A COMPARATIVE STUDY OF SELF-CONCEPT OF ABILITY BETWEEN I-NSTITUTIONALIZED DELINQUENT BOYS AND NON-DELINQUENT BOYS ENROLLED IN PUBLIC SCHOOLS Thesis for {he Degrao of Ph. D. MICHIGAN STATE UNIVERSITY David Livingstone Haarer 1964 1Hegs IIII IIIIII III IIII III IIIIII IIII III III III IIIII III IIII IIII IIII 321293 10277 This is to certify that the thesis entitled A COTTAZATIVE STUDY OF SELF-COIFCEI‘T OF :‘i'F'ILITY BETWEEN Ii‘l’fiTITU‘IOIIALIZED DELII‘IQUEIT’I‘ BOYS AND I\ION-DBLII‘IQUEI‘I‘I‘ BOYS EIIROLLED IN PUBLIC SCHOOLS presented by David Livingstone Haarer has been accepted towards fulfillment of the requirements for Ph.D. Education degree in 4%MAA11 ffilé‘fizdeK/ ajor professor Date July 21, 136).; 0-169 LIBRARY ' Michigan State University ABSTRACT A COMPARATIVE STUDY OF SELF-CONCEPT OF ABILITY BETWEEN INSTITUTIONALIZED DELINQUENT BOYS AND NON-DELINQUENT BOYS ENROLLED IN PUBLIC SCHOOLS by David Livingstone Haarer The purpose of this investigation was to make a systematic study of the relationships between self-concept of ability and classroom achievement among ninth—grade public school non-delinquent male students and ninth-grade institutionalized delinquent boys enrolled in an academic program. Further, a systematic comparison was made between the delinquent and non-delinquent students in those factors investigated relating to self-concept of ability and class room achievement. The non-delinquent sample consisted of one-hundred ninth—grade male students in one Midwestern metropolitan school system. The delinquent sample consisted of one- hundred ninth-grade male students-~fifty from a state training school and fifty from a private school for delin- quent boys. The main research instrument was the Michigan State Self-Concept of Ability Scale developed by Brookover and others. Correlational analysis and t-tests were the main statistical techniques. David Livingstone Haarer Basing the major thesis of this investigation on the phenomenological approach to learning, it was postulated that self-concept is established early in life in an inter— personal setting and modified by subsequent experience, and that the learner tends to evaluate himself as he perceives others to evaluate him, and finally, that a learner's self- concept of ability is a functionally limiting or facili- tating factor in classroom achievement. It was also postulated that certain deviations characteristic of delinquents have delimiting effects on self-concept develop- ment. The major thesis was tested in the form of five Specific hypotheses: l. The mean self—concept of ability score of non- delinquent boys is higher than the mean self-concept of ability score of delinquent boys. 2. The self-concepts of ability of delinquent and non-delinquent male students are related to their achieve- ment when intelligence is controlled. 3. The self~concepts of ability in specific school subjects of delinquent and non—delinquent boys vary from one subject to the other and from their general self-con- cepts of ability. 4. The expectations of significant others as per- ceived by both delinquent and non-delinquent boys are positively related with the students' self-concepts as learners. 5. David Livingstone Haarer The expectations of significant others as perceived by delinquent male students differ significantly from the expectations of significant others as perceived by non-delinquent male students. All of the hypotheses were found to be tenable. The major results of this investigation may be summarized as follows: 1. Non-delinquent ninth-grade male students have more positive self-concepts of ability than delinquent ninth-grade students. Self-concept of ability is significantly related to classroom achievement of delinquent and non— delinquent ninth—grade male students when the effect of measured intelligence is controlled. Self-concept of ability is weighted higher than IQ as a predictor of achievement for both ninth- grade public school male students and ninth—grade institutionalized delinquent boys. IQ alone is not a reliable predictor of classroom achievement for ninth-grade delinquent boys. Self-concepts of ability in specific school sub- jects of delinquent and non-delinquent boys vary from one subject to the other and from their general self-concepts of ability. David Livingstone Haarer Expectations of significant others as perceived by both delinquent and non-delinquent boys are positively related with the students' self- concepts as learners and with their classroom achievement. These relationships tend to be greater for non-delinquent students. The expectations of significant others as per- ceived by non-delinquent students are higher than expectations as perceived by delinquent students when parents, teachers, and peers are identified as significant others. For delinquent students, the perceived expecta- tions are consistently higher than the actual evaluations by significant teachers, houseparents, and counselors. COpyright by DAVID LIVINGSTONE HAARER 1965 A COMPARATIVE STUDY OF SELF-CONCEPT OF ABILITY BETWEEN INSTITUTIONALIZED DELINQUENT BOYS AND NON-DELINQUENT BOYS ENROLLED IN PUBLIC SCHOOLS By David Livingstone Haarer A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY College of Education 1964 ACKNOWLEDGMENTS The investigator would like to acknowledge his deepest gratitude to Dr. James M. Crowner for his assis- tance throughout the doctoral program and for his interest and suggestions in the development of this thesis. Appreciation is also expressed to Drs. Wilbur Brookover and Jean leEere for their help with the theoretical basis and methodological design, and to Drs. Henry Gottwald, William Roe, and Marian Kinget for their assistance and helpful criticism of the entire research. Further appreciation is expressed to Mr. Shailer Thomas, Research Assistant in the Bureau of Educational Research, Michigan State University for making available essential data and for his assistance with methodology. Acknowledgment is also given to Mr. Robert Lance for his priceless contribution of time and effort in assisting in the task of data gathering. The investigator is also indebted to Dr. Floyd Starr, President of the Starr Commonwealth for Boys, and Mr. Virgil Pinckney, Superintendent of the Boys Training School for their openness and assistance in gathering crucial data for the research. Gratitude is also expressed to the many staff members from both institutions for their cooperation and responses to questionnaires. ii The investigator is particularly indebted to his wife, Ann, and his three energetic sons, Carlton, Eric, and Brian, for their patience and considerations during the past year when this thesis monOpolized so many evenings and week-ends. iii TABLE OF CONTENTS ACKNOWLEDGMENTS . . . . . . . . . . VITA. O O O O O O O O O O I 0 LIST OF TABLES . . . . . . . . . . . . LIST OF APPENDICES . . . . Chapter I. II. III. IV. INTRODUCTION . General Statement of the Problem. Specific Statement of the Problem Research Hypotheses to be Tested. Importance of the Investigation Limitations of the Study . Plan and Content of this Thesis A REVIEW OF. THE LITERATURE Self- -Concept and Academic Achieve— ment . . . Self— -Concept and Delinquency . METHODOLOGY The Universe and Samples Operational Definitions of Terms and Research Instruments and Tech- niques . . Statistical Procedures RESEARCH FINDINGS . . . . . . . . Test of the Hypotheses Hypothesis