Physical factors in personality development concern those aspects of the individual's life space over which he has little control. Parental attitude toward education, the physical environment of the home, parent education, siblings, the sex of the individual, socioeconomic status of the family, family mobility, and health all are asso ciated in one way or another with the decision to remain in school or to drop out. 62 Intellectual development appears to be related to persistence in school, despite the fact that many dropouts are well above average in intellectual achievement. Evalu ation of this function is usually made in terms of intelli gence test scores, either group or individual tests. Read ing achievement may be an indicator of potential for academic success. Grade retardation has been found asso ciated with early school leaving. Longitudinal studies have shown that more dropouts are remaining in school longer, that is, there has been a higher modal dropout grade-on-leaving. Emotional concomitants in personality development involve factors such as belonging to a group, feelings of inferiority, being withdrawn or open rebellion toward one's immediate environment, personal, and social adjustment. Social interaction extends personal development to functioning in interpersonal relationships, specifically relationships with teachers, peer group interaction both within and outside the school, and to participation in extra-curricular activity. Values, attitudes and motivations for learning are inferred from overt behavior. Attitude-value systems give rise to motivations which receive expression as behavior correlates. 63 FAMILY CORRELATES Certain factors have been identified within the home environment which appear to be closely associated with persistence in school: (1) parent attitude toward educa tion, (2) parent education, (3) the emotional milieu of the home, (4) sibling influence, (5) socioeconomic status of the family, and (6) the mobility of the family. Parental Attitude toward Education Perhaps the most significant single factor associ ated with a pupil's decision to leave school before gradu ation is the attitude of his parents toward education. Brooks (1903:362) made one of the earliest statements con cerning parental attitude when he noted that indifference on the part of the parent toward education was an important factor in dropping out of school. Dynes (1914:396) observed a relationship between retention in school and favorable parent attitude toward education, concluding that the homes of persisters in school were above average, parents were active in professional callings, and parents took an active part and interest in the education of their children. Bowden (1917:48) commented: "Home environment [was] a very important factor in creating a favorable attitude" toward school. Abbott (1939:169) reported that parents interviewed in one area where they had had the opportunity of education were "free to express . . . to 64 regret and to explain the many steps that had been taken in order to encourage graduation." He concluded that parental influence undoubtedly accounted for the lower dropout rate in the area. Hollingshead (1949:331) explained that the class to which a child belonged was a significant correlate with the child who decided to remain in school. If the family regarded education lightly, was unable or refused to contribute money (usually small sums) to meet increasing financial needs, or had a "bad reputation," the pupil developed stronger motives to leave than to remain. The beliefs and perceptions the child learns from the family regarding education and school, combined with attitudes and perceptions from school experience, result in distinctly different response patterns in the school situation. The child's perceptions, together with the influence of parental attitudes . . . appear to act as agents which keep adolescents from the three higher classes in school long after legal requirements have been met and to drive many in class IV and particularly class V from the school before they have met the educational standards by law. (Hollingshead, 1949:331) Layton (1952:2) added in his discussion of a Detroit study that parental attitude toward education was an important factor in keeping children from leaving early. Kahl commented: Parents who were discontented tended to train their sons from the earliest years of grammar school to take school seriously and use education as a means to climb into the middle class. Only 65 some who internalized such values were sufficiently motivated to overcome the obstacles which faced the "common man" boy in school; only they saw a reason for good school performance and college aspirations. (1953:203) Kahl remarked that schools were more a means than an initiator of "social ascent." Sando (1952:47) repeatedly emphasized that parental understanding of the educational program and attitude toward it were probably of greater importance to the youngster than many of the other stresses already identified. Sando felt that the literature was incomplete with respect to studies of parental attitude or opinion regarding the issues and values of high school education. The Ministry of Education, London, England (1954:57) described the influences of parental attitude toward education, noting that girls leave at an earlier age than boys; when it was a matter of money, the girl was taken out of school. Boys were considered the money earn ers, and therefore were encouraged to remain in school as long as possible. The writers extended the discussion of parental attitude to include consideration of the different social assumptions which affected not only the child's parents, but the whole society in which he was brought up. They completed the discussion by stating: "The influence of the home is of profound importance." Penty (1956:42) stated in her study of reading retardation and dropout that "lack of home interest and support" was a major reason for'dropping out, giving as an 66 example the case of a girl whose group test IQ score was 69 and reading level 5-6; she was able to graduate from high school, explaining, "My mother and dad wanted me to" (1956:48). In conclusion, Penty stated that poor readers who graduated did so because of better home environment and parent interest. Snepp (1956:49-54) observed that parents of drop outs were characterized as "indifferent" in teachers' and counselors' notations on pupil records. Munoz (1957:232) found that the parents of the stay-ins were more satisfied with school, with the child's work, and with the teachers. More parents of stay-ins wanted their children to attend college than did parents of dropouts. In general, the parents of dropouts did not know the occupational goals of their children, whereas the parents of the stay-ins expected their children to enter professional occupations. Fraser (1959:57) correlated the influence of parent attitude with persistence in school. Parents' attitudes toward education were rated by the interviewer on a six- point scale. Attitude ratings correlated .391 with progress in school, parental encouragement correlated .604 with IQ scores, and .660 with progress in school. Fraser's findings indicated a positive relationship between parental encouragement, progress in school, and academic attainment. The National Education Association (1960:11-14) reported in the Research Bulletin that the most crucial influences exerted by the home were parental attitudes. The feelings of parents toward school were noted as impor tant factors in shaping the attitudes of the child. The article closed with the comment that if parents valued education, we could expect a positive (pupil) attitude toward school. McMurrin (1961:17) found that among sizable numbers of dropouts from more privileged families, there was a background of parental indifference to the worth of education. Swan (1961:79) identified "lack of parental emphasis on education" as one of eleven significant factors influencing the decision of a student to drop out. Woollatt (1961:7) felt that the most significant reasons for dropping out assigned by counselors had to do with parental attitude toward the school and family situa tions. Brooks (1962:103-108) studied the factors affecting school attendance and found significant correlations between parent attitude toward school and attendance by the child. Fink (1962) cited parent aspiration (in this instance reported by the student) and over-ageness as the two most important factors in persistence in school. Lichter (1962) discussed at length the importance of parental attitude in school adjustment, concluding: "Like parent, like child." This investigation reported that parents of potential dropouts were psychologically immature and transmitted this immaturity to their children through identification. A positive attitude on the part 68 of the mother was associated with personality improvement in the child. "Sweeping parental disapproval" had a seri ous effect upon the potential dropout, and "his feelings of worthlessness are intensified as is his discouragement" (1962:165). Many parents were identified as having charac ter disorders, although none had neuroses. Parents had strong reaction patterns to the child's problems and "responded nonconstructively by withdrawing expressions of affection and instituting handicapping systems of punish ment" (1962:254). It was reported that frequently parents of dropouts had unhealthy and distorted relationships with their children. Liddle (1962:277), in response to a question regarding the extent to which parental attitude influenced school attendance, stated that children from families where middle-class values prevailed were taught that life was a series of hurdles to be jumped if one were to succeed. Lower-class adults, particularly mothers of dropouts, did not share these values. They taught that life was a series of trying situations to be avoided if at all possible. It was more important to get by, but achievement was neither valued for itself nor viewed as a means to a better life. For the most part, these parents were disinterested in school, and in some instances were actively opposed to it. Liddle stressed the importance of parental interest and involvement as critical factors in a student's decision to drop out. Mannino (1962:193) found in a group of thirty-four lower socioeconomic families in northern Florida that a larger proportion of mothers of stay-ins showed interest in and encouraged their children to attend school than did mothers of dropouts. At the same time, he found that more mothers of stay-ins were acquainted with families who had children in college than did the mothers of dropouts. The mothers of stay-ins more often referred to the importance of early training if their children were to succeed. Medinnus (1962:100) developed a 40-item opinionnaire to evaluate parent attitudes toward education. No significant differences were found with respect to parent education (all parents had a minimum of a high school education), nor socioeconomic status. A significant difference was found between mothers' and fathers' responses. This study derives significance from the fact that it represents an attempt to quantify parental attitude. Elliott (1963) in a presentation before the Cali fornia Association of Supervisors of Child Welfare and Attendance emphasized that parents of dropouts were charac terized by strong reactions (usually negative) toward the child and frequently by distorted mental health concepts. For the most part, Elliott found, parents of dropouts lacked educational experience and were in all probability dropouts themselves. 70 McDonald (1963:437-438) conducted a study in which parents of successful and unsuccessful readers were inter viewed. Just as with the parents of dropouts and stay-ins, parental attitudes differed between treatment groups. Schreiber (1963:217) cited studies from Louisiana and New York in which some two thirds of the parents of dropouts held negative or indifferent attitudes toward the value of education. Parent Education Parent education indirectly influences the poten tial dropout through identification. Palmer (1930:94) reported a comparative study of leavers and non-leavers in Oakland (California) and reported that 43.6 per cent of the non-leavers' parents had more than an eighth grade educa tion, while only 23 per cent of the parents of dropouts had more than an eighth grade education. Rosander (1939:603) related dropping out of school to the father's educational attainment. Gragg (1949) established a correlation between education of the parent and persistence in high school. Hollingshead (1949:335) described a "direct and positive" relationship between early school leaving and the extent of parent education. The educational background of parents in the lower socioeconomic group was found to be very impor tant: "Within class IV, the family's formal educational experience is the most significant factor in an 71 adolescent's continuation in or withdrawal from school" (1949:335). In fact, 22 per cent of the class IV fathers of stay-ins had completed one or more years of high school, in contrast with 12 per cent of the fathers of dropouts in class IV; and 34 per cent of the mothers of class IV stay- ins had had some high school training, while only 12 per cent of the mothers of dropouts had had any high school experience. Hollingshead concluded: We believe that the significantly poorer family heritage, as measured by the items mentioned in the preceding paragraph, among the out-of-school class IV youth give rise to their poorer adjustment to the demands of the school in comparison with those of this class in school. (1949:337) Sando (1952:92) found that parents of dropouts had substantially less educational experience than did parents of stay-ins. The results of his study are summarized in Table 7. TABLE 7 PARENT EDUCATION Educational Experience Percentage of Parents of Dropouts Percentage of Parents of Stay-ins College 8 9 High School 73 66 Not beyond 8th grade 70 71 72 Sando concluded that a positive relationship exists between the amount of education which a parent has had and the per sistence of the child in school. Lombardi (1953:40), in a study of one local high school in the Los Angeles area, found that only one of eighty-eight parents of dropouts had completed college. Lombardi's study lacks needed detailed data; therefore, comparison of this with other studies would be meaningless. Epps and Cottle (1958:91) identified the potential dropout as having parents with a low educational level. Van Dyke and Hoyte (1958) commented that one of the factors affect ing dropouts in Iowa concerned the extent of parent educa tion. They found that parents of non-leavers had remained in school longer than had the parents of dropouts. Epps and Cottle (1958:92) found dropouts very sensitive to the educational accomplishments of their parents. Anderson and Harris (1959:28) tabulated the propor tion of children who drop out of school between the fifth grade and high school graduation in relation to education of parents. Table 8 is adapted from Table 13. Bledsoe (1959:3) computed the ratios of years of parent education between dropouts and the total school population shown in Table 9; expected frequency would be 1.00. The rather obvious conclusion is that dropping out of school is associated with parents who have less educa tion than do the parents of stay-ins, Fraser's (1959:42) 73 TABLE 8 EDUCATION OF PARENTS Education of Parents Percentage of Dropouts Boys Girls Both parents graduated from high school 10 0 Father or mother, but not both graduated 25 10 Neither parent graduated from high school 34 25 TABLE 9 YEARS OF PARENT EDUCATION Grade Completed Ratio 1-4 1.08 5-6 3.15 7 4.86 8 .92 1, 2, 3 years of high school .39 Graduated .01 Attended college No dropouts study, performed in London, England, found that children of parents with more education progressed better in school than did children of equal IQ score whose parents had had less education. The Maryland Study (1963) reported that 78.5 per cent of mothers and 80.3 per cent of fathers had, them selves, been dropouts. Completing nine years or less of formal education were 63 per cent of fathers and 56.7 per cent of mothers; completing only sixth grade or less were 30.9 per cent of fathers and 24.4 per cent of mothers. Brooks (1962:103-108), in a .study of factors affecting school attendance, found no significant correlation between parent education and persistence in school. This inconsis tency was not supported by Liddle (1962:279), who commented that parents of dropouts had less than "average'’ education, had been unsuccessful themselves in school, and failed to support and encourage education. Nachman reported in a study of Ohio dropouts that "data also indicate that the dropouts who remained in school longer had parents who com pleted more school than the parents of pupils who dropped out in grades 9 and 10" (1963:28). Home Environment Parent and sibling attitudes, as well as various aspects of the child's home living conditions, play an important role in the development of attitudes conducive 75 to persistence in school. Bowden (1917:449) identified four factors predisposing early school leaving, among them ’ ’lack of a suitable place to study." Beckstead (1940) believed that in about 42 per cent of the dropout cases the home environment was in part responsible for the pupil's leaving school before graduation. Delaney's 1950 study of Chicago youth who dropped out identified overcrowding in the home as a significant factor. This appears to be con sistent with Bowden's 1917 finding regarding the lack of adequate study space. Layton (1952) found that unsatisfactory family relationships and friction within the family greatly influ enced a pupil's decision to drop out. Sando (1952) noted that family friction often prevented a child from talking with his parents about school problems. Many parents of dropouts were unaware that their children were faced with particular problems at school. It is reasonable to believe that if a pupil could feel secure in discussing school problems with his parents, he would in all probability persist longer in school. McCreary and Kitch (1953:36) reinforced this concept by stating, "It is not a common practice for drop-outs to talk it over with the school people before they leave." Cook (1956:191) reported that 4.6 per cent of the dropouts were "needed at home." This could mean anything from a genuine demand for family assistance to a convenient 76 excuse to leave school. "Left home" was the reason given by 2.3 per cent of the dropouts in Cook's study. The coun selors reported that 28.1 per cent had left school because of "home circumstances." Again, the category was blind to the detail necessary for adequate interpretation. Snepp (1956:50) listed "need for encouragement from the home" second on his list of reasons for early school leaving. Lee (1958) identified thirteen characteristics of potential dropouts. Included on this list was the item "foreign-born parents." Lee's study was done in the Berkeley-San Francisco area in which there was a substantial population of foreign-born orientals. Children from homes with foreign-born parents often experienced serious conflict between the demands of parental cultural standards and community cultural standards. Belonging within the peer group often became a difficult task for the child. Lee also identified the broken home as an indicator of poten tial dropout. Fraser (1959) found no such correlation between "abnormal" homes and either IQ score or progress in school. Perhaps this is an American peculiarity. Swan (1961:77) included "broken home" as a factor in lack of achievement and subsequent dropout. Fraser (1959) con curred with Delaney's (1950) finding that children from large families did less well in school than did children from smaller families. Bowman and Matthews (1960:31) tabulated family size and found a positive correlation with 77 early school leaving. Family discord appears to be a significant indi cator of potential dropouts. Plott (1964) observed that the families of dropouts (1) displayed obvious conflict between parents regarding supervision of children, and (2) showed little interest in either the child* s activities or * his school progress. Sibling Influence Persistence in school is associated with the favor able attitudes and behavior of siblings toward education. Hollingshead (1949:337) stated that the mean level of grades completed by siblings was more significant than parent education. For example, 19 per cent of older sib lings dropped out at the end of the eighth grade, whereas 52 per cent who were the younger siblings of dropouts left during or before the eighth grade; 45 per cent of in-school group siblings completed high school, but only 2 per cent of the siblings of the dropouts graduated. Layton (1952:2), in a Detroit study of dropouts, commented that the attitude of siblings toward education influenced a younger child's decision to drop out. Epps and Cottle (1958:92) observed that the educational attainment of siblings was an impor tant influence on potential dropouts. Greene (1962:52-54) reported that 58 per cent of the dropouts in elementary school were from homes where siblings were also dropouts; concomitantly, 47 per cent of the high school graduates 78 . were from homes where siblings graduated. Liddle (1962: 279) noted that first-born and only children left school less often (when the factor of socioeconomic status was i controlled) because parents usually were more interested in first-born and only children and, therefore, were more concerned with their achievement. Socioeconomic Status Bowden (1917:450), writing in the early years of dropout studies, identified economic factors as one of the important considerations in early school leaving. Sando cited a study by Counts (1922) as one of the most compre hensive early studies of the vocations of fathers of dropouts, quoting Counts' conclusion: If we examine the entire high school population, we find certain occupational groups very well and others very poorly represented, in proportion to their numbers in the general population. Among the former are the five great non-labor groups with professional service occupying the most advantageous position, followed by the proprietors, commercial service, managerial service, and clerical service. At the other end of the series are the lower grades of labor with common labor almost unrepresented and fishermen, and the miscellaneous trades of machine operatives in the manufacturing or mechanical industries, occupying somewhat better positions in the order named. (1952:35) The following summary of studies reporting the relationship of father occupation to persistence in school tends to substantiate the concept that this is probably an effective index of potential dropout. The "higher level" occupations (skilled labor, professional and managerial) 79 are not immune from children who do not complete high school, and certainly there are many from lower socioeco nomic classes who do graduate. This leads us to consider additional factors involved in a youngster's decision to leave high school before graduation. Gragg (1949) and Hollingshead (1949:331) commented that the social class of the child was important in the prediction of dropouts. Hollingshead discussed this influ ence as a two-way relationship: the class and the family culture furnish one set of standards, beliefs and percep tions about the school system; on the other hand, a different set of attitudes is furnished by school board, administrators and teachers. Layton (1952:2) reported that father occupation was a determinant useful in predicting dropout. Epps and Cottle (1958:92) developed a test of attitude toward school to be used as a predictor of dropouts. Socioeconomic status of the home and father's income level were listed as indicants of potential dropout. Similarly, in Iowa, Van Dyke and Hoyte (1958) reported a relationship between early school leaving and father occupation. Tesseneer (1958:141- 153) reinforced this concept by reporting that fifteen out of twenty studies reviewed identified socioeconomic status as a correlate of early school leaving. The significance of family income was emphasized in the study by Fraser (1959) in which this factor was identified as one of the 80 four most important in the prediction of early school leaving. Professional occupations. Wind (1931:140) in a Minneapolis study reported that less than one half of one per cent of the dropouts had parents working in profes sional occupations, whereas 15 per cent of the stay-ins' parents were classed as professional. It was also noted that not only were there fewer dropouts among the children of professional parents, but they continued longer in school than children in skilled or non-skilled work classi fications. Anderson (1959:28) found that 13 per cent of boy dropouts and 3 per cent of girl dropouts were from professional families. Bledsoe (1959:4) established ratios of parent occupation of dropouts to parent occupation of total school population (expected ratio 1.00); dropouts from professional homes occurred in a ratio of .09. Nach man (1963:25) presented data showing that only 1.7 per cent of dropout fathers were classed as professional, census figures being used to determine that 10 per cent of all jobs comprised the professional category, the implication being that a high correlation exists between father occupa tion and persistence in school. Clerical and skilled labor groups. Wind (1931:140) reported that children of skilled labor groups made up 42 per cent of the dropouts and only 22 per cent of the 81 stay-ins. Anderson (1959:28) found that 32 per cent of boy dropouts and 24 per cent of girl dropouts had parents work ing in semiskilled and clerical occupations. Bledsoe (1959:4) noted that the ratio of dropout parent occupation to total school parent occupation was .06 for clerical workers and 1.02 for skilled labor (expected ratio 1.00). Semiskilled, unskilled and unemployed. Bledsoe (1959:4) established ratios of dropout parent occupation to total school parent occupation (expected ratio 1.00), and reported a ratio of 2.64 for unskilled labor and a ratio of 3.49 for unemployed parents of dropouts. The Maryland Study (1961) identified 46.4 per cent of dropout fathers in unskilled occupations, and found that 6.2 per cent of the parents were unemployed. The writers commented that more than half of the dropouts came from families in which the occupation of the head of the household was relatively unstable and in the lowest income brackets. Swan (1961:77) included inadequate income and low socioeconomic status as factors in academic retardation. Nachman (1963:25) observed that 55 per cent of dropout fathers listed semi skilled or unskilled employment, whereas the Ohio census reported only 30 per cent of all jobs were semiskilled or unskilled. Socioeconomic status independent of dropout. Abbott (1939:162) related the socioeconomic status of the 82 parent, "regardless of ethnic background," to the student's vulnerability to early school leaving, and commented, "Their dropping out of school at undesirably low levels of attainment [is] the result not so much of economic neces sity as of indifference." Aseltine (1941:174) described the effects of socioeconomic status on persistence in school and noted that in families with low and irregular income, where parent education was lacking or limited, the relationship between family occupation objectives and school experience "may be decidedly remote or entirely lacking." He concluded that family background was of greater importance in predicting retention in school than any other single factor. Such other writers as Hollings head (1949), Layton (1952), Epps and Cottle (1958), and Greene (1962) might argue that the influence of siblings was more important. Livingston (1958:197) reported in an Illinois study that employment status of the principal wage-earner was not significantly correlated with early school leaving. Dillon (1949:22) and Sando (1952:88) did not support the finding that parent occupation was associ ated with failure to persist in school. Family Mobility : - The absence of firm social ties and the disappoint ment over failure in new settings prevents the development of attitudes and feelings essential for perseverance in school. Aseltine (1941:182) found that mobility of family 83 residence was a consistent factor with students who dropped out. Layton (1952:2) was able to associate numerous changes in residence with an increased dropout rate among pupils in Detroit. The mobility of dropouts and non leavers can be shown in Table 10 adapted from Sando (1952: 101) . TABLE 10 SCHOOLS ATTENDED AND PERCENTAGE OF DROPOUTS AND STAY-INS Number of Other Schools Attended Dropouts Stay-ins 0 23 28 1 39 31 2 16 26 3 14 6 4 or more 8 9 Young (1955:22-23) reported results of an eight- year study in Tucson (Arizona) in which indications were found that frequent moving from school to school and from home to home was a major dropout factor. He commented that a combination of limited abilities and the encountering of a completely different environment will frequently discour age the student. Cook (1956:191-196), in an analysis of factors related to withdrawal from high school, noted that 84 dropouts had more school transfers than did stay-ins. Snepp (1956:49-54) found that 70 per cent of dropouts "moved frequently." Lee (1958:51) was able to include "moved more than six times between kindergarten and ninth grade" as a distinct factor in the list of thirteen items identified as significant predictors of early school leaving. Bledsoe (1959:36) reported that 9.2'per cent of dropouts had attended only one school district before quitting, while 35;2 per cent had attended two or more schools. Swan (1961:77) identified "attendance at several elementary schools" as an indicator of potential dropout. Vogel (1962:77-78) listed family mobility as a prime factor in school adjustment and concomitantly a factor in the decision to drop out. PHYSICAL FACTORS This discussion concerns differences between the number of boy and girl dropouts and some of the health correlates identified with early school leaving. Sex Most studies of dropouts indicated that sex was related to vulnerability to dropping out. Boys, as a rule, tend to drop out in larger numbers than do girls. Thorn dike (1908) reported that many more girls remained in school than boys. He found that 75 boys for every 100 girls entered high school; yet there were 60 per cent more 85 girls than boys in the senior year. Ayers (1909) reported that 57 per cent of the students in high school were girls. Palmer (1930) stated that more boys than girls left high school before graduation. Eckert (1938:38) found that 55 per cent of the dropouts were boys, whereas only 45 per cent of the stay-ins were boys. Aseltine (1941:23) found consistent but not often significant differences between the numbers of boy and girl dropouts. However, the differ ences were clearly better than chance. Gragg (1949) found boys more likely than girls to drop out. Hollingshead (1949:337) noted that more boys than girls left school before graduation. Dillon (1949:23) indicated that boys constituted 54 per cent of all dropouts, and that girls tended to complete more grades than boys. Tompkins (1951: 310) summarized federal statistics on education, noting that there were more boys than girls enrolled in the elementary school, and more girls than boys enrolled in secondary school. Sando's (1952:91) findings contradicted those of other researchers in his report that there were 43 per cent boys and 57 per cent girls among the eight high schools in Contra Costa (California). Sando offered no clear explanation for this inconsistency, other than that it might possibly be related to the short period of time covered by his study. Livingston (1958:198), similarly, found no significant difference in proportions of boy and girl dropouts. Anderson and Harris (1959:28) found that 86 there were 30 per cent boys and 17 per cent girls among the early school leavers. Bledsoe (1959:4) reported 146 boys to 101 girl dropouts in a study of small communities in Georgia. Health The factor of health as it affects early school leaving has not been a major area of investigation. Delaney (1950:22) examined Chicago children's school health records and reported that early school leavers were identi fied with faulty vision and with hearing defects. No mention was made regarding the extent of similar defects among stay-ins. Lee (1958:75) stated, "Boys and girls with physical disabilities which cannot be corrected and who are ineligible for placement in special educational programs, attend school under a continuous handicap." The following handicaps were considered sufficiently serious to predis pose dropout: uncorrected visual and hearing disabilities; cardiac, diabetic, orthopedic, post-rheumatic and post- tubercular conditions; histories of epilepsy or severe allergies; a period of home teaching due to chronic illness or. extended hospitalization. Fifty per cent of dropouts and 27.2 per cent of girl dropouts had physical conditions serious enough to constitute a significant factor in leaving school, while 39.1 per cent of potential dropouts had serious health problems. Plott (1964:109) commented 87 that many health problems were found among potential drop outs— resulting in continued absence for minor "illness.*1 ABSENCE FROM CLASS The preceding discussions have concerned those characteristics which research has demonstrated to be significantly correlated with predisposition toward early school leaving. There remains a significant area yet unexplored: that of excessive absence from school. Few youngsters quit school without "giving notice" well in advance of their actual leave-taking. The intermediate step between staying in and dropping out frequently takes the form of a period of excessive absence and truancy. The habit of excessive absence and truancy becomes self- sustaining when coupled with continued disappointment in class, rejection by classmates and the absence of positive parental attitudes toward education. One of the first manifestations of maladjustment, loss of interest and dissatisfaction with school is reflected in excessive absence from class. Constructive attitudes, values and motivation to pursue an education deteriorate rapidly in the presence of an aggravated habit pattern of school avoidance. Evidence of this condition was found by Allen (1956), Nachman (1963), The Maryland Study (1963), and others in which the greatest proportion of dropouts coincided with the expiration of compulsory 88 attendance laws. Allen (1956:1) observed that less rigor ous attendance enforcement was evidenced by a larger proportion of dropouts occurring at earlier ages. Com pulsory attendance laws, while they have an effect upon the age at which a pupil will drop out, have small effect upon the number of dropouts. Allen observed: "Enforced attendance does not make school meaningful to the student nor does it reduce his sense of failure or make him feel that he is a valued member of a group of his classmates" (1956:1) . Ewing (1913:252) was among the first to comment upon the effects of poor attendance upon academic retarda tion. He stated that one of the greatest causes of retardation in school was irregular attendance. Aseltine (1941:98) found that stay-ins were absent an average of 9.17 per cent of the time, whereas dropouts were absent 16.43 per cent of the time. Dillon stated in this connec tion : The significance of attendance data as one indication of possible vulnerability to early school leaving stood out clearly when elementary school attendance records were compared with attendance records in junior and senior high school. Three-fourths of the school leavers were in attendance 90 per cent of the time school was in session while they were in elementary school, a record which would compare favorably with attendance of the entire student body. This record dropped in junior high school where only 60 per cent of the school leavers were found to be in attendance 90 per cent of the time. . . . These findings on attendance reveal symptoms to be faced realistically and point to the need for giving serious consideration to attendance records as 89 evidence of maladjustment that may be due to a variety of underlying causes. (1949:45) Gragg (1949:252) wrote that ’’absence from school for more than one-third of the total number of school days in the year immediately prior to the time the pupil reaches the maximum age of compulsory school attendance" is suffi cient to indicate a potential dropout. Dropout absence was almost double that of graduates. Delaney (1950:22-23) observed, "Repeated absence from school is a common symptom of school leaving and . . . this pattern of absence begins as early as the fourth or fifth grade." Snepp (1956:1) found that 80 per cent of the dropouts had "poor" attendance. Lee (1958:121) noted that 20.8 per cent of the boy dropouts had "poor attendance," as did 40.9 per cent of the girl dropouts. Van Dyke and Hoyte (1958:56) found significant differences between absence rates of dropouts and non-leavers. They stated, "Absentee ism as measured by an absence ratio is a significant factor in differentiating dropouts from persisters." Longstreth (1961:7) identified a "dropout syndrome" which included a record of excessive tardiness and truancy. Savitzky (1961:8) listed irregular attendance third on a list of ten dropout characteristics. Swan (1961:77) reported "irregular attendance" as an indicator of poten tial dropout. Lichter stated: Truancy represents an active form of aggression against school requirements. It signifies that a child is so pressed and finds school so unpalatable 90 that he must escape it, even at the risk of severe penalties. (1962:102) Truancy is a logical forerunner of leaving school. Elliott (1963) emphasized that truancy for the dropout was not so much an avoidance reaction as it was a positive action to establish contact with a more suitable environment. Vogel (1962:78) listed attendance as well as family mobility as prime factors in school adjustment. Nachman (1963:20) stated that attendance deteriorated from elementary school to high school, culminating usually in grades nine through twelve. These experimenters reported that the dropouts were characterized by chronic absence and truancy, whether boys or girls. Regular class attendance appears to be an accom paniment to academic success and schpol adjustment, in that absence from class is an obvious indicant of failure and dissatisfaction with school. Research has supported the importance of good attendance and the importance of using attendance records in the identification of potential dropouts. INTELLECTUAL FACTORS Achievement within the classroom is an essential factor in experiencing success and a sense of personal worth in the school setting. Competence in performing at least as well as most of one's classmates on a standardized test of ability is intimately related to persistence in 91 school. The discussion which follows concerns those com ponents of educational experience most closely allied with general intelligence: (1) group test IQ scores, (2) read ing achievement, (3) grade retardation, and (4) grade achievement. Lecky, in a discussion of reading and arithmetic, offered a statement which has meaningful implications for all educational experience: The cause of most failures in school is not insuffi cient or inadequate instruction, but active resist ance on the part of the child. A child fails to learn because, from his point of view, the material to be learned is inconsistent for him to learn it. . . . His attitude (toward the material to be learned) will be changed if we are able to change the self-conception which underlies this viewpoint. (1945:120) Intelligence Test Scores Standardized group test scores provide one of the most common variables considered in the dropout problem. The variation in comparability introduced by different tests, administration techniques, and evaluation methods restricts generalizations. Ellis (1930:792) reported mental incapacity far down the list as a cause of dropouts. Items concerning the desire to earn money and disinterest in school preceded mental ability. Eckert (1938:52) found that the average graduate surpassed about 85 per cent of the dropouts in IQ score, but the "top per cent of the withdrawing pupils are easily the peers of the average graduate," lending credence to the concept that dropouts are a heterogeneous group. Abbott (1939:163) summarized a discussion of IQ score with the thought that IQ was impor tant because of its effects upon scholastic ability and vocational goals. Aseltine (1941:79) summarized IQ scores measured by the Terman Group Test of Mental Ability for groups of drop outs and stay-ins. The median IQ score for all dropouts was 95.5, the median for all stay-ins was 105.0. Dillon stated: Approximately 40 per cent of those for whom intelligence data were obtainable had IQ's above 95, and approximately 20 per cent were above 105. It can legitimately be assumed that the 40 per cent of the youths whose IQ's were above 95 were educable in programs of secondary education as now constituted. (1949:45) Gragg commented, "Dropping out is more frequently associ ated with those occupations which require relatively low intelligence and little training" (1949:241). He found that an intelligence, aptitude, or achievement score on a standardized test which placed the student in a lower decile among the pupils tested justified the use of such information in the prediction of dropouts. Subsequently, Gragg (1950:72) wrote that, in general, IQ score was not a reliable predictor of dropouts. Hollingshead observed: The inschool class IV1s had significantly higher IQ's than the out-of-school class IV1s, but there was no significant difference between class V's matched in the same way. However, we do not believe that the lower intelligence scores of the withdrawees in class IV are an adequate explanation for these youngsters' withdrawal from school before their class equal. (1949:335) 93 Delaney (1950:22-23) found that 46 per cent of the dropouts in the Chicago group obtained IQ scores of 100 or higher. All dropouts, regardless of IQ score, were charac terized by poor scholarship. Sando (1952:99) presented the tabulation of IQ scores for dropouts and non-leavers shown in Table 11. IQ SCORES FOR TABLE 11 DROPOUTS AND NON-LEAVERS IQ Score Dropouts Non-Leavers Below 84 38 32 85-94 24 29 95 and above 29 34 No significant difference was reported between dropouts and non-leavers per se. However, both groups obtained signifi cantly lower IQ scores than all sophomores. Hecker (1953) reported that 40 per cent of all dropouts in Kentucky had IQ scores of 95 or more. These students were considered capable of profiting from addi tional high school education. McCreary and Kitch (1953:34) presented the summary of dropouts in San Diego, exhibited in Table 12. 94 TABLE 12 IQ SCORES OF DROPOUTS AND HIGH SCHOOL GRADUATES Percentage of Percentage of High School IQ Score Range Dropouts Graduates Above average 110 - Average 90 - 109 Below average - 89 Penty (1956:13) investigated a group of sixty non reading dropouts. Three times as many poor readers dropped out as did good readers. A significant difference was obtained between poor readers who dropped out and poor readers who stayed in. Snepp (1956:49) reported that 60 per cent of the dropout group obtained an Otis Test of Mental Ability IQ score at or below 96, 23 per cent scoring 80 or lower. Munoz (1957:169) noted that the IQ scores of dropouts were more than one standard deviation below those of the stay-ins. Lee (1958:71) remarked that the mean IQ score for boy dropouts was 91.3, and 87.7 for girls; 33.3 per cent of the boys and 63.6 per cent of the girls scored below 84. Dropouts from the Iowa secondary schools (Van Dyke and Hoyt, 1958) produced a mean IQ score of 96.6 against a mean of 106.4 for the non-leavers. High IQ scores (120+) were reported for 17.6 per cent of the 12 40 34 33 50 12 95 dropouts--about the same proportion as for all students. Bienstock (no date:18) tabulated the nationwide IQ scores for dropouts given in Table 13. TABLE 13 NATIONWIDE IQ SCORES FOR DROPOUTS - IQ Score P e r c e n t a g e Under 85 SS-S9 90-109 llO-over Dropout s 31 15 48 6 Stay-ins 10 11 63 16 Wolfbein (1959:2) reported similar figures for dropouts and stay-ins: 46 per cent of the dropouts obtained IQ scores of 89 or lower, 48 per cent ranged from 90 to 109, and 6 per cent scored above 110. Woollatt (1961:6) reported data from the New York Holding Power Project, in which a mean IQ score of 99 was found for all dropouts In grades 10, 11, and 12. He explained that the dropouts at the ninetieth percentile had an IQ score equal to the median score of the total group. The IQ of the dropouts at the twenty-fifth per centile was lower than that at the tenth percentile of the total group. Of the dropouts, 7.1 per cent had an IQ score of 93. Davis (1962:799) reported findings that were 96 in contrast to those of Woollatt, in that no significant difference in IQ scores was found between treated and untreated groups of potential dropouts. The only meaning ful correlation noted was between IQ score and course marks. Lee (1962:7) found that the dropout IQ score aver aged 85 or below. In Lichter's (1962:69) sample 48 per cent of boy dropouts and 31 per cent of girl dropouts obtained IQ scores of 110 or over. These findings were consistent with information from the Maryland Study (1963) in which 49.8 per cent of the dropouts attained IQ scores of "average” or better. Nachman (1963:15) reported that dropouts in the Ohio Study obtained an average IQ score of 85, well below the national average of 104 for the Cali fornia Test of Mental Maturity. Fifty per cent of the dropouts had IQ scores of 90 or below. Reading Achievement Thomas (1954:11) commented that reading achievement is a better indicator of academic ability than is an IQ. Interestingly enough, Thomas found no relationship between reading ability and dropout. He did find, however, that grade average was related to reading ability. Penty (1956:13) conducted an extensive study of reading disabil ity and dropouts, 70 per cent of whom read at or above the - sixth grade level. A comparison of reading ability and mental age between dropouts and stay-ins showed that in a large percentage of both groups mental age exceeded reading 97 age. This would tend to confirm Thomas' conclusion that reading achievement was a meaningful indicator of the potential dropout. Snepp (1956:49) analyzed scores of dropouts on the California and Iowa Silent Reading Tests and reported that 21.4 per cent of the dropout group were three or more years retarded in reading, 26.6 per cent were two years retarded and 22.0 per cent retarded one year. Dice (1958:172) studied 433 high school seniors in Pennsyl vania, 70 per cent of whom read below the forty-fifth percentile for twelfth grade students, and 46 per cent of whom were below the forty-fifth percentile in mathematics. Lee (1958:71) identified potential dropouts as reading two or more years below grade level and being retarded in arithmetic two or more years. Bledsoe (1959:3-6) obtained mean reading comprehension scores of 7.9 for ninth and tenth grade dropouts. Greene (1962:54) identified retardation in reading "below grade level . . . with little opportunity to gain feelings of security and worthiness from school activities" as significant. Lee (1962:8) summarized nationwide research, stating that dropouts were retarded in both read ing and arithmetic two or more years. Schreiber gave an excellent statement in a recent publication: To my mind the greatest factor in school drop outs is reading retardation. It affects not only the child's attitude but also the parent's attitude toward school. All parents regardless of their cultural or economic level, or how unschooled they are, know that there is one subject the school is 98 supposed to teach their children and that is: how to read. Yet study after study has shown the aver age dropout is two years or more retarded in read ing. (1962:237) The Maryland Study (1963) showed that almost half (47.5 per cent) of the high school dropouts were reading at or below the sixth grade level. Grade Retardation and Grade Retention In this study the term ’'retardation" refers to a slowed academic achievement. The term "retention" is taken to mean non-promotion. Seymour (1940), in a rather sophis ticated statistical analysis of factors that contribute to dropping out, determined that grade retardation contributed 45 per cent variance for boys and 39 per cent variance for girls to the specific variable--in this case, "dropping out of school." Sando (1952:95) found that 65 per cent of dropouts had been retained one year prior to leaving, seven other dropouts having been retained two years. The largest number of retentions occurred in the fourth grade. A fac tor seldom taken into account concerns the relatively large number of kindergarten, first and second grade retentions, according to Sando. Cook (1956:191) reported that counsel ors indicated that 34.9 per cent of the dropouts left for reasons of "failure and retardation." Contrasted with this, 9.4 per cent of the dropouts themselves gave "failing courses" as a reason for leaving. Snepp (1956:50) stated 99 that 60 per cent of the dropouts had failed one or more grades. Lee (1958:71) noted that the dropouts had repeated or were provisionally promoted one or more semesters. Bienstock (no date:18) summarized the national statistics on grade retardation displayed in Table 14. TABLE 14 NATIONAL STATISTICS ON GRADE RETENTION P e r c e n t a g e Behind more one or years Behind two or more years Boy Girl' Boy Girl Dropouts 87 80 59 44 Graduates 33 20 8 4 Bowman and Matthews (1960:5) reported that dropouts entered school at an older age and were four times as frequently retarded as controls. Academic retardation was one of four attributes of the "dropout syndrome" developed by Longstretb (1961). Greene (1961:53) listed "retarded at least one year in elementary school" as one of seven iden tifying characteristics of dropouts. Lichter (1962:103) commented that past school failures made it exceedingly difficult for pupils to remain in school, and that "pupils became discouraged by the necessity of making up course deficiencies before going on." Lee remarked: 100 Retardation is considered one of the most relia ble measures of the probability that a pupil will not finish high school. Any pupil retarded two years by the time he reaches the seventh grade is unlikely to finish the tenth grade and has only a negligible chance of finishing high school. If a pupil is retarded as much as three years, he is not likely even to enter the ninth grade. (1962:8) Counterevidence was offered in the Maryland Study (1963) in which it was noted that nearly half (45.3 per cent) of the dropouts had not been retained in either elementary or junior high school. Regarding course fail ure, during the semester they left school, 47.5 per cent of the dropouts were failing three or more courses. This may indicate that they were "just marking time," waiting for an opportunity to leave. Plott (1964:105) wrote that all potential dropouts studied were two or more years retarded, were below average in academic achievement, were failing in arithmetic, or were retarded two or more years (according to group achievement test scores). Grade Achievement Many factors may account for the consistent rise in grade-on-leaving during the past sixty years. Better understanding of the individual and his needs, improved curricular offerings, enforced compulsory attendance laws and changes in job opportunities, together with other motivating situations, have contributed to higher grade achievement and age-on-leaving. Law (1898:40-49) studied dropouts in five wards of 101 New York City and reported that most pupils left school at the age of 14, between grades six and seven. Thorndike (1908) noted that most dropouts left between grades six and seven and between seven and eight, 14 per cent leaving at each grade. In each instance, the dropouts represented "summer loss" in that the pupils failed to re-enroll the following Fall. Bowden (1917:447) tabulated dropouts by grade level (see Table 15). TABLE 15 DROPOUT GRADE LEVEL Grade Achieved Percentage of Dropouts Freshman 28 Sophomore 13 Junior 14 Senior 45 Eaton (1922:221-222) found that more than half (52.6 per cent) of all high school dropouts occurred at the ninth grade level, in a Syracuse (New York) study. An additional 20.4 per cent left in the sophomore year, 20.4 per cent in the junior year, and 6.6 per cent of the dropouts left during the senior year. Kline (1933:611) conducted a follow-up study of 102 Thorndike's sixteen communities. He demonstrated that there had been an increase in grade achievement during the early 1930's. Dropouts were leaving before high school in the period 1918-1928, whereas in 1933 more students were dropping out in high school. Dillon (1949:26) presented the data given in Table 16 illustrating the grade achievement distribution of his three-state study. TABLE 16 DROPOUT GRADE ACHIEVEMENT Grade Achieved Drop Number outs Percentage 7 29 3 8 121 10 9 305 26 10 429 37 11 227 19 12 60 5 Snepp (1956:51) found that 17.3 per cent of the dropouts had achieved the ninth grade before leaving, 28.9 per cent reached the tenth grade, 38.9 per cent the eleventh, and 15.8 per cent left during the twelfth grade. Woollatt (1961:4) noted the New York State Holding 103 Power Project concerning dropout grade achievement shown in Table 17. TABLE 17 DROPOUT GRADE ACHIEVEMENT Number of Cumulative Grade Achieved Dropouts Retention 7 55 99.6 8 211 98.1 9 (age 16) 519 95.1 10 679 90.9 11 657 86.8 12 495 82.9 The data presented here are representative of scores of studies done during the past half century. Not only are students remaining in school longer--in many instances until the end of compulsory education--but the dropouts are occurring at higher grade levels. In the early 1900's the greatest number of dropouts occurred in the sixth and seventh grades, whereas more recent studies show that the greatest frequency of dropout occurs in the tenth and eleventh grades. EMOTIONAL CONCOMITANTS 104 The individual's characteristics are a product of nature and environment. Reed observed that a person "develops by living; by living in math, in the home, with friends, by building blocks and hotrods, and by running the 100 yard dash" (1962:2). The concern of the present study is with the individual in the school environment. Cassel and Coleman (1961:311) commented that "the adolescent remains in school, treated like a child, for a longer and longer period, while he gains social sophistication earlier and earlier." Slater (1960:7) found among a group of college students that the incorporation of curricular and personal goals into the self was a central task essential for persistence in an academic setting. Lecky (1954:120) suggested that the process of incorporation included only that which was perceived as having relationship to the self. Slater (1960:7) demonstrated that home and family experiences influence the process of incorporation, both favorably and unfavorably. He suggested that other pre conditions, such as school experiences, health, peer rela tions, and relations with adults, have a bearing on the facility with which students develop toward achieving this identification of self. Students who are confident, who experience a minimum threat of fear in new settings, who find it easy to identify with new persons, goals and values need only an opportunity to pursue an academic career. 105 Similarly, a student who feels an inadequacy regarding new ideas, new relationships, and with people will fail in the process of incorporation which facilitates persistence. Lichter (1962:247) remarked that school problems "stand out in plain view" and cannot be concealed. Consequently the student feels exposed and develops a sense of resentment, shame and a loss of self esteem. The result is that the dropout has many problems at school and at home, and life becomes overbearing. Previous research offers few studies treating with the precise nature of the dropout personality. The dropout personality is usually found to be more like than unlike that of the stay-in. Selected for discussion here are the factors of: (1) belonging, (2) feelings of inferiority, (3) withdrawn behavior, (4) acting out or rebellion, and (5) generalized evaluations considered as poor personality and social adjustment. Belonging Being a part of and having a secure position in a group is an essential characteristic for the individual's emotional well-being. Swan (1961:79) found that attendance at several elementary schools, preventing the student from developing a sense of belonging (and interrupting his educational experience during a time when the basic skills were being learned) was an important factor in persistence 106 in school. Greene (1962:54) noted that the dropout per ceived himself as not fitting in with others and was not interested in school or school social activities. Lee (1963:10) reviewed studies in which it was determined that when interest and a feeling of belonging were effected, the potential dropouts were held in school through gradua tion, Savitzky (1961:7) reported that students who dropped out tended to display withdrawal symptoms within the school setting. The Maryland Study (1963) found that many drop outs, especially those who listed lack of success as the reason for leaving school, revealed feelings of being left out. Having a feeling of belonging, a feeling of being 4 wanted and respected as a person, is a universal need, and the fact that so many dropouts show little interest in school activities may be a sign that the programs in the schools are not meeting this need. Inferiority Proficiency in learning is seriously depressed when the pupil experiences a sense of worthlessness. Faris (1928:796) found "abundant evidence" that unwise emphasis on competition produced unnecessary feelings of inferior ity. The validity of this observation is as evident today as it was in 1928. Penty (1956:43) reported that more "dropout poor readers" than "stay-in poor readers" felt inferior, were ashamed in class, and wanted to leave school because of reading difficulty. Epps and Cottle (1958:92) 107 were able to differentiate between dropouts and stay-ins by means of true-false items dealing with students' feelings about "things in general." Liddle (1962:278) noted that the dropout was characterized by a "pervasive sense of failure" as reflected on the test of semantic differential. Plott (1964:106) commented that repeated failure in read ing, arithmetic, and school in general contributed to feel ings that the problem was insurmountable, and led to feel ings of devaluation and dejection regarding school. Withdrawn Behavior In many cases, the dropout's attempts at success in the school setting, and perhaps at home, have met with repeated disappointment and open rejection. Sando (1952: 148) noted that the dropout was characterized by shyness and a withdrawn attitude that prevented him from seeking much-needed help (perhaps because of repeated rebuffs). Shanley (1961:116-117) hypothesized that the dropout "tends to be 'withdrawn,' 'self occupied,' and 'anxious'" and that this complex of personal characteristics could interfere with his capacity to perform effectively. Not unlike previous researchers, Davis (1962:800) described the drop out in terms of loss of self-esteem, and the absence of feelings of being wanted, which produced within him a sense of aloneness. Liddle (1962:277) reinforced this concept by relating that dropouts experience a sense of isolation, 108 rejection and defeat and lack a sense of acceptance of and respect for others as determined by a sentence completion test. Riendeau (1962:523) summarized research findings which defined dropout personality dimensions as: lack of personal interest in home, inability to see value in school and withdrawal from participation in school activities. These traits indicated a generalized withdrawing from school, home and society— a withdrawal.from the environ ment. The findings of Sando (1952:146), Munoz (1957:202), Ullmann (1957:264), and Huber (1963:4) regarding the drop out's feelings of being unwanted, rejected, and unliked by teachers and peers might warrant the generalization that dropouts experience more severe paranoid tendencies than do stay-ins. Rebellion Overt rebellion frequently takes the form of "act ing out" behavior, classroom misconduct, and anti-social activity. Allen (1956:12) cited Kahl (1953) who claimed that, if a student were not in rebellion (against his family for emotional reasons), he was more likely to have done well with homework and school. Bloch and Niederhoffer (1958:181) said that one cause of excessive absence resulted from a hatred of school because psychologically it kept the pupil a child, rather than allowing him to assume some aspects of adult responsibility. Liddle (1962:277) 109 noted that rebellious behavior sometimes was expressed as a striking out at everyone and everything in the adult- controlled world. Lichter (1962:247) stated that dropouts "left school because they were motivated to run away from a disagreeable situation; they did not feel impelled to run toward a definite and positive goal." The act of dropping out was an active resolution to the educational problem, an avoidance reaction. Elliott (1963) contradicted this notion with the thesis that the dropout was attempting to take action to better his situation and was therefore running toward a positive solution. Plott (1964:107) observed that the potential dropout frequently displayed hostility toward others as well as toward school. Poor Personal or Social Adjustment This term appears in the research literature to devote any constellation of acts or attitudes inconsistent with accepted (middle class) standards of behavior. The term is derived from the subjective evaluations of the examiner, teacher or counselor. Linton (1945), Davis (1950), Bloch and Niederhoffer (1958) and other social anthropologists have probed the relationships between various personality dimensions and successful adjustment within given subcultures. Adjustment is relative with respect to the cultural aspirations and mores of a partic ular subgroup. 