Hypothesis Hypothesis Hypothesis Hypothesis Ul-IZTUJIUl-J iv Page ii vi vii ._J (Dm'flUlUl l--' 10 IO 15 18 18 Other Relevant Results . . . . . Identification of Significant Others . . . . . . . Comparison Between Expected and Actual Perception of Delin- quent Students' Abilities Comparative Analysis. . Differences Between the Delin- quent and Non—Delinquent Mean Scores for the Major Variables Difference Between Delinquent and Non-Delinquent Correlated Data V. SUMMARY AND CONCLUSIONS . . . . Summary of Research Objectives Summary of Research Findings . . . Implications of Research Findings Implications for Further Research BIBLIOGRAPHY . . . . . . . . . . . . . APPENDICES . . . . . . . . . Page AS 50 54 6O 62 67 25 7A 77 00 85 David Livingstone Haarer Candidate for the degree of Doctor of Philosophy Date of Examination: July 21, 1964, 11:00 A.M., College of Education Dissertation: A Comparative Study of Self—Concept of Ability Between Institutionalized Delinquent Boys and Non-Delinquent Boys Enrolled in Public Schools Outline of Studies: Major Area - Teacher Education, Special Education Minor Area - Psychology, Professional Education Biographical Items: Birthdate - March 25, 1930, Shipshewana, Indiana Undergraduate Studies - Goshen College, B.S., 1955 M.A., 1959, Ypsilanti, Michigan Graduate Studies - Eastern Michigan University, M.A., 1959, Ypsilanti, Michigan Michigan State University, 1960-1964, East Lansing, Michigan Experience: Starr Commonwealth for Boys, Albion, Michigan Teacher of socially maladjusted, 1955-1958 Starr Commonwealth for Boys, Albion, Michigan Director of Education, l958~l960. Ingham Intermediate Board of Education, Mason, Michigan, School Diagnostician, l963-present Membership held in Michigan Society of School Psychologists, Council for Children with Behavioral Disorders, Council for Exceptional Children vi Table LIST OF TABLES Mean Self— ~Concept of Ability Scores, Stand- ard Deviations, and t- -Tests With Probability of Difference Between Delinquent and Non- Delin— quent Ninth- -Grade Male Students . . Coefficients of Correlation Between Academic Grade Point Average (GPA), Measured Intelli- gence (IQ): and Self— Concept of Ability (3-0) for Delinquent and Non- Delinquent Male Students. . Correlation Coefficients, With and Without IQ Controlled, Between General Self—Concept of Ability and GPA; and Between Self-Concept of Ability in Specific Subjects and Grades in Each Subject for Ninth— —Grade Institution— alized Delinquent Boys and Ninth- -Grade Public School Non- -Delinquent Boys. . . . Correlation Coefficients, With and Without IQ Controlled, Between General Self-Concept of Ability and Total GPA; and Between Self-Con— cept of Ability in Specific Subject and Grades in Each Subjects for Ninth—Grade Delin- quent Boys From a Private Institution and a State Training School Mean Self-Concept of Ability Scores in all Subjects and for Each of the Given School Sub- jects for Institutionalized Delinquent and Public School Non- Delinquent Male, Ninth- -Grade Students- . - . - . - - Correlation Coefficients Between General Self- Concept of Ability and Specific Self-Concepts of Ability, and Between the Various Specific Self- ~Concepts of Ability for Delinquent and Non- Delinquent Males. . . vii Page 28 31 3A 35 . 38 AO IO. ll. 12. 13. 14. Coefficients of Correlation Between Gen- eral Self-Concept and General Achieve- ment (GPA) and Specific Subject Self- Concept and GPA for Institutionalized Delinquent and Public School Non-Delinq- uent Ninth—Grade Male Students . . . . Coefficients of Correlation Between Stu— dents' General Self-Concept of Ability and the Images They Perceive Significant Others to Hold of Their Abilities for Both Delinquent and Non- Delinquent Male Ninth- Grade Students Coefficients of Correlation Between Stu- dents' Self-Concept of Ability in Specific School Subjects and the Images They Perceive Their Parents to Hold of Their Abilities in These Subjects for Delinquent and Non- —Delinquent Ninth- Grade Students . . . . - . . . . . . Coefficients of Correlation Between the Students' Grade Point Averages (GPA) and the Images They Perceive Significant Others To Hold of Their Abilities for Delinquent and Non-Delinquent Ninth-Grade Male Students. Difference in Delinquent and Non-Delinquent Male Ninth- Grade Students' Mean Perceived Images of Ability Scores Held by Three Significant Others . . . . . Percentage of Students Naming at Least One Person From Each of the Following Categories as Being "Important in Their Lives." (Gen— eral Significant Others) . . . . . . Percentage of Students Naming at Least One Person From Each of the Following Categories as Bein Concerned with ”How Well They do in School Academic Significant Others) . Difference in Mean Perceived Images of Ability Held by Three Significant Others and Actual Perception of Students‘ Abilities by Three Significant Others for Institution— alized Delinquent Male Ninth-Grade Students. viii Page 42 AA 46 47 A9 51 52 56 15. l6. 17. 18. Coefficients of Correlation Between Stu- dents' General Self-Concept of Ability and the Perceived Evaluation of Teachers, Houseparents, and Counselors; and the Coefficients of Correlation Between General Self-Concept and Actual Evalua- tion by Teachers, Houseparents, and Counselors for Institutionalized Delin- quent, Male, Ninth-Grade Students. Coefficients of Correlation Between Per- ceived Evaluations of Teachers, House- parents, and Counselors, and Actual Evalua- tions by Teachers, Houseparents, and Counselors for Institutionalized Delinquent Male, Ninth-Grade Students . . . . . Means, Standard Deviations, and t-Test Be- tween Scores for all Major Variables for the Institutionalized Delinquent and Public School Non- -Delinquent, Male, Ninth- -Grade Students . . . . . . . Comparative Coefficients of Correlation and Z Tests for the Non- Delinquent and Delinquent Students ix Page 57 59 61 63 Appendix A. B. LISTS OF APPENDICES QUESTIONNAIRES TO IDENTIFY SIGNIFICANT OTHERS . . . . . . . . . . . SELF-CONCEPT OF ABILITY SCALES PERCEIVED EXPECTATION OF SIGNIFICANT OTHERS SCALES. . . . . . . . . EVALUATION OF SIGNIFICANT OTHERS SCALE CORRELATION MATRICES OF MAJOR VARIABLES Page 85 88 95 107 110 CHAPTER I INTRODUCTION General Statement of the Problem In recent years there has been a growing interest in the role of self—concept as related to learning. At— tempts have been made to study the learning process of a child through an understanding of his perceptual field. In the perceptual frame of reference, how the child perceives his life experiences, how things seem from his point of View, are considered important variables in the learning -process. The perceptual psychologists have a new approach to the problem of intelligence. Combs and other (2, 3, 12, 13) postulate that the capacity for intelligent behavior is de- pendent upon the state of the child's perceptual field. If the child's perceptions are rich, extensive, and readily available when he needs them, he is likely to behave in an efficient, effective, "intelligent" manner. If a person is threatened, if he is confronted with situations he is unable to cope with, his perceptions are hindered, narrowed, not readily available when he needs them and he is likely to respond in ”unintelligent” ways. Assuming that human capacities are functions of perceptions, it can be hypothesized that human capacities are not as limited as we have been inclined to think. Attitudes toward self are acquired and deveIOped in an interpersonal setting (15, 27, 41). A basic contribution of Mead (27) and Cooley (15) lay in their emphasis upon the influence of the responses of others in shaping self- conceptions. It is assumed that appropriateness of behavior and self perceptions are defined and formed through the internalization of the expectations of significant other such as parents, teachers, and peers (6, 16, 41). A person will tend to evaluate himself as he perceives others evaluate him. Sullivan has stated that "the self may be said to be made up of reflected appraisals." (41:10) Brookover (6) further postulates that the learner learns to do what he deems desirable or appropriate; that the learner's functional limits of his ability to learn are determined by his own conceptions of his abilities to learn as acquired in social interaction; and that the learner "learns what he believes significant others expect him to learn in the classroom and other situations." (6:86) Granting the above, the implication for classroom learning is that the learner's self‘concept of his learning iibility is a functionally limiting or facilitating factor in maximum academic achievement and the child who has an appropriate self-concept of learning ability learns more readily than the child who has an inapprOpriate self-concept. The above postulates assume that there is a casual relation- ship between self—concept and academic achievement, although the opposite may also be true, that is, high academic achieve- ment may produce a more positive or appropriate self-concept of ability. Which of these variables serves as the primary determiner may be somewhat difficult to ascertain, except theoretically, at the present time. It is highly possible that they reinforce each other in a more or less continual cycle. Present research by the Bureau of Educational Research, Michigan State University, under the direction of Brookover and others should provide empirical evidence on the possi— bility of changing students' self-concepts of ability and thereby raising their levels of performance. The perceptual approach to learning may have direct implications in the education of delinquent youth. One of the.major differences between delinquents and non-delin— quents is in the degree of school adjustment. Kvaraceus (23) lists some of the major deviations in school which are characteristic of the delinquent population: Poor or failure marks Repeater (retarded in grade) Strong dislike and hostility for school Truancy Intent to leave school early Vague or no educational-vocational goals Motivational problem Member of special class Has attended many different schools Destroys school material and property Does not feel he ”belongs" in classroom Does not participate in volunteer extracurricular school activities (23'367) Seriously and persistently misbehaving in school. ‘ Assuming that a person's self-concept develOps in an interpersonal setting and considering that delinquents characteristically have serious deviations in home, family, and neighborhood, it is no surprise that Kvaraceus lists as one of the characteristic deviations of delinquents as "low self~concept: a 'nothing' or 'less than nothing.'" (23:36?) Deviations, characteristic of delinquents, which may have a delimiting effect on self—concept development are summarized as: Contradictory social norms in home and/or neighborhood Identified with delinquent subculture Atypical fume structure (broken home) Interpersonal relationships in home wanting Economic stress, insecurity, and/or substandard economic conditions Lack of moral conformity—-spiritual values lacking; little or no nominal church contact Criminality pattern Culture conflicts Deteriorated neighborhood residence Discipline overstrict, punitive, erratic, lax Lack of cohesiveness Supervision by mother inadequate or unsuitable<2 . 6 Affection of parents indifferent or hostile.” 3'3 7) Basic to the problem of delinquency is an understand- ing of the delinquent's school failure, lack of interest in education, and early drop-out from school. If the phenomen- ological approach to learning is accepted, there needs to be an understanding of the delinquent's self-concept of ability and its relation to school achievement. Does the delinquent's self-concept of ability differ significantly from the self-concept of ability of the non-delinquent, and if so, could this difference be an important variable in considering the differences in academic achievement and total school adjustment? Specific Statement of the Problem The specific purpose of this investigation is to extend and compliment the research currently in process by the Bureau of Educational Research at Michigan State University through (1) a systematic study of the relation- ship between self—concept of ability and classroom achieve- ment among both ninth-grade non-delinquent boys enrolled in public schools and ninth-grade institutionalized delinquent boys enrolled in an academic program, and (2) a systematic descriptive comparison of the delinquent group with the non-delinquent group in those factors investigated relating to self-concept of ability and classroom achievement. Research Hypotheses to be Tested Hypothesis l.--The mean self-concept of ability score 0f non-delinquent boys is higher than the mean self-concept Of ability score of delinquent boys. Hypothesis 2.--The self-concepts of ability of delin- quent and non-delinquent male students are related to their achievement when intelligence is controlled. Hypothesis3.--The self-concepts of ability in spe- cific school subjects of delinquent and non-delinquent boys vary from one subject to the other and from their general self-concepts of ability. Hypothesisfifl.--The expectations of significant others as perceived by both delinquent and non-delinquent boys are positively related with the students' self-concepts as learners. Hypothesis 5.--The expectations of significant others as perceived by delinquent male students differ significantly from the expectations of significant others as perceived by non-delinquent male students. In addition to these five specific hypotheses, two specific questions were also investigated. Question l.--(a) Who are the relevant significant others to whom delinquent and non-delinquent boys relate themselves in examining their behavior as learners, and (b) do significant others differ for delinquent and non—delin- quent boys? Question 2.