110 Bowden (1917:449) recognized that academic achieve ment and persistence in school were influenced by "lack of a proper attitude on the part of the student." The teacher had, accordingly, as a major responsibility the task of creating within the pupil "proper attitudes" regarding study and deportment. Cook (1956:193) reported that the majority of dropouts had "poorer personality adjustment" than did stay-ins, in the judgments of school counselors. Lee (1958:74) found that about one third of the boys and one fourth of the girls had adjustment problems severe enough to be considered a significant factor in selection as a potential dropout. Bowman and Matthews (1960:6) observed that 18 per cent of the dropouts gave "poor social adjustment" as the major reason for leaving school before graduation. Lefever (1961:110) commented that teachers' ratings of both deviant groups were less responsible, less ambitious, and less emotionally mature than the well- adjusted groups. Cassel and Coleman (1962:63) listed "poor general and personal adjustment" as a significant trait in the identification of potential dropouts. These writers commented that dropouts tended to be distrustful and resentful toward adults and usually displayed purposeless attitudes and no personal goals for achievement. Lichter (1962:267) stated that our educational program was "designed for acceleration and higher achievement [and] does not take into account emotional readiness." Ill Stimulation for maximum motivation is essential for optimal achievement, but overstimulation in the absence of emo tional readiness will inevitably have an adverse effect upon some students. Reed (1962:2), discussing the self in a school setting, remarked that "emotions interfere with maximum potential achievement." This reference to emotions would mean adverse or disruptive emotions, since there are many facilitating emotional attitudes. Elliott (1963) characterized dropouts in terms of emotional disturbance, both at home and at school. SOCIAL INTERACTION Wylie affirmed the contention that one’s self con cept influences his interaction with others. Acknowledging that there are probably many ways in which social interac tion and self concept are related, Wylie asserted that "perhaps the most obvious and important possibility is that one's self concept is shaped through interaction with others" (1961:136). In conclusion, Wylie remarked: "This theoretically crucial class of relationships between vari ables has been inadequately explored" (1961:136). Argyrus (1957:235-236) noted in research conducted in an industrial management situation that adaptive behav ior of people had a cumulative effect which reinforced itself. Norms for behavior and codes of conduct were the outgrowth of the continual re-emphasis of these adaptive 112 mechanisms and tended to maintain the adaptive behavior and make it "proper” within the specific cultural system. Reed (1962:6) wrote that "the degree to which the individual's environment accepts him and makes him feel worthwhile" strongly influences his self concept and accordingly biases his reactions to his environment. Concomitantly, the indi vidual develops an interpersonal environment of rules and regulations unique to his particular situation. Social interaction, then, must be discussed with respect to: (1) interpersonal relationships in general, (2) relationships with teachers, (3) peer relations, and (4) social interaction expressed through extra-curricular activity. Interpersonal Relationships The type of interaction developed between an indi vidual and the persons within his environment was found to be a significant correlate in his decision to drop out of school. Beckstead (1940:37) reported that 14 per cent of dropouts were anti-social, expressing neither a desire to make new friends nor to extend themselves to keep those that they had, whether adult or peer group members. In general, dropouts criticized society more severely than did the stay-ins. Dropouts were especially critical of social class differences, particularly of those more fortunate financially than themselves. Conversely, drop outs tended to be sympathetic with less fortunate 113 individuals. Lecky (1945:153) wrote that sensitiveness and inability to make friends was an outgrowth of incomplete assimilation of existing interpersonal relationships. Early childhood experiences with parents were held to be especially important in that the person no longer has someone to provide strong support in interpersonal con tacts, ’’ unconsciously demanding more recognition that commonly received he feels neglected." Hollingshead (1949:332) observed that the class IV's in school were better adjusted with respect to adult relationships than were the class IV’s out of school. Penty (1956:43) reported from a study of poor readers who remained in school that their more fortunate experiences with inter ested teachers, parents (home security), and adults in the community were associated with graduation. Snepp (1956: 49) reported that 56 per cent of dropouts were socially maladjusted but were not identified as behavior problems within the school. Bowman and Matthews (1960:4) remarked that dropouts were below class average on measures of social and interpersonal adjustment. Lefever (1961:110) presented evidence based upon a differential rating pattern which suggested "the existence of a relatively unsympa thetic interpersonal relationship environment" for the potential dropout. It was stated that this condition contributed significantly to the findings that dropouts were isolates and generally unwanted. Savitsky (1961:7) 114 characterized the potential dropout as a student with "poor personal and social adjustment." Davis (1962:800) reported that a group of counseled dropouts had significantly fewer referrals to the principal's office and for reasons which were less severe, indicating an improved interpersonal environment. Liddle (1962:276) identified the dropout as being dissatisfied with his adjustment to the school social atmosphere. Liddle felt that this below-average social adjustment was a significant factor in the dropout's deci sion to leave. Inadequate social and interpersonal rela tionships were important determinants in the identification of potential dropouts, according to Riendeau (1962:523). Nachman (1963:24) found that fewer than 4 per cent of drop outs were rated as above-average in social and interper sonal relationships, whereas more than 40 per cent rated below average. They found that a higher proportion of early dropouts received below-average interpersonal ratings than did later dropouts. Teacher Relations Differentiated from interpersonal relationships in that within the school setting the teacher assumes a particular role which is apart from other adult relation ships. Bowden (1917:48) stressed that the most important duty of the high school teacher was the creation of a suitable attitude in students. He felt that teachers who 115 were unduly strict and unthoughtful of the pupils as individuals tended to alienate students' feelings about persisting in school. Hollingshead (1949:331) wrote that the attitudes and actions of teachers tended to accelerate the pupil's decision to leave before graduation. This, he reported, was particularly evident among children from class IV and class V. On the other hand, teachers of middle-class children displayed attitudes which were con ducive to persistence in school; similarly, the pupils and parents accepted--in fact welcomed--teacher assistance. Sando (1952:140) presented considerable evidence that there were significant differences in attitude between the dropout and the stay-in on the matter of receiving adequate attention and understanding from teachers. Statistically, he reported, the dropout felt, as did his parents, that he did not receive adequate attention or feel that teachers really cared about him as a person. Lee contradicted this finding, reporting: Social acceptance of students on a basis that approximates warm friendly relations and honest respect for every pupil appears to be characteris tic of the majority of the teachers described by the potential dropouts. (1958:201) She reported 145 "approved practices" in forty-five inter views. The potential dropout mentioned: teacher's inter est in the subject 32 times; encouragement and understand ing by the teacher 23 times; and that teachers were friendly toward the subject 22 times. Conversely, Shanley 116 (1961:114) noted that the drojpout expressed difficulty in relating to the teacher and other adults. Peer Relations Age-mates and classmates have been discussed as meaningful indicators of the dropout and the potential dropout. Aseltine (1941:163) presented data which indi cated that the dropout developed a sense of being an out sider with respect to other pupils. Hollingshead (1949: 342) described the dropout as being isolated by his peers 7 from higher classes, and that this occurred as early as the second, third, and fourth grades. McCreary and Kitch (1953:39) postulated that the dropout was rejected by fellow students and looked (in vain) for friendship from teachers and adults. Counselors reported that 9.6 per cent of all dropouts left for reasons of "feeling rejected" by classmates, although none of the dropouts themselves reported this category, according to Cook (1956:195). Penty (1956:43) observed that students who had difficulty in reading and who were potential dropouts were character ized by personal insecurity, shyness, self-consciousness in a group of age-mates, and fear of being laughed at by classmates. Lee (1958:198) wrote that if potential drop outs could become involved in activities, particularly with friends, that "half the holding-power battle" was won. She commented that friendship was a powerful force in a pupil’s 117 staying in school. At the same time, "friendship" was a recurring concomitant among dropouts. Steadman (1959:274), in an interesting study of the effects of non-promotion on a group of elementary school children, commented that successful non-promotion was accompanied by the retention of many friends and the maintenance of satisfactory peer relationships. Lee (1963:11) reported studies which indi cated that dropouts gave "avoidance of age-mates" as a reason for leaving school before graduation. Liddle (1962: 278) observed that lower socioeconomic students enrolled in middle-class classrooms where achievement was stressed, produced fewer dropouts than matched lower socioeconomic groups enrolled in areas where graduation would set them apart from their peers as "different." He noted that drop outs seemed to encourage each other to drop out so that they might not be alone. This finding was consistent with Plott (1962:107) who commented that potential dropouts tended to associate with students who were known discipline problems--again, apparently, in an effort to be accepted by someone. Extra-curricular Activity Extra-curricular activities customarily refer to activities at school that occur outside of regular class time. More broadly, the term describes out-of-school activities. Sando (1952:260) commented that a number of 118 girl dropouts reported that they attended public dances because the high school students seemed too young for them. This finding would suggest that these girls were more a,dult in social development or that they sought the thrill of a grown-up atmosphere because school had become oppressive. Thomas (1954:17) described the existence of a close rela tionship between extra-curricular activity and persistence in school, stating, MNot one person who dropped out before completing the third year had engaged in even one activ ity.” Eighty-nine per cent of those who remained to graduate participated in one or more activities. Swan (1961:77) listed "doesn't take part in school activities” as one of eleven factors identified as indicants of poten tial dropouts. Included on the list was the item "going steady with older boys" which was consistent with Sando's finding that the potential dropout girl selected male companions older than herself. Cassel and Coleman (1962: 61), in a review of dropout literature, noted that the dropout did not participate in extra-curricular activities, and that dropout girls tended to date older boys. Greene (1962:53) remarked that the dropout appeared not to be interested in school or school activities. Liddle (1962: 279) offered another constructive observation by noting, as did Thomas (1954), that potential dropouts who were interested in music, athletics or something extra-curricu lar almost always graduated from high school, no matter 119 what their ability level. Lee (1963:13), in summarizing recent dropout studies, noted the general absence of participation in extra-curricular activity by dropouts and potential dropouts. Charlap-Hyman (1964:54) commented that one criterion of poor adjustment and potential dropout was ’’lack of participation in activities such as service, representation or recognition." Ristow (1964), summarizing some basic characteristics of dropouts, wrote that the potential dropout had uncomfortable feelings about school, felt out of place, "couldn't dance, didn't go to parties, and didn't know how to talk to other kids." VALUES, ATTITUDES AND MOTIVATION Every course in every subject should have as its chief end the cultivation of these attitudes of mind. As long as acquisition of items of informa tion, whether they be particular facts or broad generalizations, is the chief concern of instruc tion, the appropriation of method into the working constitution of personality will continue to come off a bad second. (Dewey, 1934:1) Ristow (1948:1) commented that the actual use of knowledge in a given situation depended upon whether the individual's attitude permitted the use of such facts. Lecky (1945:120) suggested that learning includes only those factors which are perceived as having relationship to the self. If learning is to be facilitated, adverse attitudes toward the material must be changed, and this requires a change in the self-conception which underlies 120 the attitude. Ristow (1948:2) quoted Jordan (1947) as saying: "Even more than knowledge, attitudes affect actions." Ristow continued; Unless the schools can provide the means of developing certain desirable attitudes in pupils, any extension of attendance requirements beyond the time needed to acquire basic skills in tool subjects may do more harm than good. Some authori ties have pointed out that many pupils who are compelled to attend school longer than they desire, acquire undesirable habits. (1948:2) Values, attitudes and motivation cannot be observed directly. They can only be implied from manifest behavior. Value, Linton believes: . . . may be any element, common to a series of situations, which is capable of evoking a covert response in the individual. An attitude may be defined as the covert response evoked by such an element. (1945:111) Response content is probably emotional, for the most part, but may include "... other types of responses such as anticipations." Linton (1945:113) was of the opinion that value-attitude systems vary considerably in their specific ity. This meant that the fewer factors involved in a response, the greater the specificity. Value-attitude systems, he asserted, function automatically largely below the level of consciousness. "A single system of this sort may underlie several different patterns of overt behavior, providing motivations for all of them" (1945:112). Thus, the student's value-attitude system relative to persistence in school may influence him to drop out in one instance or 121 to graduate in another. Faris (1928:801) was among the earlier writers to demand that schools train students in "moral education." Faris' recognition of the need to develop positive values and attitudes toward education was affirmed by Dewey (1937), Linton (1945), Reed (1962) and others. Aseltine (1941:188) found that the dropouts were reluctant to discuss their failure in school— only three out of 400 admitted that the work was too hard, yet 20 per cent had failed all courses, and all were well behind classmates. Admission of failure is bad for morale, and these students found it easier to claim illness, "going to work," or lack of interest, but never lack of ability to meet the demands of the classroom. Hollingshead (1949) found different social class groups to reflect distinctly different value-attitude systems in their response patterns: They [value-attitude systems] appear to act as agents which keep adolescents from the three higher classes in school long after legal requirements have been met and to drive many in class IV and particularly class V from the school before they have met the educational standards by law. (1949:331) Clique and peer group relations also play an important function in the pupil's decision to drop out; but Hollings head warns: This oft-repeated accusation must be interpreted carefully, for the child who is susceptible to with drawal from school generally associates withHjoys or girls who are~also exposecfto the same class and 122 social pressures which motivate him to leave school. TT949:3327------------------------------------- The assumption that membership in a clique causes a student to drop out, however, is unwarranted. Hollingshead noted that clique membership can be an effective factor in per sistence in school. Lee (1958:198) confirmed this point of view. Sheerer (1949:174) reported a definite and positive correlation between attitudes of acceptance and respect toward the self, and attitudes of acceptance and respect toward others. Sando (1952:249) claimed that parents of dropouts were significantly more critical of school than were parents of stay-ins. This critical attitude undoubtedly is an important factor in the development of adverse attitudes among dropouts and potential dropouts. Allen (1956:10) reinforced this concept by noting that a pupil's attitudes toward school reflected the values of the family. Even before (a pupil) starts to school, he often acquires attitudes toward it. Depending upon what he hears about it from others and how he inter prets this information, he may look forward to school with dread or eagerness or indifference. (Allen, 1956:10) Schneider and Lysgaard (1953:142-149) offered an important contribution to this discussion. These writers noted that the middle-class family displayed a pattern of behavior centered around the postponement of gratification or immediate satisfaction in certain areas. The "deferred gratification pattern" concerned the tendency to delay 123 economic independence through an elaborate process of education. The deferred gratification pattern appeared closely related to the "impulse renunciation" identified by Davis and Havighurst (1941:698-710). -The middle-class family expressed concern that they should save money by postponing and renouncing a variety of gratifications. These examiners, employing a pencil-paper self-identifica tion index, found that lower-class pupils preferred "to settle matters right away," rather than allowing time for tempers to subside. They found that middle-class families were significantly more concerned with college and special training; concomitantly, that lower-class families were more interested in obtaining jobs, entering military service and "other plans." Free spending characterized the lower socioeconomic group, whereas middle-class families were more interested in saving money and providing for further education. Being well mannered and obedient differentiated (p = .01) between the middle-class and lower-class families, in favor of the former group. Middle-class children tended toward prolonged dependence upon parents as indicated by responses to the effect that their parents were saving money for the pupil's future education. These writers concluded that lower socioeco nomic families enjoyed the privilege or advantage of yielding to impulse and reaping "pleasure" by not deferring gratifications. V 124 Interclass mobility is not without particular problems. Schneider and Lysgaard (1953:149) speculated that lower socioeconomic status families who want to "move up" through the acceptance of middle-class tendencies toward deferred gratification patterns may well engender anger, hostility, and derisive reaction: "You're a sissy to study, to save . . . to defer." Rosen (1956:204) also contended that there were differences in concern with achievement and values among different social strata. These findings were consistent with those of Hollingshead (1949:358). Rosen defined "achievement motivation" as a value-attitude system to excel, and "value orientation" as implementing achievement motivated behavior. Achievement motivation probably has its origin in parent-child interactions which are likely to be unverbalized; value orientations are perhaps acquired during cultural training "when verbal communication of fairly complex nature is possible" (Rosen, 1956:210). Remmers wrote in the foreword to Peters (1957) that "one of the well established principles of social psychology is that attitudes are learned." An implication to be drawn from this comment concerns the role of the classroom teacher in providing an appropriate atmosphere in which good attitudes of self-concept can be learned. Lee (1958: 178, 179) noted that when pupils had a positive feeling and attitude toward school, the task of completing high 125 school became easier. She felt that the statements of potential dropouts were not always based upon facts. How ever, she believed, pupils were motivated primarily by their perceptions of school situations; and, as percep tions, the opinions of potential dropouts were meaningful. Gerberich observed: The attention that has been directed toward attitudes by several nationwide surveys of public opinion illustrated the educational importance of attitudes. Attitudes are formed, crystallized, and sometimes modified or changed in the home, the church, on the playground, and elsewhere, as well as in the school. (1962:37) Greene (1962:53) noted the apparent influence of one sibling upon the value-attitude system of another when he found that 58 per cent of dropouts were from homes where siblings had dropped out; even more important was the find ing that 47 per cent of high school graduates were from homes where siblings had graduated. The implication appears to be that attitudes toward persistence in school, whether favorable or unfavorable, were the product of perceived sibling and parent value-attitude systems. Reed (1962:3) commented that perceptual psychology was more concerned with motivation and the contributing value- attitude systems than with ability per se. The term ’’ ability" refers here to a potential to achieve, whereas motivation is founded in desire, value-attitude systems, anxiety to achieve, and aspirations. Reed continued by noting that the approach to personality and behavior is 126 holistic— the individual reacting in accordance with his self concept. No panacea can be offered. The research literature appears to oversimplify the problem of creating new self concepts, inducing positive motivations, and effecting behavioral changes. Lecky (1945) and Linton (1945) both discussed in detail the individual's tendency to resist change. Lecky (1945:119) wrote that a pupil fails to learn because, from his point of view, the material to be learned is inconsistent for him to learn; thus, the cause of most failures in school is not insufficient or inadequate instruction, but active resistance on the part of the child. Linton stated: Value-attitude systems are extremely hard to extinguish. Such systems tend to survive even when their overt expressions have been inhibited in many situations and to reassert themselves with almost undirainished vigor when new situations involving the particular value factor arise. (1945:115) These two writers offer convincing evidence that, while changes in value-attitude systems and changes in the self- concept are possible, they are not probable. The early identification of potential dropouts has been emphasized throughout research literature. Two objec tives would be accomplished through early recognition: the classroom teacher would have an opportunity for indi vidual student contact, and the change of attitude-value systems could be begun at a more favorable time in the 127 personality development of the pupil. Linton commented on establishment and extinction in learning by stating that "generalized responses of the value-attitude type seem to be easy to establish in childhood but exceedingly difficult to establish in adult life" (1945:116). Two possible explanations were suggested. Adult life is reasonably complex and the assumption of new value-attitude systems requires so many readjustments that the change "is more trouble than it is worth." Similarly, the extinction of an adult value-attitude system because of its complex integration would provide a source of discomfort and a sense of futility. ATTITUDE MEASUREMENT One of the most pervasive, but at the same time most subtle, effects of attitudes on behavior both implicit (symbolic) and overt, involves their influences on and selective modifications of re sponses in particular situational settings. (Sells and Trites, 1960:109) Definitions Attitude measurement consists in assessing an individual's response to a set of social objects or situa tions, according to Green (1954:338). Assessment is accom plished by observing a sample of behavior from an attitude universe in response to a particular situation. Green referred to the particular situation that evoked the response, together with a specified set of response 128 categories, as an "item.*' Thus, within the context of the Meaning of Words Test, the instrument used in this study (cf. post, Chapter Three), the concept together with a set of bipolar adjectives separated by several scale value points would constitute an item. Sells and Trites (1960:102) wrote that the content of an attitude was determined by the responses which com prised it; and the set from which these were drawn was called an attitude universe (after Guttmann, 1950a). These writers conceptualized attitude measurement as involving sampling attitude universes and then measuring the uni verses by means of the samples. The attitude measurement is essentially a matter of selecting a scaling method by which response data can be related to attitude variables. Green (1954:335), in an earlier publication, described attitudes as latent or hypothetical variables which derived meaning from "the syndrome of response con sistencies." Response consistency implies that the observ able items used to assess attitude are mutually intercorre lated. The intercorrelation and resistance to change of attitudes has been discussed by Lecky (1945), Linton (1945), and Argyrus (1957). Green (1954:338), in designing a mathematical model for the latent attitude variable con cept, merely reversed this reasoning, saying that response items are intercorrelated because of the latent variable. It follows that if the items are related to the same underlying variable, the items must be related to each other. Definitions of attitude have been subjected to the tyranny of words. Every major writer in the field has developed a description of attitude to support a particular hypothesis. Chave wrote in the foreword to Thurson and Chave (1929:ix): "The attitudes taken by persons indicate the values discovered in their personal and social . . . experience." Thurstone (1959:216) later amplified this concept by defining attitude as the sum total of a man's inclinations and feelings, prejudice, bias, preconceived notions, ideas, fears,threats and convictions about any specified topic. Opinion was differentiated as a verbal expression of attitude--an opinion symbolized an attitude. Ebel (1938), in Ristow (1948:10), wrote that "an attitude is a stabilized mental set which expresses itself in a tendency to react to any members of a class of stimuli in the same general way." Cattell (1946:384) differentiated between true attitudes and "mere" interest measures and specialized attitudes and personality traits which he called "generalized attitudes." Attitudes were contrasted with interests, being defined not only by the object and the strength of the attitude, but also by the direction, in dynamic, affective-conative terms. MacCorquodale and Meehl (1948:96) developed a definition of attitude in terms of intervening variables. 130 Lazarsfeld (1950) discussed attitudes as latent variables. The term latent variable appears throughout the literature, however, many authors referring to Lazarsfeld as the originator. Stevens (1951:1-49) referred to atti tudes variously in terms of manifest variables, phenotypes and indicants. Thurstone (1929) would have referred to these as opinions, manifestations or symbols of attitude. Coombs (1952:145) proposed that attitudes be defined as genotypes, after K. Lewin, in which a phenomenon was described in terms of its causes. Green (1954:335) described attitudes as hypothetical or latent variables rather than as immediately observable variables. The general term latent variable describes the consistency of covariation of different responses to stimuli of the same general class. "The responses are said to covary because they are all mediated by the same hypothetical variable," according to Green (1954:336). Edwards (1957:2) defined an attitude as the degree of positive or negative affect associated with some psychological object. Thurstone (1959) referred to attitudes as "factors," without further elaboration. Gerberich (1962:132) considered "an attitude [to be] a state of readiness that exerts a directive, and some times compulsive, influence upon an individual's behavior." Stern (1963:404) summarized several aspects of attitude in an article on measuring noncognitive variables. Attitudes, 131 he commented, were socially formed, based upon cultural experience and were orientations toward others and toward objects. They were selective, providing a basis for form ing discriminations between alternative courses of action, and were characterized by consistency of response in social situations. Stern concluded with the thought that atti tudes reflected a disposition to an activity, were organi zations of incipient activities, and represented the underlying dispositional or motivational urge. Methods of Attitude Measurement Thurstone and Chave (1929) and Likert (1932) were unquestionably influential workers in the field. Thurstone was the editor of some thirty-five social attitude scales designed to measure attitudes toward a wide variety of practices, productions, organizations, institutions and racial groups. Another series of scales for the measure ment of attitudes consisted of about a dozen generalized scales designed for broader uses than are the individual Thurstone scales. The vast literature continues to grow. Cattell (1946) produced a prodigiously comprehensive summary of factor analysis contributions which covers both structure and measurement of personality and motivation. The book represents an important contribution to the field of attitude measurement. Green (1954) wrote an exhaustive chapter on measurement, presenting the logic of measurement 132 and a detailed account of the more prominent scaling tech niques. Edwards (1957) published a compendium of informa tion on techniques of attitude scale construction. Judgmental Techniques In one instance, statements of opinion are gathered concerning a given subject which judges and then rates the subject according to the favorableness or unfavorableness expressed. A variation focuses on the subject who sorts or arranges prepared statements according to a predetermined system. Psychological scaling methods. Thurstone in 1927 described what he termed- the "Law of Comparative Judgments" which provided a rationale for the ordering of stimuli along a psychological continuum. This concept made possi ble the quantitative study of subjective experiences and % values. Now, according to Edwards (1954:20), these are generally referred to as "Psychological Scaling Methods." Thurstone (1929:93) made the assumption that the distribu tion of all discriminal processes aroused by a given stimulus was normal about the modal discriminal process. This, he referred to as the Maximum Probability Type. The mean of the discriminal process was called the Scale Value, and the standard deviation as the Discriminal Dispersion. An attitude score was obtained by determining the median of the scale values of the statement with which the subject 133 agreed. Chave (Thurstone and Chave, 1929:iv) wrote that Thurstone*s major contribution was this idea of treating the ’’equally-often-noticed** difference or discriminal error as a unit of measurement for the objective description of attitude and opinion. Equal-appearing intervals. Rather than holding attitude constant as did Thurstone (1927), Edwards (1957: 101) suggested a method of equal-appearing intervals which would allow the examiner to vary the attitude and hold the scale value of the statement constant. For every point of the scale he considered the probability of endorsement of a given statement by those subjects scaled at the particu lar point. The distribution of subjects about a given point should be approximately normal with decreasing proba bilities of endorsement at scale positions above and below the point. Edwards stated that a measure of attitude as a latent variable may be considered monotonic if the proba bility of endorsement increases as the value of the latent variable increases (as in the Guttman scale). Similarly, nonmonotonic scales show an increase in probability of acceptance with an increase in the value of the latent variable, reaching a maximum value and then decreasing. Greene (1954:347) commented that judgment methods of attitude measurement were variants of the technique developed by Thurstone. A disadvantage of this method, 134 according to Green (1954:335), concerned the fact that, as the number of items increased, the number of paired com parisons became very large, since the number of paired comparisons was 1/2(K)(K-1) pairs. Methods of successive intervals. Edwards (1957: 122-124) suggested the method of successive intervals to take into account possible inequalities in the widths of the intervals on the psychological continuum. The method required a single judgment from each subject for each statement. This procedure obviated the problems of a large number of statements. The subject's responses were sorted into successive intervals by a group of judges according to the degree of favorableness-unfavorableness expressed by each statement. A frequency distribution was developed to show the number of times a statement had been placed at each of the successive intervals and the frequencies arranged in cumulative order and expressed as cumulative proportions "by multiplying each by the reciprocal of the number of judges" (1957:124). Thurstone (1959), in a current statement of the original 1929 work, introduced the term Subjective Metric as a measure of social, moral, and aesthetic values. Evaluation of scale values was in terms of the mean posi tion of the subjective continuum and by the standard deviation of its dispersion of discriminal processes. 135 j Summated ratings. The method of summated ratings, proposed by Likert in 1932, requires the subject to respond so as to indicate the magnitude of "agreement" or "disagree-, ment." The basis of this approach was a scale marked by the subject. This technique assumed that attitudes as expressed were normally distributed, and that the scale values represented equal units throughout the range. Green (1954:335) referred to these scales as unidimensional, that is, subjects with the same score had about the Same atti tude . Edwards (1954:169) commented that the method of summated ratings was simpler and easier to apply in devel oping attitude scales than was the method of equal-appear ing intervals. He offered support for the contention that, once the judgments had been secured, scale and Q values could be readily determined and that the overall effort required was not appreciably greater than that of the equal-appearing method. t Scalogram analysis. Guttman (1950) suggested another response technique. He proposed in 1944 a non metric procedure for scaling monotone attitude items. In this scale a subject who responded positively to the third item on the scale was almost certain to have responded positively on the first two. A comprehensive description of scalogram analysis is found in Guttman (1950) and in Green (1954:353-358). This technique of analysis can be applied to items with any number of response alternatives. The basic idea in this procedure is that the items can be arranged in an order so that a subject who responds posi tively to any given item also responds positively to all items of a lower order. The rank order of the items repre sents the scale of the items; subjects are ranked according to the number of positive responses. Development of a perfect Guttman scale requires careful preparation "through (tedious) testing and proper item sequence," wrote Green (1954:336). Scoring, according to Green, remains a problem for this technique. Reference was made to several studies treating assorted methods, including IBM techniques; how ever, he noted that if the number of respondents was large, the task of scoring was overwhelming. A second limitation offered by Green was that the procedure was impractical with more than ten or twenty items. Green wrote that a difficulty in this technique is that one "must first decide which response patterns to accept and which to class as error" (1954:359). Edwards (1957:205-206) mentioned three problems involved in the development of the Guttman scale. (1) If the distribution is not normal over the entire Thurstone continuum, then one would expect the majority of subjects to accumulate at the extremes of the scale on the same items. Thus, the modal frequencies for all statements 137 would be high. (2) The neutral statements would be quite poor in that they would not differentiate the middle group, and would contribute considerable error. (3) Not all statements with the same scale values of the Thurstone continuum are equally discriminating. Edwards found that statements, even though they fell within the same scale interval and with comparable Q values, still could differ considerably in their power to differentiate between high and low criterion groups. Green (1954:359) presented Lazarsfeld1s general mathematical model for the measurement of attitude in a discussion of latent structure analysis, commenting that within his definition of an attitude as a syndrome of response consistency, "the items used to assess the atti tude are mutually intercorrelated." Green discussed in detail two special cases of the general model— latent class analysis and latent distance model. Latent class analysis. According to Green (1954: 359-362), the method of latent class analysis has as its basic idea that interrelationships among particular items reflect the existence of two or more distinct subpopula tions which make up the total population of respondents. "Within each subpopulation, or latent class, it is assumed that the items are independent. However, the proportions of people who respond positively to each item will vary 138 from class to class” (1954:360). Several unsolved problems remain in latent class analysis, including item selection and statistical tests of goodness of fit, Green believes (1954:361). Latent distance model. A probability model for scalogram analysis is provided in a latent distance model devised by Green (1954:362). Green noted that this tech nique allowed for the use of imperfect Guttman scales, and that since one might want to know whether the model fitted the data within the limits of statistical fluctuation, the arc sin chi-square test was suggested as appropriate. Unfolding techniques. Unfolding techniques, described by Coombs in 1950, were scaling techniques which Green classed as "in the spirit of scalogram analysis” (1954:358). The subject-rated items, in order of prefer ence, ranged from "strongly agree” to "strongly disagree." In this model, items were viewed as having a fixed position on the latent dimension, and each subject was located on the dimension. Scoring was determined by locating the respondent's relative position of the items from his loca tion. Green observed, "Since direction is ignored, it is as if each individual had folded the scale over, using his location as a hinge” (1954:358). From a set of these rank orders, the order of items on the continuum can be deter mined. Again, Green noted, this technique suffers in 139 usefulness, as did scalogram analysis, in that no effective means has been established to treat errors. Green (1954:365) offered the critical comment that latent structure models and the unfolding technique should have been classed as scaling models rather than as scaling methods, since satisfactory practical methods of applica tion are not currently available (in 1954). Limitations in attitude measurement. This approach is not specific to the field. The limitations reviewed have been identified by the writers as significant for the particular discussion. Construction of attitude measurement devices. These devices have involved sampling a behavior universe and measuring the universe by means of the sample, as stated by Sells and Trites (1960:102). Green defined an attitude as a latent variable, its meaning having been derived from the covariation of responses in some attitude universe. A more precise definition would require a mathe matical model that related the responses (observed varia bles) to the latent variables. The major problem of measurement of attitude concerns the selection of a scaling model by which the response data could be related to the attitude variable. No system of analysis is practical if the items are not carefully worded; they must be insightful and cover the area in question. Green (1954) and Thorndike 140 (1959) specified that the attitude variable must be clearly identified, and measurement then limited to that variable. Mathematical considerations. In attitude scale development, mathematical considerations have been offered by Thurstone and Chave (1929:10) who commented that atti tudes (expressed as opinions) were represented as a point or a restricted range on an attitude continuum, and that all measurement was limited to a linear continuum. Cattell (1946:384) wrote that most existing attitude tests measured only Mthe orientation through the 180 degree place between approval and disapproval." He postulated that there were other planes important to prediction (e.g., curiosity- incuriosity, acquisitiveness-lack of acquisitiveness). Cattell made the point that the "strength" of an attitude on many published attitude scales was actually a statement of vector direction. He criticized attitude scales in general, noting that some doubtful procedures had been used in identifying attitude factors. In some instances, an examiner found simple structure with insufficient popula tion of variables, naming a factor "A" when the loading on "A" was in fact less than "B", accounting for intelligence as a general factor in one study and ignoring it in another. Cattell felt that considerable confusion arose from attempts to identify "attitude factors" in terms which were too specific for social attitudes. From the 141 relatively small contribution which the identified factors made to the variance, one could conclude that several more significant factors remained to be found when the area of attitude study was widened "which would make them corres pond better with known personality factors" (Cattell, 1946: 387) . Reliability The reliability of any scale or test device is an indication of the extent to which replications will yield similar results. If chance fluctuations cause relatively large changes in an individual's score, then any inter pretation of the score is virtually meaningless. Some authors have argued that large chance differences in scores are to be expected in attitude scales because attitudes are dynamic and subject to change. Granted that this is true, it is impossible to make meaningful predictions unless scale scores are stable. Validity The validity of any scale concerns "what" the scale measures. Green (1954:335) discussed three types of atti tude universes illustrative of the literature: (1) elic ited verbal attitudes, opinions given about something or about the self; (2) spontaneous verbal attitude, the opinions expressed by the subject in normal conversation; and (3) action attitude, observed actions with respect to 142 a professed attitude or Opinion. Green commented that (as of 1954) little measurement had been done in this area. Thurstone (1959:218) commented that since attitude scales must have been answered truthfully to be valid, they should be used only in situations that offer a minimum of pressure on the attitudes to be measured. All that can be done is to minimize the conditions that prevent subjects from tell ing the truth, or else to adjust our interpretations accordingly. Thurstone (1959:228) wrote that an attitude scale must transcend the group measured, the range of scale points must exceed the expected range of opinion. Finally, the measuring instrument must not be seriously affected in its measuring function by the object of measurement. To the extent it is affected, the validity of the instrument is impaired. Remmers (1963:362) felt that the most obvious shortcoming of the semantic differential was its apparent lack of "face validity." Interpretation Interpretation of test results is limited to an evaluation of expressed opinion. Thurstone (1929:10) remarked that "all we can do with an attitude scale is to measure the attitude actually expressed with full realiza tion that the subject may be hiding his true attitude." Attitudes change from day to day and from one situation to another. Cattell (1946:470) discussed the problem of 143 establishing that variables had the same meaning in differ ent situations, that self-ratings became distorted in different settings, and "that certain forms of behavior cannot be assessed in questionnaires, that some forms of behavior are misinterpreted or underemphasized by the external rater." SEMANTIC DIFFERENTIAL PROCESS Reviewers of the semantic differential process presented by Osgood (1957) have been quick to lump findings together, in an effort to speed accession of the basic concepts. These brief summaries leave many questions unanswered about the procedure, samples, and development of the process. An attempt is made here to condense the work of Osgood into a concise statement, so as to be able to examine the thinking and effort that has contributed to the development of this novel technique. Semantic differential has been the outgrowth of diverse experimentation that began in 1938 with investiga tions of color-music synesthesia by Karwoski and Odbert. Osgood and his associates followed with a series of research programs substantiating generalized interrelation ships between color, adjectives, and graphic visual repre sentations which quite evidently exist, not just between individuals of a given race or culture, but which are equally evident between diverse cultures. 144 "It seems clear from these studies that the imagery found in synesthesia is intimately tied up with language metaphore, and that both represent semantic relations" (Osgood, 1957:23). Stagner (1946) adapted the case of a parallelism between auditory pitch and visual size. Both Osgood and Stagner found that semantic scales fell into highly intercorrelated clusters, e.g., fair-unfair, high- low, kind-cruel. Each cluster was felt to represent the operation of a single, general evaluative factor. They noted that scales like strong-weak, realistic-unrealistic, and happy-sad were independent of this evaluative group and suggested the existence of other dimensions of the semantic framework. The development of an operational device to measure meaning depended upon the postulate of a semantic space, "a region of some unknown dimensionality, and Euclidian in character" (Osgood, 1957:25). The semantic scales, com posed of a set of bipolar adjectives, constituted a linear function "that passes through the origin of this space." Definition of the dimensions of semantic space was accom plished by factor analysis, determined by the number of orthogonal dimensions that could reasonably be extracted. The location of a concept in the semantic space defined by a set of factors is equated with the evocation by the concept of a set of component mediating reactions, direction in space being equated to what mediators are evoked (from among reciprocally antagonistic pairs) and distance from 145 the origin being equated to how intensely (with what habit strength) these are evoked. (Osgood. 1957:29-30) Osgood advanced the hypothesis that through a generalizing principle a stimulus (concept) would cause the subject to mark that scale position "whose dominant media tor component most closely matches in intensity the corres ponding component in the process associated with the concept itself" (1957:30). This hypothesis is more thoroughly presented in the discussion of the principle of congruity. The positions marked on the scales were held to be the coordinates of the concept's location in semantic space, and assumed that the coordinates in the measurement space were functionally equivalent with the components of the representational mediation process associated with this concept. Construction of a test scale required the assump tion that the intricate representational mediation paradigm comprised an absolute number of elements. Concurrence with learning theory was accomplished by accepting the obtained "point in a multidimensional space" as an indicant of mediational reaction. Osgood stated: The essential operation of measurement is the successive allocation of a concept to a series of descriptive scales defined by polar adjectives, these scales selected so as to be representative of the major dimensions along which meaningful processes vary. (1957:31) Subject variability was unavoidably restricted. Compensa- 146 tion for inadequacies of sampling was offered through meticulous attention "with the sampling of descriptive scales." The initial lists of adjectives were supplemented in subsequent factorizations with additional adjectives taken from Roget's Thesaurus. Succeeding scale samples were kept independent of previous factor results to inhibit the cumulative effects obtained through the use of former scales as criteria. Four variations in methodology were employed to assure the generalizability of the obtained factor structure. The authors stated: Since the same primary factors keep reappearing despite these modifications, we conclude that the factor structure operating in meaningful judgments is not dependent upon these variables, at least. (1957:33) The initial list of adjectives was determined by orally presenting forty nouns from the Kent-Rosanoff list as stimuli to 200 undergraduate students. The subjects were instructed to write the first descriptive adjective that came to mind. A tally of the responses revealed that nearly half of the adjectives were clearly evaluative in nature. Thurstone's "Centroid Factor Method” was applied to extract four factors. Orthogonal rotations into simple structure were performed to assure factor independence. The first rotation yielded responses of an "evaluative" nature, the second rotation identified factors which were labeled "potency," and the third rotation produced an adjective list identified as "activity." The fourth 147 rotation "accounted for less than 2 per cent of the vari ance and appeared by inspection to be a residual" (1957: 36) . Validation for the semantic differential rationale was demonstrated in two subsequent independent investiga tions. Osgood (1957:66) reported a unique study by Soloman in which the Navy sonal trainees rated "meaning" of sounds against fifty bipolar seven-step scales. Soloman, quoted by Osgood, stated: A striking thing to note in the results . . . is that even with such a limited and unusual set of concepts such as sonar sounds, the first three factors evident in the unrotated matrix may be identified as evaluative, potency, and activity. (1957:66, 68) A study was reported by Tucker in 1955 In which artists and non-artists rated abstract and representational paintings against forty bipolar scales. Again, the three factors of evaluation, potency, and activity were clearly evident. Studies of color meaning and advertised products "and the effects of color on the meaning of art revealed the opera tion of evaluative, potency and activity factors in non- linguistic situations" (1955:299-302). The semantic differential process as envisioned by Osgood is not a test device; rather, it is a process or method of measurement. Standardization is necessarily in terms of each particular study. Comparability of subjects, concepts and scales can be achieved only in a particular 148 series of investigations in which there is some measure of control. Concept selection should be consistent with the dimensions of meaning evaluated. The nature of the problem determines-concept and rating scale selection. Osgood commented that, for the most part, concepts and scales were selected on the basis of "good judgment." Ideally, selec tion would be made in areas in which one could expect substantial individual differences to appear. Poorly- defined or unfamiliar concepts will cause spurious regres sion toward the middle of the scales. Scale selection is concerned with the criteria of factorial composition, relevance to the concepts being judged, and semantic stability. The criterion of factorial composition refers to the selection of scales so as to represent each compo nent (evaluative, potency, and activity or other) uni formly. Scale selection must be relevant to the concepts rated as irrelevant ratings tend to reduce effective discrimination. A concomitant criterion, semantic stabil ity, concerns the appropriateness of the concept rated and the bipolar adjectives. Concepts of political issue, for example, should not be rated on scales appropriate to denotative judgments. Sample formats for the semantic differential were presented, together with typical instruction pages. No preference was expressed for any format, except that con venience in responding and scoring should be considered. 149 Instructions to the subjects should include: (1) a state ment of orientation to the task, (2) the relevance of the bipolar scales and the method of marking, and (3) the atti tude of the subject regarding the recording of f,first impressions.” Scoring consists in assigning a numerical value to each rating position. The obtained numbers or factor scores may then be treated statistically, summed over concepts, scales or subjects as required in the inves tigation. Meaning is assessed by the set of rating scores assigned by the subject (or the mean factor score for a group of subjects). Differences in meaning between indi viduals, scales or concepts can be expressed in terms of the geometric distance between concepts. The scale scores establish a set of coordinates fixing each of the concepts as a point in space. Each coordinate establishes the dis tance of a concept from the origin along an axis which represents one of the factors. The axes are arranged in mutually orthogonal positions and intersect at the origin, which is defined as the neutral position on the scale. The meaning of a concept is this point in semantic space, and the similarity, or difference, of any two concepts can be measured as the geometric distance between any two points. The statistic nD" has been.selected to identify this difference. It is determined by taking the difference between scores of concepts for each factor, squaring the 150 difference, summing the squares and computing the square root of the sum. The larger the ”D” the greater the distance or dissimilarity in meaning of the two concepts; conversely, the smaller the ”D” the closer in meaning are the two concepts. Osgood stated that ”the distribution of D is not known,” and probably is not normal in shape. A number of non-parametric tests of significance were suggested. The sign test, or Wilsoxon's Matched Pairs Signed.Ranks Test, could be used if an hypothesis pertained to the difference in distance between two concepts for two groups. The median test, the Mann-Whitney ”U” Test or the Wilcoxon ”T” Test, could be applied if the hypothesis pertained to two unrelated samples. Evaluation Any measuring device should be evaluated in terms of objectivity, reliability, validity, sensitivity, com parability, and utility. Osgood provided a comprehensive evaluation of the semantic differential process in terms of these criteria. The criterion of replication, objectivity or repro ducibility refers to the extent to which a test device or procedure can be duplicated in its essential elements. The semantic differential process is explicit and can be replicated, as evidenced in the consistent findings from 151 the variety of unique situations to which the procedure has been applied. Scoring of rating scales, computation of distance values, "location of concept-points in semantic space, and the production of conceptual structures, are completely objective" (1957:125). Reliability Reliability refers to the accuracy with which an instrument measures; i.e., the extent to which individual subjects will attain the same score upon repeated testing. Reramers (1963:362), discussing the semantic differential process, stated that repeated studies attest to its relia bility, and anyone "who accepts the logic of measurement and of factor analysis will be impressed with the conveni ence, power and flexibility of the device." Consistent with the three sources of variability in the semantic differential process (concepts, scales and subjects), Osgood discussed item reliability,, factor-score reliabil ity, and concept-meaning reliability. Item reliability. The replication of basic rating scores describes the item reliability of the test. The initial factor study included a duplicate item sample con sisting of twenty concepts and forty of the fifty scales at the end of the 1,000-item test. Immediate test-retest correlations across 100 subjects and forty items produced a reliability coefficient of ,85. But, as Osgood 152 commented, the correlation coefficient does not take into account the absolute score differences. Summarizing three investigations, he concluded, "For all types of items (evaluative, potency, and activity), a difference of more than two scale units can be considered significant at about the 5 per cent level” (Osgood, 1957:138). Factor-score reliability. The form of reliability referred to as factor-score reliability is derived from the i stability of the averaged scores of the (evaluative, potency, and activity) item scales. Based upon two inves tigations, Osgood (1957:139) found that an evaluative factor-score change of 1.00 or more, 1.50 for the potency factor, and 1.33 for the activity factor was significant at the 5 per cent level of confidence. Norman (1959) studied several stability character istics of the semantic differential, holding concepts and scales constant and examining variability over occasions and over comparable subject groups. He found inconsistency in individual item reliability, factor-score reliability and concept-meaning. Consistency across groups, however, was excellent. Group mean D values and group mean ratings were both highly reliable. Concept-meaning reliability. Osgood stated: Since the reliability of concept-meaning con ceived as a point in the semantic space is completely dependent upon the reliabilities of the factor scores X53 . of which it is composed, no separate estimates need to be given here. The same holds for the reliabil ity of semantic distances between concepts in the space, where the D formula is applied to paired arrays of factor scores. (1957:140) Validity "An instrument is valid to the extent that scores on it correlate with scores on some criterion of that which is supposed to be measured" (Osgood, 1957:140). Since the semantic differential was intended to measure meaning, and since no generally accepted measure of meaning is in wide use at the present time, the semantic differential process depends upon "face validity." Content validity of "face validity" refers to how well the content of the instrument samples the subject matter or situation about which conclu sions are to be drawn. Osgood (1957:141) defined "face validity" in terms of the correspondence between similari ties and differences in meaning provided by the semantic differential procedure "with the results to be expected from common sense." English and English (1958:576) add the comment: "This is hardly validity at all, though it may contribute to getting a test accepted." The selection of concepts and scales appropriate to the task offers the only assurance of content or "face" validity. Concurrent validity. Correlation with specified contemporary criteria defines concurrent validity. These criteria may be performance scores on other tests or behavior patterns. Osgood and Luria (1954) were able to differentiate effectively the personality structures of Eve White, Eve Black and Jane; a case of triple personal ity. In the same year, Reeves (1954) found significant correlations between "the evaluative locations of TAT pictures judged by subjects against the differential" and clinical judgments of the stories related about the pic tures by the same subjects. Bopp (1955) reported an investigation of schizophrenic patients not under treat ment. She found significantly greater variability upon retest. Inference regarding the validity of the semantic differential was limited by the nature of the sample and the absence of therapy. Osgood (1957:104-116) supported the semantic differential process in a study of expecta tions of political attitude. To a limited extent, these studies "support the behavioral validity of semantic measurement," the writers concluded (1957:143). Empirical validity. Empirical validity of the factor concept was provided in studies by Rowan (1954) and Wilson (1954) in which evidence based on similarity judg ments obtained in the triad method was presented. Validation for the assumption of equal intervals. Osgood credited Messick for the validation of the assump tion of equal'intervals. The analysis which involved "the psychometric method of successive intervals" (Saffir, 1937 155 Gulliksen, 1954) revealed that the variations in successive intervals were small and within the error limits of the instrument (Osgood, 1957:150). Equality of concept distribution. Messick found that positive intervals were consistently greater than were negative intervals, suggesting that concept distribution was distorted. . Osgood (1957:159) found similar results in a study involving lever movement to indicate favorable- unfavorable responses. The investigators found that the positive or favorable movements were stronger and more rapid than were negative or unfavorable responses. They concluded, however, that these distortions were insuffi cient to invalidate the assumption of equality of concept distribution. Prediction of Behavioral Response Response prediction was demonstrated by Osgood in a study of voting preference, and in two studies to determine the meaning of neutral stimuli when associated with speci fied significates. Solly and Messick (1955), cited in Osgood (1957:163-166), introduced evidence that neutral (non-linguistic) stimuli in the form of "stickmen" figures were evaluated in essentially the same manner as were linguistic qualifiers. Dodge (1955) observed that the meaning a nonsense word will acquire can be estimated from the measured meanings of the signs with which it is 156 associated. Sensitivity of the Semantic Differential Process The sensitivity of a test instrument concerns the extent to which it is capable of making "discriminations commensurate with the natural units of the material being studies" (Osgood, 1957:166). Precise differences were established by Osgood between words usually considered synonyms; predicted responses were achieved in eleven of twe1ve concept s. Comparability of the Semantic Differential Process The comparability of a test procedure is determined by the extent to which it can be applied to a variety of situations and the results meaningfully interpreted. Bas ing his assertions upon the meticulous work of Kumata and Schramm (1956) at the Institute of Personality Assessment and Research, University of California, Osgood concluded that Japanese, Korean, and American college students use the same (evaluative, potency, and activity) factors in making judgments. Suci (1960) extended the cross-cultural approach to the American Indian communities of Hopi, Hopi Tewa, Navajo, and Zuni. The semantic differential process, when adapted to monolinguals restricted to their own cultures, produced findings identical to the earlier 157 American studies, providing essential verification of cross-cultural validity and indicating subject comparabil ity. The criterion of concept comparability would require that each scale maintain the same meaning from situation to situation, regardless of the concept. Almost immediately, Osgood commented that this condition could be shown not to hold. However, he contended, if the same factors could be demonstrated with various concepts and different scales, then perhaps the criterion would be satisfied. Turning to the original Thesaurus Study, - Osgood felt that analysis of these data wduld indicate the extent to which concepts correlated across a variety of scales. Concept-scale matrices were factor analyzed by the Centroid Method and rotated by Quartimax, utilizing ILLIAC computer facilities at the University of Illinois. Blind Rotations produced correlation coefficients ranging from .21 to .99, median .49 for a sample of fifty-two scales. A difference in r of .26 was determined to be significant at the .01 level of confidence. Evident from this analysis were changes in meanings of certain concepts when different scales were used. The evaluative scales tended to vary the most, potency and activity being some what more stable. 158 The Utility or Usefulness of the Semantic Differential Process Acceptance and application by members of the pro fession indicate the usefulness of a given test device or procedure. The growing volume of literature relating to various applications of the semantic differential testifies to an increasing interest. Remmers (1963:362) wrote that the semantic differential was "a flexible, widely applica ble, test procedure and is in accord with many criteria of an acceptable measuring device." Attitude Measurement by Means of the Semantic Differential Process Attitude measurement is more than a by-product of research in semantic differential. It is the raison df etre. Osgood discussed attitudes as learned activity— a predisposition to respond--but distinguished from other readiness states in that "they predispose toward an evalu ative response" (1957:189). Accepting this idea, attitudes can be defined in discrete terms along a bipolar continuum and given direction and intensity. Direction is indicated by the favorableness-unfavorableness of the polar term toward which the response is made. Intensity of attitude is represented by the numerical scale value of the response. One purpose of the semantic differential is quantification of attitudes as mediational activity between stimulus and response, in terms of evaluative behavior. 159 Osgood justified obtaining an attitude score by summing overall evaluative ratings, noting that in every analysis a factor readily identified as evaluative had appeared. The reliability of the semantic differential process as a measure of attitude is dependent upon the reliability of evaluative measures. A mean test-retest coefficient of .91 for six evaluative scales retested after five weeks was obtained by Osgood (1957:191). Validity was established by correlating the seman tic differential process with Thurstone scales and Guttmann-type scales. Correlations between these scales were significant at the ,01 level of confidence. Osgood effectively argued the merits of semantic differential as a generalized attitude measure. He noted that it was less laborious to construct than either the Thurstone or Guttmann scales; claiming that a single set of bipolar adjectives against which selected sets of con cepts could be rated was less cumbersome than the separate "master" scales of Remmers; and, what is perhaps most important, the response mode freed the subject from answer ing specific questions or making forced decisions regarding agreement or disagreement. The flexibility in application and relative ease of construction and administration, together with demonstrated comparability, effectively supported the semantic differential process as a generaliz- able measure of attitude. 160 Prediction of behavior, whether covert or overt, is the objective of any test device or procedure. Attitudes, evaluations, expressions of meaning are sampled so as to provide a quantitative means for determining the probabil ity of certain classes of behavior. No single self- administered instrument yet devised is capable of determin ing precisely the manner in which an individual will respond in predetermined circumstances. The identification of identical attitudes does not infer similar behavior. Osgood suggested that, unless careful analyses were made of ratings on the semantic differential, differences in mean ing were not immediately apparent among the responses. Cognitive events such as are. involved in mean ing formation and change do not transpire in isolation from one another; human learning and thinking, the acquisition and modification of the significance of signs, involve continuous interac tions among cognitive events. (Osgood, 1957:199- 200) Attitude and meaning, then, can best be understood as aspects of cognitive interaction, based upon the principle of congruity. Cognitive reaction in its simplest form involves the interacting of two signs. Whenever two signs are related by an assertion, the mediating reaction characteristic of each shifts toward congruence with that characteristic of the other, the magnitude of the shift being inversely proportional to intensities of the inter acting reactions. (Osgood, 1957:200-201) An assertion is a cognitive coupling or interaction of two 161 signs brought into a particular evaluative relation elic ited by linguistic or behavioral activity. Osgood general ized regarding direction and location in cognitive interac tions by stating that whenever two signs are related by an assertion, . . . they are congruent to the extent that their mediating reactions are equally intense, either in the same (comparable) direction of excitation in the case of associative assertions or in opposite (reciprocally antagonistic) directions in the case of dissociative assertions. (1957:203) That is to say, like sources can be expected to support ideas of which we approve and to reprove ideas we are against. The establishment of congruity in response is momentary. Having served to effect a response, the source and concept signs return to their respective cognitive locations. These locations are not fixed and repeated cognitive interactions can effect semantic change in language. Attitude or change in attitude is reflected in the factor score (as an assertion) which is established by the attitude toward the source and the existing attitude toward the concept. Predictions thus generated can be applied to both these sources, and to the concepts evaluated. Discussion Kendler (1961:459) commented that, in relating the forces of the mediational process with factor analysis, Osgood and his associates have developed objective measures 162 of meaning which should ’’encourage tough-minded research in many areas of human behavior that heretofore lacked adequate means of coping with the problem of meaning.” Meaning becomes apparent only as it is expressed in atti tude or opinion, through covert or overt behavior. The semantic differential process permits the subject to express cognitive and conative meaning as overt behavior, interpretable as generalized attitude. Cross-cultural investigations (Suci, 1952; clin ical studies; Dodge, 1955; Semans, 1957; Osgood and Luria, 1954; industrial management applications; Triandis, 1959), and the direct measurement of attitudes and opinions (Moscovici, 1963) offer convincing evidence of the general- izability of this testing procedure. Consistency in find ings and statistically interpretable data support Anastasi's (1958) comment that the semantic differential process represents a ’’standardized and quantified procedure for measuring the connotations of any given concept for the individual.” Cautions in interpretation of the D statistic were offered by La Forge (1961) and Cronbach (1958). Chronbach (1958:377) commented that data to be treated statistically, and translated back into abstract propositions, must meet three conditions: (1) Concepts paired with attitudes must be explainable in terras of the attitudes, (2) the hypoth eses must satisfy the assumptions of the statistical model, 163 and (3) if generalization is implicit in the verbal state ment of the hypothesis, the phenomenon must be observed at representative points along the range of intended general izations. He emphasized that these conditions are not often met, the result being that Mwe obtain contradictory conclusions, results traceable to statistical artifacts, complicated interpretations of simple effects and conclu sions which are overgeneralizedM (Cronbach, 1958:377). Remmers (1963:362) discussed the possibility of response sets occurring in the marking of semantic differ ential protocols. He further reinforced the concept that expressions of attitude are varied by the extent to which responses are functions of different cultural and subcul tural groups. Such a problem is common to all verbal measuring devices since meaning derives from experience. In this respect, no two individuals perceive exactly the same meaning for any given concept. In practice, minor differences in factor scores are obscured within groups of• concepts and within groups of subjects. Further limitation was noted by Osgood (1957:327) in recognition that the bipolar adjectives may well not be true psychological opposites, and that they may not "fall at equal distances from the origin of the semantic space and in opposite directions along a straight line passing through the origin." The validity of the semantic differential process 164 depends to a large extent upon discretion in selection of concepts to be rated and the appropriateness of the bipolar adjectives. Validity appears to have been established, in that the process differentiates among the clustered con cepts in much the same way as most people do spontaneously. Osgood noted that further research was needed to determine the extent to which the subject was forced into unnatural judgments by virtue of the adjectives selected,.and the extent to which behavior can be predicted from test responses. Learning theorists will probably want to explore various parameters concerning the principle of congruity. Schramm (1962:261) commented upon this relation between attitudes and beliefs toward the self, and toward the concept of communication. These relationships tend toward "maximum simplicity," which is congruity within Osgood's frame of reference. Thus, any change in attitude is in the direction of consistency between attitudes toward the sub ject and toward the self. Schramm noted, as did Anastasi, that by means of the semantic differential process it is possible to make quantitative evaluations about this inclination toward congruity. The interaction of cognitive events was tested in many ways by Osgood who affirmed the predictive usefulness of semantic differential; however, he pointed out that certainly there were additional factors to be tested which affect cognitive interactions. 165 Acceptance of the procedure throughout the profes sion as an effective means for the evaluation of general ized attitude is reflected in numerous comments by Harris (1956), Cronbach (1956), Gagne (1959) and others. CURRENT PREDICTORS OF DROPOUTS Effective prediction of potential dropouts can probably be made at the present time from knowledge obtained through dissertation and other dropout research during the past thirty years. Studies by Bowden (1917), Palmer (1930), Abbott (1939), Aseltine (1941), Sando (1952), Lee (1958) and Bowman and Matthews (1960) have described remarkably similar constellations of the phenom ena associated with early school leaving. Identification requires only that these findings be applied in the school setting. The lack of widespread publication and support s. for refinement have restricted these instruments to the library shelf. Cottle's School Interest Inventory (research edi tion; Houghton Mifflin Company, 1963) is the only currently-available test specifically directed at the identification of potential school dropouts. This 150-item forced-choice test comprises five clusters of items keyed so as to differentiate: (1) boy-dropouts, (2) girl- dropouts, (3) linguistic-nonlinguistic interests between boys and girls, (4) boy delinquents, and (5) underachievers. 166 Selection of stimulus sentences was developed from item analysis and criterion group techniques based upon phi coefficients and chi-square tests. Cottle (1963:2) reported that the inventory contained eighty-one items that distinguished boy dropouts from boy stay-ins and fifty-two items that differentiated between girl dropouts and girl stay-ins for grades seven through twelve. Subjects respond in a true-false mode to stimulus sentences so as to sort the sentences into two groups. The sentences were selected to cover a great variety of con cepts: social relations, socioeconomic status, cultural level of family, educational aspirations, peer relations, self concept, goal orientation, and many others. Tests of this type are effective predictors, pro viding the subject: (1) is capable of knowing his own feelings, and interpreting them according to the stimuli, and (2) will truthfully answer each item. Controversial items regarding family behavior and relationships, or items concerning religious activity may be highly predictive with respect to dropouts, but probably have a tendency to arouse uneasiness in the subject. Since a yes-no response is called for, one might expect that the sophisticated test- taker conceivably could identify the ’’proper" response. Research findings, however, appear not to support this. Cottle (1963:2) described the current status of the scales as follows: eighty-one items identify boy 167 dropouts, fifty-two distinguish girl dropouts. Kuder- Richardson Formula 20 reliabilities were .75 for boys (N = 101, grades 8-12) and .68 for girls (N = 85, grades 8-12). Forty-five items comprised the linguistic-nonlinguistic scales. The scale was reported to distinguish between boys and girls in grades nine through twelve, but without sig nificant differences for boys and girls between given grade levels. Kuder-Richardson reliabilities were reported to be .75 (N = 303, grade 9) and .82 (N = 227, combined grades 10-12) for boys and .72 (N = 303, grade 9), and .68 (N = 227, combined grades 10-12) for girls. Boy delinquents (state industrial school placements) were differentiated by 87 of the 150 items (p less than ,01), with another set of 21 items differentiating at the .05 level of confidence. There was limited application in the identification of underachievers. Cottle (1963:2) reported a study by Hales in which nineteen items were identified as significant of male underachievers, and thirty-three significant of girl underachievers. The School Interest Inventory is in the process of further validation, and promises to become an important addition to the guidance materials available to the school counselor. 168 CHAPTER SUMMARY This chapter has presented an overview of school dropout research literature. Statistical studies were arranged according to national, regional, state, city and local Los Angeles investigations. The statistical reports have established an increasing proportion of high school graduates during the past fifty years. Direct comparisons between studies are limited by inconsistencies with respect to the grades reported on, base grade, definition of the dropout and other factors. Nationally, some 70 per cent of all pupils entering the fifth grade can now be expected to graduate from high school. Modal dropout age is sixteen to seven teen years at grade ten. Studies involving several states, or more than one large community, tended to reflect a higher proportion of high school graduates than did the national figures. Three significant regional studies were those of Thorndike (1908), Dillon (1949) and Wolfbein (1959). Additional information was available in these reports regarding grade retardation, school success, and, in some instances, atti tudes toward school. Results of state-wide surveys were essentially consistent with national and regional studies; the former reporting more detailed findings than the broader 169 inquiries. Noteworthy state-wide studies include an Iowa study by Van Dyke and Hoyt (1958), Woollatt's (1961) New York State Holding Power Project, an Ohio Study by Nachman (1963, 1964), and the Maryland State Department of Educa tion study (1963). City-wide investigations represent the bulk of dropout research. In many instances these projects were reports of dissertation research, much of which has since faded into oblivion. Restriction in sample, rigid controls and perhaps more conscientious participation by the respondents have resulted in more precise findings in a wider variety of areas. Findings begin to approach cause- and-effect relationships in the dropout problem. Almost without exception, these are cross-sectional studies; consequently, inference is limited. The dropout might be characterized as follows: 1. More probably a boy than a girl. 2. Between the ages of 16 and 17 years, in grade 10. 3. Basic intelligence (group test IQ score) probably below average, although not neces sarily so. 4. Very likely to be two or more years retarded in reading. 5. Usually retarded in grade level, one or more years. | 170 I 6. In general, dislikes school and teachers, j 7. Poor peer relationships both at school and within the community. 8. Poorer than average in attendance. 9. From a family of many children. 10. A background of attendance at several elemen tary schools (not within the regular promotion sequence). | Further definition of the dropout was provided in : terms of developmental dimensions of the individual. The discussion of physical factors treated those aspects of the individual's environment that are associated with early school leaving: 1. Parent attitude toward education; regarded as i one of the most critical factors in a student's decision to drop out of school. Brooks (1903) was one of the first writers to emphasize the importance of the attitudes of parents; even today, researchers continue to identify the influence of parent attitude as an important concomitant in the decision to drop out. i 2. The home environment in all its aspects makes a decided contribution to the decision to leave school I before graduation. 3. As early as 1917 studies revealed that the j number of children in a family, financial status, and I provision for a place to study were associated with per- | sistence in school. 4. Parent education has been positively associated ! with dropping out. Commonly, parents of dropouts were not high school graduates and many were themselves dropouts. i 5. The educational attainment of siblings has been ! shown to be related to the desire to remain in school or to I drop out. Dropouts tended to have brothers and sisters who | dropped out, I I 6. Dropouts tend to come from homes of lower i socioeconomic status in which the wage earner is semi skilled or unskilled. 7. Just as a positive correlation between family mobility and socioeconomic status has been established, a similar correlation exists between mobility and dropping out of school. 8. Finally, excessive absence from class has been j demonstrated to be a correlate of dissatisfaction with ; school and has been positively related with early school | leaving. | Intellectually, the dropout is quite the equal of ! the stay-in; however, there exists a definite tendency for j | the dropout to score lower on achievement tests than the | average of his classmates. This is consistent with the finding that most dropouts were retarded in reading and i 172 t other subjects as well as frequently being below grade for j age. Encouragingly, during the past fifty years the modal I grade-on-leaving has risen steadily from the sixth grade in |1900 to the tenth grade in 1963. The emotional concomitants of an unsatisfactory f i school experience might be anticipated. The dropout often i feels alone, as though he did not fit in, is inferior to I his peers, and is observed to be self-occupied, shy and j withdrawn. Consequently, the dropout is often character- | ized by "poor personal and social adjustment." There was an implication throughout the research ; literature that attitudes, values and motivations, whether favorable or unfavorable, toward persistence in school were ; the product of perceived parent, sibling and peer-group value-attitude systems. A discussion of attitude measurement was presented ; to illustrate some of the more prominent techniques avail- ; able to the researcher in this field. The works of ; Thurstone and Chave, Likert, and Guttman have provided the I profession with the essentially fundamental attitude j revaluation techniques which have stimulated the development I of other scaling systems. Attitudes must necessarily be | inferred from covert or overt behavior expressed as a predisposition to behave in a specified manner, or as a j 173 verbal opinion. i ! Osgood's semantic differential process represents a j ! significant departure from traditional methods of general- | ized attitude measurement. The technique proposes a unique application of three-dimensional space in the interpreta- i tion of the expressed meaning of words and phrases. The j semantic differential measures these connotations indi- j j rectly. Concepts (usually a word or group of words) are | rated along a seven-point continuum delimited by two adjec tives of opposite meaning. The task consists of making the j ^ratings rapidly so as to capitalize on first impressions. There are no right or wrong answers--only subjective rat ings are asked for. Scoring may be accomplished by factor analysis of the scale scores, grouping them according to the respective bipolar adjectives. Thus, average scale : scores can be assigned for each concept. "School” might be indirectly described at 6.0 as "fair” (on a 1 to 7 scale), 4.0 as "active” and 2.0 as "weak.” A second scoring method compares the concepts two at a time, converting the differ- ; ences between the respective ratings into a "distance ! | score" which measures the extent to which the subject ! j perceives the stimuli as similar. I Particular value appears to be found in the evalua- j tion of individual personality changes during psychother- I apy. Individual, group and cross-cultural investigations i . by Endler (1961), Smith (1963), Triandis (1959), Moscovici 174 (1963), Suci (1952) and others attest to the versatility of this technique of attitude evaluation, j There are at present no commercially available |tests or inventories in wide use which are specifically i I oriented to the identification of the potential dropout. {Items from the Illinois Inventory of Public Opinion, SRA |Youth Inventory, and the California Test of Personality | have been correlated with early school leaving by Sando |(1952), Ullmann (1957), Lee (1958) and others. None of t these investigations has produced an effective predictor of dropouts. The School Interest Inventory (research edition) by William C. Cottle (1963) represents a pioneering effort in the self-identification of potential school dropouts. {This is a carefully researched test of 150 forced-choice items keyed for the identification of boy dropouts, girl dropouts, linguistic-nonlinguistic attitudes, boy delin quent scales (differentiated from boy-dropout scales), and an underachiever scale. Currently, the inventory is under going further validation, and shows promise of becoming a {valuable addition to the materials available to the high school counselor. i I | Finally, from this r^sum£ of related research it {appears that the semantic differential process offers a |unique approach to an evaluation of the attitudes and 175 perceptions of those pupils who persist in school and those: who drop out. Sando (1952:247) and Lee (1958:10) both per ceive that ’’ other factors” are needed to explain why one j ' ! pupil drops out and why another, apparently matched on several observable characteristics, continues on in school. What may appear to be a series of vague--at times ! almost illogical--ratings that must be made, is actually a | strength of the technique, for it affords the pupil an | opportunity to respond according to the way he feels, j giving expression to his biases and impressions. Osgood ; (1957) cited several instances in which knowledge of expressed attitudes permitted the prediction of response at a better-than-chance level of confidence. Precise interpretation of clinical investigations has indicated j that the semantic differential process is a sensitive and effective technique. It is not unreasonable to expect that, among the thirty concepts selected for this study, some items will demonstrate significant differences in attitude and feeling between the stay-ins and the dropouts. | CHAPTER III i I | SOURCE OF DATA AND METHODOLOGY | ' • | This chapter describes the nature of the sample |populations with respect to (1) socioeconomic status i expressed as a rating of father occupation, (2) group test IQ scores, (3) chronological age at the time of testing iwith the semantic differential, (4) reading achievement, ;(5) number of absences from school, (6) number of days enrolled in school, and (7) age of the dropouts on leaving. i j A discussion of the development of the test instrument is i given, together with a statement concerning the research j j design and statistical treatment of the data. ! ! NATURE OF THE SAMPLE The southern California community of Torrance, in which this study was conducted, has a population which is 4.5 per cent residential, 21.5 per cent commercial, 35.9 per cent manufacturing (both light and heavy), and 62.8 per cent agricultural (commercial vegetable gardens and exten sive floriculture). Gross population in 1963 was estimated I i at 114,000 of which 1,389 persons were non-white (Negro, 176 Oriental, Mexican-American and others). The median income was $8,050 and median home value $17,800.* The school age population in the 15-to-19 age bracket was estimated at 7,715 for the 1963-1964 school year.** I | The Dropout Sample A sample population of 3,095 pupils completed the | data-gathering instrument, the Meaning of Words Test (MWT). ; The sample consisted of 1,630 tenth grade and 1,465 ! | eleventh grade pupils in three high schools. Table 18 I I describes the apportionment of the dropout sample. Follow ing the testing, and during the school year September 1, 1963 to June 30, 1964, 223 pupils were identified as drop outs. Each student was followed up at the end of sixty days to determine whether he had re-enrolled in a program I of formal education or whether he was, in fact, a dropout. ; Of the 223 so identified, 131 boys and girls provided com- ; plete MWT protocols. Approximately 8.5 per cent (11 pupils) failed to complete the test in the time allowed, or ; mutilated the test cards in such a manner that they were | meaningless. Only one pupil, identified as a dropout, was known to have been present at the time of testing, yet *The Growth and Economic Stature of the South Coast Area of Los Angeles County. 1963 (Los Angeles. Calif.: The Security First National Bank, Research Department, 1963). **Growth is Only Half the Success Story (Torrance, Calif.: Torrance City chamber of Commerce, January 1963), not paged. 178 failed to respond to the test. A substantial proportion of the reported dropouts were absent from class at the time of testing (36.6 per cent, or 48 of 131). A lesser proportion (24.4 per cent, or 32 of 131) of the dropouts enrolled after the test had been administered and left school before the close of the year. TABLE 18 DISTRIBUTION OF DROPOUT SAMPLE Analysis of Dropout Sample Number Dropouts for whom complete MWT protocols were available 131 Incomplete or destroyed protocols 11 Dropouts present in class, not tested 1 Dropouts absent from class at time of test 48 Dropouts not enrolled in school at time of test 32 Total reported dropouts 223 The Stay-in Sample The stay-ins consisted of a random sample drawn from the 2,872 protocols remaining after the dropouts had 179 been identified. Stay-ins were matched with dropouts according to sex, grade level, and group test IQ score groups (89 and below, 90 through 110, and 111 and above). Incomplete or mutilated test protocols were rejected. Supplemental Data Selected for Study Seven areas of particular interest have been evalu ated in order to describe more adequately these samples of students. Socioeconomic status. Pupils were asked to indi cate father occupation, or that of the mother if she provided the family support. Father occupations.were rated in accordance with a revision of the 1949 Warner, Meeker, Eells (1960) scale for rating occupations. Group test intelligence quotient score. The California Test of Mental Maturity, Form 57S, administered as a part of the compulsory state-wide testing program, provided the group test IQ score. This test was adminis tered during the same week the Meaning of Words Test was given, or when follow-up tests were given within one school year. j Test age. Chronological age to the nearest whole i jmonth, as of the test date, was established. Birthdates f [occurring on or before the fourteenth day of the month were predated to the first of the month; similarly, 180 birthdates occurring on the fifteenth and thereafter were postdated to the first of the month following. Reading score. The Iowa Tests of Educational Development. Forms X-3S and Y-3S (1959 edition), which were administered throughout this district do not develop a ‘‘reading achievement" score per se. An index of reading achievement was derived by averaging the standard scores of sub-tests 5 (Ability to Interpret Reading Materials in the Social Studies), 6 (Ability to Interpret Reading Mate rials in the Natural Sciences), and 7 (Ability to Interpret Literary Materials). ■ Absences. Apportionment fiscal attendance records were examined to determine the number of days each pupil was absent. For the purposes of this study, all absences were considered. Days enrolled. The apportionment fiscal attendance records, again, provided the number of days a pupil was maintained on the school enrollment record. Leaving age. Chronological age, to the nearest jwhole month, was determined for each dropout. A similar determination for stay-ins was not determined, since age differentials between dropouts and stay-ins would be evident in comparisons of "test age." 181 DESCRIPTION OF THE MEANING OF WORDS TEST Rationale i Research literature on the semantic differential |procedure substantiated its value in the measurement of |attitude as one of the major dimensions of meaning in jgeneral. The semantic differential procedure can provide ;an estimate of the way in which a person perceives himself, | others, values, and his environment* Personality dynamics and attitudes are rather well established by ages 16 to 17. jEngle (1959:212) found no significant difference in self- concept test-retest scores over a period of two years, confirming a hypothesis based upon the assumption that the I crystallization of self-concept was achieved earlier (than i jages 15-16) in development. Establishment of the fifty bipolar adjectives by i jOsgood (1957) was based upon factor analysis of the |responses of 100 college freshmen. It appears reasonable j to assume that the dimensions of attitude expressed by the standardization group are representative of the attitudes i expressed by sixteen- and seventeen-year-old high school students. Slater (1960:7) reported personality traits and factors related to persistence in college that were iden tical to those related to persistence in high school, further supporting the use of Osgood's adjective scales and the semantic differential procedure. Lecky (1945:121) 182 ! supported attitude measurement as a meaningful approach to the prediction of behavior by citing several examples which j illustrated "how little we are justified in judging the | future potentialities of the pupil by the record he has | made in the past." It would appear more meaningful to know i j where the pupil is emotionally (expressed as generalized f | attitude) than to tabulate items of past behavior. Person- ! ality, according to Lecky (1945:150), is an organization of | ideas which are consistent with one another; behavior i | becomes an attempt to maintain this unity. A marked similarity can be seen here between Lecky's concept of self-consistency and Osgood's congruity principle. Lecky commented that the nucleus of the self-consistency system | was the individual's concept of himself. Linton (1945:112) ; noted that, once established in the individual, such self I concept value-attitude systems operate automatically, and, for the most part, below the level of consciousness, i Should this be the case, one is advised to employ measuring i devices sensitive to the influences of the subconscious, i The semantic differential process appears to offer this I sensitivity. Livingston (1958) commented that there is available no generally accepted means of identifying the level of motivation of the individual student to achieve to a given ; level of education. He speculated that "each student at a given point in his educational career places a value of ! 183 r education and aspires to a particular level of education or desires to withdraw from school at the earliest opportunity” I (1958:202). The role and Importance of student motivation i j and attitude In relation to other factors require investi- | gation. Dropouts almost always indicate a dislike for | school as one of the reasons for leaving before graduation. ! This all-inclusive statement, by virtue of its very vague ness, reflects an attitude about school which cannot j readily be defined in terms of forced-choice response items | on a test, questionnaire or inventory. i Limitations of denotative items. Layton (1945:118) wrote that, as a general rule, the more specific a response, the greater the proportion of overt to covert response j elements. Conversely, the more generalized the response, ! the greater the proportion of covert to overt elements. The apparent need is for an instrument to sample covert or I broadly generalized response patterns, and to be capable of : detecting quantitative direction and intensity of value- j attitude systems. Lee (1958:180) commented that a serious limitation exists in the questionnaire, forced-choice type | of study which relies upon a student's memory and percep tions. This limitation concerns "purposive forgetting" through the suppression of unpleasant and frustrating experiences. Value-attitude systems, then, might be i expressed as feelings and hunches without needing the 184 specificity of factual documentation. Aseltlne (1941:183) reported a generalized unwill" | ingness on the part of dropouts to say anything wrong about their schools. He thought that most dropouts were unable. i | to face their own inadequacies and selected this manner of t I response as a means of not embarrassing themselves, "it is i interesting and significant to note that boys and girls ■ j adopted different subterfuges for what must have been the j basic causes of withdrawal,*' Aseltlne noted (1941:183). The reasons given for dropping out were not the causes for dropping out, but the occasions, he concluded. Edwards (1957:10) reported that the discrepancies between overt i responses to direct questioning, and the use of "secret j ballot," indicated that the subjects were quite reluctant i | to express their attitudes openly. This is consistent with | the views of clinical psychologists and psychiatrists who ; note that some subjects may not be aware of their feelings I ! about specified psychological objects. Therefore, it would ! appear that the vague, somewhat projective aspects of the semantic differential procedure would tend to discriminate i differences in attitude and feeling. In an analysis of factors related to withdrawal from high school, Cook (1956:193) observed that the use of ; such terms as "dissatisfied with school," "economic need," | "attitude of parent," and "discouraged" had little meaning i 1 j with respect to actual or real reasons for withdrawing I 185 i before graduation. He reported a definite tendency on the part of the dropout to conceal the factors of scholastic failure, grade retardation, and home and family difficul- | ties by giving as reasons for dropping out some ffrather j thinly veiled rationalization.” Lee (1958:61) found that | many of dropouts and stay-ins were rather alike in those characteristics usually ascribed to dropouts, but different i in their feelings and thinking about their school experi- ; ences, i.e., differing in their attitudes toward school. I | Lichter (1962:3) supported these impressions, remarking i i that dropping out of school was probably only a surface phenomenon representative of deeper social and emotional difficulties. Remmers (1957:1) noted that it was possible to describe "delinquency patterning” in terms of negative ; attitudes toward self, others, family, peers, school, : church, work, laws, sex, and things in general. Although : Remmers' work concerned delinquent youth, delinquency and i i i dropout have been correlated elsewhere in the literature : so closely that a cautious generalization from delinquents i ; to dropouts in this instance is not unjustified. Lee i (1963:6) reported recent literature in which youth who I i drop out of school do not readily give information regard- j i ing the reason for leaving. She stated that "many of them j i give vague and meaningless reasons for leaving or become | voluble about some Incident which was important only because it was the last in a series of dissatisfactions.” ! 186 ! Questionnaire responses, she felt, usually were biased so |as to present the dropout favorably. Nachman (1963:38), | too, described discrepancies between the Individual's | i reasons for leaving and those given by the school coun selor, the inference being that the students either failed I to recognize their problems, or were unwilling or unable |to cope with them. Consistently, the dropouts minimized I as reasons for leaving the same reasons the counselors ■ • * believed were basic. I ! I Limitations of interview techniques. Shanley i observed that the potential dropout tended to give soclally-acceptable responses in the interview situation. The contamination of this influence cannot be evaluated. I Lee (1958:180) identified "purposive forgetting" as an i 'important limitation of the interview technique. McCreary land Kitch (1953:36) stated that "it is not a common prac- i ;tlce for dropouts to talk it over with the school people I before they leave." This, again, would indicate the need \ ; for some test device which would be sensitive to the value- ^ - attitude systems that predispose dropping out. i j Limitation of connotative tests. The meaningful- |ness of the test results or their interpretation is i r ^ | dependent upon the assumption that the concepts and scales i j actually measure what is supposed to be measured, and upon the skill of the examiner in interpreting the responses. Subjective evaluation almost always incorporates a measure of personal bias which confounds real meaning. : i i | In summary, virtually every study of dropouts | reveals, one way or another, that pupils leave because of a i j dislike for school. The expressions of dislike take many | forms; most, however, are vague and without definition i because the subject is required to make forced-choice I • responses on a test or to answer questions which make no ! i I allowance for objective‘quantitative expressions of values i and attitudes. The lack of cognizance of the true reason I for an unfavorable attitude toward school supports the need for an evaluative instrument sensitive to subconscious value-attitude systems. Based upon a review of the dropout literature, the i many attitudes expressed by dropouts have been grouped ; into seven general categories which comprise the framework i of the Meaning of Words Test described later In this i : chapter. I i Bipolar Adjective Scales ' Osgood (1957:37) provided a list of fifty bipolar adjectives for which orthogonal factor loadings had been derived. Three sets of evaluative adjectives (fair-unfair, pleasant-unpleasant, sweet-sour) were selected. Three sets I j of potency adjectives (strong-weak, large-small, heavy- ! | light) and three sets of activity adjectives (active- | 188 ; ; passive, sharp-dull, fast-slow) were selected to complete j | the nine bipolar rating scales. The favorable and unfavor- I able adjectives were alternated on succeeding lines. The ; three rating scales of the first set were arranged in the order of evaluative, potency and activity with this pattern ! ; repeated twice more. The second set of rating scales was ! ! rearranged and rotated right-for-left, so as to avoid ! | response sets. The evaluative, potency, activity sequence was preserved in the second set of ratings, however. The ] seven-step rating scale was prepared according to Osgood I (1957:81). SELECTION OF CONCEPTS ! Identification of the seven areas for study was i | made from a.review of the dropout research literature: school experience, family relations, social interaction, self concept, authority relations, goal orientation, and ; moral and social values. Within each area of study, ; several concepts were identified which appeared to have ' relevance to the present investigation. The areas of i investigation and the selected concepts are explained at | this point. School Experience The importance of the five or six hours spent in ! school each day cannot be underestimated in the social and emotional development of the student.Teachers and school ! 189 | ; administrators provide father and mother figures, authority figures, and are symbols of security. The attitude of the student toward persistence in school is influenced consid- | erably by his perception of his relationships with teach- I ers, his progress in class, and his perceptions of the ! total school environment. i Teachers. Aseltlne (1941:198) found that more girl i dropouts (41 per cent) than boy dropouts (37 per cent) were ; critical of their teachers. Hollingshead (1949:331) j reported that the actions and attitudes of teachers accel- I ! erated the pupil's decision to continue in school or to leave. Pupils from class V (lower socioeconomic class families) were particularly reluctant to receive help from teachers; feeling that such help constituted "meddling” in ! their affairs. Pupils from the middle-class groups were : more likely to seek teacher help and, in turn, were more likely to remain in school longer. Sando (1952:133) stated that most dropouts felt that teachers were not interested i I in them individually, that they didn't provide understand- | ing and attention. Consistent with Hollingshead's find- | ings, Sando found that half of the dropouts responded that I a- i they were afraid of asking for help in course work and i | regarding problems in social adjustment. He noted further | that there was a significant difference between dropouts and stay-ins. Eighteen dropouts indicated that they were 190 treated fairly one-half of the time or less, while only four stay-ins indicated the same feeling. McCreary and j Kitch (1953:39) commented that dropouts reported "few,” j "none," or "one" when asked how many teachers really cared i ' • about the subject as a person. Many dropouts expressed i ; fear of asking for needed help, they reported. Cook ! (1956:195) found that school counselors had reported that 7.2 per cent of the dropouts left because of "conflicts ! with teachers." Penty (1956:40) reported that "good class- | room relations" was given as a major reason for persisting j ' in school, and was similarly identified by dropouts as a reason for liking school. She reported that many poor readers stayed in school because teachers and counselors helped them or showed an interest in the potential dropout. ! Munoz (1957:202) stated that more stay-ins than dropouts I felt that teachers were fair. Ullmann (1957:264) noted | that dropouts reported that "teachers don't understand me," i "teachers don't like me" more often than did stay-ins. i Lee (1958:200) observed that "the presence of respected ! men and women on faculties of the schools . . . was of : support to potential dropouts." Bowman and Matthews (1960: 45) found that 6 per cent of the dropouts stated that they dropped out because "teachers were unfair." Shanley (1961: ! 102) wrote: j ' - , | One can speculate that perhaps the more basic I sources of dissatisfaction in the school experience | are not the formal rules that guide student 191 I behavior, but are Instead the character of the social relationships between teacher and students, i the "quality** of academic instruction, and the attractiveness of the content of course subject j matter. (1961:102) | Liddle (1962:280) noted "disagreement with teachers" as | second most immediate reason for a decision to leave school j before graduation. Also among the long-standing reasons I < for leaving early was "isolation and rejection by teach- | ers." Many dropouts reported "when I don't know the ! answers, the teacher gets mad and the kids laugh." Lee j (1963:8) wrote that a frequent reason given by dropouts for leaving was "disapproval by teachers." Grades. Grades represent the academic scorecard left behind as a pupil progresses through school. Pond ; (1953:86) noted that "too many poor grades" was the reason ; listed by 10.8 per cent of all dropouts as the primary E cause of leaving school. Cook (1956:195) reported that 9.4 per cent of the dropout group gave "failing courses" : as the reason for early school leaving. On the other hand, ; 34.9 per cent of the school counselors reported failing grades as the primary reason for dropping out. This | apparent inconsistency reflects either the reluctance of the student to recognize the reality of the situation, to admit that he has been unable to meet the competition of i i the classroom, or he has become accustomed to failure and j j he has learned to accept the lack of success as a part of j 192 j everyday living. Lee (1958:81) stated that 3.9 per cent i | of all graduates failed two or more subjects in their last ! | year of high school, whereas 36.8 per cent of the dropouts ! failed two or more classed during their last year of ; attendance. The group studies by Bowman and Matthews j (1960:36) appeared to be better oriented with respect to | reality, with 20 per cent reporting academic failure as i | the major reason for dropping out. Liddle (1962:280) listed "failed examinations" among the three most immediate reasons for leaving school. School. Beckstead (1940:38) quoted one dropout as saying, "School takes care only of the upper half; the rest of us just drift along." This attitude was expressed by ! 30 per cent of the dropouts who evidepced negative feelings i about school. Aseltine (1941:123) concluded that the nega- 1 tive attitude of girls toward school was relatively influ ential in their decision not to persist in school. Boy i dropouts tended to be more favorable in attitude toward ] school than girl dropouts: 65 per cent of the boy dropouts i expressed a liking for school, whereas 57 per cent of the i girl dropouts expressed a liking for school. About the same proportion of boy and girl dropouts expressed a dis- I I ; like for school. Sando (1952:257) established a signifi- i | cant difference in satisfaction with school between drop- i outs and stay-ins. Thirty-five per cent of dropouts 193 ; reported satisfaction and 65 per cent reported dissatisfac tions. Among stay-ins, 86 per cent stated that they were satisfied with school and 14 per cent were not. | Cook (1956:195) disclosed that 20.9 per cent of I the dropouts marked "dislike for school" as the primary | reason for leaving. Snepp (1956:50) listed "dislike for I school" third on a list of four reasons for early school | I leaving. Ullmann (1957:264) commented that more dropouts ! | than stay-ins marked "I hate school" on an inventory of I attitude toward school experiences. Bowman and Matthews i I (1960:44) listed "Just don't like it" first among seven reasons for leaving school; 21 per cent of dropouts gave this as the most important reason for early leaving. Shanley stated: I In the assessment of factors that influence the adjustment at school, general attitudes of adoles cents toward their school experiences and the over all value that they place upon these experiences is of obvious relevance. (1961:79) | The "well-adjusted” pupil gave proportionately more favor- : able responses regarding the Importance of school than did j the potential dropouts. The Maryland Study (1960:Recora- | mendations II C, D) offered the following considerations | for further investigation: "How many pupils do we have who seem to dislike school?" and "What reasons may be given for the apparent dislike of school?" Plott (1964:103) f i remarked that dropouts expressed adverse attitudes toward school, probably fostered by rejection by classmates and ! 194 ! ! dejection over grade retention, repeated failure In class, I | and over-ageness. | Livingston (1958:200) stated that r'the dropout was I > i a student who failed to achieve basic satisfaction from his j school experience." He observed that the philosophy and | climate of the school were important factors in a pupil's decision to drop out. Bienstock (1959:17-19) and Wolfbein ; (1959:4) presented essentially identical data, reporting i that 70 per cent of all dropouts left because of "adverse school experiences." Bienstock (1959:17) reported that 38 ! per cent of boy dropouts and 32 per cent of girl dropouts ! gave this reason. Wolfbein (1959:4) noted 38 per cent of | boy dropouts and 31 per cent of girls gave "adverse school i | experiences" as the reason for leaving. Lichter (1962:246) | observed that the dropout had unsuccessful and unhappy i I school experiences and that most of them "had serious and ! i multiple school problems that Interfered substantially with the educational process." Similarly, Sando (1952:257) noted that the parents . of dropouts were less satisfied with school than were the i parents of stay-ins. Studies concerning parent attitude about school and the persistence of their children in school are notably absent from the literature. Reading. Lanier (1949:207-208), Warren (1954:185), and Snepp (1956:50) discussed inadequate reading ability 195 and associated dissatisfaction as correlates of early school leaving. Pond (1953:86), in a study of Pennsylvania dropouts from 1949 to 1953, identified “trouble in reading ! | books*' as the main reason given by 2 per cent of those I leaving school early. In contrast, Thomas (1954:12) stated that reading ability per se was not a factor related | to dropping out of school. Penty (1956) wrote a treatise J i : on reading disability and its correlation with high school i I dropouts. She found that school subjects which the dropout | disliked most were those which required proficiency in I | reading. Evidence of the association of reading disability and dropout was presented throughout the publication: 50 ! per cent of the poor readers dropped out before graduation, while 15 per cent of the ''readers*' failed to complete high ] | school, according to Penty (1956:29, 51). Contrast was also provided by Penty (1956:39), who | noted that, in response to the question, “What is your ! I general feeling toward reading?" 12 dropouts and 6 stay-ins openly expressed a dislike for reading; their comments ; were, typically: "I would like to read, if I could read i better," "I like to read now, but I disliked reading in i | high school." Penty remarked that these students were | reading material well within their grasp; magazines, sim ple books, and newspapers. Lee (1958:71) noted that the high school dropout was retarded in reading two or more years. Schreiber stated: To my mind the greatest factor in school drop- | outs is reading retardation. It affects not only j the child's attitude but also the parent's attitude | toward school. All parents regardless of their cultural or economic level, or how unschooled they | are, know that there is one subject the school is | supposed to teach their children and that is: how to read. Yet, study after study has shown the average dropout is two years or more retarded in reading. (1962:237) j Ideal teacher. Shanley (1961:120) reported that | the potential dropout group "suggested that Ineffective | teachers be^removed, e.g., 'get rid of some of the teach- | ers,' 'get rid of .' 'get rid of the very old \ 1 teachers,' . . . 