--Is there a discrepancy between how delinquent male students expect significant others to per- ceive their ability and how significant others actually Perceive their ability? Importance of the Investigation Although the comparative aspect of this investigation is primarily descriptive, it is felt that it will reveal the types of research questions that may have definite implica- tions in the educational planning of institutionalized de- linquent boys. It is believed that this study will afford concrete applications for those administrators, teachers, and related personnel concerned directly with the education of institutionalized delinquent boys. For example, if the phenomenological approach to learning is accepted that (l) the learner's self-concept of his learning ability is a functionally limiting or facilitating factor in maximum academic achievement, and (2) a learner tends to evaluate himself as he perceives others see him, then it may be possible by working through appropriate significant others (such as counselors, houseparents, and teachers) to elevate the self-concept of academic ability of institutionalized delinquent boys and consequently raise their level of classroom achievement. It may prove beneficial to identify the significant others for particular children and to work through these significant others by concentrating on ways of building more positive self-concepts of ability for delinquent students. Limitations of the Study Generalizations of the findings of this investigation must be made with extreme caution and need to be restricted to the social conditions and subjects very similar to those tested in this study. The non-delinquent male sample is not necessarily typical of all ninth-grade public school students since selective measures tended to insure the probability of a non-delinquent male sample. (Furthermore, the delinquent sample is not necessarily typical of all delinquents, since the sample was drawn from those ninth— grade delinquent boys enrolled in an academic program. This selective factor would tend to exclude mentally defective delinquents. All subjects were selected from one state in the Midwest. To generalize the findings of this study beyond these social conditions and the type of subjects‘ selected could prove misleading. Plan and Content of This Thesis, In this chapter the problem of the thesis has been introduced. Both general and specific statements of the problem have been presented. The major hypotheses and questions for consideration have been outlined. A brief discussion of the importance of the investigation was also made. Chapter II contains a selective review of the relevant literature. Chapter III deals with methodological procedures. Included in the chapter are: a brief description of the samples used in the investigation, the operational definition of terms and a description of the research instruments, a description of the statistics used to test the hypotheses, and a description of the method of comparative analysis. Chapter IV deals with the analysis of the data. Comparative data are presented and statistical tests of the hypothesis are made. This chapter provides a com- parison of the data obtained from the ninth-grade insti- tutionalized boys with that obtained from ninth-grade boys enrolled in public schools. In Chapter V the investigation is concluded. The research findings and theoretical implications are dis- cussed. CHAPTER II A REVIEW OF THE LITERATURE Self-Concept and Academic Achievement The only known research specifically devoted to the study of students' self—concept as learners in relation tetheir academic achievement has been that conducted at Michigan State University under the direction of Payne, Farquhar, and Brookover. Payne and Farquhar (30, 31) working on the assumption that a student's self—concept is a func- tionally limiting and facilitating factor in academic achievement which interacts with motivation, devised a 119-item instrument (The Word Rating List) to measure academic self-concept. Their data.indicated with a high degree of validity and reliability that it is possible to devise an objective, reliable, theory—derived measure of academic self-concept which discriminates significantly between underachieving and overachieving (high, low motivated) eleventh-grade high school students. Brookover and others (8) developed a simple, reliable, eight-question fixed-alternative (Guttman-type) scale of selfwconcept of ability which correlated approximately .57 With grade point averages of public school seventh graders. lO 11 With IQ partialled out, self—concept of ability scores still correlated VHJfll school grades at .42 for boys and .39 for girls (8:72). A cooperative research project between the U. S. Office of Education, Department of Health, Education and Welfare and the Bureau of Educational Research at Michigan State University (7) will hopefully develop procedures designed to change the self-concepts of low achieving junior high school students and thence their level of achieving. Other research, aimed more at a global self—concept rather than a self-concept of ability, has shown the relation of self-concept to scholastic performance. Two recent publications by the Association for Supervision and Curriculum Development (2, 3) emphasize the importance of selfvperception as a variable in human learning. Combs and Snygg (12, 13) have emphasized the phenomenological field as the important variable in behavior. Learning, or behavior, is considered a function of perception. Wylie (A3) reviewed several studies which provide some evidence of a relationship between self-concept and motivation to learn. Reeder, (34) Manis, (25) Helper, (20, 21) Miyamoto and Dornbusch, (29) Kipnis, (22) Burke, (11) and Videbeck (42) have reported findings indicating that an individual's self—ratings are significantly correlated with the ratings of him made by his associates. These findings lend support 12 to the Mead-Cooley symbolic interactionist framework holding that the conception of the self is an organization of socially derived and symbolically represented self- identifications (27, 15). Reader (34) using the interac- tionist framework as a basis for his analysis, found with military groups a direct relationship between self-concep- tion, the perceived generalized other, and the actual responses of others. In a study involving one hundred-one male students at the University of Illinois, Manis (25) found that the subjects' selfrconcepts were significantly influenced by their friends' opinions of them, particularly when they were perceived by these friends in a favorable light. Helper (21) found that correlations between parental evaluations and fifty-one eighth and ninth-grade students' self-evaluations tended to be small but consistently positive. Using one hundred ninty-five college students as subjects, Miyamoto and Dornbusch (29) found that the sub- jects' perceptions of the responses of others are positively related to self-conceptions. Using eighty-seven male students living together in a university dormitory, Kipnis (22) found that the subjects' self-concepts become more like the conceptions of their best friends. In an eXperi- mental study using forty-eight undergraduate students, Burke (11) found that self-concept and liking for others tended to be positively related with reactions from others. In Videbeck's (42) study with thirty students in an l3 introductory speech class, by experimentally varying the reactions of others and observing consequent changes, evidence was obtained to support the proposition that a person's organization of self attitudes are learned, and how reactions of others are perceived, play a significant role in the learning process. Reeder, (35) using grade children, found that positive feelings about the self are significantly related with good academic achievement. Stevens (40) working with college students concluded that three dimensions of the self-concept (self—insight, self-acceptance, and salience of personality traits) are related to academic achievement. In a doctoral thesis investigating the relationship between immature self—concept and certain educational disabilities, Bodwin (5) found a close association of immature self-concept with academic disability particularly in reading and arithmetic. His study further