'get young teachers,' 'get pleasant teachers.'" By inference, the dropout might regard the i ideal teacher as pleasant, strong and active. I Family Relations Intrafamily relationships appear to be associated with persistence in school and dropping out. Abusive family interactions have been associated with lack of per sistence in school. Kuhl (1953) commented that, provided i the child was not in rebellion against his family for I emotional reasons, he was more likely to do well with ! i homework and in school. Ristow (1964) observed that chil dren were able to tolerate abuse from parents, but have a very low tolerance for situations in which parents phys- I ically abuse each other. Parental discord was considered r by Plott (1964:106) to be a significant correlate of i dropping out of school. He reported that the families of | dropouts displayed conflict between parents, lack of super- j vision of the children, and little interest in the child1s i i activities at home or at school. Lichter (1962:165) j described parental disapproval as an important concomitant | in the maladaptive behavior of adolescents. Lichter com- ! mented that the child* s ability to overcome adversity in j | the school situation was undermined, which made success [ even less attainable. Disapproval fosters a sense of I j worthlessness which Intensifies the child* s anger and dis couragement. Kvaraceus (1959:77) identified **family solidarity'* as one of the focal concerns of the middle- i i class family. By inference, pupils who persist in school I would tend to come from families in which there was | harmony, high regard for each other, and a feeling of well- ! being. Home. The perception of the physical and emotional I environment of the family setting determines the pupil's : attitudes regarding "home." Dillon (1949:63, 64) noted a contrast with previous research in that 70 per cent of the I dropouts were living with both parents. Cook (1956:195) reported that 2.3 per cent of the dropouts stated that they | withdrew from school before graduation because they were i I leaving home. This might be considered an indication of an unsatisfactory home situation. Ullmann (1957:265) 198 noted that a substantially larger number of dropouts than stay-ins checked the statement, "I feel like leaving home.*’ i | Epps and Cottle (1958:92) found from an analysis of test | Items that dropouts and stay-ins could be differentiated i j clearly on statements concerning the home. The School and | Early Employment Survey (1960) indicated that 85 per cent i I of all dropouts and 8 per cent of the dropouts with an IQ score above 110 reported adverse home circumstances as the ! primary reason for leaving school before graduation. I i | Ideal parent. This concept refers to the parent Image most desired by the subject. Abbott (1939:169) noted | that students, when interviewed regarding father occupa- i tion, displayed a tendency to dignify that employment as I : much as possible, "which undoubtedly resulted in mlsclassi- ! : fication of the father on the occupational scale." Dresher I (1954:288) found that one factor associated with persist- I ence in school was "preference for a father in a skilled I occupation." Charlap-Hyman (1964:102) commented that, ! ideally, students would like parents that were more active ; in school affairs, and more interested in the student. j Parent. On the other hand, the concept "parent" was selected to identify the feelings and attitudes of the subjects toward his parents. Sando (1952:92) found posi tive correlation between stay-ins and youngsters who were | at ease in discussing school problems with their parents. 199 Schneider end Lysgaerd (1953:148) found significant differ ences between "middle” and '"working classes” in attitude | toward continued dependence upon parents. "Working class” | or lower socioeconomic children were much less dependent i j upon parents, particularly with respect to future needs. | Snepp (1956:50) noted that dropouts listed the need for | parental encouragement as second on a list of reasons for i ; leaving school early. Munoz (1957:205) supported the con- i cept that more stay-ins felt more favorable toward their parents than did the dropouts. Ullmann (1957:264) stated | that more dropouts than stay-ins reported that "parents neglect me,” and that they felt a barrier between them- | selves and their parents. Liddle (1962:279) observed that ! ! ! most parents of dropouts were disinterested in school and | in their children. This disinterest probably would affect ! the pupil's attitudes toward the parent. Mannino (1962: i 201) reported that a larger proportion of the mothers of i stay-ins showed an active interest in their children and encouraged school attendance. i I Social Interaction | j_ . \ "In our culture there is no more serious handicap than to be unable to measure up to the temperamental and social expectancies of one's associates . . .” (Bloch and Niederhoffer, 1956:153). The socially,maladjusted child experiences an unusual sensitiveness and an inability to make friends due to incomplete assimilation of existing interpersonal rela tionships. Lecky (1945:153) wrote that early childhood | jexperiences with parents are especially important, i.e., ! the adolescent no longer has someone to provide strong | support in interpersonal contacts, therefore, "unconscious- | ly demanding more recognition than he commonly receives, he j feels that he is being neglected." Argyrus (1957:235-236) ! approached this problem from an industrial management set ting, and deduced that the adaptive behavior of an indi vidual had a cumulative effect and reinforced itself. "The continual existence of these adaptive mechanisms tends to I make them norms or codes which in turn tend to maintain the i adaptive behavior and to make it proper behavior for the system." Sando (1952:137) reported that fifty dropouts (50 I per cent of the sample) felt that they needed more help i I j with social relation problems, whereas only twenty-nine stay-ins indicated that this was a significant problem. | Cook (1956:195) noted that the school counselors felt that ! 9.6 per cent of dropouts left because they felt "rejected" by classmates. Contrary to other research, Ullmann (1957: 264) found that stay-ins reported more often that they were ill at ease in social situations than did dropouts. Bowman and Matthews (1960:94) stated that 18 per cent of the drop outs reported poor social adjustment as the major reason | 201 j | for leaving school before graduation. | Llddle (1962:280) wrote that the dropout almost j always demonstrated a below-average personal and social i | adjustment, and that "this lack of adjustment seemed to be ! j self-perpetuating, catching the subject in a vicious cycle j of dislike-rejection-further dislike." Nachman (1963:23- | 24) related that 40 per cent of the dropouts were rated | below average in emotional and social adjustment. They j ! found, in addition, that the earlier a student dropped out, t j the more intense was the maladjustment. | Most people. The stimulus concept, "most people," was selected to evaluate the dropout's attitude-value system toward people in general, in this instance peers, ! adults and younger children being grouped together. No related research was found that pertained specifically to this concept. My best friends. The concept, "my best friends," ; was selected to determine how the dropout evaluated those ; individuals whom he most respected. Lee (1958:199) showed ! that, if students could be involved in activities with friends, "half the holding-power battle" was won. She commented that friendship was a powerful force in a student's decision to remain or to leave school. Deviant value standards combined with avoidance by age-mates, and disapproval from teachers might well provide adequate 202 reason for dropping out of school. Thompson and Nishimura, reported in Charlap-Hyman (1964:66), found that persons chosen sociometrically were perceived by the chooser as more similar to their own ideal self than non-chosen per sons, and that the choosers idealized their friends to the extent that the average correlation between the subject's own ideal and his ratings of his friends were significantly higher than the correlation between the subject's own self and his ideal. Classmates. The concept stimulus, "classmates," was intended to elicit attitude-value perceptions of the classroom peer group environment. "Peer isolation" and "discrimination" were listed second among reasons given by dropouts for leaving school early, according to Ho1lings- head (1949:340). Sando (1952:140) demonstrated significant differences between dropouts and stay-ins in response to questions about getting along with classmates. Fourteen dropouts and six stay-ins indicated that they could not "get along with kids." In response to the question, "Are you one of the gang?" fifteen dropouts, but only six stay- ins reported that they were not a part of "the gang." Dropouts in general were significantly more rejected than stay-ins. Sando noted that girl dropouts felt more rejected than girl stay-ins, and, although boy dropouts also felt more rejected than boy stay-ins, girls were much 203 more critical than boys of peer relationships* McCreary and Kitch (1953:38), summarizing dropout research in California, noted that twice as many dropouts as stay-ins mentioned snobbishness of classmates as a reason contributing to leaving school before graduation. Penty (1956:43) reported that twenty-one of fifty dropouts commented that other children were unfriendly, whereas only seven stay-ins so indicated. Munoz (1957:202) stated that the stay-ins enjoyed the interest of classmates more than did the dropouts. Epps and Cottle (1958:92) commented that, although both boy and girl dropouts were concerned with friendship of classmates, girls tended to be more sensitive to this. Lee (1958:153) commented that "avoid ance of age mates" frequently was given as the reason for dropping out. Livingston (1958:197) associated early school leaving with poor interpersonal relationships with class members— that the dropout did not "fit in" with classmates. Steadman (1959:276), although concerned with prob lems of retention and promotion, found that a child's success in school, whether retained or promoted, was directly related to the child's peer relations and was a significant factor in satisfactory school adjustment. Bowman and Matthews (1960:94) noted, in addition, that students perceived their potential dropout classmates as lacking in friendship qualities. Swan (1961:79) reported 204 that girl dropouts found own-age boys uninteresting and immature. Liddle (1962:280) listed as first among the "long standing" reasons for wanting to leave school early "feeling at odds with classmates/1 He noted that the drop out felt unable to compete with classmates and "for many dropouts this was combined with a feeling that teachers rejected them." Plott (1964:108) found that dropouts experienced a rejection by classmates which fostered a generalized hostility between the dropouts and the stay- ins. Grownups. Lee (1958:201) noted, "The presence of respected men and women on the faculties of the schools . . . was of support to potential dropouts interviewed." From this finding, one might infer that the dropout lacked this support and might be expected to express a less favor able attitude toward teachers. This response might be generalized to apply to all adults, in which case the drop out would tend to express an unfavorable evaluation of grownups. Shanley (1961:114) found that potential dropouts had difficulty in relating to adults1 . Self Concept Lee (1958:199) found that changes in self concept were responsible in part for the decision on the part of some potential dropouts to continue in school until gradua tion. A person will not accept changes that will destroy his self concept, according to Lecky (1945:150). The indi vidua1 will defend himself against threats to the self by not perceiving those meanings in his experience that contradict his present self concept. Defense of the self concept may take the form of lack of interest in class, truancy, and underachievement. Rogers (1947:358) wrote that one's behavior is consistent with one's self concept. Changes in attitude toward self could be made, Rogers con tinued, provided the individual experiences an appropriate atmosphere. This supports Lee's (1958:199) observation that potential dropouts could be influenced to remain in school by the presence of interested, concerned adults. Slater's (1960:7) study of a group of college dropouts presented essentially the same findings. He noted that persistence in college was associated with the incorpora tion of positive values and attitudes toward the self. A comprehensive discussion of self concept is found in Wylie (1961), to which the reader is referred for further discussion. Me. Bloch and Niederhoffer (1958:153) cited Rollo May who noted that anxiety and hostility are interrelated, one usually generating the other. Failure in the struggle for success leads to anxiety, and involves not only social contempt but, more important, self contempt and feelings of worthlessness. Abbott (1939:172) found that dropping 206 out was associated with failure to develop a sense of self worth In either school or personal achievement. Allen (1956:5, 17) noted that dropouts have a sense of failure, and perceive themselves as unsuccessful, left out, and Ignored. Penty (1956:36) wrote that more dropouts than graduates felt Inferior, were ashamed In class, and wanted to leave school because of reading disability. Ullmann (1957:265) found that considerably more dropouts than stay-ins reported: "X lack drive.*' Livingston (1958:200) noted that the dropout usually failed to find a secure place for himself in the school setting. From these find ings one might expect the dropout to feel ill at ease, anxious and insecure, the result being a low self evalua tion. Lichter (1962:165) commented that the combination of parental disapproval and increased sense of worthless ness leads to a deterioration of the potential dropout's functioning, to increasing the damage to his self concept, and to the deepening of his undesirable defenses and personality problems. Liddle (1962:280) supported this view with the observation that the dropout expressed a sense of rejection through below-average adjustment which in ‘ 'effect contributes to a feeling of isolation and defeat and leads to a lack of acceptance and respect for the school situation. Dropouts, Liddle reported, evidenced a "pervasive sense of failure" on a semantic differential test. Huber (1963:4) found that a significant proportion 207 of dropouts Indicated, "I thought I wasn't liked." The Maryland Study (1963) offered this explanation; "Children who are not succeeding in school work and who see them selves as poor students, or their school work as too diffi cult or of little interest, develop at an early age an attitude against school." The writers of the Maryland Study felt that there was considerable need for research on the self-images of children who persist, and of those who drop out from school. My school ability. The concept, "my school abil ity," extends beyond the individual's self perception to an evaluation of the self in the school setting. Aseltine (1941:202) found that only three of 400 dropouts identified academic failure as the primary reason for leaving school, whereas 20 per cent, or eighty, of the dropouts had failed all courses. This finding is consistent with the belief that pupils do not like to make unfavorable comments about themselves and their school work. Support is found also for the concept that dropouts tend to become accustomed to academic failure and assign it less significance than stay-ins. Penty (1956:36) reported that dropouts felt that it took them longer to do their school work than it took other students. Epps and Cottle (1958:92) stated that girl dropouts were more concerned with progress in school than were boy dropouts. Lefever (1961:119) wrote that 208 "'well-adjusted' students' self-perceptions of relative competence . . . might contribute significantly to a more generalized feeling of self-confidence and assurance." Generalizing from this comment, the dropout would be expected to reveal self-perceptions that are lacking in competence, self-confidence and assurance. The potential dropout group studied by Lefever (1961:113) tended to rate themselves low on "Good Grades," "Active in Clubs," and "Good in Sports." The Maryland Study (1963) reported that children who were not succeeding actually perceived them selves as poor students. How I would like to be. Ullmann (1957:265) noted that more dropouts than stay-ins indicated that they would like to be more popular. Charlap-Hyman (1964:78) noted that where a self-ideal discrepancy is indicative of lower self-regard, Wylie's observation is pertinent: It is generally conceded theoretically that a low degree of phenomenal self-regard should be indicative of, or an aspect of, or perhaps even a cause of, "maladjustment." . . . It is felt that high phenomenal self-regard may indicate (a) good adjustment; or (b) denial of problems and self rejection which are actually more serious than those of subjects who consistently admit low self- regard; or (c) unsophisticated conventionality. (1961:203) Charlap-Hyman (1964:59) offered evidence that behavior- deviant boys saw themselves less like persons they admire and would like to be. From this finding one might infer that dropouts would express greater factor score 209 differences than stay-ins with respect to the above concept. How my class sees me. The concept, "how my class sees me," relates directly to the individual's perception of the emotional and social atmosphere in the classroom. Lee (1958:199) stated that each individual views personal interrelationships in two ways: (1) as he sees the rela tionship, and (2) as he thinks others see the relationship. Aseltine (1941:176) noted that dropouts do not have a feeling of belonging, tending to be outsiders with respect to other pupils. Sando (1952:260) observed that many drop outs reported that they were only "small wheels" and that the "big wheels" had little to do with them. Sando summa rized his discussion by stating: When it is realized that boys and girls of high school age are looking for status with their peers, it becomes important that considerable time be spent studying these problems and how best to meet them. (1952:260) Penty (1956:36) stated that, although the reasons for trouble in using ideas gained from reading were about the same for dropouts and stay-ins, the dropouts were unsure of themselves, shy, afraid of being laughed at by class mates, self-conscious, and reluctant to participate because of being afraid of not being right. Livingston (1958:200) observed that dropouts do not relate successfully to the individuals in the "social system of the school." Gordon 210 (1959:131) wrote that "approved classroom behavior was another important factor in gaining and maintaining posi tion." Based upon this consideration, dropouts would be characterized by unacceptable classroom behavior. Swan (1961:77) associated excessive mobility during early school years with dropping out of school. This mobility, he believed, prevented the individual from developing any sense of belonging or acquiring close friends for any length of time. Authority Relationships Authority relationships concern any constellation of interactions in which the individual is subjugated to the wishes of another. These associations are not expressly confined to adults; however, parents, teachers and grownups in general comprise those with whom the adolescent is engaged in an authority relationship. Authority. Beckstead (1940:42) observed that 80 per cent of the dropouts felt that their employers were unfair, and expressed the feeling that they would like to exchange places with "the boss" just for a few minutes— to get even. Sando (1952:116) found that twelve dropouts and only one stay-in reported "unreasonable discipline" as a significant factor associated with dropping out before graduation. Schneider and Lysgaard (1953:149) evaluated their data in terms of being well-mannered and obedient. 211 Differences between "middle0’ and "working" groups were significant at the .01 level of confidence, the "middle" group preferring better manners than the working class, and showing preference for more obedience. Cook (1956:195) noted that 2.3 per cent of the dropouts listed "administra tive request" for disciplinary reasons as the explanation for leaving school. Bloch and Niederhoffer (1958:181) reported that the adolescent protests against authority, and objects to parental restraints because he sees this as one of the prerogatives of the mature adult. Cassel and Coleman (1962:311), in a summary of related literature, noted that the dropout evidences strong resentment toward school control and often is in difficulty with community law enforcement agencies. Greene (1962:53) wrote that the dropout experiences difficulty in relating to teachers as authority figures; the dropout, in need of help, is afraid to ask for it. Charlap-Hyman (1964:51) identified "non-acceptance of teacher authority" as first among criteria given for poor adjustment and potential dropout among a group of ninth grade boys. Rules. This concept relates to the system of written and unwritten regulations affecting behavior in social situations. Lee (1958:152) found that most poten tial dropouts were uncomfortable in rigidly controlled classroom situations; however, many expressed approval of 212 quiet classroom. Lichter (1962:247) commented that all of the dropouts were unsuccessful in adapting to some school requirement or regulation, and three-fourths had trouble in more than one area of school experience. Punishment. Sando (1952:116) found that twice as many dropouts as stay-ins felt that teachers and school officials were unfair in their administration of punish ment (’’sometimes too strict, sometimes too lax”). Dropouts were significantly more dissatisfied than stay-ins with the administration of punishment. McCreary and Kitch (1953:38) reported: "Unreasonable discipline was cited by over one- third of the dropouts.” Especially disturbing to the drop outs is any variation in disciplinary practices. Nachman (1963:20) noted that discipline was a frequent problem for the dropout, and that 27 per cent of boy dropouts and 11 per cent of girl dropouts identified "discipline” as the primary reason for leaving school before graduation. Goal Orientation Sells and Trites (1960:109) stated that the poten tial dropout whose goals are well oriented have set up dominant motives that develop a generalized momentum which causes compatible lesser motives to flow along (and over rides some lesser incompatible motives) with the dominant motive. "The effect on behavior is augmented dominant drives and more consistent behavior.” This point of view 213 is not inconsistent with Osgood's principle of congrulty. Future. Schneider and Lysgaard (1953:146) found that the "working" class tended to discount the future, taking cash now, yielding to Impulse, taking "pleasures" and not deferring any Satisfaction that can be had now. The Implication is that the stay-in would value "future" satisfactions more highly than would the dropout. Rosen (1956:208) stated that future-oriented individuals believe "that planning and present sacrifices are worthwhile, or morally obligatory, in order to insure future gains." Argyrus (1957:233) commented that individuals who were experiencing failure and frustration often expressed short time perspectives. Therefore, they did not value the future as highly as did those who were adjusted to the school environment. Ullmann (1957:264) noted that more dropouts than stay-ins indicated, "I see no future for myself." Kvaraceus and Hiller (1959:77) described the deferment of immediate pleasures and gains for future goals as one of the fen focal concerns of middle-class youth. According to this finding, the stay-ins would value the concept "future" more highly than would the dropouts. Bowman and Matthews (1960:62) listed reasons given by potential dropouts for remaining in school: 14 per cent of the potential dropouts indicated that they needed all the education they could get, even to going to 214 college. Liddle (1962:280) wrote that the dropout not only felt at odds with his classmates; he also felt that he was an academic and social failure, and would continue to be one. Collage. Schneider and Lysgaard (1953:146) com mented on the "marked pursuit of education" which was associated with the middle-class students who persisted in school. Moreover, the "working class" students preferred "a job," "military service," and "other plans" to college. Slianley (1961:89) found an interesting inconsistency among responses of potential dropouts. Regarding the question, "When do you plan to leave school?" 3 per cent indicated: "before high school graduation"; 33 per cent indicated "after high school graduation"; and 46 per cent indicated "after college." The writers suggested that with respect to the large dropout population among California high school students, these figures probably represent contami nation due to the desire to provide socially acceptable answers to the interviewer. A job. Hollingshead (1949:358-359) stated that among the class IV and class V children, groups from which the largest proportion of dropouts occurred, "an essential conditioning factor in his childhood is the desire to grow up and to get a job"--and this means withdrawal from school. Dillon (1949:50) reported that 62 per cent of the 215 early school leavers gave as the primary reason for leav ing: preferred to work, needed money to buy clothes. Hecker (1953) noted repeatedly that the desire to work was frequently given as a reason for dropping out. Schneider and Lysgaard (1953:146) commented that the "working class" preferred working, military service and "other plans" to college or special training of some sort. Cook (1956:195) found that 39.6 per cent of the dropouts gave "going to work" as the primary reason for leaving school. Penty (1956:43) stated that school counselors reported one of the primary reasons for dropping out to be "more enjoyment from work outside of school." Rosen (1956:208) coimnented that students who persist in school believe that planning and present sacrifices are worthwhile, even obligatory, in order to assure future successes. By inference, the poten tial dropout would be expected to give a higher evaluation to leaving school and obtaining a job. Bowman and Matthews (1960:57) observed that 16 per cent of the dropouts gave "financial need" as the reason for leaving school early; and that 32 per cent of the stay-ins remained in school in order to be able to obtain a better job upon graduation. Lombardi (1960:28) found that "work" was first in frequency of response given for dropping out. Savitzky (1961:14) listed "occupation mindedness" as a significant correlate of early school leaving. Swan (1961:79) found that among boy dropouts, the desire for acquisition of a car led to 216 needing money and a job to support It. For this treason, the dropout would value "a job" more highly than would the stay-in. Lichter (1962:247) commented that, although the dropouts often talked of "a job," this was vague and mean ingless, and unrealistic for most. Liddle (1962:280)' listed "seeking work" as the third most immediate reason for leaving school. Graduating. None of the dropout literature reviewed pertained specifically to the concept of "graduat ing." Schneider and Lysgaard (1953:146) found a signifi cant difference between middle and lower socioeconomic groups regarding their attitudes toward completing high school. Bowman and Matthews (1960:60) reported that 14 per cent of stay-ins persisted because "they never had any other idea," and 6 per cent "wanted a high school diploma." By inference, students who dropped out would be expected to value graduation less than quitting school. Shanley (1961: 90) highlighted the unrealistic aspirations of potential dropouts in their finding that 33 per cent stated they expected to quit after high school, and that 46 per cent expected to quit school after college. Only 3 per cent felt that they would quit before high school graduation. The writers commented that this discrepancy was probably a contamination due to the desire of the subjects to give socially acceptable responses before the interviewer. 217 Quitting school. Dresher (1954:288) listed "desire to complete high school" as an Important Item in the hold ing power of the school. Concomitantly, a lack of desire to complete school would characterize the dropout, who would value "quitting school" more highly. Munoz (1957: 193) found significant differences between dropouts and stay-ins, in that all dropouts had said they would like to drop out as soon as possible. Ullmann (1957:264) reported that more dropouts than stay-ins indicated: "I wish I could quit school." Lee (1958:158-161) offered an apparent contradiction, in that the potential dropouts interviewed were not considering quitting school. This finding lacks validity because the examiner firmly restricted the inter views to those aspects about school which the subject liked. Cohen (1963) noted that potential and actual drop outs "come from families and neighborhoods where education is unimportant and quitting school is the normal pattern." Moral and Social Values Moral and social values are the attitude-value systems which predispose an individual to a given feeling or behavior. The connotation of such values usually takes the form of the truth or rightness or wrongness of a thought or act with respect to some eternal verity. Trying hard, toinoz (1957:200) found that dropouts rated the concept "working hard" less favorably than did i 218 stay-ins. Kvaraceus and Miller (1959:77) identified ’’responsibility" as one of the ten focal concerns of the middle-class society. An inference suggested by this finding is that the lower-class society, from which most dropouts appear to come, would not be concerned with "responsibility" and consequently would value "trying hard" less than would the middle-class group. Savitzky (1951:7) listed "lack of ambition and incentive" sixth in a group of ten dropout characteristics. The Maryland Study (1963) suggested that children who do not succeed see their school work as difficult and of little interest. Cheating. Schneider and Lysgaard (1953:146) noted that lower socioeconomic pupils had a marked tendency to disavow self-discipline, to want to take the easy way out. Argyrus (1957:24) pointed out that an individual develops adaptive behavior in maintaining self-integration which may manifest itself in excessive absence, cheating, and lowered achievement levels. This finding comes from a study of industrial management problems; however, if appears reason able to generalize the disclosure to regular students and to dropouts. By inference, the dropout would value cheat ing more positively than would the stay-in. Something easy. Schneider and Lysgaard (1953:146) noted that the lower socioeconomic classes evaluated the concept "something easy" more highly than did the middle 219 classes. The implication is that the potential dropout would value more highly "something easy" than would the stay-ins. Woollatt (1961:8) found that dropout boys of above-average intelligence were characterized by "lack of effort." Again, by inference, the dropout population would value "something easy" more highly than would the stay-in. Something important. Kahl, cited in Allen (1949: 12), commented that if the individual is convinced of the importance of school, through parent attitude and peer group values, there is an increased probability of success in academic experiences. Relative to this point, the dropout would probably express lower evaluations of this concept than would the stay-in. Penty (1956:41) found that the dropout listed "importance of the subject" last among four reasons for liking school. Shanley (1961:81) noted that the "well-adjusted" gave proportionally more favorable responses regarding the importance of school. a Money. Ellis (1903:792), in an early study of dropouts^ found that the desire to make money was frequently mentioned as the primary reason for leaving school early. Hollingshead (1949:359) listed "money" first among the reasons given by dropouts for leaving school. One should note that "economic need" in the Hollingshead study con cerned families in a farming community among tenant farm 220 children where money was not readily available. Schneider and Lysgaard (1953:146) found that the "workers” with which the majority of potential dropouts were Identified, preferred spending money to saving it. This, again, rein forces the suggestion that the "workers" have rather short- range goals, living here-and-now without concern for the future. Allen (1956:13) related socioeconomic status to financial need and potential dropout. He noted that the lower socioeconomic family placed a high value upon finan cial independence. Snepp (1956:51) observed that first among reasons given for dropping out of school was the need for money. Epps and Cottle (1958:92) found that boy dropouts were more concerned with working and earning money than were girl dropouts, and that both dropout groups were more concerned with money than the stay-ins. Swan (1961: 77) reported that among boy dropouts, the desire for and the acquisition of a car led to the need for money and, of course, a job. In contradiction, Liddle (1962:279) be lieved that money was no longer a critical factor in drop ping out. He noted that it may be of secondary importance, particularly among the lower socioeconomic group which wants the economic status symbols that money can buy, but know they can’t have them if they remain in school. Success. Livingston wrote regarding the role of success in school persistence: 221 It has been common to generalize that students must find some measure of "success feelings" if they are to remain in school. This study suggests that the necessary ingredient may be more than mere success. The program that the student follows must be relevant to his goals. His goals, in turn, are related to taTs home environment, his natural abil ity and interests, and the stimulation that the school provides. (1958:200) Bowman and Matthews (1960:64) found no significant inter group distinctions regarding the factor of money. Woollatt (1961:8) wrote that girl dropouts identified lack of success in academic courses as the chief reason for drop ping out. Liddle (1962:277) observed that among the lower socioeconomic groups it was more important to "get by." Achievement was valued neither for itself nor as a means to a better life. He found that dropouts felt that they were academic failures and would continue as failures with out hope of success. In this regard, the dropout would regard "success" less favorably than would the stay-in. THE MEANING OF WORDS TEST Format The Meaning of Words Test (Appendix A-l) was pre pared on IBM cards with one set of bipolar scales (nine seven-step rating scales) on each side of the card. Fifteen cards comprised a deck which included the thirty concepts. Basic Identification data were entered by the student on the envelope in which the cards were placed (Appendix A-3). This information consisted of the pupil's 222 name, sex, date of testing, school, grade, blrthdate, and father's occupation. In addition, a four-digit student number was assigned, for subsequent identification of the test protocol. Instructions to the student were prepared and inserted in the envelope together with the cards. At the time of test administration, the pupil was instructed to read silently while the test administrator read the same instructions aloud. Administration The examiner scheduled meetings with the teachers and counselors in each school who were designated as administrators of the compulsory state-wide group tests. The purpose of the research was described and the concept of semantic differential briefly explained. Each teacher was given a sample packet of cards, identical to those to be given to the students; and administration of the test was reviewed carefully. In addition, each teacher received a page of instructions, including instructions to the student (Appendix A-4), so that a measure of comparability of administration could be achieved. The Meaning of Words Test was administered along with the tests required as a part of the compulsory state wide group testing program. This procedure was selected so as not to attract unnecessary attention to the project. Public reaction against "testing” in southern California 223 has given rise to many problems within the schools. For this reason, it was necessary to include in the teacher's instructions to the students a statement to the effect that any pupil, who, for religious reasons, could not participate in any but the required state-wide tests was asked not to take the Meaning of Words Test. Compulsory state-wide testing required time during four school days to complete. The administration of the Meaning of Words Test was left to the discretion of the individual test supervisor. The unique nature of the Meaning of Words Test, and its relatively short adminis tration time (about thirty minutes), permitted its use as a "break” in the testing routine. Some of the supervisors reported that the test was given at the close of the overall program; others gave it during the routine when time was available. No control * over exact time was possible concerning the day nor hour of administration. An examination of the test protocols indicated that the majority of pupils were able to complete the Meaning of Words Test in the time allowed; however, no information is available concerning possible test situa tions in which the pupils were fatigued, hurried, anxious or uneasy because of a previously administered group test. Contamination introduced by these factors cannot be evaluated. 224 RESEARCH DESIGN Treatment Groups The sample from which the dropouts were identified consisted of all boys and girls in grades ten and eleven In three high schools of one district who discontinued regular class attendance before the end of the school year.- A fol- lowup investigation at the end of sixty days determined those who had re-enrolled in regular classes or who were participating In some form of formal education. A further determination was made at the end of the school year to eliminate others who had re-entered programs of formal education. A random sample of stay-ins was drawn from the remaining students who had satisfactorily completed the Meaning of Words Test. The stay-in sample was matched on grade level, sex, and group test IQ score (within five IQ score points). Independent Variables Data were obtained from two sources: (1) Pupils provided information regarding father occupation, birth- date, sex of student and grade level. (2) Group test intelligence score, index of reading achievement, number of absences and number of days enrolled in school were obtained from school records. Socioeconomic status scores were established from 225 father occupation as given by the student at the time of testing. Occupation was rated according to the 1960 revi sion of Warner, Meeker and Eells1 (1949) Revised Scale for Rating Occupations. Group test intelligence scores were obtained from the California Test of Mental Maturity, Form 57S, administered at the time of or within one year of testing. The subject's age in months at the time of testing was determined to provide an index for comparison as well as to indicate the extent of over-ageness for grade. An evaluation of reading achievement was obtained as the average standard score of subtests 5 (Reading in Social Science), 6 (Reading in Natural Science), and 7 (Interpretation of Literary Materials) of the Iowa Tests of Educational Development (1959), Form X3-S or Y3-S, given at the time of testing or within one school year. The procedure of averaging the standard scores was indi cated, since no other current index of reading achievement was available for this sample population. The number of days absent from school, for all reasons, as well as the number of days enrolled was deter mined from the subject's apportionment fiscal record card. Leaving age, calculated to the nearest whole month, was established for each dropout. 226 Dependent Variables Evaluative, potency and activity factor scores were calculated for each of the thirty concepts. Each set of factor scores was treated in a 2x2x2 analysis of covari ance design with group test IQ score as the covariant. The basic statistical model consisted of three main effects: treatment group (dropout, stay-in), grade level (tenth, eleventh), and sex (boy, girl). Control Variables The variables of grade level, sex of student, and group test IQ score were controlled in matching the dropout with the stay-in sample. Students in classes for the mentally retarded were not tested. Only students who were regularly enrolled in grades ten and eleven were included in the study. Administration of the Meaning of Words Test was controlled to an extent in that the examiner met with those teaching personnel responsible for administering the state wide compulsory group testing program to explain the research project and to discuss administration of the MWT in detail. Written instructions were prepared for each testing supervisor in addition to the written instructions provided with each copy of the MWT. An attempt to avoid calling unnecessary attention to the research project was made by scheduling the MWT 227 during the time of compulsory testing. In this way it might more likely be considered r,just another test,” with attendant acceptance and lack of emotional tension. Uncontrolled Variables Many factors were not amenable to control. Neither the time of day nor the day of the week that the MWT was given could be controlled nor could it be specified. There was no way of regulating the standardized test that pre ceded or followed administration of the MWT. Pressures or tensions resulting from this circumstance cannot be evalu ated. Time allowed for administration was not uniformly regulated. Situations preceding and following the MWT could not be regulated. Clusters of incomplete protocols indicate the possibility of a foreshortened test period in certain classrooms. The advent of lunch or going home would probably influence attentiveness to the task at hand. No assurance was offered that the prepared instruc tions were presented as suggested. Limited assistance was permitted in the definition of concepts or adjectives; however, the extent and the effects of such assistance cannot be evaluated. 228 STATISTICAL TREATMENT OF DATA Null Hypothesis Non-directional null hypotheses were advanced for all independent and dependent variables. Significance Level Alpha equal to .05 was required for the rejection of the null hypothesis. Two hundred sixty-two pupils were selected for participation in this study. INDEPENDENT VARIABLES Socioeconomic Status Data expressed as father occupation rating scores were treated by the median test for differences in central tendency after Siegel (1956:111). Appropriateness. The obtained information was at best ordinal in level of measurement; sample sizes were not equal. The median test was chosen as no assumptions must be made regarding normality of distribution, homosce- dasticity, nor homogeneity of variance. The obtained statistic was distributed as chi square with one degree of freedom, offering convenience in interpretation. Limitations. The seven categories of the occupa tion rating scale could not be tested simultaneously by this method. Scores were dichotomized at the median with 229 some attendant loss of Information. Additionally, the power efficiency of the median test decreases with increased sample size; however, with samples varying from sixty-four to sixty-eight subjects, the loss in power effi ciency was not sufficient to invalidate inferences drawn from these findings. The assumption was made that the compared populations had identical distributions, without reference to the population distribution. Group Test IQ Score. Test Age, Reading Achievement Index, Number of Absences, Days Enrolled, and Leaving Age These data provide information of at least the interval level of measurement permitting means, standard deviations and t _ ratios to be computed. Information obtained from recognized standardized group intelligence tests (CTMM) and standardized achieve ment tests (ITED) provide scores that may reasonably be regarded as interval data. Appropriateness. Each sample of information men tioned above represents a continuous distribution of scores. Group test IQ scores and reading scores are at best interval data, while test age, number of absences, days enrolled and leaving age provide data at the ratio level of measurement. Limitations. Parametric statistics assume that 230 these are independent, homoscedastic samples drawn from normally distributed populations with equality of variance. Sample size in each instance was minimal for the satisfac tory application of parametric methods. DEPENDENT VARIABLE Responses to the Meaning of Words Test Primary consideration in the application of para metric statistics must be given to the level of measurement of the obtained information. Osgood (1957:146, 159) cited work of Messick (1955) which investigated the scaling properties of the semantic differential process by means of the method of successive intervals after the process developed by Saffir (1937) and Gulliksen (1954). Although the scaling intervals of the semantic differential process were not identical, ’’the deviations from equal intervals are small and . . . within the error limits of the instru ment,” according to Osgood (1957:150). In this respect, the obtained scores may reasonably be considered to achieve the interval level of measurement. Osgood (1957:78) observed that, since the bipolar scales were ’’ neither perfectly aligned with factors nor perfectly reliable,” they suggested the use of a sample of closely related scales to represent each factor, producing an average score which they felt was more representative 231 and more reliable than individual scale scores. This sug gestion to treat the MWT responses as summed factor scores has been followed. Appropriateness. The statistics of mean, standard deviation and F ratio 'appear warranted in that the data may be considered to meet the requirements of interval level of measurement and the assumptions of independent samples from normally distributed populations of equal variance. The robust nature of the F ratio statistic with respect to unequal variance, together with the greater stability of summed factor scores, provides reasonable defense for the selected procedure. Limitations. As indicated, parametric statistics assume that independent samples have been drawn from normally distributed populations with equality of variance. The dropout sample probably is not representative of the general school population. Because the stay-ins were matched with the dropouts on three characteristics, this sample could not be regarded as an independent, random sample from the general population. Analysis of Covariance Design To test mean differences for statistical signifi cance, a 2x2x2 general linear factorial analysis of covariance computer program, that allows for unequal 232 numbers of cases in the cells (modified BIMD 05V),* was utilized so that both the main effects and the interactions of treatment groups (i.e., dropout or stay-in), grade level (tenth or eleventh grade), and sex of the student (boy or girl) could be evaluated. Identical cell sizes were main tained throughout between dropouts and stay-ins. This design also compensated statistically for the differences in intelligence scores between groups. Recent research (Rosenthal, 1965) demonstrated that the factor of IQ score differentiated treatment groups in a similar study only in the extremes. Covariance technique provides for ancillary information so that the influence of the IQ score is eliminated from the estimates of treat ment effects, thereby increasing the precision of these estimates. Limitations. Information treated in an analysis of covariance must meet the requirements of all parametric statistics. The properties of independent, homoscedastic samples with equality of variance drawn from normally distributed populations are at best approximated in most behavioral research. *Tbe writer is grateful to the Western Data Processing Center, Health Sciences Computer Facility, for statistical computations. Hypotheses A nondirectional null hypothesis was advanced for all Independent and dependent variables. In this 2x2x2 analysis of covariance design, seven research hypotheses were generated. Main Effects The statistical model provides for three main effects: dropouts and stay-ins; tenth and eleventh grades and boys and girls. The covariance procedure answers the question of whether or not a subject's deviation from the sample mean can be considered to be a function of one or more factors and an error term, corrected with respect to some ancillary measure. First research hypothesis. This hypothesis states that the mean summed factor scores on the MWT, when cor rected for IQ score, will differ significantly between dropouts and stay-ins when averaged over grade level and sex of the student. Second research hypothesis. The second hypothesis holds that the mean summed factor scores on the MWT, when corrected for IQ score, will differ between subjects in grade ten and grade eleven, when averaged over dropouts and stay-ins, and boys and girls. 234 Third research hypothesis. The third research hypothesis maintains that the mean summed factor scores, corrected for IQ score, will differentiate between boys and girls, averaged over the conditions of dropout and stay-in and grade level. First Order Interactions Research hypotheses drawn from first order interac tions postulate that significant differences between the two levels of each main effect, when corrected for IQ score, are not independent of the effect of the third factor. Fourth research hypothesis. Significant differ ences in mean summed factor scores according to this hypothesis, mean that the condition of dropout or stay-in is not independent of grade level, averaged over sex of the student. Fifth research hypothesis. This hypothesis states that differences in mean summed factor scores between s . tenth and eleventh grade subjects are not independent of the conditions of dropout or stay-in, averaged over sex of the student. Sixth research hypothesis. Significant differences in mean summed factor scores, according to this hypothesis, demonstrate that boy and girl groups are not independent 235 o£ the conditions of dropout and stay-in, averaged over grade level. Second Order Interactions Significant interactions between the three main effects, taken together, mean that the dropout and stay-in mean summed factor scores are not the same for boys and girls of the tenth and eleventh grades. CHAPTER SUMMARY This chapter has presented a description of the source of data and the methodology of treatment. The sample of high school dropout and stay-in pupils was drawn from three southern California high schools. A random sample of stay-ins was matched by grade level, sex of student and IQ score. Independent variables relating to socioeconomic status, group test intelligence score, an index of reading achievement, test age, number of absences, days enrolled in school and leaving age were tested for significant differences between groups. Construction of the Meaning of Words Test was explained in detail. Each item selected for study was documented and the rationale for inclusion established. The format and administration of the MWT was presented and discussed. The research design of the study was outlined and 236 discussed. Statistical procedures were evaluated with respect to appropriateness and limitations for the data. Independent variables and the dependent variable were noted and considered with respect to the selected statis tical treatment. Cpntrol and uncontrolled variables were identified and evaluated with respect to their influence on the study. Concluding the discussion was a statement regard ing the hypotheses postulated by this research design. CHAPTER IV FINDINGS This chapter presents in detail the statistical findings of the study. Items descriptive of the sample have been tabulated and treated to appropriate statistical measures. The selected descriptive variables are socio economic status rating, group test IQ score, chronological age at the time of testing, an index of reading achieve ment, total number of days absent, total days enrolled, and age of the dropout at the time of leaving school. Responses to the Meaning of Words Test are pre sented within the framework of seven additional areas of investigation: school experiences, family relationships, social interaction, self concept, authority relationships, goal orientation, and moral and social values. Attitudes were expressed as the ratings of thirty stimulus concepts on nine bipolar adjective scales. The rating scores were treated in a computer-based analysis of covariance program. 237 238 Variable 1.0. Socioeconomic Status Rating The frequency distribution of rating scores assigned to father occupation, displayed in Appendix B, reveals that the families represent middle-income groups consistent with Warner, Meeker and Eells' (1960) Revised Occupation Scale, classes 4 and 5. A combined median rating of 4 was obtained for both treatment groups. Sample medians are reported in Table 19, from which general con sistency can be observed. Significance tests for the independence of the median occupation rating scores, reported in Table 20, describe two sample differences. Tenth grade stay-in boys' fathers had higher occupation ratings than fathers of tenth grade dropout boys. Simi larly, the tenth grade boys' fathers were rated higher more frequently than were the tenth grade girls' fathers. As may be seen in Table 19, median occupation ratings for fathers of dropouts, as a group (a median of 5), were lower than occupation ratings for fathers of stay-ins (a median of 4). No corresponding differences were evidenced between samples of the tenth and eleventh grade pupils nor between boys and girls, as is tabulated in Table 20. TABLE 19 SUMMARY OF AVERAGES OF SELECTED DESCRIPTIVE VARIABLES t o ► » ' < d na Q 4) < d o ■u u Q E E-*W * « V i X 4) 0) V O JG s T3 S U 6 t-l . •H S 4) s (d O to 4t *0 H C O •H l£-*nl 0 0) 0) S i -O •U H U bflcd* 60 <5 M Q ) s (d bo O < 3 to S < U H cd a s a o 0S.S H iH & a ■ H S U p C O 4 -1 H 4J •a c d | i 55OT TJ O o . to id s < d 4J J S S S u < y O H £ £ £ £ o H (M O 4) eoco C d ^ *J ( 0 SOU 4) c OSS) H (0 CO £ 2 5 J5 > ( 0 cd n * m os 0) J3 w *J c PSA OtIS a _ Otl’ O U 6 0 s d < cd DROPOUTS Tenth Grade Boys Girls 34 33 5.0 4.0 95.21 97.82 16-1 15-10 8.21 10.55 31.09 34.91 90.74 103.79 34.26 33.63 16-4 16-3 Eleventh Grade Boys Girls 33 31 5.0 5.0 99.00 101.74 17-0 16-5 11.15 11.39 24.48 24.90 87.48 95.65 27.98 26.03 17-3 16-7 STAY-INS Tenth Grade Boys Girls 34 33 3.5 4.0 99.94 98.85 15-6 15-4 15.00 14.46 9.53 9.24 178.00 177.70 5.35 5.19 O k Eleventh Grade Boys Girls 33 31 4.0 5.0 100.76 101.16 16-4 16-4 15.18 16.45 10.85 11.26 178.00 175.00 6.09 6.43 ♦Figures represent group means. Raw score data included in Appendix B. ”t” ratios for significance of mean differences tabulated in Table 21. N3 VO 240 TABLE 20 MEDIAN TEST FOR INDEPENDENCE OF SOCIOECONOMIC STATUS RATING* Grade Level 10 11 N Chi p N Chi p Square Square Dropout8 and Stay-ins Boys 34 11.64 .001 33 1.72 Girls 33 .06 — 31 .00 Boys and Girls ( Dropouts 67 5.55 .02 64 .00 Stay-ins 67 1.92 -- 64 .91 Total Sample • --------- — Square £ Dropouts x • - Stay-ins 5.16 .05 Tenth Grade x Eleventh Grade 0.47 Boys x Girls 0.02 ♦Interpreted as chi square with one degree of freedom (Siegel, 1956:111). 241 Variable 2.0. Group Test Intelligence Quotient Scores Examination of Table 19 indicates that group test IQ scores are essentially within the normal range for both treatment groups. Mean IQ scores range from a low of 95.21 for tenth grade boy dropouts to a high of 100.76 for eleventh grade boy stay-ins. Tests for significant differ ences between sample means reveal no differences with respect to the IQ variable: Exhibited in Table 21 are ,ft” ratios for mean differences between dropouts and stay-ins for selected variables; in Table 22 are "t** ratios for mean differences between boys and girls for selected variables; and in Table 22 are analysis of variance of group test IQ scores. Variable 3.0. Age at Time of Testing Chronological age at the time of testing is dis played for all subjects in Table 36, Appendix B. Differ ences in mean age between dropouts and stay-ins are presented in Table 21. It is seen that, with the single exception of eleventh grade girls, all differences were significant at the .01 level of confidence. Tenth grade dropout boys averaged sixteen years one month in age at the time of testing, while the stay-in tenth grade boys averaged fifteen years six months. Tenth grade dropout girls averaged fifteen years ten months of age and tenth grade stay-in girls had a mean age of fifteen years four TABLE 21 ? , t" RATIOS FOR MEAN DIFFERENCES BETWEEN DROPOUTS AND STAY-INS FOR SELECTED DESCRIPTIVE VARIABLES Grade 10 Grade 11 Variable Boys N=68 P Girls N»66 P Boys N«66 P Girls N»62 P Socioeconomic Status * * * * Group Test IQ Score 1.518 — .351 .557 — .202 m mm Test Age 4.185 .01 3.559 .01 5.685 .01 .718 tm Reading Score 5.344 .01 3.601 .01 3.669 .01 3.967 .01 Absences 4.393 .01 6.112 .01 4.421 .01 5.204 .01 Days Enrolled Leaving Age 12.175 ■ b « .01 9.598 .01 12.343 .01 10.382 .01 *See Table 20 for tests of independence. N5 243 TABLE 22 "t" RATIOS FOR MEAN DIFFERENCES BETWEEN BOYS AND GIRLS FOR SELECTED DESCRIPTIVE VARIABLES Variable Dropouts p Stay-ins Tenth Grade (N»67) Socioeconomic Status * * Group Test IQ Score .942 — .333 Test Age 1.195 — 1.071 Reading Score 2.096 .05 .432 Absences .628 -- .130 Days Enrolled 1.242 — 1.470 Leaving Age .749 Eleventh Grade (^64) Socioeconomic Status * * Group Test IQ Score .908 — .131 Test Age 3.837 .01 .049 Reading Score .197 — 1.099 Absences .091 -- .158 Days Enrolled ^799 — 1.048 Leaving Age 3.769 .01 *See Table 20 for tests of independence. 244 months. Eleventh grade dropout boys averaged seventeen years, whereas the stay-in eleventh grade boys were sixteen years four months of age, on the average. Dropout and stay-in eleventh grade girls had the same average age at * the time of testing— sixteen years four months. Eleventh grade boy dropouts were significantly older than eleventh grade girl dropouts. The dropout boys averaged exactly seventeen years and the dropout girls sixteen years five months of age. None of the other groups differed in average age at the time of testing, as can be seen in Table 22. Variable 4.0. Index of Reading Achievement An index of reading achievement generated from an average of the standard scores of subtests 5, 6, 7 of the ITED, Form X-3S and Y-3S (1959 edition), demonstrated differences significant at the .01 level of confidence between all samples of dropouts and stay-ins, as is pre sented in Table 21. Table 19 revealed that tenth grade dropout boys had a mean reading index of 8.21 (falling below the second per centile of the standardizing population*) while the mean score of tenth grade stay-in boys was 15.00 (essentially equivalent to the ninetieth percentile of the standardizing population*). Tenth grade dropout girls had an average *E. F. Lindquist, The Iowa Tests of Educational Developments General Manual, Forms X-3S and Y-3S, pp. 32, 33. 245 reading index of 10.55 (tenth percentile*) and the tenth grade stay-in girls averaged 14.46 (eighty-fifth percen tile*). The eleventh grade dropout boys averaged 11.15 in reading score (below tenth percentile*), while comparable ^ 4r stay-in boys earned a mean score of 15.18 (sixty-fifth percentile*). Similarly, the eleventh grade dropout girls achieved a reading index of 11.39 (below the tenth percen tile*), and the stay-ins earned an average score of 16.45 (almost the ninetieth percentile*). Table 22 presents the differences between boys and girls for the dropouts and stay-ins at each grade level. Reading score differences were significant only between dropouts boys and dropout girls (p = .05). Dropout girls achieved an average reading index of 10.55, while dropout boys achieved an index of 8.21, a difference significant at the .01 level of confidence. Tenth grade stay-in boys and girls did not differ significantly. Neither eleventh grade dropout boys and girls nor eleventh grade stay-in boys and girls differed in mean reading index. Eleventh grade drop out boys averaged 11.15 and girls averaged 11.39. Eleventh grade stay-in boys averaged 15.18 and girls averaged 16.45. *E. F. Lindquist, The Iowa Tests of Educational Development. General Manual, Forms X-3S and Y-3S, pp. 32, Variable 5.0. Total Days Absent Table 23 illustrates that dropouts were absent from school, for all reasons, from two to three times as often as were stay-ins. Tenth grade boy dropouts were out of school an average of 31.09 days during the year; tenth grade boy stay-ins were absent 9.53 times. Tenth grade girl dropouts were out an average of 34.91 days; tenth grade stay-in girls were absent 9.24 days. Eleventh grade boy dropouts were absent for an average of 24.48 days throughout the year, whereas the matched group of boy stay- ins were absent 10.85 days. Eleventh grade girl dropouts were out an average of 24.90 days, while their stay-in counterparts were absent 11.26 days. Table 21 displays the significance of the differences between dropouts and stay- ins in mean number of days absent from school. In each instance, differences were significant at the .01 level of confidence. Differences between average number of absences for boys and girls for the treatment groups at the tenth and eleventh grade levels are reported in Table 22. No signif icant differences are noted between boys and girls of either treatment group or for either grade level. Variable 6.0. Total Number of Days Enrolled in School Dropouts were enrolled in school about half as many days as the stay-in group. Table 19 reveals that tenth 247 TABLE 23 PERCENTAGE OF MEAN NUMBER OF DAYS ABSENT Mean Days Enrolled Mean Number of Days Absent Percentage of Mean Days Absent DROPOUTS Tenth Grade Boys (N**34) 90.74 31.09 34.26 Girls (N»33) 103.79 34.91 33.63 Eleventh Grade Boys (N®33) 87.48 24.48 27.98 Girls (N«31) 95.65 24.90 26.03 STAY-INS Tenth Grade Boys (N*34) 178.00 9.53 5.35 Girls (N«=33) 177.70 9.24 5.19 Eleventh Grade Boys (N«33) 178.00 10.85 6.09 Girls (N®31) 175.10 11.26 6.43 248 grade boy dropouts were enrolled for an average of 90.74 of the 178 days in the school year. Tenth grade dropout girls remained on the school rolls for an average of 103.79 days; eleventh grade dropout boys were enrolled for an average of 87.48 days; and eleventh grade girl dropouts for an average of 95.65 days. Stay-ins, who completed the school year in a program of regular high school classes, were enrolled an average of 177 of the 178 days that school was in session. The mean differences in days enrolled given in Table 21 are significant at the .01 level of confidence between dropouts and stay-ins at both tenth and eleventh grade levels. Data presented in Table 22 reveal no signif icant differences between boys and girls at either grade level for dropouts or stay-ins. Percentage of absence. Table 23 gives descriptive information regarding the proportion of time dropouts and stay-ins were actually in school. The mean number of absences is corrected with respect to the mean number of days enrolled. Tenth grade boy dropouts were absent 34.26 per cent of the time and stay-in tenth grade boys 5.35 per cent of the time. Tenth grade girl dropouts were absent 33.63 per cent of the time while tenth grade girl stay-ins were absent 5.19 per cent of the time enrolled in school. Eleventh grade boy dropouts were absent from school 27.98 per cent of the time enrolled and eleventh grade boy 249 stay-ins were out of school 6.09 per cent of the time enrolled. Eleventh grade dropout girls were out of school 26.03 per cent of the time enrolled and eleventh grade stay-in girls were out of school 6.43 per cent of the time. Leaving Age of Dropouts Table 19 revealed the average age of leaving school of tenth grade boy dropouts to be sixteen years four months and tenth grade girl dropouts to be sixteen years three months. Eleventh grade boy dropouts left at an average age of seventeen years three months and the eleventh grade girl dropouts left at an average age of sixteen years seven months. Tenth grade dropout boys and girls did not differ significantly in leaving age as seen in Table 22. Eleventh grade boy dropouts were significantly older than eleventh grade girl dropouts at the .01 level of confidence. Analysis of Variance of Group Test IQ Scores Group test IQ scores were examined in a 2x2x2 analysis of variance design. The F ratios reported in Table 24 Indicate that when the slopes of the regression lines for the dropouts and stay-ins (each dependent varia ble was plotted against the covariate) were tested no significant differences were found. This absence of interaction affirms the appropriateness of the group test IQ score as a covariate. 250 TABLE 24 ANALYSIS OF VARIANCE OF GROUP TEST INTELLIGENCE QUOTIENT SCORES BETWEEN TREATMENT GROUPS, GRADE LEVELS AND SEX OF THE STUDENT Probability Source of Variance F Ratio Level* Dropout x Stay-in 1.004 - - Tenth x Eleventh Grade 2.562 — Boy x Girl .622 • m Treatment Group x Grade Level .612 -- Treatment Group x Sex .731 -- Grade Level x Sex .103 -- Treatment Group x Grade Level x Sex .049 — *p.05 « 3.89 with 1 and 254 degrees of freedom. 251 Group Test IQ Score as a Correlate of the Dependent Variables Table 25 reveals those areas In which group test IQ score Is a correlate of the factor score ratings of partic ular concepts. Significant F ratios Indicate a greater- than-chance probability that subjects with a given IQ score will respond In a predictable manner. The evaluative factor scores were found to be cor related with IQ score on the following concepts: Factor £ Factor £ Reading .05 Graduating .01 Ideal teacher .05 Quitting school .01 My school ability .01 Cheating .01 College .01 Success .05 Potency factor scores were associated with IQ st these concepts: Factor £ Factor £ Classmates .05 Authority .05 Ify school ability .01 College .05 How ray class sees me .01 Quitting school .01 Activity factor scores were related to IQ i score following concepts: Factor E Factor £ Grades .05 College .01 Reading .01 Graduating .05 Ideal teacher .01 Quitting school .01 My school ability .01 Cheating .05 Authority .05 Something easy .05 252 TABLE 25 INTELLIGENCE QUOTIENT AS A CORRELATE OF CONCEPT RATINGS ON SELECTED FACTOR SCALES FOR THE TOTAL SAMPLE POPULATION OF DROPOUTS AND STAY-INS Evaluative Factor Potency Factor Activity Factor Concept F Ratio P* F Ratio P E Ratio P School Experiences Teachers 0.07 - 0.04 • 0.19 - Grades 3.41 - 3.14 - 5.34 .05 School 1.46 - 1.40 - 0.61 - Reading 4.32 .05 2.99 - 6.91 .01 Ideal teacher 5.24 .05 1.44 “ 7.23 .01 Family Relationships Home 0.66 0.83 0.53 Ideal parent 0.01 - 0.00 - 0.37 mm Parents 0.52 — 2.09 - 0.07 — Social Interaction Most people 1.06 * 2.45 — 0.73 -» My best friends Classmates 1.02 - 0.33 tm 0.86 mm 0.04 - 4.54 .05 0.74 - Grownups 0.71 — 0.05 - 0.00 - Self Concept Me 9.45 .01 15.48 .01 11.30 .01 How I would like to be 0.63 - 1.58 - 1.61 - How my class sees me 0.56 - 7.42 .01 2.56 - *p.05 = 3.89, p.01 =6.76 with 1 and 253 degrees of freedom. Note: Five of the eleven differences significant at the .05 level of confidence might reasonably be expected to have occurred by chance among the ninety comparisons, whereas only one difference at the .01 level of confidence might be considered to have occurred by chance. 253 TABLE 25 (continued) INTELLIGENCE QUOTIENT AS A CORRELATE OF CONCEPT RATINGS ON SELECTED FACTOR SCALES FOR THE TOTAL SAMPLE POPULATION OF DROPOUTS AND STAY-INS Evaluative Potency Activity Factor Factor Factor F F F Concept Ratio p Ratio p Ratio p Authority Relationships Authority 0.67 - 5.80 .05 3.84 .05 Rules 2.78 - 0.02 - 2.11 - Punishment 0.95 - 0.20 - 1.64 - Goal Orientation Future 3.06 •» 1.91 _ 3.23 at College 11.51 .01 6.50 .05 12.58 .01 A job 2.76 - 0.06 - 0.25 - Graduating 7.69 .01 1.84 mm 4.19 .05 Quitting school 13.27 .01 7.38 .01 18.24 .01 Moral and Social Values Trying hard 0.24 0.69 — 0.24 Cheating 14.85 .01 1.06 - 5.38 .05 Something easy 1.51 - 0.97 - 5.69 .05 Something important 0.01 - 0.34 - 0.00 - Money 2.64 - 0.30 - 2.26 - Success 5.47 .05 1.06 - 2.70 - B B B B B i B R B i B B . 254 The evaluative, potency and activity factor scores were associated with IQ score for the following concepts: my school ability, college, and quitting school. Responses to the Meaning of Words Test The statistical model for this study consisted of a 2x2x2 linear factorial analysis of covariance design in which the eight groups were compared on the mean sunmed evaluative (E), potency (P), and activity (A) factor scores for all thirty concepts. The concepts have been classified according to the following areas for investigation; School Experiences Family Relationships Social Interaction Self Concept Authority Relationships Goal Orientation Moral and Social Values The findings expressed as F ratios for each concept are presented according to these seven areas. Mean differ ences significant at the .05 and .01 levels of confidence are reported. 255 Area of Investigation 7.0. School Experiences Concept 7.1. Teachers* The F ratios reported in Table 26 clearly indicate significant differences between dropouts and stay-ins in their ratings of the concept "teachers.*1 Stay-ins rated this concept higher in evalua tion (E), as more potent (P), and as more active (A) than did dropouts. No significant differences were found between students in grades ten and eleven on any of the three scales. Boys, as a group, differed significantly from girls on all three scales, as is reported in Table 27. Girls rated teachers more favorably in evaluation, as more potent, and as more active than did boys. The first order interaction of treatment groups and grade levels failed to produce any significant differences. Neither treatment group nor grade level, when compared with boys and girls, produced significant differences. The three-way interaction of treatment groups, grade level and sex yielded a difference significant at the .05 level of confidence only on the activity factor scale. An examina tion of the mean activity factor scores in Table 35 reveals that the eleventh grade girl stay-ins rated the concept of "teachers" as more active than did other groups. 256 TABLE 26 ANALYSIS OF DROPOUT AND STAY-IN MEAN SUMMED FACTOR SCORES OF CONCEPTS RATED AGAINST SELECTED FACTOR SCALES* Evaluative Potency Activity Factor Factor Factor Concept F Ratio P F Ratio P F Ratio P School Experiences Teachers Grades School Reading Ideal teacher 146.91 98.71 139.05 43.90 5.09 .01 30.01 .01 64.46 .01 59.70 .01 14.84 .05** .27 .01 .01 .01 .01 85.18 61.24 108.74 26.20 1.32 .01 .01 .01 .01 Family Relationships Home Ideal parent Parents 56.76 2.97 57.46 .01 .01 22.91 .96 33.09 .01 .01 38.82 .90 39.37 .01 .01 Social Interaction Most people My best friends Classmates Grownups 81.97 3.24 113.88 77.81 .01 .01 .01 48.40 4.72 71.27 33.58 .01 55.09 .05** .36 .01 77.12 .01 41.75 .01 .01 .01 Self Concept Me My school ability How I would like to be How my class sees me 79.26 103.60 18.14 74.16 .01 .01 .01 .01 22.33 69.08 .26 12.99 .01 .01 .01 58.68 100.25 5.93 47.34 .01 .01 .05* .01 *Analysis of covariance design with group test IQ score as covariate, p.05 = 3.89, p.01 =» 6.76 with 1 and 253 degrees of freedom. **Since five F ratios, significant at the .05 level of confidence, could be expected to occur by chance, these four *’.05" differences may reasonably be considered to have occurred by chance. 257 TABLE 26 (continued) ANALYSIS OF DROPOUT AND STAY-IN MEAN SUMMED FACTOR SCORES OF CONCEPTS RATED AGAINST SELECTED FACTOR SCALES Evaluative Potency Activity Factor Factor Factor F F F Concept Ratio p Ratio ■p Ratio p Authority Relationships Authority 112.42 Rules 123.94 Punishment 73.48 Goal Orientation Future 27.07 College 56.57 A job 7.26 Graduating 48.26 Quitting school 85.50 Moral and Social Values Trying hard 76.19 Cheating 43.72 Something easy .04 Something important 64.84 Money .14 Success 41.95 01 14.76 .01 49.38 .01 01 24.92 .01 83.63 .01 01 5.51 .05**39.27 .01 .01 19.,23 .01 24.84 .01 .01 33.,76 .01 55.84 .01 .01 2,.76 - .56 - .01 27,,36 .01 40.82 .01 .01 30..58 .01 81.99 .01 01 58.75 .01 72.49 .01 01 11.90 .01 13.62 .01 - 19.56 .01 .02 - 01 37.50 .01 59.79 .01 - 2.59 - 2.47 - 01 22.96 .01 27.97 .01 258 TABLE 27 ANALYSIS OF TENTH AND ELEVENTH GRADE MEAN FACTOR SUM SCORES OF CONCEPTS RATED AGAINST SELECTED FACTOR SCALES* Concept Evaluative Factor F Ratio p Potency Factor F Ratio i Activity Factor F Ratio p School Experiences Teachers 0.37 - 0.02 - 0.69 Grades 1.49 - 0.01 - 0.00 School 0.18 - 0.07 - 0.01 Reading 1.17 - 4.32 .05 3.21 Ideal teacher 0.10 - 1.36 - 1.36 Family Relationships Home 3.76 .05 0.00 - 3.39 Ideal parent 0.00 - 0.00 - 0.34 Parents 0.07 - 0.29 - 0.01 Social Interaction Most people 1.39 - 0.11 - 0.20 My best friends 0.14 - 0.68 - 0.39 Classmates 0.92 - 0.39 - 0.14 Grownups 0.33 - 0.20 - 2.06 Self Concept Me 0.07 - 0.31 - 0.82 My school ability 0.78 - 1.39 - 0.01 How I would like to be 0.07 - 0.02 - 0.23 How my class sees me 0.53 * * 4.64 .05 0.13 *Analysis of covariance design with group test IQ score as covariate. p.05 « 3.89, p.01 *= 6.76 with 1 and 253 degrees of freedom. Note: Since five F ratios, significant at the .05 level of"confidence, could be expected to occur by chance, these three differences may reasonably be considered to have occurred by chance. 259 TABLE 27 (continued) ANALYSIS OF TENTH AND ELEVENTH GRADE MEAN FACTOR SUM SCORES OF CONCEPTS RATED AGAINST SELECTED FACTOR SCALES Concept Evaluative Factor F Ratio p Potency Factor F Ratio i Activity Factor F Ratio p Authority Relationships Authority 2.37 Rules 0.20 Punishment 0.06 Goal Orientation Future 0.22 College 0.79 A job 1.22 Graduating 0.21 Quitting school 0.13 Moral and Social Values Trying hard 0.00 Cheating 0.01 Something easy 0.13 Something important 0.49 Money 0.13 Success 0.79 0.61 1.00 1.17 0.19 1.15 1.56 0.89 0.21 0.09 0.12 0.32 0.06 0.74 0.85 0.01 0.59 1.89 0.10 0.03 0.10 0.00 0.02 0.31 0.27 0.00 0.75 0.08 0.51 260 Concept 7.2. Grades. Dropouts and stay-ins per ceived grades differently, the difference being significant at the .01 level of confidence. Stay-ins rated grades higher on the evaluation, potency and activity scales than did the dropouts. Tenth and eleventh grade students, as is reported in Table 27, did not rate the concept of grades differently nor did the boys and girls rate the concept differently. Interaction effects significant at the .05 level of confidence were found between the treatment groups when compared with grade level on the potency factor scale. Examination of Table 49 (Appendix D) reveals that tenth grade pupils regarded the concept of "grades” as more potent than did the eleventh grade pupils. The difference between dropouts and stay-ins for tenth and eleventh grade pupils was greater for the eleventh graders. Eleventh grade stay-ins rated "grades" as more potent than did the dropouts or the stay-ins of either grade level. No signif icant differences were found between either the treatment groups or grade levels, when compared with the sex of the student on any of the factor scales. The three-way comparison did not indicate any significant interaction effects. Concept 7.3. School. Table 27 indicates no differences in ratings of the concept of "school" between 261 grade levels on any of the factor scales. Similarly, no differences were found in the responses of boys or girls. No significant differences were found when the treatment groups were compared with different grade levels or with the boys or girls. No differences were found between grade level and sex of the student nor between the groups treated in a three-way interaction. Concept 7.4. Reading. The concept of reading was perceived differently by the dropouts and the stay-ins at the .01 level of confidence on all factor scales. The stay-in rated reading more favorably in evaluation, potency and activity factor scales than did the dropouts as seen in Table 49. Table 27 reveals a difference between grades ten and eleven that is significant at the .05 level of confi dence; however, since only three such differences showed up in ninety possible occurrences, this difference may reason ably be considered to have occurred by chance. Boys and girls differed significantly in their ratings of the concept ’‘reading." Data given in Table 51 reveal that girls rated reading more favorably on the evaluation factor, at the .01 level of confidence, and as more potent and more active than did the boys. Differences were significant at the .05 level of confidence. Treatment groups when compared with grade level 262 revealed differences significant at the .05 level of con fidence on the potency factor. Tenth grade stay-ins consistently rated "reading” as more potent than did other groups. However, when treatment groups or grade level were compared with the sex of the student, no significant differences were exhibited. The three-way interaction, given in Tables 34 and 35, reveals a difference in potency factor responses signif icant at the .05 level of confidence. Stay-ins in both tenth and eleventh grades rated this concept as more potent than did the dropouts. Tenth grade boy stay-ins rated the concept of "reading" significantly more potent than did the other groups. Concept 7.5. Ideal teacher. Table 26 reveals that only the evaluation factor differentiated the dropouts from the stay-ins with respect to their attitudes toward the concept of an "ideal teacher." Stay-ins evaluated the concept "ideal teacher9* more favorably than did dropouts, as is seen in Table 49 (Appendix D). No significant differences were found between grade levels for this concept c The potency factor scale failed to differentiate between boys and girls; however, Table 51 (Appendix D) reveals that girls rated the concept higher in evaluation and as more active, the differences being significant at 263 tbe .01 level of confidence. Treatment groups when compared with grade level and with sex of student, as well as when grade levels were com pared with sex of student, failed to reveal any significant differences in the perception of this concept. The three-way interaction effect, tabulated in Tables 34 and 35, reveals a difference significant at the .05 level of confidence for this concept. Tenth grade boy and eleventh grade girl stay-ins evaluated the concept more favorably than did the dropouts. There were no significant interaction effects between the boys and girls of either grade level. Area of Investigation 8.0. Family Relationships Concept 8.1. Home. Attitudes toward the concept "home" differentiated significantly, at the .01 level of confidence, between dropouts and stay-ins on the evalua tive, potency and activity factor scales. The stay-ins, reported in Table 49, rated the concept more favorably on evaluation, more potent and more active than did the drop out group. Only the evaluative factor rating revealed a significant difference, at the .05 level of confidence, between tenth and eleventh grade pupils. Tenth grade pupils rated the concept of "home59 higher than did the 264 eleventh grade pupils. No differences In attitude between boys and girls toward the concept of "home" were Indicated. No differences were Indicated between treatment groups when compared In terms of grade level, sex of stu dent, or when grade levels were compared with the sex of the student. The three-way Interaction failed to identify any differences in the attitudes of treatment groups, grade levels or sex of the student toward the concept. Concept 8.2. Ideal parent. Attitudes expressed on this concept failed to differentiate between dropouts and stay-ins on any of the three factor scales. Nor were there differences found between grade levels on this factor. Table 26 and Table 51 Of Appendix D present evidence that girls rated the concept more favorably on the evaluation factor (p »= .01) than did the boys. Boys, however, rated the concept as more potent than did the girls, the differ ence being significant at the .01 level of confidence. Treatment groups when compared with grade levels and sex of student and grade levels when compared with sex of student failed to identify any significant differences in expressed attitude. The three-way interaction also failed to identify any differences in attitude. 265 Concept 8.3. Parents* Attitudes expressed toward this concept clearly differentiate dropouts from stay-ins at the .01 level of confidence on the evaluative, potency and activity scales. This is displayed in Table 26. Table 49 in Appendix D reveals that stay-ins rated the concept more favorably on the evaluative factor as more potent and as more active than did the dropouts. Differ ences in the responses of tenth and eleventh grade pupils were not significant. The attitudes expressed by boys and by girls toward the concept of "parents" were not signifi cantly different. Treatment groups, when compared in terms of differ ent grade levels, displayed no significant differences. Table 32 presents evidence that when treatment groups are compared with sex of the student, there is a significant difference. In this instance, stay-in boys and girls evaluated more favorably and rated as more active the con cept of "parents" than did the dropouts. Since only four of the ninety possible differences were significant at the .05 level of confidence, this difference may reasonably be considered to have occurred by chance. Grade levels, when compared with sex of the student, failed to identify any significant interactions. The three-way comparison did not reveal any significant interaction effects. 266 Area of Investigation 9.0. Social Interaction Concept 9.1. Most people. Dropouts and stay-ins, as may be seen in Table 26, differed significantly at the .01 level of confidence on all three factor scales. Table 49 of Appendix D displays evidence that the stay-ins expressed more favorable evaluative attitudes, more potent attitudes and more active attitudes than did the dropouts. Attitudes expressed toward the concept of "most people" were not distinguished significantly between tenth and eleventh grade pupils. Girls, seen in Table 51 of Appendix D, evaluated the concept more favorably than did the boys. Similarly the girls rated the concept as more active than did boys. The potency factor scale failed to differentiate between boys and girls. Treatment groups, when compared on the bases of grade level and sex of the student, and when two grade levels were compared in terms of sex of student, failed to exhibit significant differences of expressed attitude. The three-way interaction failed to indicate any significant differences. Concept 9.2. My best friends. Responses to this concept differentiated the dropouts from the stay-ins (as exhibited in Table 49 of Appendix D) only on the potency factor. Dropouts regarded the concept as more potent than 267 did stay-ins. The evaluative and activity factor scales failed to differentiate these groups. No significant differences were noted between expressed attitudes of tenth and eleventh grade pupils on this concept. Boys and girls differed significantly at the .01 level of confidence on evaluative and potency factors. As is summarized in Table 51 of Appendix D, girls rated the concept more favorably on the evaluative factor than did the hoys; whereas the boys rated the concept as more potent than did the girls. Responses to the activity factor scale did not differen tiate boys from girls. Treatment groups when compared with grade level exhibited an interaction effect on the evaluative and activity factor scales, significant at the .05 level of confidence. Combined mean factor scores presented in Table 30 disclose that tenth-grade stay-ins rated this concept more favorably and as more active than did the eleventh grade stay-ins or the tenth or eleventh grade dropouts. The potency factor ratings did not identify any interaction between the treatment groups when compared on the basis of grade level. Treatment groups and grade levels, when compared in terms of sex of the student, were not differen tiated by attitudes expressed toward this concept. The three-way interaction did not identify any significant differences between groups. 268 Concept 9.3. Classmates. Table 26 reveals that attitudes expressed about the concept, "classmates," differentiated between dropouts and stay-ins at. the .01 level of confidence on the evaluative, potency and activity factor scales. Without exception, as is seen in Table 49, Appendix D, the stay-ins evaluated the concept more favor ably, as more potent and as more active than did the drop outs. No significant differences were exhibited between grade levels. Boys and girls, as is reported in Table 28, t were differentiated at the .05 level of confidence on the evaluative factor and the activity factor. The potency factor did not differentiate boys from girls. Girls evalu ated the concept more favorably and rated it as more active than did the boys, as may be seen in Table 51, Appendix D. Treatment groups, when compared in terms of grade level and sex of student, and grade levels when compared on the basis of sex of student, produced no significant inter action . The three-way comparison did not indicate any interaction between groups. Concept 9.4. Grownups. The expressed attitudes toward the concept of "grownups" differentiated between dropouts and stay-ins at the .01 level of confidence. In Table 49 of Appendix D, it is reported that stay-ins evaluated this concept more favorably, as more potent, and 269 TABLE 28 ANALYSIS OF BOY AND GIRL MEAN FACTOR SUM SCORES OF CONCEPTS RATED AGAINST SELECTED FACTOR SCALES* Evaluative Potency Activity Factor Factor Factor Concept F Ratio P F Ratio P F Ratio P School Experiences Teachers Grades School Reading Ideal teacher 15.10 2.71 3.39 9.18 13.99 .01 .01 .01 6.60 0.69 0.71 4.89 0.04 .01 a m .05 4.43 0.32 2.72 4.46 9.53 .05 .05 .01 Family Relationships Home Ideal parent Parents 1.51 17.89 2.56 .01 1.05 9.16 2.47 .01 a m 1.66 1.15 2.34 Social Interaction Most people My best friends Classmates Grownups 5.68 8.61 4.48 3.14 .05 .01 .05 0.12 10.48 1.56 0.71 .01 7.26 0.67 6.30 2.17 .01 .05 Self Concept m Sty school ability How I would like to be How my class sees me 1.77 7.84 18.59 0.92 .01 .01 2.90 0.80 51.78 1.16 .01 0.25 3.92 1.76 0.15 .05 *AnaIysis of covariance design with group test IQ score as covariate. p.05 “ 3.89, p.01 ® 6.76 with 1 and 253 degrees of freedom. 270 TABLE 28 (continued) ANALYSIS OF BOY AND GIRL MEAN FACTOR SUM SCORES OF CONCEPTS RATED AGAINST SELECTED FACTOR SCALES* Evaluative Factor Potency Factor Activity Factor Concept F Ratio P F Ratio P F Ratio P 3.87 .05 0.37 1.28 5.31 .05 0.66 - 4.52 .05 0.00 0.06 2.40 12.06 .01 2.63 3.21 1.84 - 0.54 - 0.57 - 3.88 .05 0.67 - 0.22 - 4.60 .05 0.32 0.73 - 4.01 .05 0.23 - 1.60 - Authority Relationships Authority Rules Punishment Goal Orientation Future College A job Graduating Quitting school Moral and Social Values Trying hard 1.07 - 1.34 - 0.85 - Cheating 8.76 .01 1.15 - 5.15 .05 Something easy 0.14 - 9.18 .01 1.29 - Something important 7.08 .01 0.53 M 4.51 «■ Money 0.12 0.03 0.42 - Success 10.04 .01 0.69 t e a 0.00 - 271 as more active than did the dropouts. Grade level and sex of the student did not produce significant differences. Treatment groups and grade levels, when compared on the basis of sex of student, failed to indicate significant differences. Grade level, when compared in terms of sex of ^ student, did not differentiate between groups. The three-way interaction produced no significant differences. Area of Investigation 10.0. Self Concept Concept 10.1. Me. Dropouts and stay-ins were differentiated by their responses to the factor of "my self," as may be seen in Table 26 and in Table 49 of Appendix D. In each instance the stay-ins rated the con cept more favorably on the evaluative factor, as more potent and as more active than did the dropouts. Differ ences were significant at the .01 level of confidence. Differences between the tenth and eleventh grade and sex of the student were not significant. Treatment groups, when compared in terms of grade level (see Table 29 and Table 30), revealed interactions significant at the .05 level of confidence between evalua tive factor scores; and an interaction significant at the .01 level of confidence between activity factor scores. Tenth grade stay-ins rated the concept more favorably on 272 TABLE 29 TREATMENT GROUPS AND GRADE LEVELS: TWO-WAY INTERACTION EFFECT, F RATIOS FOR EVALUATIVE, POTENCY AND ACTIVITY FACTOR SUM SCORES* Evaluative Factor Potency Factor Activity Factor Concept F Ratio P F Ratio P F Ratio P School Experiences Teachers 0.02 0.21 0.32 Grades 0.04 - 5.57 .05 0.53 - School 0.29 - 0.80 - 0.29 - Reading 2.09 - 5.75 .05 1.57 - Ideal teacher 2.31 — 0.10 - 2.04 - Family Relationships Home 0.00 - 0.48 0.74 mm Ideal parent 2.55 - 3.14 - 2.39 Parents 0.01 - 0.56 - 0.82 — Social Interaction Most people 2.28 - 0.57 2.70 mm My best friends 5.25 .05 1.08 - 5.64 .05 Classmates 1.47 - 0.09 - 1.66 - Grownups 0.01 - 0.02 — 0.02 mm Self Concept Me 4.74 .05 0.87 - 7.37 .01 My school ability 4.15 .05 1.54 - 5.47 .05 How I would like to be 10.20 .01 4.80 .05 4.75 .05 How my class sees me 0.01 - 0.16 - 1.63 - *Analysis of covariance design with group test IQ score as covariate. p.05 ® 3.89, p.01 “ 6.76 with 1 and 253 degrees of freedom. 273 TABLE 29 (continued) TREATMENT GROUPS AND GRADE LEVELS: TWO-WAY INTERACTION EFFECT, F RATIOS FOR EVALUATIVE, POTENCY AND ACTIVITY FACTOR SUM SCORES Evaluative Potency Activity Factor Factor Factor Concept F Ratio P F Ratio P F Ratio P Authority Relationships Authority 0.08 0.25 mm f t 0.50 Rules 0.00 - 0.68 0.12 - Punishment 0.14 - 1.52 - 1.10 - Goal Orientation Future 3.41 3.19 m 4.00 .05 College 1.92 - 1.57 - 0.47 - A job 2.21 - 3.04 - 0.20 - Graduating 5.39 .05 3.57 - 4.62 .05 Quitting school 1.38 — 0.63 mm 0.40 — Moral and Social Values Trying hard 1.69 — 1.63 mm 1.07 Cheating 0.20 - 0.23 - 0.00 - Something easy 1.18 - 0.04 - 0.56 - Something important 0.09 - 1.05 - 0.09 - Money 0.01 - 0.70 - 6.05 .05 Success 6.14 .05 6.73 .01 5.35 .05 / TABLE 30 TREATMENT GROUP B7 GRADE LEVEL TWO-WAY INTERACTION EFFECT: COMBINED MEAN SCORES FOR SELECTED CONCEPTS AND FACTOR SCALES FOR DROPOUTS AND STAY-INS COMPARED WITH TENTH AND ELEVENTH GRADE LEVELS* Dropouts Stay-Ins Factor -------------------- --------------- Concept Scale Grade 10 Grade 11 Grade 10 Grade 11 P Grades Potency 10.80 9.96 13.09 13.61 .05 Reading Potency 11.31 11.58 14.33 12.61 .05 My best friends Evaluative 17.12 18.16 18.68 17.78 .05 Activity 16.69 17.88 17.86 17.16 .05 Ms Evaluative 10.71 11.76 17.26 16.00 .05 Activity 11.63 12.52 16.98 15.40 .01 Jfy school ability Evaluative 9.98 11.01 16.53 15.33 .05 Activity 9.64 10.96 15.52 14.77 .05 How I would like to be Evaluative 17.81 10.80 20.20 19.73 .01 Potency 14.63 15.45 15,26 14.31 .05 Activity 18.40 19.11 19.70 19.31 .05 Future Activity 7.43 8.14 11.38 10.30 .05 Graduation Evaluative 5.27 5.84 11.34 9.33 .05 Activity 6.65 7.52 11.75 10.24 .05 Money Activity 6.44 7.64 6.74 5.93 .05 Success Evaluative 4.42 5.35 8.90 7.97 .05 Potency 7.02 8.33 10.55 9.69 .01 Activity 10.20 7.33 6.41 9.30 .05 *Means are reported for only the treatment group and grade levels producing n> significant interaction effects as indicated in Table 29. ^ 275 the evaluative factor scale, as well as more active, than did the eleventh grade stay-ins or the tenth or eleventh grade dropouts. Potency factor scores failed to reveal a significant interaction. A comparison of treatment groups and sex of the student, and grade levels compared with sex of the student also failed to produce significant interac tions. The three-way comparison, similarly, failed to identify any interaction effects. Concept 10.2. My school ability. Differences between dropouts and stay-ins on all factor scores, signif icant at the .01 level of confidence, were revealed in Table 26. Stay-ins evaluated as more favorable than did dropouts, and described as more potent and more active their attitudes toward this concept. Responses to this concept did not differentiate between grade levels. Boys and girls, however, expressed attitudes that were signifi cantly different, as is seen in Table 26 and in Table 51 of Appendix D. Girls evaluated their school ability more favorably than did boys, a difference significant at the .01 level of confidence; and girls rated this concept as more active than did boys, the difference also being significant at the .05 leyel. The potency factor ratings failed to differentiate the groups. Treatment groups, when compared in terms of grade 276 level, produced significant interactions on two of the three factor scales. Tenth grade dropouts evaluated the concept lower and as less active than did the eleventh grade drop outs, or either tenth or eleventh grade stay-ins, as is displayed in Table 30. The evaluative factor ratings pro duced an interaction significant at the .05 level of confidence; the activity factor ratings produced an inter action significant at the .01 level of confidence. The potency factor ratings did not produce a significant inter action. Sex of the student when compared with the treat ment groups and grade level evidenced no significant interactions. The three-way comparison produced no significant interactions. Concept 10.3. How I would like to be. The expressed attitudes toward this concept produced signifi cant differences between dropouts and stay-ins only on the evaluative factor (see Table 49 of Appendix D). The difference was significant at the .01 level of confidence. On the activity factor, the difference was significant at the .05 level. The stay-ins rated this concept more favor ably on the evaluative scale and as more active than did the dropouts. The potency factor failed to reveal any significant differences. No significant differences were exhibited between grade levels. Boys and girls were 277 differentiated in their responses to this concept on the evaluative and potency factors (p = .01), as is displayed in Table 51, Appendix D. Responses on the activity factor scale did not differentiate between boys and girls. Significant two-way interactions are identified in Tables 29 and 30 for the evaluative, potency and activity factor scales when treatment groups were compared for grade level. The evaluative factor produced an interaction significant at the .01 level of confidence, the potency and activity factor interactions being significant at the .05 level pf confidence. Eleventh grade stay-ins (reported in Table 30) rated the concept, "How I would like to be," more favorably on the evaluative factor scale and as more potent than did the tenth grade stay-ins or the dropouts at either grade level. Tenth grade stay-ins rated this concept as more active than did the eleventh grade stay-ins or the dropouts at either grade level. Treatment groups when compared with sex of the student indicated interaction significant at the .01 level of confidence between responses to the potency factor scale. Although the dropout and stay-in boys rated the concept as more potent than did the dropout and stay-in girls (see data in Table 32), the difference between responses of the girls produced the interaction effect. The evaluation and activity factor ratings exhibited no 278 TABLE 31 TREATMENT GROUPS AND SEX OF THE STUDENT: TWO-WAY INTERACTION EFFECT, F RATIOS FOR EVALUATIVE, POTENCY AND ACTIVITY FACTOR SUM SCORES* Evaluative Potency Activity Factor Factor Factor F F F Concept Ratio p Ratio p Ratio p School Experiences Teachers 0.01 - 2.21 - 0.13 - Grades 0.47 - 1.18 - 1.24 - School 0.99 0.08 - 3.75 - Reading 0.23 - 0.54 - 0.18 - Ideal teacher 0.06 0.02 0.21 - Family Relationships Home 1.15 - 3.35 - 2.25 - Ideal parent 0.19 0.00 - 1.62 - Parents 4.52 i n o a 2.62 - 5.96 .05 Social Interaction Most people 0.01 ao» 2.56 — 0.00 _ My best friends 0.63 - 3.41 - 0.01 - Classmates 0.79 - 0.23 - 0.26 - Grownups 0.00 — 0.22 — 0.13 - Self Concept Me 0.38 — 1.10 . » 0.54 My school ability 1.70 - 1.60 - 3.01 - How I would like to be 0.23 - 8.80 .01 0.29 - How my class sees me 0.05 - 2.27 - 1.35 - ^Analysis of covariance design with group test IQ score as covariate. p.05 = 3.89, p.01 » 6.76 with 1 and 253 degrees of freedom. 279 TABLE 31 (continued) TREATMENT GROUPS AND SEX OF THE STUDENT: TWO-WAY INTERACTION EFFECT, F RATIOS FOR EVALUATIVE, POTENCY AND ACTIVITY FACTOR SUM SCORES Evaluative Potency Activity Factor Factor Factor , F F F Concept Ratio p Ratio p Ratio p Authority Relationships Authority 3.14 - 0.20 - 0.93 - Rules 1.36 - 0.37 - 0.47 - Punishment Goal Orientation 1.88 0.74 0.51 Future 5.60 .05 0.22 3.30. College 0.34 - 0.01 - 0.01 - A job 0.09 - 0.03 - 0.13 - Graduating 0.26 - 0.59 - 0.57 - Quitting school 0.15 — 0.49 0.06 — Moral and Social Values Trying hard 1.85 • 0.07 1.21 Cheating 0.08 - 0.37 - 0.00 - Something easy 6.93 .01 0.00 - 4.94 .05 Something important 0.48 - 1.17 - 0.35 - Money 0.21 - 0.21 - 1.09 - Success 0.06 0.08 0.54 TABLE 32 TREATMENT GROUP BY SEX OF THE STUDENT TWO-WAY INTERACTION EFFECT: COMBINED MEAN SCORES FOR SELECTED CONCEPTS AND FACTOR SCALES FOR DROPOUTS AND STAY-INS COMBINED WITH BOYS AND GIRLS* Factor Dropouts Stay-ins Concept Scale Boys Girls Boys Girls P Parents Evaluative 9.40 12.16 16.63 16.10 .05 Activity 9.99 12.29 14.85 14.38 .05 How I would like to be Potency 15.91 14.16 16.93 12.64 .01 Future Evaluative 6.85 6.67 11.95 8.33 .05 Something easy Evaluative 6.51 7.69 7.50 6.54 .01 Activity 8.20 19.53 18.94 8.55 .05 *Means are reported for only the group and sex of the student effects producing significant interactions as indicated in Table 31. Note: Interaction effects significant at the .05 level of confidence were obtained for the concepts: "parents”— evaluative and activity factor scales; "future”— evaluative factor scale and "something easy'*--activity factor scale. Since five significant inter action effects could have possibly occurred by chance out of the 90 comparisons, these interaction effects might reasonably be considered to have occurred by chance. ^ 281 f interaction. Grade level, when compared with sex of stu dent, produced no significant interaction. The three-way comparison reported in Tables 31 and 32 reveals an interaction effect significant at the .05 level of confidence for the evaluative factor. The potency and activity factors produced no significant interaction. In general, the stay-ins rated the concept more favorably than did the dropouts, the tenth grade dropout boys rating the concept significantly lower than other groups. Concept 10.4. How my class sees me. Dropouts and stay-ins expressed attitudes differentiating these groups on the evaluative, potency and activity factors at the .01 level of confidence. Table 27 presents data to indicate that tenth and eleventh grade pupils were differentiated on the potency factor, but this difference may reasonably be expected to have occurred by chance. Boys and girls were not differentiated by any of the factor scales. Treatment groups, when compared on the bases of grade level and sex of student, and grade level, when com pared in terms of sex of student, failed to reveal any significant interactions. Table 34 and 35 summarize results of a three-way interaction in potency rating scores significant at the .05 level of confidence. A tendency existed for the stay-ins to rate this concept more favorably than did dropouts. The 282 TABLE 33 GRADE LEVEL AND SEX OF THE STUDENT: TWO-WAY INTERACTION EFFECT, F RATIOS FOR EVALUATIVE, POTENCY AND ACTIVITY FACTOR SUM SCORES* Evaluative Potency Activity Factor: Factor Factor F F F Concept Ratio p Ratio p Ratio p School Experiences Teachers 1.38 - 0.01 - 0.00 - Grades 1.91 - 0.00 - 0.28 - School 0.56 - 3.00 - 1.53 - Reading 0.11 - 0.11 - 0.07 - Ideal teacher 0.54 - 2.07 — 1.22 - Family Relationships Home 0.01 — 0.52 — 0.19 _ Ideal parent 0.04 0.14 - 0.18 - Parents 0.54 - 2.46 - 0.02 - Social Interaction Most people 0.80 40 1.27 0.72 _ My best friends 0.80 - 0.00 - 0.96 - Classmates 0.32 A 0.03 - 0.05 - Grownups 0.05 - 0.05 - 0.15 - Self Concept Me 0.16 - 2.45 — 0.52 _ Jfy school ability 2.03 - 2.15 - 0.61 - How I would like to be 0.28 - 0.13 - 0.14 ■» How my class sees me 0.16 - 0.76 - 0.01 *Analysis of covariance design with group test IQ score as covariate. p.05 « 3.89, p.01 « ■ 6.76 with 1 and 253 degrees of freedom. 283 TABLE 33 (continued) GRADE LEVEL AND SEX OF THE STUDENT: TWO-WAY INTERACTION EFFECT, F RATIOS FOR EVALUATIVE, POTENCY AND ACTIVITY FACTOR SUM SCORES Evaluative Potency Activity Factor Factor Factor F F F Concept Ratio p Ratio p Ratio p Authority Relationships Authority 0.64 - 0.00 - 0.18 Rules 0.04 - 0.69 - 0.11 Punishment 0.08 - 1.03 - 0.71 Goal Orientation Future 0.02 - 0.66 - 1.05 College 1.06 - 0.05 - 1.20 A job 0.26 - 2.89 - 1.09 Graduating 0.06 - 1.16 - 0.02 Quitting school 0.11 - 0.07 - 0.00 Moral and Social Values Trying hard 1.12 - 0.02 - 0.21 Cheating 0.17 - 0.78 - 0.00 Something easy 0.26 - 0.74 - 0.14 Something important 3.76 - 2.14 - 3.20 Money 0.55 - 0.28 - 0.25 Success 0.33 - 0.07 - 0.01 284 TABLE 34 TREATMENT GROUPS AND GRADE LEVEL AND SEX OF THE STUDENT: THREE-WAY INTERACTION EFFECT, F RATIOS FOR EVALUATIVE, POTENCY AND ACTIVITY FACTOR SUM SCORES* «" i t — r ~ r — t — L- 1 ■ in in Evaluative Potency Activity Factor Factor Factor Concept F Ratio P F Ratio P F Ratio P School Experiences Teachers 1.80 0.01 4.35 .05 Grades 0.10 - 2.80 - 0.46 - School 0.04 - 0.39 - 0.38 - Reading 1.13 - 4.84 .05 1.00 Ideal teacher 5.11 .05 0.09 ■ » 3.28 - Family Relationships Home 0.60 _ 0.02 • 0.74 — Ideal parent 0.67 - 1.51 - 0.10 - Parents 0.10 0.13 - 0.17 — Social Interaction Most people 0.28 0.00 — 0.12 < _ My best friends 0.85 0.69 - 0.05 - Classmates 0.67 - 0.38 - 0.25 - Grownups 0.19 0.07 e » 0.09 - Self Concept Me 0.34 3.11 - 0.87 _ My school ability 0.41 3.58 ; - 0.02 - How I would like to be 4.82 .05 0.03 - 0.51 - How my class sees me 0.04 6.48 .05 0.63 - *Analysis of covariance design with group test IQ score as covariate. p.05 « 3.89, p.01 «= 6.76 with 1 and 253 degrees of freedom. 285 TABLE 34 (continued) TREATMENT GROUPS AND GRADE LEVEL AND SEX OF THE STUDENT: THREE-WAY INTERACTION EFFECT, F RATIOS FOR EVALUATIVE, POTENCY AND ACTIVITY FACTOR SUM SCORES* Evaluative Factor Potency Factor Activity Factor Concept F Ratio P F Ratio p F Ratio P Authority Relationships Authority 0.05 - 0.70 0.04 - Rules 0.00 - 0.06 - 0.19 - Punishment Goal Orientation 0.08 0.41 - 0.06 Future 1.50 — 3.29 2.03 College 2.22 0.11 1.33 - A job 0.00 - 1.66 0.01 - Graduating 1.53 - 0.88 0.54 - Quitting school 1.56 “ 0.02 0.93 c * > Moral and Social Values Trying hard 0.00 — 0.03 0.00 — Cheating 4.45 .05 4.69 .05 7.92 .01 Something easy 0.57 - 0.13 0.47 - Something important 0.11 - 0.23 0.60 - Money 0.54 - 0.08 0.02 - Success 4.13 .05 1.10 2.53 ' TABLE 35 TREATMENT GROUP BY GRADE LEVEL BY SEX OF THE STUDENT: THREE-WAY INTERACTION EFFECT-- MEAN SCORES FOR SELECTED CONCEPTS AND FACTOR SCALES FOR DROPOUTS AND STAY-INS COMPARED WITH TENTH AND ELEVENTH GRADE PUPILS AND WITH BOYS AND GIRLS* Dropouts Stay-ins Factor Grade 10 Grade 11 Grade 10 Grade 11 Concepts Scale Boys Girls Boys Girls Boys Girls Boys Girls P Teachers Activity 8.53 10.88 9.64 9,58 14.38 13.97 13.76 14.55 .05 Reading Potency 10.35 12.27 11.76 11.39 14.38 14.27 12.15 13.71 .05 Ideal teacher Evaluative 16.09 19.21 18.39 18.65 19.00 19.52 18.85 19.97 .05 How I would like to be Evaluative 16.68 18.94 18.52 19.35 20.12 29.27 19.03 20.42 .05 How my class sees me Potency 10.65 12.06 11.60 11.10 12.68 12.67 12.61 14.35 .05 Cheating Evaluative , 17.91 18.39 18.03 19.71 12.06 15.61 14.48 14.48 .05 Potency 14.56 13.55 14.55 15.87 11.85 13.09 12.82 12.16 .05 Activity 16.24 15.42 15.91 17.23 11.97 15.27 13.82 13.29 .01 Success Evaluative 4.62 4.21 5.94 4.77 10.56 7.24 8.09 7.84 .05 *Means are reported for only the treatment group, grade level and sex of the student comparisons producing significant interaction effects as indicated in Table 33. r o o o Ov 287 tenth grade dropout boys, however, rated the concept as significantly less potent than did the other groups. Area of Investigation 11.0. Authority Relationships Concept 11.1. Authority. Responses to the concept of ’ ’ authority" differentiated between dropouts and stay-ins at the .01 level of confidence. In Table 49, Appendix D, " it is seen that the stay-ins expressed attitudes which evaluated the concept of authority more favorably, as more potent and as more active than did the dropouts. The tenth and eleventh grade groups were not differentiated on this concept. Boys and girls were differentiated by their evaluative factor scores, the difference being significant at the .05 level of confidence. Table 51, Appendix D, dis closes that boys tended to evaluate the concept of author ity more favorably than did girls. Neither the potency nor the activity factor scores revealed any differences between groups. Treatment groups and grade level, when compared in terms of sex of student, failed to exhibit any significant interactions. The three-way interaction revealed no significant differences of response. Concept 11.2. Rules. Attitudes expressed as responses to the concept "rules” differentiated between 288 dropout and stay-in groups at the .01 level of confidence. Data presented in Table 49 reveal that stay-ins evaluated the concept more favorably and rated it as more potent and as more active than did the dropout group. The tenth and eleventh grade samples were not differentiated on any of these factors. Boys and girls were differentiated at the .05 level of confidence on the evaluation and activity factor scales. The potency factor failed to reveal any significant differences. As is seen in Table 29, boys reported attitudes that were higher in evaluation and more active than did girls. Neither treatment groups, when compared in terms of grade level and sex of student, nor grade level, when com pared on the basis of sex of student, revealed significant interaction effects. The three-way interaction produced no significant differences in response. Concept 11.3. Punishment. Responses to the stimu lus concept of punishment revealed differences between dropouts and stay-ins on the evaluative and activity factors that were significant at the .01 level of confi dence, and differences significant at the .05 level of confidence for the potency factor. Table 49 of Appendix D reveals that stay-ins evaluated the concept of punishment more favorably and as more potent and more active than did 289 dropouts. Neither tenth and eleventh grade samples, nor the boys and girls, were differentiated by their responses to this concept. Treatment groups, when compared in terms of grade level and sex of student, and grade levels, when compared on the basis of sex of student, failed to produce any significant interactions. The three-way interaction indicated no significant interaction effects. Area of Investigation 12.0. Goal Orientation Concept 12.1. Future. Dropouts and stay-ins were differentiated at the .01 level of confidence by their responses on the evaluative, potency and activity factor scales. Table 49, Appendix D, reveals that stay-ins con sistently expressed attitudes of greater evaluation, potency and activity than did the dropouts. At tenth and eleventh grade levels, responses did not differentiate between levels. Boys evaluated the concept of "future” more favorably than did girls, but neither the potency nor the activity factors produced significant differences according to data summarized in Table 29 and in Table 51 of Appendix D. Treatment groups, when compared in terms of grade levels, revealed an interaction that was significant at 290 the .05 level of confidence for the activity factor. In Table 31, it is revealed, both tenth and eleventh grade stay-ins responded to this concept more favorably than did dropouts; the tenth grade stay-ins rated the concept "future” as substantially more active than did any of the other groups. Neither the evaluative nor the potency factors produced a significant interaction. Treatment groups, when compared on the basis of sex of student, produced an evaluative factor interaction significant at the .05 level of confidence. The tenth grade stay-in boys evaluated this concept more favorably than the other groups; however, since so few significant interactions were evidenced in this comparison (Table 28) these interactions may reasonably be attributed to chance. The potency and activity factor interaction effects were not significant. Grade levels, when compared in terms of sex of student, did not yield any significant interactions. The three-way interaction was not significant for any of the factors rated against this concept. Concept 12.2. College. Responses to the stimulus concept, "college," differentiated between dropouts and stay-ins at the .01 level of confidence for the evaluative, potency and activity scales. Stay-ins, according to data displayed in Table 49, Appendix D, evaluated "college" more favorably than did dropouts, and regarded it as more 291 potent and more active. Responses to this concept did not differentiate between the tenth and eleventh grade pupils, nor between boys and girls. Treatment groups, when compared on the basis of grade level and sex of student, and grade level, when com pared in terms of sex of student, failed to reveal any significant two-way interaction effects. The three-way interaction did not identify any significant differences. Concept 12.3. A job. Dropouts and stay-ins were differentiated significantly at the .01 level of confidence by responses on the evaluative factor scales. Neither the potency factor nor the activity factor revealed any differ ences between groups. Dropouts evaluated the concept of "a job" more favorably than did stay-ins. Responses of tenth and eleventh grade pupils failed to identify differ ences between groups. Boys evaluated the concept of "a job" more favorably than did girls, a difference that was significant at the .05 level of confidence, as is seen in Table 30 and in Table 51 of Appendix D. Treatment groups, when compared in terms of grade level and sex of student, and grade levels, compared on the basis of sex of student, did not yield any significant interaction effects. The three-way interaction did not produce any significant interaction effects. 292 Concept 12.4. Graduating. The expressed attitudes regarding the concept of "graduating" differentiated between dropouts and stay-ins at the .01 level of confi dence on the evaluative, potency and activity scales. Stay-ins rated the concept more favorably on all three factors than did dropouts. Tenth and eleventh grade pupils revealed no significant differences in their responses to this concept. Boys evaluated the concept of "graduating" more highly than did girls. The potency and activity factors did not differentiate between these groups. Treatment groups when compared with grade levels yielded an interaction effect significant at the .05 level of confidence on the evaluative and activity factors. The combined mean potency factor scores did not reveal signif icant interaction effects. Table 31 reveals that the concept, "graduation," was rated more favorably on the evaluative scale and as more active by the-tenth grade stay-ins than by other groups. In general, ratings of the stay-ins were consistently more favorable than those of the dropouts on all factor scales. Treatment groups when com pared with sex of the student and grade level when compared with sex of the student did not yield significant interac tions. The three-way interaction did not yield significant interaction effects. 293 Concept 12.5. Quitting school. Dropouts revealed significantly greater evaluation, potency and activity factor ratings than did stay-ins for this concept, as may be seen in Table 49, Appendix D. Differences were signifi cant at the .01 level of confidence. Tenth and eleventh grade pupils were not differentiated by their responses to this concept. Table 51 of Appendix D reveals that girls evaluated the concept of "quitting school" more favorably than did boys, a difference significant at the .05 level of confidence. Treatment groups when compared with grade levels and with sex of the student and grade levels when compared with sex of the student did not yield significant interac tions. The three-way interaction, similarly, was not significant for any of the three factors. Area of Investigation 13.0. Moral and Social Values Concept 13.1. Trying hard. This concept elicited responses that were significantly different (at the .01 level of confidence) between dropouts and stay-ins on all three factor scales. Stay-ins evaluated the concept more favorably, regarded it as more potent and more active than did the dropouts, as is seen in Table 49 of Appendix D. 294 Grade levels were not differentiated on this concept, nor were the boys and girls differentiated. Treatment groups, when compared on the basis of grade levels and sex of student, and grade levels, when compared in terms of sex of student, exhibited no signifi cant interactions. The three-way interaction identified no significant interactions. Concept 13.2. Cheating. Dropouts and stay-ins revealed differences significant at the .01 level of confi dence toward this concept. Dropouts, as is seen in Table 49, Appendix D, rated the concept of "cheating" more favor ably on the evaluative factor, and as more potent and more active than did stay-ins. Tenth and eleventh grade pupils were not differentiated in their attitudes toward the concept. Girls evaluated the concept of "cheating" more favorably than did boys, and regarded it as more active. This is exhibited in Table 51 of Appendix D. Differences were significant at the .01 and .05 levels of confidence, respectively. Treatment groups, when compared in terms of grade level and sex of student, and grade levels, when compared in terms of sex of student, revealed no significant interactions. Three-way interactions were significant for the 295 evaluative (p « .05), potency (p « .05) and activity (p = .01) factors. Not evident as a significant interaction in * the two-way comparisons, the concept of "cheating'1 produced Interaction effects when considered in a three-way compari son. Stay-ins in each instance offered less favorable opinions of this concept than did dropouts. This consist ency of response may account in part for this effect. Concept 13.3. Something easy. The concept of "something easy" differentiated dropouts from stay-ins at the .01 level of confidence, but only on the factor of potency. The factors of evaluation and activity failed to reveal any significant differences. The dropout sample regarded the concept of "something easy" as more potent than did stay-ins. Tenth and eleventh grade pupils were not differentiated by responses to this concept. Boys and girls were differentiated only on the potency factor, girls regarding the concept as more potent than did boys. The difference was significant at the .01 level of confi dence . Treatment groups, when compared in terms of grade levels, revealed no significant interactions. When they were compared on the basis of sex of student, treatment groups exhibited significant interactions for the evalua tive and the activity factors. The interactions were significant at the .01 and .05 levels, respectively. 296 Table 31 reveals that the activity factor score produced an interaction effect significant at the .05 level of confi dence. This effect may reasonably be considered to have occurred by chance, as was mentioned previously. Grade level, when compared in terms of sex of student, revealed no significant interaction effects. The three-way comparison revealed no significant interactions. Concept 13.4. Something important. Attitudes expressed by dropouts and stay-ins differentiated between groups at the .01 level of confidence for each of the three factors. Table 49 of Appendix D discloses that stay-ins gave this concept a higher evaluation, and rating it as more potent and more active than did dropouts. Tenth and eleventh grade pupils were not differentiated by responses to this concept. Boys and girls were differentiated on the evaluative factor, at the .01 level of confidence. Boys evaluated the concept, ^'something important, more favor ably than did girls, as is seen in Table 51 of Appendix D. Neither the potency factor nor the activity factor revealed a significant difference. Treatment groups, when compared on the .bases of grade level and sex of student, and grade levels, when compared in terms of sex of student, revealed no signifi cant interaction effects. 297 The three-way comparison failed to identify any significant interactions. Concept 13.5. Money. Dropouts and stay-ins were not differentiated by responses to the concept of "money" on any of the three factor scales. Tenth and eleventh grade pupils were not differentiated, nor were boys and girls significantly different in their responses to this concept. Treatment groups, when compared in terms of grade level, revealed a significant interaction (p » .05) for the activity factor. Combined means given in Table 30 indicate that the eleventh grade dropout rated this concept as more active than did the other groups. The sex of the student, when considered in relation to treatment groups and to grade levels, failed to identify any significant interac tion effects. The three-way comparison produced no significant interactions. Concept 13.6. Success. Dropouts and stay-ins were clearly differentiated by responses to the concept of ’’success.'* Stay-ins regarded the idea of "success” more favorably, and rated it as more potent and more active than did dropouts, as is seen in Table 49, Appendix D. Tenth and eleventh grade samples were not differentiated by responses to this concept. Boys evaluated the concept of 298 "success" more favorably than did girls, as is reported in Table 51 of Appendix D, the differences being significant at the .01 level of confidence. Interactions significant at the .05 level were found for the evaluative and activity factors when groups were compared in terms of grade level. Interactions significant at the .01 level of confidence were found for the potency factor. Interactions were also found between dropouts and stay-ins in tenth and eleventh grades. Tenth and eleventh grade stay-ins rated the concept more favor- « ably on the evaluative and potency scales than did their dropout counterparts; however, tenth grade dropouts rated the concept as more active than did the other groups (tenth grade stay-ins and eleventh grade dropouts and stay-ins)-- an interesting reversal of what might have been expected. These data are summarized in Table 30. Sex of student, when compared in terms of both treatment groups and grade levels, failed to reveal significant interaction effects. The three-way comparison exhibited an evaluative factor interaction significant at the .05 level of confi dence. The potency and activity factors yielded no significant evidence of interaction. 