suggests that the more mature the self-concept the greater the facility in academic learning. His research group con- sisted of three hundred elementary students. One hundred had reading disabilities, one hundred had arithmetic dis- abilities and the remaining one hundred had no educational disabilities. Bruck and Bodwin (10) conducted a pilot study investi- gating the relationship between self—concept and the presence or absence of underachievement. Sixty children with average l4 IQ and evenly divided by sex were grouped into two sections: thirty children had no learning difficulties; thirty were underachievers. The Self-Concept Scale of the Machover Draw—A-Person Test (ACS—DAP), a projective tech- nique adopted and validated as a quantified measure of self-concept, was administered. The results indicated a positive and significant relationship between educational disability and immature self-concept. However, no cause and effect relationship between educational disability and immature self-concept was claimed. Davidson and Lang, (16) working with 89 boys and 114 girls attending fourth, fifth, and sixth grades in a New York public school, found that children's perceptions of their teachers' feelings toward them related to self- perception, school achievement and behavior. In the study at Michigan State University, Brookover (9) had also found that a student‘s self-concept of ability is positively related to the image he perceives significant others hold of him when parents, teachers and peers are identified as significant others. Related to these studies is evidence provided by Staines (39) that teachers, in their role as significant others, have influenced positive changes in self~perceptions when there was a definite and consistent attempt to help children explore and build self-concepts. 15 Self—Concept and Delinquency There is no known research dealing specifically with the self-concepts of delinquents as learners and the consequent relationship with achievement. However, there are several studies which have theoretical relevance to this investigation. In summarizing the studies of the Gluecks, Healy and Bronner, Kvaraceus, Merrill, Sheldon and others, and Wattenburg, Kvaraceus (23) has noted one of the significant characteristics of delinquents as "low self-concept: a 'nothing' or 'less than nothing.'” Research conducted at Ohio State University by Reckless and others (32, 33, 18, 19, 24) has provided both theoretical and empirical evidence suggesting that a socially appropriate self-concept is "insulation" against delinquency. They propose that an appropriate or inappro- priate self—concept is the basic component steering youth away from or propelling toward delinquency. They further postulate that this ”insulation" is both reflected in and is a reflection of the definitions of significant others in the lives of the non-delinquents. Balester, (4) in an experimental study of self- concept and juvenile delinquency, found significantly different self—concept scores between delinquents and non-delinquents. Using a Q sort (a personality inventory in which the subject sorts a considerable number of state- ments into piles that represent the degrees to which the 16 statements apply to him) he found that all individuals tend to have positive self-concept scores but the difference lay in relative positiveness, that is, maladjusted persons also had positive scores but of a lesser magnitude than adjusted persons. Deitch, (9) using the Tennessee Department of Mental Health Self-Concept Scale, likewise found that the self-concept scores significantly differentiated between delinquent and non-delinquent boys and that the magnitude of positive self-concept scores had some relationship to adjustment when delinquents were compared with non—delin- quents. Although delinquents as a group are underachievers it might be questioned whether their self-concepts of ability are realistic. In an experimental study, Amos (1) found that there was no difference between delinquent and non-delinquent boys in the accuracy with which they esti- mated their academic ability. Since the academic perfor- mance of delinquents is consistently lower than for non- delinquents, and if Amos' findings hold true for other populations, lower estimates of self-concept of ability might be predicted for delinquents than for non-delinquents. An investigation by Selden (38) suggested that con- tinued failure experiences, such as school failures, typi- cally encountered by delinquents are related to less favorable self-evaluations and lowered aspirations. Delinquent boys were found to differ in the direction 17 expected on such self-structure variables as self-concept, self-ideal, concept of others, and self-satisfaction. The studies reviewed above indicate the association of self-concept with performance and behavior. Although only two studies, the Payne and Farquhar, and Brookover studies, focused on the self-concept of ability as related to learning and academic achievement, other studies cited offer general support for the proposition that self—concept is related to academic performance and to the expectations of significant others. CHAPTER III METHODOLOGY The Universe and Samples The universe or population for the investigation consisted of ninth-grade male delinquent and non-delinquent students enrolled in an academic program in the urbanized and industrialized Midwest. Although the major focus of the analysis was intended to be upon the institutionalized delinquent boys in that universe, the hypotheses were tested using both delinquent and non-delinquent male subjects. The non-delinquent male sample (N=lOO) was drawn at random from ninth-grade students in one Midwestern metropolitan school system meeting the criteria discussed below. This school system has a public school enrollment (of approximately twenty—eight thousand. In order to be Ireasonably sure of a non-delinquent sample, most of these Students met the criteria of having been in the school Esystem for five years (i.e. since the fourth grade) and all had.been.enrolled in the system for at least two years. EStudents who did not have ninth—grade IQ scores available Were excluded from the study. The Califnrnia_flhsjuru;IEnial l8 l9 Maturity, a group intelligence test, had been administered during the ninth grade. The sample of ninth-grade institutionalized delinquent boys (NleO) was drawn from two sources. Fifty boys were selected from each of two residential institutions designed for the custody and treatment of delinquent boys. The first sample of fifty boys was selected from ninth- grade students enrolled in an academic program at a state training school located in the same metropolitan area as the non-delinquent sample. All ninth-grade students en— rolled in an academic program at this institution and present during a given period at a given date were used as subjects. Fifty-one subjects were thus selected, but since one was called out for a visiting permit during the period of questionnaire administration, he was eliminated from the sample, leaving a N of 50. The second sample of fifty delinquent boys was selected from all ninth-grade students enrolled in a private institution for delinquent boys located approximately fifty miles from the source of the other delinquent and non- delinquent samples. Since only forty-seven students were enrolled in the ninth grade at the private institution on the date selected, the first three new students to enroll in the ninth grade were also included as subjects, making the sample total N=50 to match the number selected from the state training school. 