299 CHAPTER SUMMARY This chapter has presented in detail the data gathered in the study. Seven selected variables provided a descriptive background of the sample. Father occupation ratings indicated that the sample represented Occupation Scale Class 4 (after Warner, Meeker and Eells, 1960) . Group test IQ scores ranged from 95.21 to 101.16. The dropout sample had an average IQ of 98.77 while that of the stay-in sample was 100.18. The difference is not statistically significant. Dropouts were generally older than the stay-in group, the chronological ages at the time of testing being 16 years 4 months and 15 years 10 months, respectively (p = .01). An index of reading achievement was generated from three subtests of the ITED test battery. Dropouts earned an average reading index of 10.33 score points, while stay-ins earned an average of 15.27 score points (p = .01) for all groups. Dropouts were absent from school almost three times as often as were stay-ins. The dropouts averaged 28.87 days of absence annually, in contrast to 10.22 days of absence for the stay-ins during the school year (p = .01). The number of days each dropout was enrolled varied from 91 to 104, the difference being significant at the .01 level of confidence. The mean number of days enrolled was 96.92 for the 178-day school year. Stay-ins were in school 300 from 175 to 178 days during the school year, the average being 177.18 days. The percentage of mean”days-absent varied from 26.03 to 34.26 for the dropouts, and from 5.19 to 6.43 for the stay-ins. The average leaving age of all dropouts was 16 years 7 months. A distribution of the dropout sample based upon the sex of the student and his grade level revealed that in the tenth grade there were 34 boys and 33 girls, in the eleventh grade 33 boys and 31 girls. The dependent variables consisted of 270 ratings made by each subject. Nine bipolar adjective scales repre sented three evaluative factor scales, three potency factor scales and three activity factor scales (after Osgood, 1957) . Thirty stimulus concepts representing seven areas of investigation were selected for study. A computer-based program generated F ratios in an analysis of covariance design. Group test IQ score was the covariate. Attitudes toward school experiences were assessed in terms of ratings of the stimulus concepts "teachers," "grades," "school," "reading," and "ideal teacher." In general, the stay-in sample rated these concepts more favorably (p = .01), as more potent (p = .01), and as more active (p » .01) than did the dropout sample. Feelings about family relationships were expressed by rating the concepts "home," "ideal parent," and 301 "parents." Again, the stay-in group regarded these con cepts more favorably (p - .01), as more potent (p - .01) and as more active (p » .01) than did the dropouts. Social interactions were evaluated in terms of responses to the concepts "most people,” "my best friends," "classmates," and "grownups." As one might anticipate, the stay-ins tended to rate these concepts more favorably (p ■ .01), as more potent (p = .01), and as more active (p “ .01) than did the dropouts. The concept "my best friends" differentiated the groups only on the potency scale (p “ .05). Attitudes toward the self were assessed as responses to the concepts "me," "my school ability," "how I would like to be,” and "how my class sees me." Gener ally, the stay-in reported a better self concept than did the dropout. They reported more favorable evaluative attitudes (p = .01), more potent attitudes (p » .01), and more active attitudes (p «= .01) . Authority relationships were reported as ratings of the concepts "authority," "rules," and "punishment." Stay-ins evaluated this area more favorably (p * .01), as more potent (p ■* .01), and as more active (p = .01) than did dropouts. Dropouts tended to express negative atti tudes toward these concepts through less favorable ratings. Goal orientation was related to educational evidence, for the most part. The concepts of "future" and 302 "college" were rated more favorably (p » .01) by the stay- in group, whereas the dropouts reacted more positively to the concepts of "a job" (p * ,01) and "quitting school" (p «= .01) . Attitudes toward selected moral and social values were evaluated in terms of responses to the concepts "trying hard," "cheating," "something e^sy," "something important," "money," and "success." The concepts of "trying hard," "something important," and "success" were rated by the stay-in group as more favorable (p = .01), as more potent (p « = > .01), and as more active (p = .01). The dropouts, on the other hand, rated as more favorable the concepts of "cheating" (p = .01) and "something easy" (p = .01). Differences in responses to the concept "money" were not statistically significant, but were in the antici pated direction. In the following chapter attention is directed to an evaluation and an integration of these findings with the results of previous research, and viewpoints expressed in the literature. Unique findings are identified, and recom mendations .developed regarding further study. CHAPTER V SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS Every researcher in education knows that research is often a two-edged blade. With one edge he can cut deeply and revealingly into unexplored areas of human behavior. With the other he can carve up empirical evidence to fit snugly with his preconceived ideas. A. J. Tannenbaum (1962:69) SUMMARY The world of work demands ever-increasing knowledge and skill because many simple and menial tasks are being performed by machines, or with machines that require opera tors with specialized training. Increased longevity and improved general health have swelled the number of experi enced and older workers throughout the labor market, constricting the availability of jobs for youth. The youngster who fails to complete a regular high school education begins at a distinct disadvantage in the world of work. Educators now .have means at their dis- posal--the knowledge and techniques for coping with the academically clumsy pupil. Those observable characteris tics which set the potential high school dropout apart from his peers are symptoms of conscious or subconscious 303 304 motivations which eventually take form on the occasion of leaving school before graduation. The decision to leave school is confirmed long before the occurrence. Attitudes concerning school expressed by dropouts have been evaluated repeatedly in research literature. The dropout is discontent with school, feels rejected by others, and is withdrawn from school social life. His interests are vague, ill-defined concerns that usually take form in immediate^gratifications— a job, even without pay; truancy, malingering— almost anything to avoid school and its frustrating defeats, isolation and boredom. The pupil's tolerance for academic frustration, and his feel ings about teachers and school in general find genesis in the home. The attitude of parents and siblings toward school constitutes a powerful influence upon the desire to persist in a formal educational setting. Educators have available sufficient information to deal adequately with the "dropout problem." Studies carried out as long as sixty years ago identified some of the basic factors associated with early school leaving and offered remedial techniques that are valid today. The purpose of this study is to identify critical attltudinal variables in the motivational patterns of drop outs and stay-ins among middle-class Caucasians by means of the semantic differential process. Seven descriptive variables were selected to define 305 the sample: socioeconomic status rating, group test IQ score, age at time of testing, an index of reading achieve ment, number of absences during the school year, number of days enrolled in school, and the age of the dropouts at the time of leaving school. Thirty stimulus concepts were selected in seven areas of investigation: school experiences, family rela tionships, social interaction, self concept, authority relationships, goal orientation, and moral and social values. Nine bipolar pairs of adjectives were selected from Osgood (1957:37) representing three evaluative, three potency and three activity factor scales. A computer-based general linear factorial analysis of covariance program provided for an examination of both the main effects and the interactions of treatment groups (dropouts and stay- ins, tenth and eleventh grades, boys and girls) in a 2x2x2 statistical model. Group test IQ score was selected as the covariate. The population from which the sample was drawn consisted of 3,095 tenth and eleventh grade pupils in three suburban high schools in southern California. All pupils took the Meaning of Words Test at the beginning of the Fall semester, 1963. After the dropouts had been identified, a group of stay-ins matched on grade level, sex, and group- test IQ score group (89 and below, 90 to 110, and 111 and above) were selected. The following digest summarizes the major findings 306 of the study. The first part concerns the descriptive variables followed by an evaluation of responses to the Meaning of Words Test considered in terms of each of the seven areas of investigation. The group of dropouts in this study may be consid ered as a total population from the district or as a non- random sample of "dropouts from high school." Since the group cannot be considered to be representative of any known population, the findings reported do not permit generaliza tions beyond this sample. SUMMARY OF MAJOR FINDINGS Description of the Dropout The average dropout in this study is the product of our middle-class society (H^ 1). Boys do not drop out in greater numbers than do girls. This effect may well be related to restriction of- the sample population and to the one-time sampling procedure. The dropout is older (H-^ 3) than his classmates. This over-ageness suggests the possi bility of grade retardation or retention. The dropout performs at a lower level of achievement, for whatever reason, than does the stay-in (H^ 2). Concomitantly, the dropout is a poor reader (H^ 4). Attention to the need for remedial reading classes for those pupils performing sub stantially below grade level would appear essential to reduce the number of pupils who become dissatisfied with 307 school. Absence from class (H^ 5) represents an active form of aggression against school and Is a logical antecedent to early school leaving. Dropping out of school evidences running away, a technique of avoidance. Close attention to the attendance habits of pupils through the services of adequate child welfare and attend ance personnel in each school can be an effective influence upon the potential dropout to remain in school. Personal maturity can be developed through contact with understanding counselors, and satisfying school expe riences. Pupils who remain to graduate do so because of satisfying experiences associated with continuing in school, and encouragement from parents, teachers and peers. Attitudes of the Dropout The dropout appears, frequently, to have developed a sour and sullen outlook regarding his environment and those about him. Very probably these feelings have been fostered by repeated upbraidings, punishment and penalties for acts of which he may not always be aware. The discon tent, distress and dissatisfaction with himself, with his parent, school and environment can be demonstrated dis tinctly. Help must be individualized for each student in accordance with the particular circumstances that create 308 the emotional problems and school difficulties. Area of investigation 7. School experiences. The dropout regarded school with dissatisfaction, as an unpleasant and threatening situation. He regarded teachers with suspicion and considered them unfair, weak and dull. Area of investigation 8. Family relationships. The attitudes expressed by the dropout toward his parents and home reflected unpleasantness, lack of respect and diminished vitality. The disagreeable attitudes expressed are probably in part the subject's reactions to his own influences upon the family. Problems in intra-family relationships do not exist in a vacuum; it is essential therefore to involve parents in any treatment effort. Attentive, skilled child welfare personnel can, through interest and understanding, assist in effecting beneficial changes in family relationships. Area of investigation 9. Social interaction. Just as with the family, deteriorated interpersonal relation ships characterized the dropout. The dropout found little companionship among his classmates. He felt left out, and, in fact, avoided by others. The dropout evaluated his best friends favorably, as did the stay-in, which merely demon strates that we all like those whom we regard as friends. The involvement of a potential dropout in satisfy- 309 ing interpersonal contacts with an understanding teacher, counselor or child welfare worker who can provide unquali fied faith and confidence can be an effective influence in the child's decision to remain in school; perhaps in spite of continued academic discouragement. Area of investigation 10. Self concept. The drop out is not insensitive to his own attitudes, nor to the attitudes of others around him. Being surrounded with depreciation, discouragement and dissatisfaction at nearly every turn produces an inevitable sense of worthlessness and defeat within the individual. Construction of an adequate self image requires satisfying interaction with one's environment. Successful experiences in’ the classroom, in the community, and with other people are essential factors in developing a healthy self concept. Changes in self concept seem to be indicated, if the potential dropout is to graduate. Public school can begin to assist the potential dropout by taking the time to be available to him, and experimenting to find ways of providing small academic successes and satisfying interper sonal relationships. Area of investigation 11. Authority relationships. Because to the dropout authority almost always means unpleasantness, if not punishment, he reacts with negative feelings toward all authority relationships. Imposed 310 restrictions and external discipline tend to induce atti tudes of aggression and resentment toward the source of the control, as well as toward the self. Authority is no threat to those who possess self control and some form of inner discipline. Self discipline can be developed within an individ ual when he feels that he is a part of the school environ ment, when he feels that he belongs at home and is accepted in the community. Good attitudes toward authority and external control develop after the individual has begun to establish habits of self control. Area of investigation 12. Goal orientation. The dropout does not view the future with optimism. He is certain that, regardless of what happens, it will be unfair, unpleasant and lacking in vitality. On the other hand, the dropout favors the idea of quitting school and obtaining a job. He abhors the thought of attending coliege--even graduating from high school holds no value for him. At the same time, the stay-ins regard the future with some optimism, favor graduation from high school and college, and place less value on obtaining work at this time. Similarly the stay-in reacts negatively to the idea of quitting school. This desire to escape from the pressures of formal education is expressed in searching for another form of 311 activity; something that he can do (with his hands) is financially rewarding, and a job from which he can derive the acceptance and appreciation of others. Abundant opportunity exists within the public school structure today to provide the potential dropout with meaningful work experience programs, vocational educa tion, apprenticeship programs, and full-time employment in association with continuation education classes. Current California legislation compelling the establishment of continuation education classes will undoubtedly affect a number of pupils who would otherwise be dismissed from school. Area of investigation 13. Moral and social values. The pervasive attitude of gloom and dissatisfaction expressed by' the dropout is extended to the assessment of value systems. With the exception of a supporting attitude where cheating is concerned, the dropout expresses little t sufferance for accepted middle-class value-attitude systems. Value-attitude systems are learned. They result from the constellation of emotional and social factors continuously impressed upon the individual. We develop value systems in accordance with the expectations of those whom we respect and in a manner consistent with those from whom we seek approbation. Acceptance of good middle-class values cannot be accomplished by legislation. First the 312 pupil allies himself, literally or figuratively, with a teacher, counselor or student whom he wishes to emulate. If the person selected possesses acceptable middle-class values, the proselyte will accept and practice habits of an approved value-attitude system. Discussion The findings regarding factors of socioeconomic status rating, group test IQ score, reading achievement index, and proportion of absences were in accordance with ideas expressed in the literature, in the expected direc tion, and consistent with anticipated outcomes. The rather distressing characterizations drawn from these data cannot be generalized too broadly. These atti tudes have been expressed as forced-choice responses to discrete stimuli in a formalized setting. These are the reported attitudes of a small group of youngsters who were compelled to rate selected stimuli concepts against speci fied bipolar adjective scales. The ratings of dropouts were statistically different from those of stay-ins. Described in the study are some of those attitudes associ ated with early school leavers— offering a usable predic tive device for those who would attempt to implement these -findings within the public school setting and to conduct further research. CONCLUSIONS These conclusions are based upon a sample of 131 Caucasian tenth and eleventh grade pupils in three suburban Los Angeles high schools who discontinued regular class attendance prior to the completion of the 1963-1964 school year. * Generalizations regarding the high school students of the district are restricted to the extent that only two grade levels were included in the final study, and that three of four high schools'participated in the research. Research Hypotheses Regarding Selected Descriptive Variables Hypothesis 1. Socioeconomic Status Rating. Dropouts, it was anticipated, would come from families of lower socioeconomic status rating than .would pupils who staved in school. (Finding: &L 1.0 accepted.) The rating of family social and economic conditions as a predictor of successful completion of high school education has resulted in some indecision and contradic tion. Selected literature reviewed in this work support this contention. Although several studies have not been consistent with the argument that dropouts tend to come from lower socioeconomic environments, the majority of studies support this generalization. Median occupation ratings for dropouts were 314 generally lower than those for stay-ins; this finding is in line with those reported in the majority of related studies. Hypothesis 2. Group Test IQ Score. Group test IQ scores were expected not to differ significantly between dropout and stay-in groups. TFiFdihg: ^"'2.0 accepted.)------------ There exists little inconsistency in the literature that high school dropouts score lower than stay-ins on group intelligence tests, academic achievement tests, and over-all classroom performance. Individual reports support or renounce IQ as a predictor of persistence in school; however, there is a firm tendency to associate lower-than- average test scores with early school leaving. No signifi cant differences were found in this study between the group test IQ scores of dropouts and stay-ins. The dropout scores deviated from the stay-in scores in the expected direction, averaging about five points lower. Hypothesis 3. Age at Time of Testing. Pupils who later dropped out of school were expected to be older at the time of testing than were pupils who remained in school. (Finding: 3.0 accepted.) Dropouts, on the average, leave school during the tenth or eleventh grade at an age of seventeen years. Boy dropouts are usually reported to be older by a year or so than are girl dropouts. Since dropouts in this study were on the average six months older than stay-ins, this 315 hypothesis Is consistent with the findings. Dropouts aver aged sixteen years four months, and stay-ins fifteen years ten months at the time of testing. Age differences between dropouts and stay-ins were statistically significant for boys and girls in the tenth grade, and for eleventh grade boys. Boys and girls differed in age only in the eleventh grade sample. Although the dropout boys were six to seven months older than the girls, differences, for the most part, were not significant. Leaving age averaged sixteen years ten months for boys and sixteen years five months for girls. The findings of this study support the results of previous research; however, comparability is lacking, in that the factors of guidance, counseling, and child welfare services which directly influence persistence in school cannot be evaluated from one study to another; in fact, great differences in pupil personnel practices exist between schools of the same district. Hypothesis 4. Index of Reading Achievement. Dropouts were expected to receive lower read ing index scores than stay-ins. (Finding: 4.6 accepted.) Perhaps more than any other single estimate of academic success, an index of reading ability has been repeatedly identified as a convincing correlate. Dropouts (averaging below the tenth percentile of the standardizing 316 population) differed significantly (p «= .01) from stay-ins who scored at about the ninetieth percentile. Differences between grade levels and between boys and girls were not evident, except that tenth grade girls rated well above (p *= .05) tenth grade boys. Hypothesis 5. Total Number of Absences. It was expected that dropouts would be absent from school a greater number of times than would"" stay-inst (Finding: 5.0 'accepted'.')' Excessive absence from school was one of the earli est characteristics to be correlated with early school leaving. Reasons given by the student and by school per sonnel are important considerations in efforts to amelio rate the generalized dissatisfaction that pupils can only express through school avoidance. Regardless of the reasons given, pupils who leave high school before gradua tion almost always establish an impressive record of absences. This study found that dropouts were absent from school, regardless of reasons given, nearly three times (an average of 33 days) as often as were the stay-ins who averaged ten days (p => .01) . Hypothesis 6. Total Number of Days Enrolled. It was expected that dropouts would be enroTTed in school for fewer days during the school year than would stay-ins. (binding; H^ 6.0 accepted.) The dropout was enrolled for about one-half as many 3X7 days (an average of 95 days) as the stay-ins, who averaged 177 days (p = .01). Since this study concerned only those pupils who were enrolled at the time of testing, late enrollees and transfers into school after the test date were not considered. When the number of days of absence was corrected with respect to the number of days enrolled, the difference between dropouts and stay-ins in percentage of time spent attending school was made even more evident. Dropouts were out of school from 26 to 34 per cent of the time, whereas stay-ins were absent from 5 to 6 per cent of the days enrolled in school. Statistical Design Responses to the Meaning of Words Test were treated in a 2x2x2 general linear factorial analysis of covariance computer based program. Group test IQ was selected as the covariate. When group test IQ scores were treated in an analysis of variance program to identify any significant interaction effects, none was noted. This finding substan tiated the validity of utilizing the group test IQ score as an appropriate covariate. An analysis of variance of the factor scale ratings and group test IQ scores revealed eight concepts rated on the evaluative factor scales, six concepts rated on the potency factor scales and ten concepts rated on the activ ity factor scales that were associated. Because recent 318 research (Rosenthal, 1965) has indicated that IQ score is correlated with certain factor scores, it was expected that further analysis of these data might be a productive effort. The differences in average concept ratings reported here are corrected for disparity in the group test IQ score thereby eliminating this factor from the estimate of vari ance . Null Hypotheses Advanced for the Mean Summed Factor Scores Hypothesis 7. School Experiences. The hypothesis of no difference between dropout and stay-in In attitudes toward school experiences is rejected. Five stimulus concepts were selected to reflect the attitudes of the pupil toward his total school experience. Attitudes, expressed as opinions, were forced into a nine- scale, seven-step response mode that offered limited free dom and variability to sample the feelings of the pupils. The four rather concrete concepts of “teacher," "grades," "school," and "reading" consistently differen tiated dropouts from stay-ins. As might be anticipated, the stay-ins rated these concepts more favorably than did the dropouts. The concept "ideal teacher" was less effec tive as a discriminator; however, girls rated the ideal teacher more favorably (positive evaluation) and as more 319 active (active, smart, fast) than did, the boys. At the same time, boys regarded the ideal teacher as stronger, larger and heavier (i.e., more potent) than did girls. These findings are consistent with what one would expect. The pupil who is predisposed to early school leaving has been reported throughout the literature as being dissatisfied, discouraged and generally unhappy with the prospect of attending school. Hypothesis 8. Family Relations. The hypothesis of no difference between dropout and stay-in in expressed attitude toward family relationships is rejected. Clinical studies and more intensive individualized research projects have been able to probe attitudes of the individual toward his family environment. The sensitivity to "testing” in the geographic area where this research was conducted demanded that the examiner select concepts with less emotional involvement than would be associated with such items as "father," "mother," "brother," "step-father" and similar definitive relationships. As it was, the generalized concepts of "home" and "parents" were rated more favorably by stay-ins. Dropouts tended to devalue these concepts, rating them either as of neutral or of less positive value. The concept "ideal parent" did not distinguish between dropout and stay-in groups; however, girls rated 320 the concept as fairer and more active than did boys. Hypothesis 9. Social Interactions. The hypothesis of no difference between drop out and stay-in in attitude toward social situa tions is rejected on the basis of the findings. Four stimulus concepts were selected to explore feelings about peer relations and relations with other individuals. Again, somewhat vague and indefinite concepts had to be agreed upon to avoid the tensions that arise from the inclusion of terras that might be considered too per sonal. The concepts "most people," "classmates," and "grownups" were rated differentially by dropouts and stay- ins. Dropouts and stay-ins, alike, rated their best friends favorably. Girls tended to rate their best friends more favorably than did boys, but boys rated their best friends as stronger, larger and heavier than did the girls. These are not surprising results, as one would expect a person to rate his best friends favorably. The concept "classmates" was rated favorably by stay-ins, and perceived quite negatively by dropouts. The differences in rating contributed to a large statistical differential (p = .01). This concept might well be included in any replication of this study. Hypothesis 10. Self Concept. The hypothesis of no difference between dropout and stay-in in expressed attitude regarding the self image is rejected" The image a person develops regarding himself in his environment is critically important to effective per formance. One who develops a sense of accomplishment, of well-being and satisfaction will perform any task more efficiently than the person who has acquired a negative attitude toward himself. Universally, the high school student who has left school before graduation has developed a less than satisfactory self image. Assessments by direct and indirect means have resulted in the same findings. Dropouts and stay-ins in this study rated them selves consistently according to different patterns, each pattern so distinct that a favorably-expressed self concept usually characterized a stay-in, with little chance for error in judgment. Both dropouts and stay-ins exhibited a positive awareness of their own ability to perform success fully in the classroom, the stay-ins consistently rating their performance higher and as stronger than did the drop outs. One might adduce from this that dropouts probably had a rather realistic perception of their level of per formance, as seen by themselves and as they felt that others perceived it. Changes in self concept are undoubt edly essential before one can expect the potential dropout to persist to high school graduation. 322 Hypothesis 11. Authority Relationships. The hypothesis of no difference in attitude toward authority relations is rejected Reactions to the concepts of "authority," "rules," and "punishment" were as one might have expected. Pupils who frequently and continuously invite punishment from authority figures would be inclined to regard such concepts with less favor than would those who do not suffer such experience. In this study, dropouts rated the concepts related to authority significantly lower than did stay-ins. Ratings given to these concepts were sufficiently differen tiated to justify the inclusion of this type of item in subsequent research. The hypothesis is not tenable as stated. Hypothesis 12. Goal Orientation. The hypothesis of no difference in expressed attitude toward goal orientation is rejected. The divergent aspirations of dropouts and stay-ins were clearly revealed. Dropouts were distinctly more interested in "a job" and "quitting school" than were stay- ins. Stay-ins, correspondingly, responded to the concepts of "college" and "graduating" more favorably. Ratings of the concept "future" by dropouts were decidedly negative when compared with the ratings given by stay-ins. The dropout perceived the future as "unpleasant," "sour" and as "weak" and "unfair." Conversely, the stay-in 323 expressed a strong, optimistic attitude toward the future. There was little indecision among dropouts or stay-ins about their immediate wishes for the future. Hypothesis 13. Moral and Social Values. The hypothesis of no difference in attitude toward moral and social values is rejected- ! Six stimulus concepts were selected as representa tive of some of the more vague, less well-defined attitudes of the dropout. A comparison of the responses of the drop out with those of the stay-in indicated that the early school leaver did, in fact, regard the concepts of "trying hard," "something important," and "success" as less valu able than did the stay-in. Attitudes toward the moral issue of honesty and integrity were evaluated through responses of the two groups to the concept of "cheating." Whereas the stay-in rated this concept neutrally or slightly positively, the dropout expressed the attitude that cheating was an acceptable mode of behavior. Group Test IQ Score as a Correlate of the Dependent Variables Certain responses to the Meaning of Words Test were correlated with group test IQ score in frequencies greater than can be attributed to chance alone (Table 25). Three evaluative factor scores (reading, ideal teacher, and success) correlated at the .05 level of confidence; and five other factors (my school ability, college, graduating, 324 quitting school, and cheating) correlated at the .01 level. Three concepts (classmates, authority, and college) correlated with potency factor scores at the .05 level of confidence; and three others (my school ability, how my class sees me, and quitting school) correlated at the .01 level of confidence. Five concepts (grades, authority, graduating, cheating, and something easy) correlated at the .05 level of confidence with activity factor scores; and' five other concepts (reading, ideal teacher, my school ability, col lege, and quitting school) correlated at the .01 level of confidence with group test IQ scores. Location of the differences between means would require the grouping of subjects according to specified mean IQ ranges and treating them according to the procedure of Duncan's New Multiple Range Test (Kramer's modification for unequal N's) as given in Edwards (1960:137). Discussion The findings of this study can be generalized only to a limited extent to the high school population from which the sample was drawn. These responses are the rat ings of those pupils who completed usable test protocols. An evaluation of the responses of the other dropouts who failed to complete the Meaning of Words Test, or who mutilated the test materials could conceivably have provided different response patterns. The seven descriptive characteristics were not inconsistent with previous research. The dropouts of this sample represented only the tenth and eleventh grades of three schools in one suburban high school district. Had the twelfth grade classes participated in the study, the average dropout age would have been higher. Attitudes expressed as concept ratings on this test instrument were sufficiently varied to differentiate the dropouts and the stay-ins at the .01 level of confidence, and in many instances to differentiate boys and girls. Grade levels were not differentiated. Dropouts expressed essentially negative or unfavor able attitudes toward school experiences, family relation ships, perception of social interaction, the self image, authority relationships and middle-class moral and social values. RECOMMENDATIONS This investigation gives rise to several considera tions, with respect to the implementation of changes in attitude on the part of high school dropouts and potential dropouts. At the time a student leaves school, the deci sion is already firmly established and little can be done to restrain him. Changing attitudes which have developed throughout the life of the student will require a 326 multidimensional approach. Carefully conceived programs to develop areas of ability and to avoid areas of sensitivity are essential, for such programs must enable the student to experience success in school and in social interaction; most importantly, perhaps, he must develop a favorable self image. School Experiences 1. Pupils who express a marked dislike for school, who feel that it is unpleasant and unfair, usually have experienced prolonged failure and alienation in the class room. Traditional stereotyped classroom experiences are inappropriate for the potential dropout. These pupils require small-group, perhaps individual, learning experi ences in which they can discover the pleasures of success ful learning. Only then will their attitudes improve toward the learning experience. 2. Teachers selected for these guidance-oriented instructional programs must be sensitive to the expressed, and sometimes unexpressed, attitudes and feelings of students. Family Relationships 3. The attitudes of parents and siblings toward education exert a powerful influence upon the pupil's decision to continue in school. It is recommended that parents be involved in a program of orientation to 327 education with their dropout children. 4,. Family counseling and family involvement in satisfying home-school experiences is recommended to reduce tension-producing situations. Social Interaction 5. The dropout is often described as an isolate, withdrawn from his peers. Responses to this test instru ment confirm this characterization. Recommendations for counteraction would consider involvement of the potential dropout in at least one activity through which he can establish favorable interpersonal relationships. 6. Clubs and other extra-curricular school activ ities are not always the answer. Student-to-student contacts are to be desired in instances where the potential dropout is shy and insecure in group settings. A single friend on campus can be a positive factor in a student's decision to remain in school. Self Concept 7. Chronic frustration, defeat and isolation have created within the dropout a sense of worthlessness and a sense of dissatisfaction with himself. Therapeutic coun seling together with specialized educational experiences are recommended. The establishment of a favorable self concept will accompany other favorable changes in attitude toward others and toward one's environment. 328 Authority Relations 8. Dropouts appear to be unusually sensitive to authority figures, rules and regulations. Positive approaches to discipline rather than punitive measures are recommended. Counseling to develop better self-control will reduce the need for external controls. 9. Family counseling is recommended when there are indications that the family authority relationships are important factors in the pupil's school adjustment. Goal Orientation 10. Consistent with previous recommendations, a positive orientation toward the future requires effective counseling and small instructional groups wherein the individual has an opportunity to explore possible goals with respect to his own interests. 11. It is recommended that the dropout and the potential dropout be afforded an opportunity to explore the world of work through school-associated work situations. Part- or full-time employment combined with a program of continuation education is recommended as an effective means of assisting a pupil define his goals. Moral and Social Values 12. The establishment of new value systems may be accomplished by emulating another person's behavior or by behaving in such a manner as to be accepted by a desired 329 group. Value systems Inconsistent with the dominant value system can be replaced by precept and example. Counseling, discussion in class, and peer group reactions can effec tively promote positive changes in one's system of values. RECOMMENDATIONS FOR FURTHER STUDY 1. The concept that attitudes are learned is accepted throughout the literature. Therefore, special classroom environments might be established in order to create or to provide for the development of new attitudes toward the self. 2. Throughout this study one is impressed that self concept is an integral factor in nearly every area of investigation. Further study of the dropout is indicated in which an evaluation of the self concept is taken as the covariant in a statistical design similar to the one used in this study. 3. Restriction of the sample may have Intensified these findings. A sample more representative of the general population and including all high school grades would be worthy of study. 4. Longitudinal studies are necessary to verify the predictive validity of any test procedure. High school pupils might be studied over a period of at least four years to determine changes in attitude with maturity, and the effects of attempts at remediation. 330 5. Studies concerning elements of creativity and the dropout are conspicuously absent in the literature. Research on the various dimensions of creativity might prove worthwhile. BIBLIOGRAPHY BIBLIOGRAPHY Abbott, John L. "An Enrollment and Drop-out Study of Six Representative Secondary Schools in Lot Angeles." Unpublished doctoral dissertation, University of Southern California, 1939. Allen, Charles M. Combating the Dropout Problem. Chicago: Science Research Associates, Inc., 1956. Anderson, J. E., and D. B. Harris. A Survey of Children’s Adjustment Over Time: A Report to the People of Nobles County. Minneapolis: Institute of Child Development and Welfare, University of Minnesota, 1959. Argyris, Chris. Personality and Organization— The Conflict Between System and the Individual. New York: Harper and Brothers, 1957. Aseltine, John. "A Comparative Study of Certain General Characteristics, School Experiences, and Later Life Adjustments of Selected High School Drop-outs." Unpublished doctoral dissertation, University of Southern California, 1941. Ayer, Fred C. "Studies in Administrative Research: Progress of Pupils." Department of Research Bulletin No. 1. 1924, pp. 94-ITT Ayers, Leonard P. "Laggards in Our Schools," Clearing House. 36:530, 1909. Beckstead, Rex C. "A Comparative Case Study of Fifty Male High School Graduates and Fifty Male High School Drop outs with Regard to Employment, Social Attitudes, after School Behavior, and the Major Problems of Each." Unpublished Master1s thesis, University of Southern California, 1940. Berston, H. M. "The School Dropout Problem," Clearing House. 35:4:207-211, 1960. Bienstock, Herbert. Factbook on the School Dropout in the World of Work. New York: U. S. Department of labor, Bureau o£ Labor Statistics, 1961. 332 333 Bienstock, Herbert. "Realities of the Job Market for the High School Dropout." New York: U. S. Department of Labor, Bureau of Labor Statistics, May 1963. (Mimeo graphed .) Bledsoe, Joseph C. "An Investigation of Six Correlates of Student Withdrawal from High School," Journal of Educa tional Research. 53:5, September 1959. Bloch, H. A., and A. Niederhoffer. The Gang. New York: The Philosophical Library, 1958. Bowden, A. 0. "Investigation of Some of the Causes of Elimination from High Schools in South Dakota," School and Society. 6:146:447-450, October 13, 1917. Bower, Eli. A Process for Early Identification of Emotion ally Disturbed Children, 27:6:107-108. Sacramento: California State Department of Education, 1958. Bowman, P. H., and C. V. Matthews. Motivations of Youth for Leaving School. U.S. Department of Health, Educa tion, and Welfare, Office of Education, Cooperative Research Program, Project No. 200. Quincy, 111.: University of Chicago, Quincy Youth Development Project, September 1960. Brooks, E. E., J. Buri, E. A. Byrne, and M. C. Hudson. "Socioeconomic Factors, Parental Attitudes, and School Attendance." Social Work. 7:4:103-108, 1962. California Test Bureau. Summary of Investigations No. 3: California Test of Mental Maturity.' Division of Professional Services. Los Angeles: California Test Bureau, 1956. Cannaday, Frank W. A Study of Dropouts in Arkansas Public Elementary and Secondary Schools 1940-41— 1960-61. Little Rock: State Department of Education, 1962. Caplow, Theodore. The Sociology of Work. Minneapolis: University of Minnesota Press, 19547 Cassel, R. N., and J. C. Coleman. "A Critical Examination of the School Dropout, Reluctant Learner, and Abler Non-college Student Problem," Bulletin of the National Association of Secondary School PrincipalsT 46:277:60- 65, ’ m i :---- ----- — Cattell, R. B. Description and Measurement of Personality. New York: World Book Co., 1946. 334 Charlap-Hyman, Arthur. "Self Concept and Social Class Correlates of Contrasting Behavioral Subcultures among Ninth Grade Boys." Unpublished doctoral dissertation, University of Southern California, 1964. Cohen, A. K. Delinquent Bovs, the Culture of the Gang. Glencoe, 111.: The Free Press, 1955. Cohen, Eli. "The Drop-out Problem--A Growing Educational Concern Today," National Association of Secondary School Principals Bulletin. 264:279. April 1961. _______. "Youth Opportunities--Unlimited?" Presentation made before the California Association of Supervisors of Child Welfare and Attendance, Southern Section, March 2, 1964, Los Angeles. Combs, Arthur W., and Donald Snygg. Individual Behavior. a Perceptual Approach to Behavior™ New York: Harper and Brothers, 1959. Cook, Edward S., Jr. "An Analysis of Factors Related to Withdrawal from High School Prior to Graduation," Journal of Educational Research, 50:191-196. November t s w . ------------------------ -------- Coombs, C. H. "Psychological Scaling without a Unit of Measurement," Psychological Review, 57:145-158, 1950. _______ . "A Theory of Psychological Scaling," Engineering Research Institute, Bulletin No. 34. Ann Arbor: University of Michigan tress, 1952. Cooper, Sophia. "Employment of June, 1960 High School Graduates," Special Labor Report No. 15, Monthly Labor Review, May 1961, pp. 463-470. Coplein, Leonard E. "Techniques for Study of Dropouts," Clearing House. 36:526-530, 1962. Cottle, William C. "The School Interest Inventory," Psychological Reports. 9:66, 1961. _______. The School Interest Inventory: Preliminary Manual" Boston: HoughtonMif£lin Co., 1963. Cronbach, L. J. "Processes Affecting Scores on 'Under standing of Others' and 'Assumed Similarity,'" Psychological Bulletin. 52:177-194, 1955. 335 Cronbach, L. J. "Proposals Leading to Analytic Treatment of Social Perception Scores." In R. Tagiuri and L. Petrullo (eds.), Person Perception and Interpersonal Behavior. Stanford, Calif.: Stanford University Press. T 5 35 " : ------------------ y _______ . Essentials of Psychological Testing. 2d ed. New York: riarper and Brothers, 19&0. , and G. C. Gleser. "Assessing Similarities between Profiles," Psychological Bulletin. 50:456-474, 1953. Ctramings, Robert E. "Index of Adjustment and Values," Journal of Educational Research. 53:5:197-198, January 19 f > 0. Davis, Allison. Social-Class Influences upon Learning. Cambridge, Mass.: Harvard University Press, 1950. Davis, A., and R. J. Havighurst. "Social Class and Color Differences in Child Rearing," American Sociological Review. 11:698-710, December 1941. ‘ Davis, Donald A. "An Experimental Study of Potential Drop outs," Personnel and Guidance Journal, 60:9:799-802. 1962. ■ Delaney, John F. "That Vacant High School Seat," American School Board Journal, 21:22-23, November 1950. Dewey, John. "The Supreme Intellectual Obligation," Science Education. 18:1-4, February 1934. Dice, Kathryn L. "Unmet Needs of High School Students," Education Leadership. 26:172, December 1958. Dillon, Harold J. Early School Leavers: A Major Educa tional Problem. Publication No. 401. New York: National Child Labor Committee, October 1949. Dresher, Richard H. "Factors in Voluntary Drop-outs," Personnel and Guidance Journal, 32:287-289. January T 954T -------- ------------ ---------------- Drews, E. M., and J. E. Teahan. "Parental Attitudes and Academic Achievement," Journal of Clinical Psychology. 13:328-332, 1957. Eaton, H. T. "The Scholarship of Pupils Who Left School," School and Society, 16:221-222, August 19, 1922. 336 Eckert, Ruth, and T. 0. Marshall. When Youth Leave School: Report of the Regents Inquiry. New York: McGraw-Hill Book Co., Inc., 1938. Edwards, Allen L. Techniques of Attitude Scale Construc tion. New York: Appleton-Century-Crofts, tnc., 1957. _______ . Experimental Design in Psychological Research. New York: Holt, Rinehart and Winston, i960. Elliott, Delbert. "The Dropout.*' Address given before the California Association of Supervisors of Child Welfare and Attendance, San Diego, California, April 19, 1963. Ellis, A. C. "The Percentage of Boys Who Leave the High School and the Reasons Therefore," NEA Proceedings, 1903, pp. 792-798. ------------- Endler, N. A. "Changes in Meaning during Psychotherapy as Measured by the Semantic Differential," Journal of Counseling Psychology. 8:105-111, 1961. Engel, Mary. "The Stability of the Self-concept in Adolescence," Journal of Abnormal Social Psychology, 58:211-215, 19¥T. ---------- ------ English, H. B., and A. C. English. A Comprehensive Dic tionary of Psychological and Psychoanalytical Terms. New York: Longmans, Green and Co.” 1958. Epps, Margaret W., and W. C. Cottle. "Further Validation of a Dropout Scale," Vocational Guidance Quarterly, 7:90-93, Winter 1958-357 Ewing, E. F. "Retardation and Elimination in the Public Schools," Educational Review, 46:252-272, October 1913. Faris, Ellsworth. "The Sociologist and the Educator," American Journal of Sociology, 33:796-801, January 1928. Fink, Donald D. "The Efficiency of Certain Criteria in Predicting School Dropout." Unpublished doctoral dissertation, Michigan State University, 1962. First, Joan M. "Dropouts," Michigan Education Journal, 38:6:250-253, November 1960. Flynn, Henry J., Norma Saunders, and Robert Hoppock. "A Course for Drop-outs," Clearing House, 28:486, 1954. 337 Foster, Emery M. "Survival Rates of Pupils." Circular No. 193. Washington, D.C.: U. S. Office of Education, Federal Security Agency, 1943. Fraser, Elizabeth. Home Environment and the School., Publications of the Scottish Council for Research in Education XLIII. London: University of London Press, Ltd., 1959. Gekoski, N., and Solomon Schwartz. "Student Morality and Related Factors," Journal of Educational Research. 54:5 :192-194, January 1961. Gerberich, J. R. Measurement and Evaluation in the Modern School. New York: Ibavid McKay Co7, Inc., 1962. Glasser, William. "Reality Therapy." Presentation before the California Association of Supervisors of Child Welfare and Attendance, Southern Section, May 8, 1964. Goldberg, Miriam L. A Report on Recent Research in the Field of the Academically Talentetfl Horace Mann- Lincoln Institute of School Experimentation, Fall 1958. Gordon, C. Wayne. The Social System of the High School. Glencoe, 111.: The Free' Piess, 1957.------ --------- Gragg, William L. "School Leavers Can be Spotted in Junior High School," Clearing House. 25:72, October 1950. Green, Bert F. "Attitude Measurement." In G. Lindsey, Handbook of Social Psychology. Cambridge, Mass.: Addison-Wesley Publishing Co., Inc., 1954. Green, Joseph (ed.). "The Clearing House," Clearing House, 36:173, November 1961. Greene, Bert I. "Dropouts and the Elementary School," National Elementary School Principal. 42:2:52-54, November 1962. The Growth and Economic Stature of the South Coast Area of Los Angeles County. 1963. Los Angeles, Calif.: The ~ Security First National Bank, Research Department, 1963. Growth is Only Half the Success Story. Torrance, Calif.: Torrance City Chamber ol Commerce, January 1963. 338 Guttman, Louis. "The Problem of Attitude and Opinion Measurement." In S. A. Stouffer, tfeasurement Predic- tion. Princeton, N.J.: Princeton University Press. T93?L "The Basis for Scalogram Analysis." In S. A. Stouffer, Measurement of Prediction. Princeton, N.J.: Princeton University Press, 1950. Hand, Harold C. "Do School Costs Drive Out the Youth of the^Poor?" Progressive Education, 28:89-93, January Hecker, Stanley. "The Early School Leaver in Kentucky," Bulletin of the Bureau of School Services, College of Education, University of Kentucky, Lexington, 25:4, June 1953. Herrold, Kenneth F. "Personal and Environmental Factors Related to Individual Change." Research Project, State Department of Public Instruction, Raleigh, N.C., and U. S. Office of Education, Division of Vocational and Technical Education, Washington, D.C., 1964. Hightower, Howard W., et al. "Mystery of Elementary Drop out," PhiJOelta^^gpan, 38:2:62-64, November 1956. Hodges, Allen, and Bruce Balow. Learning Disabilities in Relation to Family Constellation, University of Minnesota, Psycho-Education Clinic, 55:1:41-42, September 1961. Holbeck, E. S. "Seven Ways to Help Prevent Dropout," Nation's Schools, 45:35-36, May 1950. Hollingshead, August B. Elmtown's Youth. New York: John Wiley and Sons, Inc., 1949. Horst, P. "A Method for Determining the Absolute Value of a Series of Stimulus Situations," Journal of Educa tional Psychology, 23:418-440, 193¥1 Hoyt, Kenneth B. "The Counselor and the Dropout," Clearing House. 36:515, 1962, Huber, Mildred. "Profile of a Dropout," California Educa- tion. 1:3:3-6, 1963. Hull, J. Dan, and H. H. Cummings. "How Fare American Youth in 1962?" School Life, 44:4:13, January-February 1962. 339 Jersild, A. T. "Self Understanding in Childhood and Adolescence,” American Psychologist, 6:122-126, 1951. Kahl, Joseph A. "Educational Aspiration of 'Common Man' Boys," Harvard Educational Review. 23:186-203, 1953. Kitch, Donald E. "Guide for a Follow-up Study of School Dropouts and Graduates," California Guidance Bulletin No. 13. page 9. SacramentoCalifornia State Depart- ment of Education, January 1950. Kline, E. J. "Significant Changes in the Curve of Elimina tion Since 1900," Journal of Educational Research, 26:608-616, April IV5T.--------------------------- Kvaraceus, W. C., and W. B. Miller. Delinquent Behavior: Culture and the Individual. Part 1. Washington, D.C.: National Education Association, 1959. La Forge, R. Some Properties of the D^ Statistic. ONR Contract NQtoR-1844(ll), Separate ‘ Technical Report 2, July 1961. La Masa, K. C., Jr. "Identifying the Potential High School Drop-out." Unpublished Master’s project, University of Southern California, 1963. Lanier, A. J. "A Study of Student Withdrawals," Journal of Educational Research, November 1949, pp. 207-2087 law, Frederick H. "The Age at Which Children Leave School," Education Review, 15:40-49, 1898. Layton, W. K. "Special Services for the Drop-out and the Potential Drop-out," The American Child. May 1952, p. 5. Lazarsfeld, P. F. "The Logic and Mathematical Foundation of Latent Structure Analysis." In S. A. Stouffer et al., Measurement and Prediction. Princeton, N.J.: "Princeton University Press, 1950. Lecky, Prescott. Self-Consistency: A Theory of Personal ity. New York: Island tress, 1945. ’ Lee, Beatrice C. School Dropouts. NEA Research Division, NEA Publications’ ! Washington, D.C.: The Association, April 1963. 340 Lee, Margaret B. T. ,fFactors Associated with School Reten tion in a Group of Predicted Drop-outs.** Unpublished doctoral dissertation, University of California, Berkeley, 1958. Lefever, D. W., F. J. Shanley, and G. Adams. "Preliminary Analysis of School Records and Behavior Data for Matched Groups of Aggressive, Well-adjusted and Under achieving Boys.** Research Report No. 8, Youth Studies Center, University of Southern California, 1961. (Mimeographed.) Lehman, C. 0. "Persistency of High School Students in Ohio, 1920-1924," Educational Research Bulletin. 5:50- 53, 59, February 3, 192b. Lichter, S. 0., E. B. Rapien, F. M. Seibert, and M. A. Sklansky. The Drop-outs. New York: The Free Press of Glencoe, 19 ”” Liddle, Gordon P. "Psychological Factors Involved in Drop ping Out of School," High School Journal. 45:276-280, April 1962. Linton, Ralph. The Cultural Background of Personality. New York: Appleton-Century-Crofts, Inc., 1945. Livingston, A. H. "High-school Graduates and Dropouts-- A New Look at a Persistent Problem,** School Review, 66:195-203, Summer 1958. Lombardi, Robert A. "A Drop-out Study at Burbank High School.'* Unpublished Master*s project, University of Southern California, 1960. Longstreth, L. E., Fred Shanely, and Roger Rice. "Evalua tion of a Work-Study Program for Potential Drop-outs.*' Research Paper No. 9, Youth Studies Center, University of Southern California, 1962. (Mimeographed.) Los Angeles City School Districts. "Transfer, Entrants, and Dropouts in Los Angeles City Secondary Schools, 1961-1962." Research Report No. 252, April 1963, Evaluation and Research Section, Los Angeles City School Districts, Los Angeles, California. . "The Dropout Who Returns to Graduate from Adult High School.*' Research Report No. 253, April 1963, Evaluation and Research Section, Los Angeles City School Districts, Los Angeles, California. 341 MacCorquodale, K., and P. E. Meehl. ”0n a Distinction between Hypothetical Constructs and Intervening Varia bles,” Psychological Review. 55:95-107, 1948. MacDonald, Dorothy P. ”0n Investigation of the Attitudes of Parents of Unsuccessful and Successful Readers,” Journal of Educational Research. 56:8:437-438. April m --------------------------------- Mannino, Furtune V. “Family Factors Related to School Persistence,” Journal of Educational Sociology. 35:193- 202, January 1962. Maryland State Department of Education. “Our Dropouts: The Maryland Cooperative Study of Dropouts.” Baltimore: June 1963. McCreary, W. H., and D. E. Kitch. "Now Hear Youth,” Bulletin of the California State Department of Educa tion. 22:9:69. October 1953. McMurrin, S. M. “Drop-out Problem,” Wisconsin Journal of Education. 94:17, October 1961. McNemar, Q. "Opinion-Attitude Methodology.” Psychological Bulletin. 43:289-374, 1946. Medinnus, Gene R. “The Development of a Parent Attitude toward Education Scale," Journal of Educational Research. 56:2:100-103, October 1962. _______ . "Parental Q-Sort Descriptions of Five-Year-Olds as Predictors of First Grade Adjustment,” Journal of Educational Research. 57:2:100-103, October 1963. Mehling, R. A. “A Simple Test for Measuring Intensity of Attitudes,” Public Opinion Quarterly. 23:576-578, 1959. Miller, Leonard M. "The Dropout: Schools Search for Clues to His Problem," School Life. 45:7, May 1963. Miller, W. B. "Lower Class Culture as a Generating Milieu of Gang Delinquency,” Journal of Social Issues, 14:3:5- 3.9, 1958. Ministry of Education. Early Leaving: A Report of the Central Advisory Council for Education (England). London: Her Majesty's Stationery Office, 1954. 342 Mitchell, J., and E. Clague. From School to Work. Washington, D.C.: U. S. Department of Labor, March 1960. Moscovici, Serge. "Attitudes and Opinions." In P. R. Farnsworth, Annual Review of Psychology. Vol. 14. Palo Alto, Calif.: Annual Reviews, Inc., 1963. Munoz, Rosalio. "Differences in School Behavior between Two Groups of Boys in a Los Angeles High School." Unpublished doctoral dissertation, University of Southern California, 1957. Nachman, L. R., R. F. Getson, and J. G. Odgers. Pilot Study of Ohio High School Drop Outs 1961-1962. Columbia, Ohio: State department of Education, February 1963. National Education Association of the United States. "High School Drop-Outs," NEA Research Bulletin. 38:11-14, February 1960. ________ "School Drop-outs." NEA Research Memorandum, Vo!. 36. Washington: NEA Research Division, August 1961. "School Dropouts: a Symposium," NEA Journal. 5T?5:51-59, May 1962. ----------- New York State Education Department. Improvement of Hold ing Power through a Continuous Study of Youth Tn 'School. Albany, N.Y.: University of the State of New Vork Ihress, 1952. Osgood, C. E. "Report on Development and Application of the Semantic Differential." Unpublished paper, Insti tute of Communications Research and Department of Psychology, University of Illinois, 1954, and Z. Luria. "A Blind Analysis of a Case of Multiple Personality Using the Semantic Differential," Journal of Abnormal Social Psychology, 49:579-591, 1954. _______ , George J. Suci, and P. H. Tannenbaum. The Meas- urement of Meaning. Urbana: University of Illinois Res~T937':----- , and P, H. Tannenbaum. "The Principle of Congruity in the Prediction of Attitude Change," Psychological Review, 62:42-55, 1955. 343 Palmer, Emily G. Pupils Who Leave School, Berkeley, Calif.: Division of Vocational Education, 1930. Penty, Ruth C. Reading Ability and High School Drop-outs. New York: Bureau of Publications, Teachers College, Columbia University, 1956. Peters, J., S. "Socio-economic Egocentrism in Delinquents and Non-delinquents," Purdue University Studies in Higher Education. 85:1-21, 1957. Phillips, Frank M. Statistical Summary of Education 1927- 1928. U. S. Department of Interior, Office of Educa tion, Bulletin No. 3. Washington, D.C.: Government Printing Office, 1930. Pickwell, F. G., and A. F. Winkelbelch. "Elimination from the Public Secondary Schools of the United States," School Review. 26:18-24, January 1918. Plott, Curtis E. "Identification of Potential Dropouts at Mulholland Junior High School." Unpublished Master’s project, University of Southern California, 1963. Plunkett, M. L. Youth--Its Employment and Educational Out look. Studies-in Unemployment, U. S. Senate Committee. Printed by the 86th Congress, February 2, 1960. Pond, Frederick L. "Pennsylvania Study of Drop Outs and the Curriculum," Bulletin of the National Association of Secondary School Principals, 37:81-82, March 1£53. Putnam, J. F. "Information about Dropouts: Terms and Computations." School Life. 45:7:24-33, May 1963. Queen, Stuart A. "Foreword." In C. Wayne Gordon, The Social System of the High School. Glencoe, 111.: The Free Press, 1957. Reed, H. J. "The Self in a School Setting." Paper pre sented before ASCD Convention, Las Vegas, March 1962. (Mimeographed.) Remmers, H. H. "Rating Methods in Research on Teaching." In N. L. Gage, Handbook of Research on Teaching. Chicago: Rand McNally and Co.,"T963. (ed.). (1) Studies in Attitudes. Studies in Higher Education No. 26. Lafayette, Ind.: Purdue Uni versity, 1934. (2) Further Studies in Attitudes. Series II. Studies in Higher Education No. 31. 344 Lafayette, Ind.: Purdue University, 1936. (3) Further Studies in Attitudes. Series III. Studies in Higher Education No. 34. Lafayette, Ind.: Purdue University, 1938. Reynolds, R. W. (ed.). "Elementary Education," California Education. 1:14, March 1964. Richards, Kelly C., II. "A Three Year Study of School Withdrawals in a New Orleans, Louisiana Junior High School." Unpublished Master's project, University of Southern California, 1963. Riendeau, A. J. "Facing up to the Dropout Problem," Clearing House. 36:523, 1962. Ristow, L. W. "A Test for Scientific Attitude in Consumer Buying." Unpublished Master's thesis, University of Southern California, 1948. . Dropouts. Report submitted to the Staff, Los Angeles County Superintendent of Schools, Los Angeles. Available in the Files, Division of Research and Guidance, 1964. (Typewritten.) Rogers, C. R. "The Organization of Personality," American Psychologist. 2:358-368, 1947. "Perceiving, Behaving, Becoming, a New Focus for Education," ASCD 1962 Yearbook, A. W. Combs, Chairman. Washington, D.C.: Association for Supervision and Curriculum Development, 1962, Rosander, A. C. "The Economic Stratification of Youth and Its Social Consequences," Journal of Educational Research. 32:592-604, April 1939. Rosen, Bernard. "The Achievement Syndrome," American Sociological Review, 21:203-211, April 1956. Rosenfeld, Carl. "The Employment of Students, October 1961," Monthly Labor Review, 85:6:635-642, June 1962. Rosenthal, A. 0. "A Comparative Analysis of the Perceived Meaning of Selected Variables for Achieving and Under achieving High School Students as Measured by the Semantic Differential." Unpublished doctoral disserta tion, University of Southern California, 1965. 345 Rowan, Helen (ed.). “Creativity,*1 Carnegie Corporation of New York Quarterly. 9:3:7, July 1961. Rubenstein, H. “Psycholinguistics.** In P. R. Farnsworth, Annual Review of Psychology. Vol. II. Palo Alto, Calif.: Annual Reviews, Inc., I960. Sando, Rudolph F. “A Comparative Study of Early School Leavers." Unpublished doctoral dissertation, Univer sity of California, Berkeley, 1952. Savitzky, C. “Introduction to a Program for Possible Dropouts," High Points. 43:5-15, November 1961. Schiffman, Jacob. “Employment of High School Graduates and Dropouts in 1961.*' Washington, D.C.: U. S. Department of Labor, Bureau of Labor Statistics, May 1962. Schneider, L., and S. Lysgaard. "The Deferred Gratifica tion Pattern,** American Sociological Review. 18:142- 149, 1953. “School and Early Employment Experience of Youth: A Report on Seven Communities,” U. S. Department of Labor Bulletin No. 1277. August i960. Schramm, W. "Mass Communication." In P. R. Farnsworth, Annual Review of Psychology. Vol. XIII. Palo Alto, Calif.: Annual Reviews, Inc., 1962. Schreiber, Daniel. “The School Dropout— Fugitive from Failure," Bulletin of the National Association of Secondary School Principals, 467274:237, fefay 1962. (ed.). The School Dropout. Washington, D.C.: National Education Association, 1964. Schroder, H. M., and D. E. Hunt. "Failure-Avoidance in Situational Interpretation and Problem Solving,** Psychological Monographs, Vol. 71, No. 3. Whole No. 4jsv p": fc, ■fesr; Sells, S. B., and D. K. Trites. "Attitudes," Encyclopedia of Educational Research. 3rd ed. Ed. C. W. Harris. New York: The Macmillan Co., 1960. Seymour, H, D. "The Characteristics of Pupils Who Leave School Early." Unpublished doctoral dissertation, Harvard University, 1940. 346 Shanley, Fred J., William R. Larson, H. L. Myerhoff, Roger E. Rice. D. Welty Lefever, and Iangdon E. Longstreth. ''Comparative Study of Factors Influencing the School Adjustment of Adolescents." Research Paper No. 2, Youth Studies Center, University of Southern California, 1961. (Mimeographed.) Shaw, M. C. "Need Achievement Scales as Predictors of Academic Success," Journal of Educational Psychology. 52:283-285, 1961. --------------------- Sheerer, E. T. "An Analysis of the Relationship between Acceptance of Self and Acceptance of and Respect of Others in Ten Counseling Cases," Journal of Consulting Psychology. 13:169-175, 1949. Slater, J. M. "Influence on Students1 Perception and Per sistence in the Undergraduate College," Journal of Educational Research. 54:1:3-8, September i960. Smith, Harry P. "Syracuse Youth Who Did Not Graduate." Syracuse Board of Education, 1950. (Mimeographed.) Smith, R. G. "Validation of a Semantic Differential," Speech Monographs, 30:50-55, March 1963. Snepp, Daniel W. "Can We Salvage the Dropouts?" Clearing House, 31:49-54, September 1956. Sobkow, Edward A. "A Study of the Dropout Problem." Unpublished Master's project, University of Southern California, 1964. Social Dynamite. The Report of the Conference on Unem- ployed Out-of“School Youth in Urban Areas, May 24-26, 1961. Washington, D.C.: National Committee for Children and Youth, 1961. Sorenson, Mourits A. "Low Ability Dropouts versus Low Ability Graduates," Personnel and Guidance Journal. 39:144-145, October I960. Steadman, E. R. "15 Who Were Promoted," Elementary School Journal. 59:271-276, 1959. Stern, G. G. "Measuring Noncognitive Variables in Research on Teaching." In N. L. Gage, Handbook of Research on Teaching. Chicago: Rand McNally and Co., 1963. 347 Stevens, S. S. ^Mathematics, Measurement, and Psycho physics.” In S. S. Stevens (ed.), Handbook of Experimental Psychology. New York: John Wiley and Sons, Inc., 1951. Stice, Glenn. Talent Losses Before High School Graduation. Princeton, N.J1 .: Educational testing Service, January 1960. Strang, Ruth. The Adolescent Views Himself. New York: McGraw-Hi11 Book Co., Inc., 1957. Suci, G. J. ”A Multidimensional Analysis of Social Atti tudes with Reference to Ethnocentrism.” Unpublished doctoral dissertation, University of Illinois, 1952. ”A Comparison of Semantic Structures in American Southwest Culture Groups,” Journal of Abnormal and Social Psychology. 61:25-30, 1960. Swan, Leonard. ”Early Recognition of School Dropouts Helps Canton,” School Management. 5:77-79, November 1961. Tannenbaum, A. J. Adolescent Attitudes toward Academic Brilliance. New York: Bureau of Publications, Teachers College, Columbia University, 1962. Taylor, Florence. Why Stay in School? Chicago: Science Research Association, 1949. ~ Thomas, Robert J. ”An Empirical Study of High School Drop outs in Regard to Ten Possible Related Factors,” Journal of Educational Sociology. 28:11-18, September 1954. Thorndike, E. L. The Elimination Of Pupils from School. Bulletin No. 4, 1 9 0 Department of the Interior, Bureau of Education. Washington, D.C.: Government Printing Office, 1908. Thorndike, R., and E. Hagen. Measurement and Eyaluation in Psychology and Education. New York: John Wiley and Soisrtn?., "i933"r — Thurstone, L. L. Multiple Factor Analysis. Chicago: University of Chicago Press, 1947. Measurement of Values. Chicago: University of — Chicago Press, 1959. ’ Thurstone, L. L., and E. J. Chave. The Measurement of Attitude. Chicago: University of Chicago Press, 1929. Tompkins, E. E. "How Can the School Reduce the Number of Early School Leavers?" National Association of Second ary School Principals Bulletin! 35:307-310, March 1951. Topetzes, John, and Johp Ivanoff. "Drop-out; How Can the School Help Him?" Catholic School Journal, 62:35-36, February 1962. Torgerson, W. S. "Scaling and Test Theory." In P. R. Farnsworth, Annual Review of Psychology. Vol. XII. Palo Alto, Calif.: Annual Reviews, Inc., 1961. Triandis, H. C. "Differential Perception of Certain Jobs and People by Managers, Clerks, and Workers in Indus try," Journal of Applied Psychology. 43:221-225, 1959. Tunnel, James W. "What Research Says about the School Dropout Problem." Lubbock: West Texas School Study Council, 1963. Ullmann, Charles A. "Teachers, Peers and Tests as Predic tors of Adiustment," Journal of Educational Psychology, 48:257-267, 1957. U. S. Department of Commerce, Bureau of the Census. "Current Population Reports, Population Estimates. Illustrative Projections to 1980 of School and College Enrollment in the United States," Series P-25, No. 232, June 22, 1961. _______ . "Current Population Reports, Population Charac teristics. School Enrollment: October 1961," Series P-20, No. 117, July 11, 1962. Newspaper Release. Los Angeles Times, March 20, XM4. Van Dyke, L. A., and K. B. Hoyt. The Dropout Problem in Iowa High Schools. Iowa City: State University of Iowa, 1958. Vogel, Anita. "How to Check Dropouts: Six Case Histories," School Management, 5:77-78, November 1961. Warner, W. L., M. Meeker, and K. Eells. Social Class in America. New York: Harper and Brothers, I960. — 349 Warren, D. "Who are Most Likely to Drop Out?" School Science and Mathematics. March 1954, p. 185. "Why Do Boys and Girls Drop Out of School and What Can We Do about It?” (Editorial), School Life, 32:136-137, June 1950. Williams, P. V. "School Dropouts," NEA Journal. 52:2:11- 12, February 1963. Wind, Kate. "A Comparative Personnel Study of Retained and Eliminated Pupils in Junior High Schools." Unpublished Master’s thesis, University of Minnesota, 1931. Wirt, R. D., and A. L. Wirt. "Psychotherapeutic Processes." In Q. McNemar, Annual Review of Psychol- ogy. Vol. XIV. Palo Alto, Calif.: Annual Reviews, Inc., 1963. Wirtz, Willard. "National Trends and Goals for Youth." Governor's Conference on Youth Employment, Washington, D.C., June 1962. Wolfbein, S. L. The Transition from School to Work: A Study of the School Leaver. Washington, D.C.: U. S. Department of Labor, Bureau of Labor Statistics, Division of Manpower and Employment, March 23, 1959. (Mimeographed.) Woody, W. H., and C. L. Cushman. "A Study of Continuance and Discontinuance," Journal of Educational Research. 30:3:183-187, November 1936. Woollatt, Lorne H. "Why Capable Students Drop Out of High School," Bulletin of the Association of Secondary School Principals^ 45:7:1-8, November 196L. ... Worchel, P. "Adaptability Screening of Flying Personnel, Development of a Self-Concept Inventory for Predicting Maladjustment." School of Aviation Medicine, U.S.A.F., Report No. 56-62, 1957. Wyatt, Rose M. "Approaches to the Dropout Problem." Unpublished Master's project, University of Southern California, 1963. Wylie, Ruth C. The Self Concept. Lincoln: University of Nebraska Press, 1961. Young, J. M. "Can Counseling Reduce Dropouts?" Clearing House, 30:22-23, September 1955. APPENDICES APPENDIX A TEST INSTRUMENT: THE MEANING OF WORDS TEST Instructions to Test Supervisors Memorandum to Students Regarding Exemption from Taking the Meaning of Words Test The Meaning of Words Test Packet Envelope Instructions to Students Format of Meaning of Words Test APPENDIX A 1. INSTRUCTIONS TO TEST SUPERVISORS Instructions for Administering the Meaning of Wbrds Test Tou have been handed an envelope containing 15 cards. Care fully remove them from the envelope, look at them to be sure they are numbered from 1 to 15 on the front, and from 16 to 30 on the back. If you find either more or less than 15 cards, return the envelope with the cards to the teacher for a new packet of cards. On the face of the envelope are blanks for you to fill in. Print your last name, first name and initial In the first blank; next, mark whether boy or girl, and today's date. Fill in the name of your school, circle your grade and fill in the month, day and year of your birthdate. For father's occupation, answer as carefully as you can. For example, if your father is an engineer, describe the kind of work he actually does, not what company he works for. The ratings you are to make must be accurate, honest and according to directions given. There are no right or wrong answers. The results of this test will not affect your grades, go into your record, or affect your school progress. Mark the frotit and then the back of each card in order. That is, begin with card number 1; mark it front and back, then go on to card number 2, front and back, and so on through 15. Mark each bubble from end to end with a single line. Make no marks between the bubbles, gjhen you have finished, examine the cards and erase any marks made between the bubbles. Check to see that you marked only one bubble in each row. After you finish checking both sides of the cards, put them in order beginning with card 1 on top and ending with card 15 on the bottom. Then place the deck of cards in the envelope. Leave the folded page of mimeographed instructions out of the envelope; it will be collected separately. Are there any questions? Now, read aloud the INSTRUCTIONS TO STUDENT while the students read silently. 352 353 APPENDIX A 2. MEMORANDUM TO ALL STUDENTS REGARDING EXEMPTION FROM TAKING THE MEANING OF WORDS TEST TO: TEST ADMINISTRATORS The MEANING OF WORDS TEST is not a part of the mandatory State testing program. Therefore, please read the following statement to your class before distributing the MWT envelopes: ANYONE WHO, FOR RELIGIOUS REASONS, CANNOT PARTICIPATE IN OTHER THAN ROUTINE SCHOOL TESTS IS REQUESTED NOT TO TAKE THE MEANING OF WORDS TEST. APPENDIX A 3. MEANING OF WORDS TEST PACKET ENVELOPE RDS TEST MEANING GIRL ROY LAST DATE C I R C L E O N E MONTH DAY YEAR SCHOOL 10 12 C IR C L E G R A D E BIRTH DATS FATHER’S OCCUPATION. J I I I i f - k LSAVS TH13 LINE BLANK 354 APPENDIX A 355 4. INSTRUCTIONS TO STUDENTS* INSTRUCTIONS: To be read silently while examiner reads aloud. This is a test to measure the meaning of words. Mark each item according to what it means to you. At the top of each card, both front and back, is a word or group of words. Under each of them is a set of scales. Fill in ONE bubble on each line. If you feel that what you are rating is VERY CLOSE to one end of the scale you should fill in the bubble like this: GOOD $ () () () () () () BAD OR GOOD O 0 0 77 0 0 • BAD If you feel that what you are rating IS CLOSE to one end or the other then fill in the bubble like this: LITTLE () () $ () () () () BIG OR LITTLE 0 0 0 77 # 0 0 BIG f . If you feel that what you are rating is as close to one end of the scale as the other, then fill in the middle bubble. IMPORTANT 1. Each item should be rated separately. Some times you may feel as though you have had the same item before; this will not be the case, so do not look back through the cards. Do not try to remember how you rated the other items. 2. Fill in one bubble on each line, both front and back of the card. 3. Please work carefully because we want your true impressions, but do not spend too much time on any one item. It is your first impressions that are most important. *For taking the Meaning of Words Test (included with deck of IBM cards). 336 APPENDIX A 5. FORMAT OF MEANING OF WORDS TEST Card Number 1 (Front of card) TEACHERS FAIR c=> c=> c => UNFAIR WEAK <==> C- TJ £=> STRONG ACTIVE <=> e==> c=> c = PASSIVE UNPLEASANT :=> < z= > c=> PLEASANT LARGE c^> SMALL DULL “e==> e = Z > c = SHARP SWEET c^> C Z = > SOUR LIGHT C=> <=> c^> HEAVY FAST < Z = > c=> SLOW « smianMtiNiittnsti i n m a & n ■ n » n nnMHntJS*«aeoM««i«i»«nitODH0BHfflB(! qao*«0enttaT*M7ii»Haaf>naa U W T P L A T E N O . 1 (Back of card) SOUR HEAVY SLOW PLEASANT SMALL SHARP UNFAIR STRONG PASSIVE HOW MY CLASS SEES ME SWEET LIGHT FAST UNPLEASANT LARGE DULL FAIR WEAK ACTIVE tfi«ft9T*tntint!MnROiiB0iiiizit<iia0M!9&»aa*sana»a«tec!«ea<7«i4»»«ittq9<8iifTa»aiiaG>Mcoaffa»ivnttni|ia!gniiii« t M t H I H l i t e r P L A T E N O . Z 1 357 APPENDIX A 5. FORMAT OF MEANING OF WORDS TEST r Card Number 2 (Front of card) FAIR WEAK ACTIVE UNPLEASANT LARGE DULL SWEET LIGHT FAST GRADES UNFAIR STRONG PASSIVE PLEASANT SMALL SHARP SOUR HEAVY SLOW ifi4««Tii«tittt>i4fS!infft»Bftno*ssnaaa]iaa»is8ifa3io«ae«4aaa«09tiBS)aaairQs«p»!tft>«HaffMMii!i7i7iMi>nniti*« w « L l 4 i t t U W T P L A T E N O . 1 (Back of card) AUTHORITY SOUR HEAVY SLOW PLEASANT SMALL SHARP UNFAIR STRONG PASSIVE SWEET LIGHT FAST UNPLEASANT LARGE DULL FAIR WEAK ACTIVE f i M i n i tanniiHt»MONfianziPMfiaaananaaMfiaaaa4i<i «««««<? a a a a a o H a B n B a e n D n a n p a a a a n n a n B n n B n B t a i t 8 4 i 1 6 U l t f T P L A T E N O . 2 it APPENDIX A 5. FORMAT OF MEANING OF WORDS TEST 358 Card Number 3 (Front of card) FAIR WEAK ACTIVE UNPLEASANT LARGE DULL SWEET LIGHT FAST SCHOOL UNFAIR STRONG PASSIVE PLEASANT SMALL SHARP SOUR HEAVY SLOW M L M i l t M W T P L A T E N O . 1 (B ack o f c a r d ) RULES SOUR HEAVY SLOW PLEASANT SMALL SHARP UNFAIR STRONG PASSIVE SWEET LIGHT FAST UNPLEASANT LARGE DULL FAIR WEAK ACTIVE U U T PLATE NO. 2 tit 359 APPENDIX A 5. FORMAT OF MEANING OF WORDS TEST Card Number 4 (Front of card) FAIR WEAK ACTIVE UNPLEASANT LARGE DULL SWEET LIGHT FAST 118 lltll W M L I4IC B READING UNFAIR STRONG PASSIVE PLEASANT SMALL SHARP SOUR HEAVY SLOW P L A T E N O . l (Back of card) SOUR HEAVY SLOW PLEASANT SMALL SHARP UNFAIR STRONG PASSIVE PUNISHMENT SWEET LIGHT FAST UNPLEASANT LARGE DULL FAIR WEAK ACTIVE I t t 1 1 4 1 1 f t M t o T p l a t e n o . z 360 APPENDIX A 5. FORMAT OF MEANING OF WORDS TEST Card Number 5 (Front of card) IDEAL TEACHER FAIR WEAK ACTIVE UNPLEASANT LARGE DULL SWEET LIGHT FAST ? i t t n i t z n H t t M f f t t M a n ItW LM IPt M UNFAIR STRONG PASSIVE PLEASANT SMALL SHARP SOUR HEAVY SLOW n a a a a f t a a a t i t t a u a a n a a a a a a a a a t t a a a t T a o a a s v a a a t i B i t i M a a n a a f f n n n M n a n a a a PLATE NO. 1 £ (Back of card) FUTURE SOUR <=> C = ^ > < = = > £ = = > SWEET HEAVY <=5 eZ? < = = • c=^ LIGHT SLOW c^> c=^ e=> FAST PLEASANT < = = = > eZ> c^> C = ^ UNPLEASANT SMALL = £ = > < = = » c=> C = 2 LARGE SHARP c ^ = > c=> < = z > DULL UNFAIR < z = > c=> CZ? <== FAIR STRONG C => C=> c=> e== WEAK PASSIVE <=> <=> ACTIVE o a t t « M i r i B T P L A T E N O . £ / ' 3<a I M t APPENDIX A 5. FORMAT OF MEANING OF WORDS TEST Card Number 6 (Front of card) HOME FAIR CT CT> cr> cr cr cr cr UNFAIR WEAK CT> <rr CT cr cr c = STRONG ACTIVE cT> cr cr cC CT PASSIVE UNPLEASANT c r = > cr cr CT PLEASANT LARGE CT CT cr cr cr <== SMALL DULL = = = > cT> cr CT cr CT SHARP SWEET < C = > CT cr cr CT e = SOUR LIGHT CT cr cr CT CT HEAVY FAST = = = > <cr cr cr cr cr SLOW i 9 » i i i i awMtiNnti tsar no naanaaon boms an M t t f c G f c a «««»?«« one DHaHOnoisntitoMcsttiTMeaKTiTtnwr U W T P L A T E N O . 1 (Back of card) COLLEGE SOUR cT CT CT cr cr < r r > rr SWEET HEAVY er CT cr cr cr cr cr LIGHT SLOW cr CT cr cr err cr cr FAST PLEASANT :=> CT CT cr err cr err UNPLEASANT SMALL :=> CT cr cry cr rr> <rr LARGE SHARP = = = - CT cr cr cr < r r > cr DULL UNFAIR cr err cr rr < r = > cr FAIR STRONG cr cT cr cr cr cr WEAK PASSIVE CT cT cr < r r > cr cr ACTIVE t i i M i r i iintttfitttiRtrnig8nitBM!sazraa*>TaiiMfljins&«<iea«a«tt«MiiHttis)MBtfiJttn*BiataMa«ir«a»nivi9»i»*rrBJ99 U t o T P L A T E N O . Z 312 APPENDIX A 5. FORMAT OF MEANING OF WORDS TEST Card Number 7 (Front of card) FAIR WEAK ACTIVE UNPLEASANT LARGE DULL SWEET LIGHT FAST IDEAL PARENT <==> C =3 UNFAIR STRONG C = = > C = = PASSIVE C => <== PLEASANT SMALL C= SHARP SOUR <== HEAVY SLOW iiJ4a»?ii«tinti*4ntinttoa»i&&MB9nBa8«ia&Mti*t?»a«i't<!a«4«tttta«&ii«Kjiiti9DaM0neit)«4a«neattnn7Tnrinnnnaa p a t M l t a U » T P L A T E N O . 1 c a r d ) SOUR c=> A JOB < = = > <=> c = = > SWEET HEAVY e = = > e=> e = LIGHT SLOW = = = > C = Z > c=> FAST PLEASANT c^> < = z > C=> c=? cZ? < = = = UNPLEASANT SMALL c=> < = z > e Z = > LARGE SHARP c=> < ^ = > DULL UNFAIR cZ? < z = > c = ^ > <== FAIR STRONG c^> < Z = > c^> <=> cZ? C = = WEAK PASSIVE c=> <=> c=> c^> CZ> ACTIVE C M tLt4lfl UWT PLATE NO. 2 363 APPENDIX A 5* FORMAT OF MEANING OF WORDS TEST Card Number 8 (Front of card) FAIR WEAK ACTIVE UNPLEASANT LARGE DULL SWEET LIGHT FAST PARENTS <=> UNFAIR STRONG PASSIVE PLEASANT SMALL SHARP SOUR HEAVY SLOW t!i4«9Vtintiai!Htiitnt9!intiBBflasfT8a»nQ&ttBS9aae«a«)«4i<itt«afliTtit)tttattBi»a0iietoMHai7»0»n!!T)rtn>ifnaiia n L H l i l M W T P L A T E N O . 1 (B ack o f c a r d ) SOUR HEAVY SLOW PLEASANT SMALL SHARP UNFAIR STRONG PASSIVE GRADUATING SWEET LIGHT FAST UNPLEASANT LARGE DULL FAIR WEAK ACTIVE n i i M i i *wntiiswti*ovt**Br»MMHarasaiiaa«a»iTa»a«««*4««<T«a»»ii!nMO*rn©0notJN»«*P(»a*Tinn»»Bn!iTiB B U T T P L A T E N O . 2 vttt 3*4 APPENDIX A 5. FORMAT OF MEANING OF WORDS TEST Card Number 9 (Front of card) FAIR WEAK ACTIVE UNPLEASANT LARGE DULL SWEET LIGHT FAST MOST PEOPLE UNFAIR STRONG PASSIVE PLEASANT SMALL SHARP SOUR HEAVY SLOW n i L M l i i M W T P L A T E N O . 1 QUITTING SCHOOL (Back of card) SOUR HEAVY SLOW PLEASANT SMALL SHARP UNFAIR STRONG PASSIVE . i n t i t M » t t n t * t a n a B O W o » r b n w a n n a f i a n 00* 1 1 * « « « « * « * » n o o M « i M P B O M H I N T P L A T E N O . 2 SWEET LIGHT FAST UNPLEASANT LARGE DULL FAIR WEAK ACTIVE •teoMtfatfttanntiitMttttnana is APPENDIX A 5. FORMAT OF MEANING OF WORDS TEST Card Number 10 365 (Front of card) FAIR WEAK ACTIVE UNPLEASANT LARGE DULL SWEET LIGHT FAST M7 BEST FRIENDS c = UNFAIR e = STRONG C = > e== PASSIVE < = = : PLEASANT < = ^ > C => c = SMALL C^> < = ^ > < £ = > c = SHARP c = SOUR c=? HEAVY CZ> C =5 SLOW aim M W T B u n a s an ns»B»&ii»aaao44«(*on«!«grT6enaanBDa»it!fcMtiacT«ica7innT]!tni*n?i!3o P L A T E N O . L (Back of card) SOUR HEAVY SLOW PLEASANT SMALL SHARP UNFAIR STRONG PASSIVE TRYING HARD <£=> e=^ SWEET LIGHT FAST UNPLEASANT LARGE DULL FAIR WEAK ACTIVE i f » 4 i i v i i a t i t t t » M a i B t ? t i t 9 a i i a n i i B f l 0 t m a i f a a M & & i ? a a « f t « a « a * 4 7 « « i M i i i a i i a i i t > i t A f t t t t i a t « i m « r c a 0 i « n 7 Y 8 » i » B f f i i i » v t m L l M t O l i f e T P L A T E N O , 2 X 366 APPENDIX A 5. FORMAT OF MEANING OF WORDS TEST Card Nuaber 11 (Front of card) CLASSMATES FAIR c=> < = = > c=> UNFAIR WEAK = = = > < = = > c^> < = = > <=3 STRONG ACTIVE < = = > c = = > PASSIVE UNPLEASANT £ = = > PLEASANT LARGE < = = > SMALL DULL e = = > <=> C = = > SHARP SWEET = => c = Z > e = = > c=> c = SOUR LIGHT sr> < z = > < = = > c=> c=? e = HEAVY FAST < = = > < = = = > < Z = > SLOW tmitffiMn nift9tiBftfli)MaBfvsa»niia»aaJ7s»aae««:«*fl«aBMBBtiaBnBit0»fatiMaa0attiaTiiini«nMftaio& W W T P L A T E M O . 1 (B ack o f c a r d ) CHEATING SWEET LIGHT FAST UNPLEASANT LARGE DULL FAIR WEAK ACTIVE U t a T P L A T E N O . 2 SOUR HEAVY SLOW PLEASANT SMALL SHARP UNFAIR STRONG PASSIVE APPENDIX A 5. FORMAT OF MEANING OF WORDS TEST Card Nunfrer 12 (Front of card) FAIR WEAK ACTIVE UNPLEASANT LARGE DULL SWEET LIGHT FAST GROWNUPS UNFAIR STRONG PASSIVE PLEASANT SMALL SHARP SOUR HEAVY SLOW MHMii t mft*nMTi?»nttQHi'DBHaBna9an&nMStts&a«4iae4<e«M)<i«aitai!MaaBtt0attetoMaaawN*nniiAiiiinaiifl BHLMlit MffT PLATE NO. 1 (Back of card) SOMETHING EAST SOUR <=> HEAVY SLOW PLEASANT cZ? SMALL CZ> SHARP = = = > UNFAIR STRONG PASSIVE cZ> n « n H i > n t > n a 9 i i J D H B a c B B o n n SWEET LIGHT FAST UNPLEASANT LARGE DULL FAIR WEAK ACTIVE UttT PLATE NO. Z 368 APPENDIX A 5. FORMAT OF MEANING OF WORDS TEST Card Number 13 (Front of card) FAIR WEAK ACTIVE UNPLEASANT LARGE DULL SWEET LIGHT FAST ME UNFAIR STRONG PASSIVE PLEASANT SMALL SHARP SOUR HEAVY SLOW M K t t i i • i i i ) t i n H t i n n » a s n a i » H a B n i > a a i i t > 8 M B i a » a a o i ) a < i « « a n 4 i « i t « » D H B n n i > n i i t > i i b M a i a n H a n r < n n # n i i n n B * M L X I I I I I B T P L A T E N O . 1 (Back of card) SOMETHING IMPORTANT SWEET LIGHT FAST UNPLEASANT LARGE DULL FAIR WEAK ACTIVE wlMiM It 1 ST PLATE NO. 2 SOUR HEAVY SLOW PLEASANT SMALL SHARP UNFAIR STRONG <=> PASSIVE ==> < = = > alii 369 APPENDIX A 5. FORMAT OF MEANING OF WORDS TEST Card Number 14 (Front of card) MY SCHOOL ABILITY FAIR G G> C G G G G G G G G G UNFAIR WEAK G G G G c G G G C TJ STRONG ACTIVE cG ? G G G G G G G G C G PASSIVE UNPLEASANT == > G G g G G G g G G G C G PLEASANT LARGE G G G G G G G G g g G G G G SMALL DULL G G G G G G g G G G G=- SHARP SWEET G G G G G G G G G G G G SOUR LIGHT G G G G C c ? G G G G HEAVY FAST e= 2 G G G G c G g g G G SLOW t i i i i i i i i a n a a i t t i n o t i a a i t iowosnD»B>io»»iBnnt)ii«nnnjM««o<i*B«iHo«Bao«oc: tmbMBBnaannTtiTMaMnaiia B H L M I t t M V T P L A T E M O . 1 MONEY (Back of card) SOUR HEAVY SLOW PLEASANT SMALL SHARP UNFAIR STRONG PASSIVE I I I « q • I « in ittttitta a tta f ta iiB a a a a o a f la iia tta a a n a a a a a a a a a a a a tr a ttiitia a f ia a a tta e tta a e s a a T in tta n a f f a n a B 9 L I 4 1 I B u f e T P L A T E N O , 2 c^> SWEET LIGHT FAST c = UNPLEASANT c = LARGE e== DULL G G FAIR G G WEAK G G ACTIVE slF 370 APPENDIX A 5. FORMAT OF MEANING OF WORDS TEST Card Number 15 (Front of card) FAIR WEAK ACTIVE UNPLEASANT LARGE DULL SWEET LIGHT FAST HOW I WOULD LIKE TO BE UNFAIR STRONG PASSIVE PLEASANT SMALL SHARP SOUR HEAVY SLOW m u m i h U W T P L A T E N O . 1 (Back.of card) SOUR HEAVY SLOW PLEASANT SMALL SHARP UNFAIR STRONG PASSIVE SUCCESS SWEET LIGHT FAST UNPLEASANT LARGE DULL FAIR WEAK ACTIVE irittttHffttBowQann&Mfi&ffMaafla&iiSttBsaativtiMfKMiaairtiEinttiisniiaaBiiteBiciHtfviiitTinnftaHfliaftS « * 1 8 M I B ( 4 W T P L A T E n o . e RV APPENDIX B TABULATION OF DATA FOR SEVEN SELECTED DESCRIPTIVE VARIABLES 372 TABLE 36 FREQUENCY DISTRIBUTION OF FATHER OCCUPATION RATING* Rating Group DROPOUTS Tenth Grade Boys N-34 0 1 3 7 17 4 2 Girls N-33 2 6 4 8 10 0 3 Eleventh Grade Boys N=33 3 2 3 5 16 3 1 Girls N-31 2 3 3 5 17 1 0 STAY-INS Tenth Grade Boys N«34 5 6 6 5 9 3 0 Girls N«33 2 5 4 12 9 1 0 Eleventh Grade Boys N»33 2 4 7 9 9 1 1 Girls N-31 0 6 2 8 13 2 0 *Warner, Meeker and Eells' Revised Occupation Scale (1961).! TABLE 37 GROUP TEST INTELLIGENCE QUOTIENT SCORES Median Minimum Maximum Range Mean Standar Deviation DROPOUTS Tenth Grade Boys N®34 95.00 64.00 118.00 Girls N»33 99.00 72.00 121.00 Eleventh Grade Boys N®33 99.00 68.00 134.00 Girls N-31 103.00 84.00 121.00 STAY-INS Tenth Grade Boys N®34 102.00 68.00 133.00 Girls N-33 99.00 74.00 128.00 Eleventh Grade Boys N«33 104.00 75.00 118.00 Girls N»31 101.00 74.00 121.00 54.00 49.00 66.00 37.00 65.00 54.00 43.00 47.00 95.21 97.82 99.00 101.74 99.94 98.85 100.76 101.16 11.23 11.45 13.64 10.11 14.29 12.37 11.99 12.36 373 TABLE 38 AGE IN MONTHS AT TIME OF TESTING Median Minimum Maximum Range Mean 'StandarS' Deviation DROPOUTS Tenth Grade Boys N«34 191.00 179.00 212.00 Girls N«33 189.00 180.00 223.00 Eleventh Grade Boys N-33 203.00 192.00 216.00 Girls N=31 196.00 182.00 219.00 STAY-INS Tenth Grade Boys N®34 186.00 174.00 202.00 Girls N=33 185.00 176.00 192.00 Eleventh Grade Boys N»33 195.00 188.00 205.00 Girls N»31 196.00 187.00 211.00 33.00 43.00 24.00 37.00 28.00 16.00 17.00 24.00 193.03 190.73 204.48 197.23 186.26 184.91 196.15 196.10 7.44 8.30 7.52 7.59 5.80 4.40 3.78 4.37 374 TABLE 39 MEAN STANDARD SCORE OF SUBTESTS 3, 6, 7 OF THE IOWA TESTS OF EDUCATIONAL DEVELOPMENT Median Minimum Maximum Mean StanTarcT* Deviation DROPOUTS Tenth Grade Boys N®34 Girls N*33 Eleventh Grade Boys N»33 Girls N»31 STAY-INS Tenth Grade Boys N«*34 Girls N»33 Eleventh Grade Boys N«33 Girls N»31 8.00 10.00 10.00 12.00 14.00 14.00 16.00 16.00 0.00 0.00 5.00 0.00 6.00 6.00 6.00 9.00 20.00 19.00 23.00 20.00 27.00 27.00 26.00 25.00 20.00 19.00 18.00 20.00 21.00 21.00 20.00 16.00 8.21 10.55 11.15 11.39 15.00 14.46 15.18 16.45 4.78 4.34 4.56 5.16 5.66 4.48 4.36 4.88 375 DROPOUTS Tenth Grade Boys N=34 Girls N-33 Eleventh Grade Boys M=33 Girls N»31 STAY-INS Tenth Grade Boys N=34 Girls N«33 Eleventh Grade Boys N=33 Girls N»31 TABLE 40 TOTAL ABSENCES DURING PERIOD OF ENROLLMENT Median Minimum Maximum Range Mean 26.00 34.00 29.00 15.00 7.00 6.00 10.00 6.00 0.00 0.00 2.00 4.00 0.00 0.00 0.00 0.00 120.00 80.00 54.00 90.00 41.00 37.00 32.00 51.00 120.00 80.00 52.00 86.00 41.00 37.00 32.00 51.00 31.09 34.91 24.48 24.90 9.53 9.24 10.85 11.26 Standard Deviation 27.24 22.20 15.88 20.78 8.77 9.45 7.84 12.51 376 TABLE 41 TOTAL NUMBER OF DAYS ENROLLED IN SCHOOL Median Minimum Maximum Range Mean Standar Deviation DROPOUTS Tenth Grade Boys N-34 80.00 13.00 171.00 Girls N*=33 108.00 14.00 178.00 Eleventh Grade Boys N=33 89.00 16.00 168.00 Girls N=31 98.00 23.00 170.00 STAY-INS Tenth Grade Boys N«*34 178.00 178.00 178.00 Girls N«33 178.00 173.00 178.00 Eleventh Grade Boys N=33 178.00 178.00 178.00 Girls N«31 178.00 88.00 178.00 158.00 164.00 152.00 147.00 0.00 5.00 0.00 90.00 90.74 103.79 87.48 95.65 178.00 177.70 178.00 175.10 41.79 44.22 42.13 39.53 0.00 1.19 0.00 15.90 377 TABLE 42 DROPOUT LEAVING AGE IN MONTHS Median Minimum Maximum Range Mean Standard Deviation Tenth Grade Boys N=34 Girls N=33 195.00 194.00 182.00 180.00 217.00 225.00 Eleventh Grade Boys N«=33 207.00 196.00 221.00 Girls N«31 200.00 188.00 224.00 35.00 45.00 25.00 36.00 196.47 195.03 207.94 201.09 7.61 8.13 7.01 7.53 378 APPENDIX C TABULATION OF MEAN SUMMED FACTOR SCORES FOR DROPOUTS AND STAY-INS, TENTH AND ELEVENTH GRADES, AND BOYS AND GIRLS WHEN TESTED ON THE EVALUATIVE, POTENCY AND ACTIVITY SCALES 380 A P P E N D I X C T A B L E 6 3 MEAHS AMD STANDARD DEVIATIONS OF SUMMED EVALUATIVE FACTOR SCORES FOR TENTH CRADE DROPOUTS AND STAT-INS D r o p o U t B ■' ■ —— ■ ■ — S t a y - I n s C o n c e p t B o y s G i r l s B o y s G i r l 8 M s s n S . D . M e a n . S . D . K e a n S . D . M e a n S . D . S c h o o l E x p e r i e n c e s ; T e a c h e r s 7 . 2 6 6 . 3 6 9 . 6 5 6 . 2 6 1 6 . 2 6 2 . 6 8 1 6 . 8 2 3 . 1 5 C r a d e a 8 . 3 8 3 . 9 2 9 . 1 5 6 . 2 5 1 6 . 1 2 3 . 9 6 1 3 . 6 7 3 . 2 8 S c h o o l 6 . 3 5 3 . 6 9 9 . 6 3 6 . 6 0 1 6 . 3 5 6 . 6 0 1 5 . 0 9 5 . 2 1 R e a d i n g 9 . 2 6 5 . 1 0 1 1 . 6 1 5 . 6 7 1 5 . 0 3 6 . 5 5 1 6 . 1 2 6 . 1 9 I d e a l T e a c h e r 1 6 . 0 9 6 . 3 7 1 9 . 2 1 2 . 3 2 1 9 , 0 0 3 . 2 5 1 9 . 5 2 2 . 3 6 F a a i l l v R e l a t i o n s h i p s : H o n e 1 0 . 5 0 6 . 1 6 1 2 . 8 8 6 . 6 3 1 7 . 3 5 6 . 0 1 1 6 . 8 5 6 . 3 6 I d e a l P a r e n t 1 8 . 0 0 2 . 3 9 1 9 . 6 5 1 . 5 2 1 9 . 2 6 2 . 2 0 1 9 . 9 6 1 . 3 7 P a r e n t s 9 . 0 9 5 . 5 8 1 2 . 6 7 6 . 9 6 1 6 . 6 1 6 . 5 0 1 6 . 2 6 6 . 5 3 S o c i a l I n t e r a c t i o n : H o s t P e o p l e 9 . 1 2 5 . 0 1 9 . 5 8 6 . 6 6 1 6 . 6 8 3 . 3 6 1 5 . 8 8 3 . 2 8 M y B e a t F r i e n d s 1 7 . 0 9 2 . 6 1 1 7 . 1 5 3 . 8 8 1 7 . 9 6 2 . 6 7 1 9 . 6 2 1 . 7 1 C l a e a o a t a a 9 . 0 3 5 . 1 6 9 . 8 2 6 . 2 5 1 5 . 6 5 3 . 0 5 1 6 . 6 1 3 . 1 3 G r o w n u p s 9 . 1 5 6 . 9 7 9 . 7 9 6 . 7 0 1 6 . 1 8 6 . 3 6 1 5 . 6 2 3 . 6 9 S e l f C o n c e p t : H e 1 0 . 0 6 5 . 1 5 1 1 . 3 6 5 . 1 0 1 7 . 3 8 2 . 8 7 1 7 . 1 5 2 . 3 6 M y S c h o o l A b i l i t y 8 . 9 7 6 . 2 6 1 0 . 9 9 6 . 6 3 1 6 . 7 6 3 . 0 6 1 6 . 3 0 2 . 8 6 H o w T W o u l d L i k e t o H e 1 6 . 6 8 3 . 3 5 1 8 . 9 6 2 . 5 6 2 0 . 1 2 1 . 6 0 2 0 . 2 7 1 . 2 3 H o w M y C l a s s S e e s H a 8 . 6 6 2 . 9 6 8 . 1 2 2 . 9 0 1 3 . 3 8 6 . 9 0 1 2 . 3 6 6 . 1 6 A u C h o r l t v R e l a t i o n s h i p s : A u t h o r i t y 1 0 . 6 1 3 . 9 9 1 0 . 0 6 3 . 8 3 1 7 . 3 2 3 . 3 5 1 6 . 7 3 3 . 8 8 R u l e s 1 1 . 6 6 6 . 3 2 1 0 . 7 6 3 . 8 2 1 7 . 8 2 2 . 7 6 1 5 . 9 1 3 . 6 6 P u n l e h n c n t 1 6 . 5 9 2 . 8 2 1 5 . 0 9 2 . 6 3 1 8 . 2 6 2 . 5 5 1 7 . 9 1 2 . 7 9 C o s t O r i e n t a t i o n ; F u t u r e 6 . 1 5 2 . 9 6 6 . 5 6 6 . 0 9 1 2 . 7 6 6 . 7 1 8 . 3 9 5 . 0 7 C o l l e g e 7 . 3 5 3 . 8 3 8 . 7 3 6 . 7 3 1 6 . 5 6 6 . 5 1 1 2 . 6 7 6 . 9 5 A J o b 6 . 6 6 2 . 6 6 5 . 7 9 2 , 7 0 5 . 7 9 2 . 3 3 5 . 6 5 2 . 2 7 G r a d u a t i n g 5 . 1 8 3 . 2 0 5 . 3 6 2 . 9 8 1 2 . 7 1 5 . 7 9 1 0 . 0 0 5 . 9 6 Q u i t t i n g S c h o o l 1 8 . 7 6 6 . 0 6 1 8 . 9 7 3 . 0 8 9 . 9 7 6 . 3 6 1 2 . 2 7 6 . 9 0 M o r a l a n d S o c i a l V a l u e s ; T r y i n g H a r d 7 . 8 5 3 . 6 2 8 . 6 8 3 . 6 9 1 3 . 3 5 6 . 6 0 1 2 . 7 6 3 . 6 1 C h e a t i n g 1 7 . 9 1 6 . 6 1 1 8 . 3 9 2 . 7 2 1 2 . 0 6 6 . 9 1 1 5 . 6 1 5 . 7 2 S o s w t h l n g E a s y 6 . 2 9 3 . 2 7 6 . 9 7 3 . 5 0 7 . 5 3 2 , 7 6 6 . 7 6 3 . 6 6 S o n s t h i n g I n p o r t a n t 7 . 6 6 3 . 8 6 6 . 9 6 3 . 3 1 1 1 . 5 0 6 . 1 1 1 1 . 2 7 3 . 5 0 H o n e y 5 . 3 5 3 . 6 3 5 . 5 8 2 . 6 3 6 . 9 6 2 . 2 1 5 . 6 6 3 . 7 3 S u c c e s s 6 . 6 2 2 , 6 0 6 . 2 1 2 . 2 6 1 0 . 5 6 6 . 6 7 7 . 2 6 6 . 1 5 381 m m a a c T A B L E 4 4 H E A M 3 A M ) S T A N D A R D D E V I A T I O N S O r S U M M E D P O T E N C Y T A C T O R S C O R E S F O B T E N T H C 1 A S E D R O P O U T S A M D S T A Y - I N S D r o p o u t s S t a y - In * C O M i p t S o y # G i r l * S o y s G i l l * H M D S . D . K e a n S . D . H e n S . D . M e a n S . D . S c h o o l t a p a r l a n c e s : T e a c h e r a 1 0 . 2 9 2 . 6 8 1 1 . 6 1 3 . 3 2 1 2 . 4 4 2 . 8 4 1 2 . 8 5 1 . 6 6 G r a d e s 1 0 . 7 9 3 . 3 3 1 0 . 8 2 2 . 4 0 1 2 . 7 4 2 . 8 3 1 3 . 4 5 2 . 1 1 S c h o o l 1 0 . a s 3 . 7 3 1 2 . 5 5 4 . 2 1 1 5 . 3 8 3 . 5 8 1 6 . 1 2 2 . 9 3 R a i d i n g 1 0 . 3 5 3 . 0 5 1 2 . 2 7 3 . 6 4 1 4 . 3 8 3 . 3 4 1 4 . 2 7 3 . 1 1 I d e a l t a a c h a c 1 4 . 5 0 2 . 9 3 1 3 . 8 8 4 . 5 7 1 4 . 6 2 4 . 4 6 1 3 . 7 0 3 . 4 6 E a s i l y R e l a t i o n s h i p s : H o a a 1 1 . 5 6 3 . 8 2 1 1 . 4 8 4 . 3 8 1 4 . 2 1 3 . 5 3 1 2 . 6 1 3 . 0 7 I d a a l P a r e n t 1 4 . 5 3 2 . 9 3 1 3 . B 2 4 . 2 2 1 5 . 4 7 3 . 6 0 1 8 . 3 0 2 . 3 0 P a r e D C a 1 1 . 2 1 3 . 2 9 1 0 . 8 2 3 . 9 1 1 4 . 3 5 4 . 0 4 1 2 . 0 9 2 . 7 4 S o c i a l I n t e r a c t i o n : M o s t P e o p l e 1 0 . 1 5 3 . 0 1 1 0 . 1 5 1 . 6 5 1 3 . 3 5 2 . 4 4 1 2 . 2 7 2 . 0 8 B e e t F r i e n d s 1 5 . 1 5 2 . 5 6 1 4 . 1 8 3 . 3 8 1 5 . 0 6 2 . 9 9 1 3 . 4 2 2 . 6 7 C l a s e i a a t a a 1 0 . 3 8 2 . 9 1 1 0 . 7 0 2 . 0 1 1 3 . 0 9 2 . 3 0 1 3 . 7 0 2 . 6 6 C r o w n u p a 1 0 . 6 8 3 . 5 9 1 1 . 0 6 2 . 7 9 1 3 . 1 5 4 . 0 0 1 3 . 6 7 2 . 8 7 S e l f C o n c e p t : , M e 1 1 . 1 8 3 . 6 8 1 0 . 3 3 3 . 1 6 1 4 . 1 8 3 . 5 6 1 2 . 2 4 3 . 7 0 M y S c h o o l A b i l i t y 9 . 7 1 2 . 7 9 1 1 . 1 8 3 . 3 3 1 5 . 5 9 3 . 2 0 1 3 . 7 3 3 . 8 8 R o w I M o u l d L i k a t o B e 1 5 . 5 6 2 . 6 4 1 3 . 7 0 3 . 4 0 1 7 . 5 0 3 , 3 6 1 3 . 0 3 3 . 5 3 H o w M y C l a s s S e e * M e 1 0 . 6 5 3 . 3 1 1 2 . 0 6 2 . 7 4 1 2 . 6 8 2 . 7 6 1 2 . 6 7 2 , 4 6 A u t h o r i t y R e l a t i o n s h i p * : A u t h o r i t y 9 . 2 9 4 . 0 2 9 . 3 6 3 . 8 1 1 2 . 2 4 4 . 2 9 1 1 . 4 5 4 . 1 6 R u l e s 9 . 5 6 3 . 6 2 9 . 1 8 2 . 7 0 1 3 . 0 6 3 . 7 4 1 1 . 7 3 4 . 0 2 P u n l a h a K n t 1 1 . 0 3 3 . 9 3 9 . 9 1 3 . 8 0 1 1 . 0 0 4 . 9 1 1 1 . 2 1 5 . 1 2 G o a l O r i e n t a t i o n : F u t u r e 8 . 1 5 3 . 5 5 9 . 0 3 3 . 2 1 1 2 . 1 8 3 . 7 5 1 0 . 5 5 3 . 5 1 C o l l e g e B . 6 8 4 . 7 2 8 . 5 2 3 . 8 9 1 2 . 8 5 3 . 2 9 1 1 . 9 4 1 . 8 6 A d o b 8 . 1 5 3 . 4 7 9 . 9 4 3 . 2 7 . 8 . 9 1 2 . 9 9 9 . 2 1 2 . 9 8 G r a d u a t i n g 7 . 4 7 3 . 6 9 9 . 1 8 2 . 6 4 1 1 . 8 2 4 . 0 1 1 1 . 6 1 3 . 7 4 Q u i t t i n g S c h o o l 1 5 . 2 1 5 . 2 2 1 4 . 7 5 3 . 8 2 1 1 . 2 6 4 . 1 7 1 1 . 5 8 3 . 6 0 M o r a l a n d S o c i a l V a l u e s : T r y i n g B a r d 8 . 8 5 2 . 8 7 8 . 6 4 2 . 7 5 1 3 . 0 0 3 . 7 5 1 2 . 3 0 3 . 4 2 C h e a t i n g 1 4 . 5 6 4 . 5 6 1 3 . 5 5 4 . 3 7 1 1 . 8 5 3 . 6 7 1 3 . 0 9 3 . 9 2 S o u t h i n g E a s y 1 2 . 0 3 5 . 1 9 1 4 . 2 7 3 . 9 3 9 . 9 1 2 . 9 8 1 1 . 8 2 4 . 5 3 S o u t h i n g I m p o r t a n t 8 . 2 4 3 . 5 0 8 . 3 6 3 . 4 0 1 1 . 3 2 3 . 2 6 l i . 8 5 3 . 1 8 M o u s y 9 . 0 3 3 . 8 6 9 . 3 3 3 . 4 7 7 . 6 5 3 . 0 2 8 . 0 9 3 . 6 9 S u c c e s s 6 . 8 8 2 . 8 4 7 . 1 5 3 . 1 5 1 1 . 2 1 3 . 3 8 9 , 8 8 3 . 3 0 382 A F F C N D U C T A B L E 4 3 KZAKS AMD STAM DAW DEVIATIONS OF S lf ltB D A C T IV IT T FACTOR SOORES FO R TEXTH GRADE D IO TOUTS AMD STA Y- IH S D r o p o u t a S t a y - I n * C O B C * p t R o y * C i r l a B o y * G i r l * M e a n S . D . M e a n S . D . M e a n S . D . H a a n S . D . S c h o o l E x p e r i e n c e * : T e a c h * r a 8 . 3 3 2 . 9 7 1 0 . 8 8 3 . 7 6 1 4 . 3 8 2 . 4 6 1 3 . 9 7 3 . 3 0 G r a d e * 9 . 6 2 3 . 4 5 9 . 5 2 2 . 8 5 1 3 . 5 6 3 . 4 0 1 3 . 0 9 3 . 0 4 S c h o o l 8 . 3 0 4 . 1 7 1 0 . 7 0 4 . 1 5 1 4 . 4 4 4 . 0 6 1 5 . 3 3 4 . 2 9 R e a d i n g 9 . 8 8 3 . 6 4 1 1 . 6 1 4 . 8 2 1 4 . 3 8 4 . 0 9 1 4 . 9 1 4 . 6 1 I d e a l T a a c h a r 1 3 . 6 5 3 . 8 8 1 8 . 6 1 2 . 2 8 1 8 . 0 3 3 . 3 0 1 8 . 5 8 2 . 1 5 F a a i l y R e l a t i o n a h l p a : B o a a 1 0 . 7 9 4 . 1 9 1 2 . 6 4 4 . 7 8 1 6 . 0 0 3 . 3 8 1 5 . 0 6 4 . 0 5 I d e a l F a r e n t 1 6 . 7 9 3 . 1 3 1 8 . 0 0 2 . 7 2 1 8 . 3 8 3 . 5 5 1 8 . 3 0 1 . 9 9 F e r e a c * 1 0 . 1 8 4 . 2 1 1 2 . 4 3 5 . 7 2 1 4 . 5 9 4 . 5 9 1 4 . 2 4 4 . 4 3 S o c i a l I n t e r a c t i o n : M o a t F a o p l e 9 . 8 2 3 . 8 7 1 0 . 4 8 3 . 5 5 1 3 . 8 5 3 . 5 1 1 4 . 9 1 3 . 1 4 M y B e a r F r i e n d * 1 6 . 7 4 3 . 0 7 1 6 . 6 4 3 . 7 1 1 7 . 8 8 2 . 7 5 1 7 . 8 5 2 . 1 7 C l a a a a a t e e 9 . 5 3 4 . 2 4 1 0 . 7 9 3 . 3 1 1 4 . 3 8 3 . 4 8 1 5 . 8 2 3 . 4 8 G r o v n u p a 1 0 . 0 9 3 . 7 4 1 0 . 6 7 3 . 6 8 1 3 . 0 9 4 . 5 1 1 4 . 4 2 3 . 6 6 S e l f C o n c e p t : H e 1 1 . 2 1 4 . 5 6 1 2 . 0 6 4 . 0 0 1 7 . 5 3 2 . 7 3 1 6 . 4 2 1 . 8 7 M y S c h o o l A b i l i t y 8 . 7 9 2 . 6 7 1 0 . 4 8 4 . 3 8 1 6 . 0 3 3 . 4 3 1 5 . 6 1 3 . 4 5 D o w I M o u l d L i k e C o B e 1 8 . 0 6 2 . 4 0 1 8 . 7 3 2 . 1 2 1 9 . 7 6 1 . 8 1 1 9 . 6 4 1 . 7 3 R o u M y C l a a a S c a a M a 8 . 7 4 3 . 7 4 9 . 8 8 2 . 5 1 1 4 . 0 6 4 . 1 4 1 2 . 9 1 4 . 3 5 A u C h o r l t T R e l a d o o a h i p a : A u t h o r i t y 9 . 6 5 3 . 2 3 9 . 9 7 3 , 4 1 1 4 , 7 4 3 . 6 9 1 3 . 5 8 4 . 5 9 R u l e * 1 0 . 4 4 3 . 2 2 9 . 8 8 3 . 2 4 1 5 . 5 3 3 . 7 9 1 3 . 6 7 3 . 9 5 F u n l a h a e o t 1 2 . 4 4 3 . 3 2 1 2 . 3 0 2 . 8 2 1 4 . 6 5 4 . 2 6 1 5 . 4 5 3 . 9 8 G o a l O r i e n t a t i o n : F u t u r e 7 . 1 8 3 . . 2 6 7 . 6 7 4 . 2 1 1 3 . 2 1 4 . 1 0 9 . 5 5 4 . 7 0 C o l l e g * 8 . 0 0 4 . 5 5 9 . 1 8 4 . 6 2 1 3 . 9 4 4 , 3 3 1 3 . 0 3 4 . 2 4 A J o b 7 . 0 0 2 . 9 5 7 . 3 9 3 . 0 9 7 . 0 3 2 . 3 0 7 . 0 9 2 . 7 4 G r a d u a t i n g 6 . 4 1 3 . 6 5 6 . 8 8 3 . 1 5 1 2 . 5 9 4 . 5 9 1 0 . 9 1 4 . 6 7 Q u i t t i n g S c h o o l 1 7 . 8 2 4 . 7 6 1 7 . 6 1 3 . 3 1 1 0 . 6 8 4 . 8 7 1 2 . 5 8 5 . 4 7 M o r e l a n d S o c i a l V a l u e * : T r y i n g B a r d 8 . 5 3 3 . 9 3 8 . 8 5 3 . 4 3 1 3 . 5 9 3 . 7 7 1 2 . 8 2 3 . 6 4 C h e a t i n g 1 6 . 2 4 5 . 1 3 1 3 . 4 2 3 . 5 3 1 1 . 9 7 3 . 6 4 1 5 . 2 7 4 . 5 5 S o a a t h l n g g a a y 7 . 9 1 3 . 4 3 9 . 1 6 3 . 4 4 8 . 7 3 3 . 0 3 8 . 9 1 3 . 8 0 S o a a t h l n g t a p o r t a n t 7 . 8 2 1 . 7 5 7 . 8 5 2 . 9 6 1 1 . 5 6 3 . 9 7 1 1 . 2 1 4 . 0 8 M o n e y 6 , 0 3 3 . 8 4 6 . 8 5 2 . 8 8 6 . 6 5 2 . 8 2 6 , 8 2 3 . 2 5 S u c c e a * 3 . 7 6 3 . 1 3 7 . 0 6 3 . 6 2 1 0 . 9 4 4 . 0 3 9 . 4 5 3 . 5 5 A P P E N D I X C 383 T A B L E 4 6 W A N S AND STANDAtD D E V IA TIO N S OF SUMMED EVALUATIVE FACTOR SCORES FOR ELEVENTH GRADE DROPOUTS AND S T A T -1N S D r o p o u t a S t a y - 1 n a C o n c e p t B o y a G i r l s B o y s G i r l s H e a n S . D . M e a n S . D . N a a n S . D . H t a n S . D . S c h o o l E x p e r i e n c e s : T e a c h e r s 7 . 3 0 4 . 4 3 8 . 9 7 4 . 6 3 1 3 . 3 0 3 . 5 6 1 5 . 9 0 3 . 4 1 G r a d a a 7 . 7 2 3 . 5 1 9 . 2 6 4 . 6 8 1 2 . 7 6 3 . 3 0 1 3 . 9 4 3 . 5 6 S c h o o l 7 . S B 4 . 3 4 8 . 7 4 4 . 9 0 1 4 . 7 3 4 . 2 5 1 4 . 8 1 4 . 2 4 R a a d l a R 1 0 . 3 0 6 . 0 1 1 1 . 3 9 4 . 7 6 1 3 . 3 0 3 . 6 6 1 5 . 7 1 4 . 0 8 I d a a l T a a c h a c 1 8 . 3 V 3 . 5 5 1 8 . 6 5 3 . 4 4 1 8 . 8 5 2 . 6 8 1 9 . 9 7 1 . 9 1 F a a i l y R e l a t i o n s h i p s : B o a a 9 . 8 5 6 . 9 6 1 0 . 6 8 6 . 1 9 1 5 . 9 4 3 . 9 8 1 5 . 9 7 4 . 9 4 I d e a l F a c a o c 1 8 . 6 7 2 . 2 6 1 9 . 6 8 1 . 8 3 1 8 . 8 2 ' 2 . 4 3 1 9 . 8 7 1 . 8 0 F a t a n t a 9 . 7 0 6 . 4 2 1 1 . 6 5 6 . 7 5 1 6 . 8 5 4 . 0 6 1 5 . 9 4 4 . 8 0 S o c i a l I n t e r a c t i o n : M o a t P a o p l e 9 . 6 7 5 . 3 9 1 1 . 9 0 - 6 . 0 6 1 3 . 9 7 2 . 9 8 1 5 . 7 7 3 . 0 7 M y B a i t F r i e n d s 1 7 . 3 6 3 . 4 4 1 8 . 9 7 2 . 0 1 1 6 . 9 1 3 . 2 1 1 8 . 6 5 2 . 8 8 C l a a a a a t c a 9 . 3 3 5 . 6 7 1 1 . 9 4 5 . 9 9 1 5 . 1 8 2 . 9 4 1 6 . 1 9 3 . 1 9 G r o v n u p a 8 . 2 7 5 . 0 5 9 . 8 7 6 . 0 9 1 3 . 7 6 4 . 0 7 1 4 . 9 7 4 . 3 1 S a i l C o n c e p t : M a 1 1 . 3 0 6 . 3 1 1 2 . 2 3 6 . 1 6 1 5 . 6 7 3 . . 6 1 1 6 . 3 2 3 . 4 5 M y S c h o o l A b i l i t y 9 . 6 8 5 . 1 8 1 2 . 1 3 5 . 5 3 1 4 . 5 5 3 . 2 5 1 6 . 1 0 2 . 9 5 H o w I l l o u l d L i k e t o B e 1 8 . 5 2 2 . 2 9 1 9 . 3 5 2 . 5 1 1 9 . 0 3 3 . 1 0 2 0 . 4 2 2 . 0 5 H u m M y C l a a a S a a a M e 8 . 6 7 2 . 3 5 8 . 4 8 2 . 6 6 1 3 . 4 2 5 . 6 7 1 3 . 1 0 5 . 3 1 A u t h o r i t y R e l a t i o n a h l p a : A u t h o r i t y 1 0 . 9 4 4 . 0 2 1 1 . 2 6 3 . 8 0 1 7 . 3 6 4 . 5 7 1 5 . 9 0 4 . 8 4 R u l e a 1 1 . 3 8 4 . 0 0 1 1 . 1 0 3 . 8 9 1 7 . 8 2 3 . 9 2 1 6 . 0 3 4 . 6 2 F u a i a b a a a t 1 4 . 6 4 3 . 0 2 1 5 . 0 3 3 . 4 8 1 8 . 7 0 2 . 4 7 1 7 . 8 4 3 . 1 0 C o a l O r i e n t a t i o n : F u t u r e 7 . 5 5 4 . 0 2 6 . 8 1 2 . 8 3 1 0 . 9 7 4 . 8 3 8 . 2 6 6 . 4 3 C o l l e g a 9 . 2 7 4 . 6 2 8 . 1 6 3 . 5 6 1 3 . 4 5 5 . 2 4 1 2 . 5 8 5 . 8 5 A J o b 7 . 5 5 3 , 1 6 6 . 5 8 2 . 7 0 5 . 9 4 3 . 6 7 5 . 2 3 1 . 8 7 G r a d u a t i n g 6 . 4 5 4 . 8 2 5 . 2 3 2 . 5 5 9 . 8 5 6 . 1 8 8 . 8 1 6 . 3 1 Q u i t t i n g S c h o o l 1 7 . 4 8 4 . 6 0 1 9 . 4 8 3 . 2 3 L I . 8 2 7 . 4 2 1 2 . 3 5 7 . 2 3 M o r a l a n d S o c i a l V a l u e s : T r y i n g H a r d 9 . 0 6 3 . 1 5 8 . 7 1 3 . 6 3 1 3 . 3 6 3 . 8 0 1 1 . 6 5 4 . 2 2 C h e a t i n g 1 8 . 0 3 3 . 8 4 1 9 . 7 1 2 . 9 9 ' 1 4 . 4 8 5 . 1 4 1 4 . 4 8 5 . 6 6 S o a t t t h l n g E a s y 6 . 7 3 3 . 2 0 8 . 3 2 4 . 3 8 7 . 4 8 3 . 8 1 6 . 3 2 2 . 9 7 S o m t h l n g I a p o r t a n t 8 . 2 1 3 . 0 3 5 . 5 5 2 . 5 9 1 1 . 7 6 4 . 0 6 1 0 . 0 0 5 . 6 0 M o n e y 5 . 5 8 2 . 8 1 5 . 6 1 3 . 4 5 5 . 5 2 3 . 3 0 4 . 8 7 2 . 3 3 S u c c e s s 5 . 9 4 3 . 9 3 4 . 7 7 2 . 5 3 8 . 0 9 4 . 6 2 7 . 8 4 4 . 5 8 384 a p f e k d i i c T A B L E 4 7 M E A N S A N D S T A N D A B D D E V I A T I O N S O F S U H S D F O T E N C T F A C T O R S C O R E S ' F O R E L E V E M T N G R A D E D R O P O U T S A N D S T A Y - I N S D t o p o u t * S t a y - la* C o n c e p t B o y * G i r l * B o y a G i r l * M e a n S . D . M e a n S . D . M e a n S . D . N a a n S . D . S c h o o l E K p a r l a n c o a : T a a c h a t * 1 0 . I S 3 . 2 2 1 1 . 5 2 2 . 5 7 1 2 . 6 7 1 . 8 7 1 2 . 9 7 2 . 0 7 C r a d a a 9 . 1 5 3 . 3 5 - 1 0 . 7 7 3 . 5 8 1 3 . 6 7 2 . 8 0 1 3 . 5 5 3 . 3 5 S c h o o l 1 2 . 2 7 4 . 3 0 1 1 . 5 5 3 . 7 8 1 5 . 6 4 3 . 3 6 1 5 . 0 6 3 . 2 2 R e a d i n g 1 1 . 7 6 4 . 5 6 1 1 . 3 9 4 . 0 1 1 2 . 1 5 3 . 3 9 1 3 . 7 1 3 . 6 1 I d e a l T e a c h e r 1 3 . 6 7 4 . 3 9 1 4 . 0 6 2 . 7 8 1 3 . 2 4 2 . 7 5 1 3 . 9 0 2 . 6 0 F t a l l y R e l a t i o n a h l p a : R o a e 1 0 . 7 3 3 . 9 4 1 1 . 4 8 4 . 0 6 1 4 . 1 5 2 . 2 8 1 3 . 1 3 3 . 4 8 I d e a l F a r a n t 1 5 . 7 0 2 . 6 7 1 4 . 0 6 2 . 8 2 1 4 , 6 4 2 . 6 4 1 3 . 5 8 2 . 2 6 P a r e n t * 1 0 . 2 7 4 . 1 3 1 0 . 8 4 3 . 0 2 1 3 . 7 9 2 . 2 0 1 3 . 1 0 2 . 1 8 S o c i a l I n t e r a c t i o n : H o a t P e o p l e 1 0 . 0 6 3 . 6 8 1 0 . 8 4 3 . 4 9 1 2 . 7 9 1 . 6 7 1 2 . 4 8 2 . 1 2 M y B e e t F r i e n d * 1 4 . S 2 3 . 8 7 1 4 . 7 4 2 . 6 8 1 4 . 3 0 3 . 0 3 1 2 . 3 2 2 . 7 5 C l * e * a a t e a 9 . 9 7 3 . 4 6 1 0 . 6 8 2 . 6 0 1 2 . 8 2 2 . 3 0 1 3 . 0 0 2 . 2 0 c r o v n u p * 1 1 . 1 5 4 . 1 5 1 1 . 0 6 3 . 5 6 1 3 . 3 6 2 . 6 1 1 3 . 8 4 2 . 1 0 S e l f C o n c e p t : M a 1 1 . 6 7 3 . 8 8 1 0 . 2 6 2 . 6 6 1 2 . 8 8 2 . 6 8 1 3 . 2 6 2 . 4 8 M y S c h o o l A b i l i t y 1 0 . 3 6 3 . 6 7 1 1 . 0 6 3 . 3 9 1 4 . 0 0 2 , 8 9 1 4 . 4 8 3 . 3 8 H o w I M o u l d L i k e t o R e 1 6 . 2 7 2 . 6 1 1 4 . 6 2 3 . 6 1 1 6 . 3 6 3 . 2 1 1 2 . 2 6 3 . 1 0 H o w M y C l e a t S e e l H a 1 1 . 6 0 2 . 7 8 1 1 . 1 0 2 . 8 8 1 2 . 6 1 3 . 1 1 1 4 . 3 5 3 . 2 1 A u t h o r i t y R e l a t i o n s h i p * ; A u t h o r i t y 9 . 9 7 3 . 7 3 9 . 3 5 3 . 3 1 1 1 . 6 4 5 . 0 5 1 1 . 9 7 5 . 1 5 R u l e * 9 . 0 3 3 . 5 3 9 . 3 5 2 . 6 8 1 1 . 5 2 5 . 3 3 1 1 . 3 5 5 . 3 1 P u n 1 a b s e n t 9 . 0 6 3 . 4 8 9 . 6 5 4 . 4 7 1 1 . 0 0 5 . 2 0 1 1 . 4 5 5 . 1 7 F u t u r e 9 . 4 5 2 . 7 4 8 . 5 8 2 , 9 4 1 0 . 8 5 3 . 7 4 1 0 . 5 2 3 . 3 8 C o l l e g e 8 . 6 1 3 . 8 1 8 . 3 9 3 . 4 4 1 1 . 1 2 4 . 3 1 1 0 . 8 7 4 . 5 7 A J o b 1 0 . 0 9 3 . 5 9 9 , 8 1 2 , 6 4 8 . 7 6 3 . 4 9 9 . 0 0 2 , 7 3 G r a d u a t i n g 9 . 2 7 3 . 5 5 9 . 3 9 2 . 8 0 1 1 . 2 4 3 . 6 9 1 1 . 1 3 4 . 2 7 Q u i t t i n g S c h o o l 1 4 . 7 3 3 . 6 1 1 3 . 9 7 4 . 2 5 1 1 . 7 0 - 4 . 6 1 1 1 . 9 4 3 . 8 7 M o r a l a n d S o c i a l W m t : T r y i n g B a r d C h a t t i n g S o a a t h l n g E a s y S o a a t h l n g I m p o r t a n t N o n a y S u c c e e s 9 . 5 8 3 . 1 7 9 . 1 3 3 , 4 1 1 2 . 4 8 3 . 4 3 1 1 . 8 7 3 . 6 6 1 4 . 5 5 3 . 9 6 1 5 . 8 7 3 . 2 1 1 2 . 8 2 3 . 9 8 1 2 . 1 6 3 . 4 7 1 2 . 9 1 4 . 3 0 1 4 . 0 3 3 . 6 9 1 0 . 3 9 2 . 9 3 1 1 . 8 7 4 . 3 8 9 . 2 4 3 . 0 7 7 . 7 4 3 , 1 4 1 1 . 0 9 3 . 3 8 1 0 . 8 4 4 , 8 0 9 . 3 3 4 . 4 6 8 . 9 0 4 . 3 6 8 . 6 7 3 . 9 9 8 . 9 4 3 . 9 8 8 . 3 9 4 . 3 1 8 . 2 6 3 . 3 4 9 . 6 1 1 . 7 3 9 . 7 8 4 . 5 9 AFFIMDIX c 385 T A B L E A S HEAK8 AES ST A H D A U D EVIATION S O F SUMMED A C T IV IT Y FACTOR SCORES FOR I LEVS MTU GRADE DROPOUTS AND S T A Y -IN S D r o p o u t * S t a y - I n * C o o c a p t » ° y G l r l a B o y * C t r l * M a a n S . D . M e a n S . D , K e a n S . D . H a a o S . D , T e a c h e r a 9 . 6 4 4 . 6 6 9 . 5 3 3 . 7 6 1 3 . 7 6 3 . 1 4 1 4 . 5 5 3 . 4 4 G r a d * * 9 . 6 1 4 . 3 6 1 0 . 3 9 3 . 3 7 1 3 . 1 8 2 . 5 0 1 2 . 5 5 3 . 1 8 S c h o o l S . S B 3 . 4 4 1 0 . 2 9 4 . 1 5 1 5 . 4 5 4 . 0 6 1 4 . 6 5 4 . 1 8 R e a d i n g 1 0 . 3 6 5 . 3 9 1 0 . 9 4 4 . 1 6 1 2 . 4 8 4 . 8 9 1 4 . 1 6 4 . 5 9 I d e a l T e a c h e r 1 7 . 3 0 4 . 0 8 1 7 . 4 8 3 . 6 3 1 7 . 4 5 3 . 2 3 1 8 . 1 0 2 . 1 8 F a w l l y R e l a t i o n a h l p a : B o m 1 0 . 5 5 4 . 7 1 1 1 . 6 8 4 . 6 9 1 4 . 2 7 3 . 4 7 1 4 . 3 9 3 . 9 5 I d e a l P a r e n t 1 7 . 6 4 3 . 9 2 1 8 . 2 3 1 . 9 2 i 7 . 9 1 2 . 4 4 1 7 . 6 8 2 . 2 1 F a r * a t * 9 . 6 1 3 . 9 3 1 2 , 1 3 4 . 6 7 1 5 . 1 2 3 . 0 3 1 4 . 5 2 3 . 9 6 S o c i a l I n t e r a c t i o n : M o a t P e o p l e 9 . 7 3 4 . 6 7 1 1 . 5 8 4 . 1 7 1 2 . 4 5 2 . 5 9 1 4 . 1 0 3 . 0 3 M y B o a t F r i e n d * 1 7 . 4 5 3 . 7 8 1 8 . 3 2 2 . 3 2 1 6 . 8 5 2 . 8 5 1 7 . 4 B 2 . 4 6 C l a c i u t e a 1 0 . 0 9 4 . 5 6 1 1 . 7 4 4 . 3 7 1 4 . 0 3 3 . 3 3 1 4 . 9 0 3 . 4 0 G r o w n u p * 9 . 4 2 4 . 1 5 1 0 . 0 0 4 . 1 7 1 2 . 4 8 3 . 6 3 1 3 . 2 3 3 . 7 2 S a i f C o n c e p t : M e 1 2 . 3 3 5 . 3 7 • 1 2 . 7 1 5 . 3 6 1 5 . 2 7 3 . 7 0 1 5 . 3 2 2 . 8 2 M y S c h o o l A b i l i t y 9 . 8 2 4 . 3 4 1 2 . 1 0 4 . 4 2 1 4 . 5 2 2 . 9 2 1 5 . 0 3 3 . 6 0 B o w I W o u l d L i k e t o B e 1 6 . 9 4 2 . 4 6 1 9 . 2 9 2 . 0 5 1 9 . 1 5 3 . 1 4 1 9 . 4 8 1 . 9 4 B o w M y C l a a a S e e * H e 9 . 4 8 2 . 7 7 1 0 . 1 6 3 . 0 8 1 2 . 9 4 4 . 8 4 1 2 . 8 7 4 . 7 3 A u t h o r i t y R e l a t i o n s h i p * : A u t h o r i t y 1 0 . 2 7 3 . 4 9 9 . 9 4 3 . 4 0 1 4 . 3 3 5 . 8 6 1 2 . 9 7 5 . 3 1 R u l e s 1 0 . 5 8 3 . 6 7 1 0 . 0 3 2 . 4 4 1 5 . 4 2 4 . 7 9 1 4 . 4 2 4 . 1 0 P u n i t h o a u t 1 0 . 7 9 3 . 1 3 1 1 . 6 1 3 . 6 8 1 4 . 1 2 4 . 6 3 1 5 . 4 3 4 . 3 2 G o a l O r i e n t a t i o n : F u t u r e 7 . 8 5 3 . 0 1 8 . 4 2 3 . 6 1 1 0 . 5 8 4 . 4 9 1 0 . 0 3 6 . 0 0 C o l l a g e 9 . 0 6 4 . 6 4 8 . 0 3 3 . 0 2 1 3 . 1 5 4 . 9 8 1 2 . 6 5 5 . 2 1 A J o b 7 , 7 0 3 . 4 5 7 . 2 5 2 . 5 4 7 . 3 6 3 . 7 6 6 . 7 1 2 . 2 8 C r a d u a t l o g 7 . 5 2 4 . 7 8 7 . 5 2 3 . 2 4 1 0 . 5 2 3 . 6 4 1 0 . 0 6 3 . 5 7 Q u i e t i n g S c h o o l 1 7 . 6 1 3 . 9 1 1 8 . 2 6 3 . 5 3 1 2 . 2 4 5 . 8 0 1 2 . 6 8 5 . 6 3 M o r a l e n d S o c i a l V a l u e * : T r y i n g B a r d 9 . 4 8 3 . 2 0 9 . 3 5 3 . 0 8 1 3 . 5 5 3 . 7 5 1 2 . 3 3 3 . 8 9 C h e a t i n g 1 5 . 9 1 4 . 4 4 1 7 . 2 3 3 . 7 4 1 3 . 8 2 4 . 5 8 1 3 . 2 9 4 . 2 0 S o a a t h l n g E a s y 8 . 4 8 3 . 2 5 9 . 8 7 4 . 2 2 9 . 1 5 3 . 8 7 8 . 1 9 3 . 0 8 S o M t h l o g l a p o r t a n t 6 . 1 2 3 . 0 5 5 . 9 7 2 . 8 0 1 1 . 6 7 3 . 4 1 1 0 . 5 3 4 . 7 9 H o n e y 7 . 3 9 3 . 3 1 7 . 9 0 3 . 6 0 6 . 1 2 3 . 1 6 5 . 7 4 2 . 4 4 S u e c e a a 7 . 1 5 3 . 6 4 7 , 3 2 2 . 9 9 9 . 0 3 4 . 6 3 9 . S B 4 . 2 6 APPENDIX D MEANS AND STANDARD DEVIATIONS OF SUMMED FACTOR SCORES FOR THE TREATMENT GROUPS, GRADE LEVEL AND SEX OF THE STUDENT 387 APPENDIX D T A B U 4 9 KEANS AMD STANDARD DEVIATIONS O F SUMMED FACTOR SCORES FOR CONCEPTS RATED A G AINST SELECTED SCALES I T DROPOUTS AND S T A Y -IN S C o n c e p t D r o p o u t * S t a y - I n * Evaluative P o t e n c y Activity E v a l u a t i v e P o t e n c y Activity M e a n S . D . H a a n S . D . M e a n S . D . K e a n S . D . M a a n S . D . M e a n S . D . 8 . 2 2 4 . 4 6 1 0 . 8 8 3 . 0 0 9 . 6 4 3 . 8 6 1 4 . 5 5 3 . 2 9 1 2 . 7 3 2 . 1 4 1 4 . 1 6 3 . 0 6 8 . 5 6 4 . 1 0 1 0 . 3 8 3 . 2 2 9 . 7 7 3 . 5 1 1 3 . 6 2 3 . 5 2 1 3 . 3 4 2 . 7 8 1 3 . 1 0 3 . 0 1 7 . 8 0 4 . 4 0 1 1 . 8 0 4 . 0 0 9 . 5 0 4 . 0 5 1 4 . 7 4 4 . 4 8 1 5 . 5 6 3 . 2 6 1 4 . 9 7 4 . 1 0 1 0 . 6 2 5 . 3 6 1 1 . 4 4 3 . 8 5 1 0 . 6 9 4 . 5 2 1 5 . 0 3 4 . 2 1 1 3 . 6 3 3 . 4 3 1 3 . 9 8 4 . 5 7 1 6 . 0 6 3 . 6 6 1 4 . 0 3 3 . 8 3 1 7 . 2 4 3 . 6 4 1 9 . 3 2 2 . 6 0 1 3 . 8 7 4 . 5 0 1 8 . 0 4 2 . 7 6 S c h o o l E x p e r i e n c e * : T e a c h e r * G r a d e * S c h o o l R e a d i n g I d e a l T e a c h e r F a i a l l v R c l a t l o o l h l p g . H o n e I d e a l P a r e n t P a r e n t * 1 0 . 9 8 1 8 . 9 3 1 0 . 7 5 6 . 4 9 2.11 6 . 5 0 1 1 . 3 1 1 4 . 5 3 1 0 . 7 9 4 . 0 0 3 . 2 6 3 . 5 8 1 1 . 4 0 4 . 5 9 1 7 . 6 5 3 . 0 3 1 1 . 1 2 4 . 7 6 1 6 . 5 4 1 9 . 4 7 1 6 . 3 7 4 . 2 9 2.02 4 . 4 2 1 3 . 5 3 3 . 1 5 1 4 . 2 4 2 . 8 9 1 3 . 3 4 2 . 9 9 1 4 . 9 5 1 8 . 0 8 1 4 . 6 6 3 . 7 3 2 . 6 0 4 . 0 0 S o c i a l I n t e r a c t i o n : H o e t P e o p l e M y B e e t F r i e n d * C l e e f l a a t e * G r o w n u p * S e l f C o n c e p t ; H e M y S c h o o l A b i l i t y B o w I U o u l d L i k e t o B e B o w C l * * * S e e * H e A u t h o r i t y R e l a t i o n a h l p a ; A u t h o r i t y R u l e * F u n l a h a e n t C o a l O r i e n t a t i o n : F u t u r e C o l l e g e A J o b G r a d u a t i n g Q u i t t i n g S c h o o l M o r a l a n d S o c i a l V a l u e * : 1 0 . 0 3 1 7 . 6 2 9 . 9 9 9 . 2 5 11.21 1 0 . 3 5 1 8 . 3 4 8 . 4 6 10.66 11.22 1 4 . 8 3 5 . 3 1 3 . 1 3 5 . 3 3 5 . 1 7 5 . 6 6 4 . 9 9 2 . 8 7 2 . 7 0 3 . 8 8 3 . 9 7 3 . 2 0 1 0 . 2 9 1 4 . 7 3 1 0 . 4 3 1 0 . 9 8 1 0 . 8 7 1 0 . 5 6 1 5 . 0 5 1 1 . 3 5 9 . 5 0 9 . 2 8 9 . 9 2 3 . 1 9 3 . 1 4 2 . 7 7 3 . 5 0 3 . 3 9 3 . 3 0 3 . 1 9 2 . 9 4 3 . 6 9 3 . 1 3 3 . 9 3 1 0 . 3 3 4 . 1 0 1 7 . 2 7 3 . 3 0 1 0 . 5 1 4 . 1 6 1 0 . 0 6 3 . 9 0 1 2 . 0 6 4 . 8 0 1 0 . 2 6 4 . 3 0 1 8 . 7 4 2 . 2 8 9 . 5 5 3 . 0 7 9 . 9 5 3 . 3 4 1 0 . 2 4 3 . 1 5 1 1 . 7 9 3 . 2 7 1 5 . 0 6 1 8 . 2 2 1 5 . 9 0 1 4 . 5 7 1 6 . 6 4 1 5 . 9 3 1 9 . 9 5 1 3 . 0 7 1 6 . 3 4 1 6 . 9 2 1 8 . 1 8 3 . 2 3 2 . 7 5 3 . 0 8 4 . 0 5 3 . 1 3 3 . 1 0 2 . 1 5 4 . 9 8 4 . 2 6 3 . 8 2 3 . 7 1 1 2 . 7 3 2 . 1 0 1 3 . 8 1 3 . 0 0 1 3 . 1 5 2 . 3 5 1 3 . 5 0 2 . 9 6 1 3 . 1 4 3 . 1 9 1 4 . 4 5 3 . 3 8 1 4 . 8 5 3 . 9 2 1 3 . 0 5 2 . 9 4 1 1 . 8 2 4 . 6 0 1 1 . 9 3 4 . 6 1 1 1 . 1 6 5 . 0 2 1 3 . 8 2 1 7 . 5 2 1 4 . 7 8 1 3 . 3 0 1 6 . 2 1 1 5 . 3 1 1 9 . 5 1 1 3 . 2 1 1 3 . 9 2 1 4 . 7 7 1 4 . 9 0 3 . 1 6 2 . 5 7 3 . 4 4 3 . 9 1 2 . 9 4 3 . 3 5 2.20 4 . 4 7 4 . 8 9 4 . 1 8 4 . 3 2 6 . 7 6 3 . 5 1 8 . 8 0 3 . 1 2 7 . 7 6 3 . 5 2 1 0 . 1 4 5 . 5 3 1 1 . 0 4 3 . 6 1 1 0 . 8 7 4 . 9 8 8 . 3 7 4 . 2 2 8 . 5 5 3 . 9 4 8 . 5 7 4 . 2 5 1 3 . 3 4 5 . 1 2 1 1 . 7 2 4 . 0 4 1 3 . 2 1 4 . 6 4 6 . 5 9 2 . 7 9 9 . 4 8 3 . 3 2 7 . 3 4 3 . 0 0 5 . 6 1 2 . 6 1 8 . 9 7 3 . 0 2 7 . 0 5 2 . 8 0 5 . 5 6 3 . 4 8 8 . 8 1 3 . 2 6 7 . 0 7 3 . 7 4 1 0 . 3 8 6 . 1 4 1 1 . 4 5 3 . S B 1 1 . 0 5 5 . 1 4 1 8 . 6 6 3 . 8 2 1 4 . 6 7 4 . 2 3 1 7 . 8 2 3 . 8 8 1 1 . 5 8 6 . 9 4 1 1 . 6 1 4 . 0 2 1 2 . 0 2 5 . 4 3 T r y i n g B a r d 8 . 5 2 3 . 5 0 9 . 0 5 3 . 0 2 9 . 0 5 3 . 4 1 1 2 . 8 0 3 . 9 8 1 2 . 4 3 3 . 5 4 1 3 . 0 9 3 . 7 4 C h e a t i n g 1 8 . 4 9 3 . 5 8 1 4 . 6 1 4 . 0 9 1 6 . 1 8 4 . 2 6 1 4 . 1 4 5 . 4 4 1 2 . 4 8 3 . 7 5 1 3 . 5 8 4 . 3 7 S o a a t h l n g E a e y 7 . 0 5 3 . 6 3 1 3 . 2 9 4 . 3 6 8 . 8 4 3 . 6 1 7 . 0 4 3 . 2 6 1 0 . 9 8 3 . 8 1 8 . 7 6 3 . 4 3 S o a a t h l n g I m p o r t a n t 7 . 0 6 3 . 3 4 8 . 4 0 3 . 2 8 7 . 4 7 3 . 2 3 1 1 . 1 5 4 . 3 5 1 1 . 2 8 3 . 6 6 1 1 . 2 6 4 . 0 4 H o n a y 5 . 2 3 3 . 1 1 9 . 1 5 4 . 2 6 7 . 0 2 3 . 4 5 5 . 2 4 2 . 9 8 8 . 3 2 3 . 6 6 6 . 3 4 2 . 9 2 S u c c a a a 4 . 8 9 2 . 6 9 7 . 6 6 3 . 4 6 6 . 8 5 3 . 3 8 8 . 4 5 4 . 6 2 1 0 . 2 3 3 . 7 7 9 . 7 6 4 . 1 3 388 A P P E N D I X D T A B L E 5 0 MEANS AMD STANDARD DEVIATIONS OF SUMMER FACTOR SCORES FOR CONCEPTS RATED AGAINST SELECTED SCALES BT TENTH AND ELEVENTH GRADE STUDENTS T e n t h G r a d e E l e v e n t h G r a d e E v a l u a t i v e P o t e n c y A c t i v i t y E v a l u a t i v e P o t e n c y A c t i v i t y C o n c e p t M e a n S . D . M e a n S . D . M e a n S . D . K e a n S . D . M e a n S . D . M e a n S . D , S c h o o l g x p e r l e n c e e ; T e a c h e r . 1 1 . 6 6 4 . 8 3 1 1 . 7 9 2 . 8 2 1 1 . 9 3 3 . 9 2 1 1 . 3 6 5 . 2 3 1 1 . 6 1 2 . 6 9 1 1 . 8 8 6 . 3 6 G r a d e . 1 1 . 3 3 4 . 6 0 1 1 . 9 5 2 . 9 1 1 1 . 6 5 3 . 6 7 1 0 . 6 6 6 . 5 6 1 1 . 7 7 3 . 7 5 1 1 . 6 3 3 . 6 7 S c h o o l 1 1 . 1 0 5 . 7 7 1 3 . 7 2 6 . 1 7 1 2 . 2 3 6 . 9 5 1 1 . 6 5 5 . 6 9 1 3 . 6 6 6 . 0 3 1 2 . 2 3 6 . 8 6 B e a d i n g 1 2 . 9 9 5 . 5 0 1 2 . 6 1 3 . 6 6 1 2 . 6 9 6 . 7 1 1 2 . 6 6 5 . 0 8 1 2 . 2 6 3 . 9 5 1 1 . 9 7 6 . 9 3 I d e a l T e a c h e r 1 8 . 6 6 3 . 6 6 1 6 . 1 8 3 . 9 8 1 7 . 7 0 3 . 1 9 1 8 . 9 5 2 . 9 9 1 3 . 7 1 3 . 1 9 1 7 . 5 8 3 . 3 2 F a n t l y R e l a t i o n a h l p a : R o m e 1 4 . 3 9 6 . 0 3 1 2 . 4 7 3 . 8 5 1 3 . 6 2 4 . 5 5 1 3 . 1 0 6 . 2 4 1 2 . 3 7 3 . 7 0 1 2 . 7 1 4 . 4 9 I d e a l P a r e n t 1 9 . 1 6 2 . 0 3 1 4 . 3 0 3 . 4 1 1 7 , 8 7 2 . 9 4 1 9 . 2 4 2 . 1 4 1 4 . 5 2 2 . 6 9 1 7 . 8 6 2 . 7 2 P a r e n t s 1 3 . 5 9 6 . 1 8 1 2 . 1 2 3 . 7 4 1 2 . 9 6 5 . 0 0 1 3 . 5 3 6 . 2 7 1 2 . 0 0 3 . 3 0 1 2 , 8 3 4 . 4 5 S o c i a l I n t e r a c t i o n : H o s t P e o p l e M y B e e t F r i e n d s C l a s s m a t e s G r o w n u p s S e l f C o n c e p t : H e H y S c h o o l A b i l i t y H o w I M o u l d L i k e t o B e H o w H y C l a e s S e e s H e A u t h o r i t y R e l a t i o n s h i p s : A u t h o r i t y R u l e s P u n i s h m e n t C o a l O r i e n t a t i o n : F u t u r e C o l l e g e A J o b G r a d u a t i n g Q u i t t i n g S c h o o l M o r a l a n d S o c i a l V a l u e s ; 1 2 . 3 1 5 . 0 6 1 1 . 6 9 2 . 8 8 1 2 . 2 6 6 . 0 9 1 2 . 8 0 5 . 0 5 1 1 . 5 6 3 . 0 5 1 1 . 9 6 3 . 9 7 1 7 . 9 0 2 . 8 9 1 6 . 6 6 2 . 9 5 1 7 . 2 8 2 . 9 9 1 7 . 9 6 3 . 0 3 1 6 . 0 6 3 . 2 3 1 7 . 5 2 2 . 9 2 1 2 . 7 7 5 . 1 8 1 1 . 9 6 2 . 8 5 1 2 . 6 2 6 . 6 1 1 3 . 1 3 5 . 3 3 1 1 . 6 1 2 . 9 6 1 2 . 6 7 6 . 3 3 1 2 . 1 3 5 . 1 3 1 2 . 1 2 3 . 5 5 1 2 . 0 1 6 . 2 6 1 1 . 6 9 5 . 5 7 1 2 . 3 5 3 . 6 0 1 1 . 2 7 6 . 1 8 1 3 . 9 9 5 . 2 0 1 1 . 9 9 3 . 7 6 1 6 . 3 1 6 . 3 6 1 3 . 8 7 5 . 6 3 1 2 . 0 2 3 . 1 6 1 3 . 9 5 6 . 6 1 1 3 . 2 5 5 . 0 1 1 2 . 5 6 3 . 9 8 I"* . 7 2 6 . 6 5 1 3 . 0 3 6 . 9 9 1 2 . 6 7 - 3 . 7 6 1 2 . 8 6 6 . 3 6 1 8 . 9 9 2 . 7 2 1 6 . 9 7 3 . 6 6 1 9 . 0 6 2 . 1 2 1 9 . 3 1 2 . 5 7 1 6 . 9 0 3 . 5 1 1 9 . 2 1 2 . 6 2 1 0 . 6 1 6 . 6 1 1 2 . 0 0 2 . 9 1 1 1 . 6 0 6 . 2 9 1 0 . 9 2 6 . 8 2 1 2 . 6 0 3 . 1 9 1 1 , 3 6 6 . 2 0 1 3 . 1 6 6 . 8 0 1 0 . 5 9 6 . 2 1 1 1 . 9 9 6 . 3 2 1 3 . 8 6 5 . 1 1 1 0 . 7 3 6 . 6 5 c* CO r* 6 . 9 3 1 3 . 9 9 6 . 6 8 1 0 . 8 6 3 . 8 6 1 2 . 3 9 6 . 2 1 1 6 . 1 5 6 . 9 7 1 0 . 3 1 6 . 6 6 1 2 . 6 3 6 . 6 6 1 6 . 6 6 3 . 3 5 1 0 . 7 9 6 . 6 6 1 3 . 7 1 3 . 8 6 1 6 . 5 5 3 . 6 6 1 0 . 2 8 6 . 6 5 1 2 . 9 8 6 . 3 8 8 . 6 7 6 . 9 6 9 . 9 8 3 . 7 9 9 . 6 1 6 . 6 8 8 . 6 2 6 . 9 0 9 . 8 6 3 . 2 9 9 . 2 1 6 . 6 7 1 0 . 8 3 5 . 3 2 1 0 . 5 0 6 . 3 6 1 1 . 0 6 5 . 0 6 1 0 . 6 8 5 . 2 9 9 . 7 5 6 . 1 8 1 0 . 0 3 6 . 9 9 5 . 6 7 2 . 2 6 9 . 0 6 3 . 2 0 7 . 1 3 2 . 7 6 6 . 3 6 3 . 0 2 9 . 6 1 3 . 1 5 7 . 2 7 3 . 0 6 6 . 3 2 5 . 6 2 1 0 . 0 1 3 . 9 6 9 . 2 0 6 . 7 9 7 . 6 0 5 . 6 6 1 0 . 2 6 3 . 6 7 8 , 9 1 5 . 0 6 1 6 . 9 8 6 . 5 8 1 3 . 2 0 6 . 5 6 1 6 . 6 6 5 . 5 6 1 5 . 2 6 6 . 6 6 1 3 . 0 8 6 . 2 6 1 5 . 1 9 5 . 6 7 T r y i n g B a r d 1 0 . 6 1 6 . 6 8 1 0 . 7 0 3 . 7 3 1 0 . 9 5 6 . 3 0 1 0 . 7 1 6 . 1 2 1 0 . 7 7 3 . 6 5 1 1 . 1 9 3 . 8 9 C h e a t i n g 1 5 . 9 8 5 . 1 6 1 3 . 2 6 6 . 1 9 1 6 . 7 2 6 . 5 1 1 6 . 6 6 6 . 9 9 1 3 . 8 6 3 . 8 9 1 5 . 0 5 6 . 6 8 S o m e t h i n g E a . y 6 . 8 9 3 . 2 3 1 1 . 9 9 6 . 6 5 1 8 . 6 8 3 . 6 1 7 . 2 1 3 . 6 5 1 2 . 2 8 6 . 0 3 8 . 9 2 3 . 6 3 S o M t h l n g I m p o r t a n t 9 . 2 9 6 . 2 1 9 . 9 6 3 . 6 6 9 . 6 1 6 . 0 7 3 . 9 1 6 . 5 6 9 . 7 6 3 . 8 3 9 . 1 0 6 . 1 5 H o n e y 5 . 3 7 3 . 0 8 6 . 5 2 3 . 8 3 6 . 5 8 3 . 1 9 5 . 6 0 3 . 0 1 8 . 9 6 6 . 1 6 6 . 7 9 3 . 2 3 S u c c e a a 6 . 6 7 6 . 3 1 8 . 7 8 3 . 6 2 8 . 3 0 6 . 0 9 6 . 6 7 6 . 1 7 9 . 0 1 6 , 0 2 8 . 3 1 6 . 0 0 389 A P P E N D I X D T A B L E S I KEANS AND STAND AND DEVIATIONS OF SUMMED FACTOR SCORES FOR CONCEPTS RATED AGAINST SELECTED SCALES BT ROTS AND GIRLS ________________________________B o y a ________________________________ ______________________________C 1 r I » _____________________________ E v a l u a t i v e P o t e n c y A c t i v i t y E v a l u a t i v e P o t e n c y A c t i v i t y Concept ■— .... .— -.I. — ..— i *— — — — — ■ ■ i i - — ■ - ■ "» M e a n S . D . M e a n S . D . M e a n S . D . M e a n S . D . K e a n S . D . M e a n S . D . S c h o o l E x p e r i e n c e s : T e a c h e r . 1 0 . 5 4 A . 9 9 1 1 . 3 9 2 . 9 0 1 1 . 5 7 A . 2 0 1 2 . 2 8 4 . 9 3 1 2 . 2 3 2 . 5 3 1 2 . 2 5 A . 0 7 G r a d e . 1 0 . 7 0 A . 5 8 1 1 . 5 9 3 . 5 0 1 1 . A 9 3 . 9 1 1 1 . 5 0 4 . 5 3 1 2 . 1 5 3 . 1 6 1 1 . 3 8 3 . 3 9 S c h o o l 1 0 . 7 5 5 . 6 1 1 3 . 5 3 A . 2 2 1 1 . 7 A 5 . 0 5 1 1 . 8 2 5 . 6 1 1 3 . B A 3 . 9 7 1 2 . 7 5 4 . 7 0 R e a d i n g 1 1 . S B 5 . 3 5 1 2 . 1 6 3 . 8 5 1 1 . 7 8 A . 8 2 1 3 . 7 1 5 . 1 0 1 2 . 9 2 3 . 7 2 1 2 . 9 1 A . 7 8 I d e a l T e a c h e r 1 3 . 0 7 3 . 6 7 1 A . 0 1 3 . 8 2 1 7 . 0 1 3 . 6 9 1 9 . 3 A 2 . 5 6 1 3 . 8 8 3 . 4 1 1 8 . 2 0 2 . 6 2 Family Relationahlpa: H o m e I d e a l P a t e n t p a r e n t ! S o c i a l I n t e r a c t i o n : M o a t P e o p l e M y B e a t F r i e n d * C l e * d « a t e a G r o w n u p * S e l f C o n c e p t : M e M y S c h o o l A b i l i t y . H o w I W o u l d L t V e t o B e H o w M y C l a a a S e e s M e 1 3 . 4 2 6 . 2 7 1 2 . 6 6 3 . 7 3 1 8 . 6 9 2 . 3 3 1 5 . 0 8 2 . 9 9 1 3 . 0 1 6 . 2 9 1 2 . 4 1 3 . 8 5 1 1 . 8 6 A . 9 2 1 1 . 5 9 3 . 1 3 1 7 , 3 2 2 . 9 3 1 4 . 8 3 3 . 1 1 1 2 . 3 0 5 . 3 3 1 1 . 5 7 3 . 0 8 1 1 . 3 4 5 . 2 8 1 2 . 0 8 3 . 7 8 1 3 . 6 0 5 . 5 1 1 2 . 4 8 3 . 6 2 1 2 . 4 5 5 . 1 7 1 2 . 4 2 3 . 9 7 1 8 . 5 8 2 . 2 9 1 6 . 4 2 3 . 0 1 1 1 . 0 1 4 . 7 7 1 1 . 8 8 3 . 0 7 1 2 . 9 1 4 . 5 5 1 4 . 1 1 6 . 0 3 1 7 . 6 9 3 . 3 1 1 9 . 7 3 1 . 6 2 1 2 . 4 2 4 . 6 3 1 4 . 1 3 6 . 1 1 1 1 . 4 7 4 . 0 8 1 3 . 2 6 5 . 1 1 1 7 . 2 3 3 . 1 2 1 8 . 5 4 2 . 8 4 1 2 . 0 0 4 . 4 7 1 3 . 6 2 5 . 1 1 1 1 . 2 7 4 . 2 6 1 2 . 5 2 5 . 3 7 1 4 . 0 9 4 . 8 3 1 4 . 2 7 5 . 0 9 1 2 . 2 9 4 . 7 0 1 3 . 8 7 4 . 7 2 1 8 . 9 8 2 . 5 3 1 9 . 7 4 2 . 2 0 1 1 . 3 1 4 . 4 9 1 0 . 5 1 4 . 4 4 1 2 , 1 7 3 . 7 8 1 3 . 4 5 4 . 5 2 1 3 . 6 9 3 . 0 1 1 8 . 0 5 2 . 2 2 1 1 . 7 0 3 . 1 3 1 3 . 3 8 4 . 8 0 1 1 . 4 3 2 . 7 8 1 2 . 7 7 3 . 8 9 1 3 . 6 7 2 . 9 8 1 7 . 5 6 2 . 7 7 1 2 . 0 2 2 . 7 0 1 3 . 3 1 4 . 1 7 1 2 . 4 0 3 . 1 3 1 2 . 0 9 4 . 1 6 1 1 . 5 2 3 . 2 5 1 4 . 1 8 4 , 1 0 1 2 . 6 1 3 . 7 6 1 3 . 3 0 4 . 4 5 1 3 . 3 8 3 . 4 6 1 9 . 2 8 1 . 9 6 1 2 . 5 4 3 . 0 1 1 1 . 4 5 3 . 9 8 A u t h o r i t y R e l a c l o n a h l p * ; A u t h o r i t y 1 4 . 0 1 5 . 1 6 1 0 . 0 8 4 . 4 0 1 2 . 2 5 4 . 7 3 1 2 . 9 7 4 . 7 0 1 1 . 5 3 4 . 2 5 1 1 . 6 2 4 . 5 0 R u l e * 1 4 . 6 6 4 . 8 9 1 0 . 8 0 4 . 3 6 1 2 . 9 9 4 . 5 8 1 3 . 4 5 4 . 6 8 1 0 . 4 0 3 . 9 3 1 1 . 9 9 4 . 0 1 P u n l a h a e n t 1 6 . 5 4 3 . 3 1 1 0 . 5 3 4 . 4 5 1 3 . 0 1 4 . 1 1 1 6 . 4 7 3 . 5 1 1 0 . 5 5 4 . 6 6 1 3 . 7 1 4 . 1 2 C o a l O r i e n t a t i o n : F u t u r e C o l l e g e A J o b G r a d u a t i n g Q u i t t i n g S c h o o l M o r a l a n d S o c i a l V a l u e a ; T r y i n g H a r d C h e a t i n g S o m e t h i n g E a s y S o m e t h i n g I m p o r t a n t M o n e y S u c c e a a 9 . 3 5 4 . 9 0 1 0 . 1 5 3 . 7 4 1 1 . 1 6 5 . 3 9 1 0 , 3 2 4 . 3 9 6 . 4 2 2 . 9 8 8 . 9 7 3 . 4 1 8 , 5 5 5 . 8 5 9 , 9 4 4 . 0 7 1 4 . 5 0 6 . 7 7 1 3 . 2 2 4 . 7 1 1 0 . 9 0 4 . 4 7 1 0 . 9 8 3 . 7 3 1 5 . 6 1 5 . 1 8 1 3 . 4 4 4 . 1 6 7 . 0 1 3 . 2 6 1 1 . 3 1 4 . 0 8 9 . 1 2 4 . 2 0 9 . 9 7 3 . 5 0 5 . 3 4 3 . 0 0 8 . 6 6 4 . 1 2 7 . 3 1 4 . 5 4 9 . 0 2 3 . 8 9 9 . 7 1 4 . 4 1 7 . 5 0 4 . 7 8 1 1 . 0 4 5 . 2 2 1 0 . 5 4 5 . 2 0 7 . 2 7 3 . 1 2 5 . 7 6 2 , 4 2 9 . 2 6 5 . 2 4 7 . 3 6 5 . 1 3 1 4 . 5 8 5 . 7 6 1 5 . 7 7 6 . 4 0 1 1 . 2 8 4 . 2 9 1 0 . 4 1 4 . 1 2 1 4 . 4 8 4 . 7 4 1 7 . 0 5 4 . 8 9 8 . 5 7 3 , 3 8 7 . 0 9 3 . 6 2 9 . 7 9 3 . 9 6 8 . 4 6 4 . 4 7 6 . 5 4 3 . 2 9 5 . 4 2 3 . 1 0 8 . 2 2 4 . 3 1 6 . 0 0 3 . 7 9 9 . 6 7 3 . 3 3 8 . 9 0 4 . 7 1 9 . 9 4 4 . 1 8 1 0 . 7 4 4 . 7 9 9 . 4 9 2 , 9 0 7 . 1 2 2 . 6 5 1 0 . 3 3 3 . 5 3 8 . 8 4 4 . 5 3 1 3 . 0 6 4 . 0 6 1 5 . 2 7 5 . 2 6 1 0 . 4 8 3 . 6 5 1 0 . 8 4 3 . 9 0 1 3 . 0 5 3 . 9 5 1 5 . 3 0 4 . 1 9 1 3 . 0 0 4 . 2 5 9 . 0 4 3 . 6 5 9 . 7 1 4 . 0 0 8 . 9 1 4 . 2 4 8 . 0 1 3 . 8 5 6 . 8 3 3 . 1 2 8 . 0 6 3 . 7 5 8 . 4 0 3 . 7 4