20 It was felt that neither of these institutions was alone representative of the institutionalized delinquent population and both had unique selective factors operating. The state training school tends to have delinquent boys as a last resort. That is, many of these boys have been in previous placement such as foster homes or private institu— tions, or they may tend to come from families of lower socio-economic areas where other provisions are less likely available. In contrast, the private institution used in this study tends to select delinquent boys who do not appear to need extensive and intensive psychiatric care and who show some promise of benefitting from the residential care offered at this institution. The private residential school tends to get more maladjusted boys from ”better” homes, or from homes of higher socio-economic status. The average length of stay is somewhat longer at the private than atthe state institution. Operational Definitions of Terms and Research Instruments and Techniques Self-Concept of Ability and/or Academic Self-Concept refers to the image or idea one has of himself in respect to his ability for academic achievement, and/or the eval- uation of the learner's capability, as a learner, as ex- pected from a certain person or group of persons. For the purpose of this study, the term was operationalized as the responses of a subject to the Michigan State Self-Concept 21 of Ability Scale,l a simple, reliable eight-question fixed-alternative scale (Guttman-type scale) with reproduci- bility of .95 for males and .96 for females as shown in a scalogram analysis made with 1,050 seventh-grade students. Reliability of the self-concept of ability scale determined by Hoyts' method was .82 for males and .77 for females. In validation studies on the same study group, predicted grade point average correlated with actual grade point average .70 for females and .71 for males. (9) ggneral SelfirConga t of Ability is the same as the self-concept of ability as described above. This term is used in contrast to specific self-concept of ability defined below. It is operationally defined as the score obtained by the learner on the Michigan State Self-Concept of Ability Scale, an eight'question, fixed-alternative scale designed to measure the subjects‘ self-concepts of ability in academic endeavors.2 Specific Self-Concept of Ability and/or Self-Concept of Abilityfiin Specific Subjects refers to the perception of ability within a given subject matter field or activity area. Operationally it is defined as the score obtained from the subjects' responses to the eight-item fixed-alternative self-concept of ability scale, asked with a change of ref- erence to specific subject areas. For the non-delinquent 1See Appendix B. 2See Appendix B. 22 sample, only reference to specific academic subjects was made, i.e., English, mathematics, social studies, and science.3 In addition to dealing with the four specific academic sub- jects, the same scale was applied to the areas of shop courses and physical education activities for the delinquent samples."I A Positive Self-Concept of Ability refers to those phenomenological perceptions of competence in general scholastic ability and/or in a given subject matter field and is operationally defined as a high score on the Michigan State Self-Concept of Ability Scale. Intelligence for both delinquent and non-delinquent samples was operationalized as recent scores obtained on standardized intelligence tests. The entire non-delinquent sample had been administered the California Test of Mental Maturity during the ninth grade. The averagescfl?verbal and non-verbal scores were utilized. Full scale scores obtained on a Wechsler Intelligence Sgalewfor Childr§g_or a Wechsler Adult Intelligence Scale (for children age sixteen and older)were available for the delinquent samples. Achievement, for the purposes of this investigation, was operationally defined as the average of a subject's school grades for the first semester of the ninth grade. ~ v v Y 7 1.. 3'See Appendix B. “See Appendix B. 23 The students' grades in the four basic subjects-~English, mathematics, science, and social studies-~were used in calculating this average (GPA). For the delinquent sample a second average was obtained by including non-academic courses in the calculation. Perceived Expectations of Significanththers refers to the images a student perceives significant others hold of his ability. Operationally it refers to a subject's responses to a series of questions designed to elicit the subject's perceived expectations as he believes significant others evaluate him.5 Protests indicated that the persons used in this study (parents, favorite teachers, and best friends) are most frequently mentioned by students as being important in their lives. In addition to parents, favorite teachers, and best friends, delinquent boys indicated on a pretest that counselors and houseparents were also important in their lives and were therefore in— cluded in the study as significant others. Significant Others' Perception of Student's Ability refers to the actual evaluation, made by other people im- portant in the student's life, of his abilities. Opera- tionally this refers to the responses of significant others on the Evaluation of Significant Others Scale, a scale Specifically designed for this study to be used in conjunction k v f fir w s—v —— 5See Appendix C. 24 with the delinquent sample to see if there is a discrepancy between how delinquents expect significant others to per- ceive their abilities and how significant others actually perceive their abilities. The scale is a paraphrased version of the Self-Concept of Ability Scale. Teachers, houseparents, and counselors of the delinquent boys were administered this scale.6 Statistical Procedures Several standard statistical techniques were utilized to analyze the data in this investigation. A one-tailed t-test, testing the difference of means, was used to test Hypothesis 1.7 The null hypothesis of no difference be- tween population means (i.e.,;il=;12) was tested. Hypothesis 2 and 3 were tested through the use of correlational analysis. Both the zero order (r) Pearson product moment correlation coefficient and the first order partial (r12.3) were utilized in these tests. The zero order correlation coefficients (r) were computed at the Michigan State University Computor Laboratory on the CDC 3600 using the CORE routine to calculate means, standard deviations and simple correlations. h; _.___v_ ._T 6See Appendix D. 7The formula for this test was: 25 In order to determine whether the degrees of correlation were statistically and significantly different from zero, the t—test of significance was applied to correlation coefficients. The statistical procedure as outlined by McNemar (26:145-146) was employed.8 The partial correlation coefficient (r12.3) indicates the degree of correlation between two variables which would exist provided variation in a third variable were controlled. The statistical procedure as given by McNemar (26:165-167) was employed to compute the partial correlation.9 To test Hypothesis 3 both correlational analysis and t-tests were employed. The t-test technique was used to test Hypothesis 5. In considering further questions .raised in Chapter I, expectation and count, t-tests, or correlation analysis were employed depending upon the nature of the question. 8The following formula was used to determine the standard error: 1 dr'r = V N—l The obtained r was then divided by this standard error to secure any%}value with which to enter the normal probability table. If-ggr is greater than 2.58 (.01 level of significance) we can conclude with a fairly high degree of sureness that the true or universe value of r is likely to be greater than zero. A 3%,.greater than 1.96 is significant at I‘ the .05 level. 9Formula used in computing partial correlation coefficients: r12.3 : r12 - r13 r23 y _*2 ‘y _ 2 26 A major phase of this study called for a systematic comparison of the results obtained from the two major samples investigated and in some instances between the two delinquent sub—samples. The two statistical techniques utilized in this comparative analysis were: one, a technique discussed by McNemar (26:147-148) to test for significant differences between obtained correlation coeffi- cients,10 and two, the t-test, as described above, to test for the significant difference between means. 10This test utilizes an r to z transformation for Ilandling sampling errors for r. In obtaining the standard earror of the difference between the two r's, both r's are 'transformed into 2's, and the standard error of the difference 'between the two z's is obtained by: l l le “Z2 =YNl-3 +N2 - 3 Uflue value of z = Zl - 22 was looked up in the normal 0’21 " Z2 probability table. CHAPTER IV RESEARCH FINDINGS Tests of the Hypotheses The presentation of the statistical analysis below is based on data obtained from the three samples described in Chapter III: one hundred ninth-grade male, non-delin- quent, public school students in a Midwestern urban- industrialized social setting, fifty ninth-grade students in a state training school for delinquent boys, and fifty ninth-grade students in a private residential school for delinquent boys. Hypothesis (l).--The mean self-concept of ability score of non-delinquent boys is higher than the mean self—concept of ability score of delinquent boys. A one-tailed t-test, testing the difference of means, Inas used to test Hypothesis 1. Table 1 lists the obtained Ineans, standard deviations, ”t's," and probability of (difference between the delinquent and non-delinquent mean Escores for self-concept of ability. The evidence presented in Table 1 indicates that tune proposed hypothesis is tenable, that non-delinquent 27 28 Ho. ma.m- Hm.: Ne.mm mm.a 0H.em Aooanzv Hobos ea. HH.H- Hm.a es.mm ma.a mm.sm Aomuzv ooocm moo. me.m- Hm.: Se.mm mm.m mm.mm Aomuzv oom>anm moocogommfio p .Q .w new: .Q .m new: mdsogm do ooHuz ocosacaaoa hoaaanmnogm asomc psoSSCHHancoz mucoUSpm was: oomnwucpcfiz peoSUCHHancoz .I new ocosvcfiaoo comzpom oocosommflm mo moHHprnogm Spas mummbnp one acoapmH>oQ Unaccepm qmosoom zuHHHQ¢ mo pdoocooanom csozuu.a mamae 29 male students in a public school setting tend to have more positive self-concepts of academic ability than institu- tionalized delinquent boys. Stated otherwise, the evidence seems to indicate that institutionalized delinquent boys have a lower perception of competence in general scholastic ability than public school non-delinquent boys. For the two groups of delinquent boys investigated, the probability of difference is greater between non-delinquent boys and delinquent boys from the private institution than delinquent boys from the state training school. This may be because all boys at the private institution have an academically oriented program, whereas, the academic program at the state training institution is stressed less. Thus boys from the latter institution who are enrolled primarily in an academic program may have a spuriously high self-concept of academic ability when comparing themselves with other boys in the institution who are not enrolled in an academic program. Hypothesis (2).--The self-concepts of ability of delinquent and non-delinquent male students are re- lated to their achievement when intelli- gence is controlled. This hypothesis was tested through the use of corre- lational analysis. Both the zero order (r) Pearson product moment correlation coefficient and the first order partial (P12.3) were utilized in testing this hypothesis. Simple 30 correlation coefficients were obtained between grade point average (GPA) and intelligence (IQ), between GPA and general self-concept of ability scores (S-C), and between S-C and IQ for each of the samples investigated. Likewise, partial correlations were obtained between GPA and IQ with 8-0 controlled, between GPA and 8-0 with IQ controlled, and between S-C and IQ with GPA controlled. The relevant coefficients of correlation are presented in Table 2 with and without the effect of the third variable controlled. Of particular significance to the testing of this hypothesis are the significant correlations between S-C and GPA, with and without the effect of IQ controlled. Data presented in Table 2 shows that even with the effect of IQ controlled the correlation coefficients between 8-0 and GPA, for all samples investigated, are positive and significant. The correlation coefficients between 8-0 and GPA with IQ con- trolled are .42 for delinquent boys from the private institution, .34 for delinquent boys from the state training school, .39 for the total delinquent sample and .63 for the non-delinquent public school sample. The data obtained from each of the samples lends support to the second hypothesis that the self-concepts of ability of both delin- quent and non-delinquent boys is positively related to classroom achievement when the effects of IQ are controlled. 31 TABLE 2.--Coefficients of Correlation Between Academic Grade Point Average (GPA), Measured Intelligence (IQ), and Self- Concept of Ability (S-C) for Delinquent and Non—Delinquent Male Students# #The multiple correlation coefficients (r1.23) among GPA, IQ, and S-C were .77 for the non-delinquent sample, Correlation Coefficients Variables No Variable Third Variable Group Correlated Controlled Controlled Delinquent - Private (N=50) GPA - IQ .24 8-0 .14 GPA - S-0 .45* IQ .42* S-C - IQ .26 GPA .17 Delinquent ~ State (N=50) GPA - IQ .01 S-C -.04 GPA - S-C .33* IQ .34* 8-0 - IQ .16 GPA .17 Delinquent - Total (N=100) GPA - IQ .14 S-C .07 GPA - S-C 041* IQ 039* S-C - IQ .19 GPA .14 Non-Delinquent (N=100) GPA — IQ .58* 8-0 .23* GPA - 8-0 .75* IQ .63* S-C - IQ .57* GPA .24* .42 for the total delinquent sample, .34 for the state training school delinquent sample, and .47 for the private institu- tion delinquent sample. *P< .05 for the test that r and r12.3=0. 32 Of further significance, upon examination of Table 2, are the low correlations between GPA and IQ with the effect of 8-0 controlled, and the low correlations between 8-0 and IQ when GPA is partialled out. For both delin- quent samples and the total delinquent group, these ob- tained correlations are not significantly different from zero. Even in the case of the non-delinquent sample, these correlation coefficients are low when compared with the correlation coefficient of .63 between GPA and 8-0 with the effect of IQ partialled out. These findings are con- gruent with the findings of Brookover (9:38) and Morse (28:38) and give further evidence that the self-concept of ability scale measures a different variable than the IQ measures and that self-concept is an independent predictor of classroom achievement when measured by grade point aver- age. In the case of the delinquent samples, IQ was not a good predictor of GPA. When self-concept was partialled out the relationship between IQ and GPA was practically nil and not significantly different from zero. Comparison of the partial correlation coefficients between IQ and GPA (with the effect of 8-0 controlled) with the multiple correlation of IQ and 8-0 with GPA, indicates that the correlation increases from .07 to .42 for the total delinquent sample, and from .23 to .77 for the non- delinquent sample. The multiple correlations reported have beta weights of .06 for IQ and .41 for self—concept of 33 ability among the total delinquent sample and .22 for IQ and .62 for self-concept among the non—delinquent sample. Thus self-concept of ability is weighted higher than IQ as a predictor of achievement for both ninth-grade public school male students and ninth-grade institutionalized delinquent boys. These findings are consistent with those reported by Morse (28:39) in his similar comparison of Negro and Caucasian subjects. The correlations between specific self-concept of ability in school subjects and grades in each subject, with and without the effect of IQ partialled out, are shown in Table 3. With the one exception of the correlation coefficient obtained between specific self-concept of ability in science and science grade, for the delinquent samply only, the correlations between specific self- concepts and corresponding grades in each subject are significant with and without the effect of IQ controlled. Other than the exception noted, the data presented in Table 3 is comparable to the data in Table 2 and lends further support to the hypothesis that self-concept is related to classroom achievement when intelligence, as measured by standard IQ tests, is controlled. Table 4 gives a further breakdown of correlation coefficients between specific self-concepts of ability and grades in corresponding subjects, with and without the effect of IQ controlled, for both sub-samples of the 34 o cum 0 NHL” one s been once one son mo.v.m * *mm. *mm. scone .om .scm s o-m .om .sra *am. *am. cease oocm s elm dorm *mn. mo. *Sm. OH. compo oocofiom a cum cocofiom *0:. *HM. *Nw. *mm. enema .pm .oom w Dim .um .oom *mm. *mm. *m:. *mm. memos smflawcm B oum cmfiawcm som. *nm. *mm. *mm. cease .eooz s o-m .eooz *mm. *mm. *mw. *H:. «mm Hence a onm ngocoo ooan ooHuz OOHuz ooan .UCHHancoz .UCHHoQ .UCHHancoz .UCHHoQ nonnaogsoo moanmfimm> ooHHogpcoo ooaaosucoo 3H noflz dH psocpfiz mucofiofimmooo coapmHogpoo .mhom pcosdcfiaomucoz Hoozow oaansm oemguncpcfiz one whom pcooUcHHoQ ooNHHm -coaosoaoncH coono-eocaz non coofiosm room on access one nooonosm oaoaooom ca moaaang mo pdoocooimfiom coozoom new “dam 6cm thHHn< go pdoocoosmaom Hmnococ coozpom .UoHHogocoo 0H escapas new no“: .mucow0fimmooo coaumamssoonu.m mqmge 35 .oumdfiu one s one coop one poo 8.le l. *am. *mm. *mm. *ms. oooso .om .srm s o-m .m .sem Hm. *mm. mm. sam. compo ooem s o-m oocm mo. we. as. ma. oooso oocoHom s o-m cocoaom *mm. *oa. *mm. *ma. oooso .om .oom s o-m .om .oom so. *sm. so. soc. condo enaawsm s o-m enaamcm tom. *mm. som. *mm. oosno .cooz s o-m .cosz sam. *ma. *mm. sma. «do Hsooe s o-m Hosocoo omuz omnz omuz omuz opbpm oum>Hgm ouspm opm>Hgm oopwaongoo mofipwfimm> eoaaogpcoo UoHHooucoo 0H Sufiz dH psocpaz mpcoaoammooo coHpmHoggoo Hoozom wcflcfimge bosom m one ceapspameH opm>asm m Scum whom pcosocfiaom ocmgwunpcfiz hoe poohpsm comm ca noooco one ncoowosm oaouooom ca soaaaoa do oooocoo-oaom coospom new mHpHmom who: one qogoom pdoocoonmaom on» Lozmfim ogev .mpcoUSBm onshwucpcfiz «was: unmodcfiaomncoz Hoonom oaansm one ocosocfiaom poNHchoHpSmeH pom mpoofinsm Hoonom co>Ho on» we comm coo one nuoonoom Hat on nonoom soHHHoe no oooocoo-ofiom some--.m mamas 39 score was significantly lower than the mean mathematics and science self-concept of ability scores, as well as lower than the mean general self-concept of ability score. The lower means in English and social studies corroborates the findings reported by Brookover (9: 43-44) and lends support to his thinking that cultural factors may be involved, i.e, mathematics and science may be more cul- turally defined as masculine than are English and social studies. The coefficients of correlation between general and specific self-concepts of ability, and between the various combinations of specific self-concepts of ability are shown in Table 6. While the correlations between general and specific academic self-concepts of ability, and correlations between some of the specific academic self'concepts of ability are expectedly high, they are significantly lower than the reliability coefficients for the self-concept measure. (See Chapter III, p. 18) In studying the various correlations in Table 6 in reference to the delinquent samples, it is noted that the correlation coefficients between the following listed variables are either low, negative, and/or not significantly different from zero: all variables correlated with self-concepts of ability in shop courses and physical education for the total delinquent sample, and all specific self-concepts of 40 TABLE 6.--Correlation Coefficients Between General Self- Concept of Ability and Specific Self—Concepts of Ability, and Between the Various Specific Self-Concepts of Ability for Delinquent and Non-Delinquent Males Correlation Coefficients Delinquent Non- Delinquent Variables Correlated Private State Total N=5O N=5O N=100 N=lOO General S-C & S-C Math. .75 .48 .60 .68 General S-C & S-C English .68 .35 .55 .64 General S-C & S-C Soc. St. .78 .46 .63 .67 General S-C & S-C Science .69 .34 .56 .71 General S-C & S-C Shop Courses .01 .29 .17 General S-C & S-C Physical Ed. -.13 .14 .04 3'0 Math. & S-C English .49 -.20 .14 .44 S-C Math. & S-C SOC. St. .56 .02 .26 .38 8-0 Math. & S-C Science .53 .25 .38 .53 S-C Math. & S-C Shop Courses -.05 .16 .10 S-C Math. & S-C Phy. Ed. -330 -.10 -.13 3‘0 English & S'C SOC. St. .75 .05 .43 .48 S-C English & S-C Science .54 .16 .39 .49 S-C English & S—C Shop Courses —.02 -.ll -.04 S-C English & S-C Phy. Ed. .03 -,03 .03 8-0 Soc. St. & S-C Science .52 .23 .39 .47 S-C Soc. St. & S-C Shop Courses .00 .41 .21 S-C SOC. St. & S-C Physical Ed. .ll .12 .12 8-0 Science & S-C Shop Courses .04 .03 .06 S-C Science & S-C Phy. Ed. -.04 .02 .03 S-C Shop Courses & S-C Phy. Ed. .36 .06 .26 4l ability correlated with the self-concepts of ability in English and mathematics for the state training school delinquent sample. The data presented in Table 6 suggests that the specific self-concept of ability scales measure different variables than the general self—concept of ability scale measures, and that each of the various specific self-concept of ability scales measure different variables. To gain further support for Hypothesis 3, coefficients of correlation were obtained between general S-C and general achievement (GPA) and specific subjects S-0 and GPA. These correlations appear in Table 7. If the general 8-0 of ability scale is functionally different from the specific self-concept of ability scales, then general S-C should prove to be a better predictor of general achievement than specific S-C. For each of the samples listed on Table 7, the correlation coefficients between general S~C and total GPA