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The Effects Of Delay In Knowledge Of Results On The Amount Learned In Teaching Machine Programs Of Differing Cue Content

PDF

A Methodological Investigation Of Affect Response Bias

PDF

An Analysis Of The Role Of The School Psychologist In The State Of California

PDF

A Study Of The Effectiveness Of Teaching Methods Of Study To Selected High School Freshmen

PDF

The Rehabilitation Of School Dropouts Through An Intensive Summer School Program

PDF

The Relationship Of Creative Thinking Ability, Intelligence And School Performance

PDF

Validity Concomitants Of Various Scoring Procedures Which Attenuate The Effects Of Response Sets And Chance

PDF

Behavioral Seriousness And Impulse-Control Balance In Delinquency

PDF

To Attend Or Not To Attend College: Some Factors In The Decision Of Qualified High School Graduates

PDF

Prediction Of Success In Training Among Electronics Technicians

PDF

A Study Of The School Achievement And Adjustment Of Children From One-Parent Homes

PDF

The Stability Of The Self-Concept In Junior College Students

PDF

The Connotative Perceptions Of Data Processing By Public School Personnel

PDF

An Analysis Of The Effectiveness Of Orientation Techniques In Elementary Schools In Southern California As Reported By Selected Teachers And Principals

PDF

A Study To Compare The Effectiveness Of Individual And Group Counseling Approaches With Able Underachievers When Counselor Time Is Held Constant

PDF

Upperclassmen As Academic Advisers To Freshmen In An Undergraduate College: An Experiment

PDF

An Empirical Comparison Of The School And College Ability Tests And The American Council On Education Psychological Examination

PDF

Non-Cognitive Factors Related To Achievement Which Can Be Ascertained From Freshman Autobiographies

PDF

An Experimental Study Of Self-Confrontation In Counseling

PDF

A Comparison Of Students' Evaluation Of Guidance Services Under Varying Plans Of Organization

Asset Metadata

Creator Strem, Bruce Elwood (author)

Core Title An Attitude Survey Of High School Dropouts By Means Of The Semantic Differential Process

Degree Doctor of Philosophy

Degree Program Education

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag education, educational psychology,OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Metfessel, Newton S. (committee chair), Carnes, Earl F. (committee member), Lefever, David Welty (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-222866

Unique identifier UC11359993

Identifier 6611592.pdf (filename),usctheses-c18-222866 (legacy record id)

Legacy Identifier 6611592.pdf

Dmrecord 222866

Document Type Dissertation

Rights Strem, Bruce Elwood

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA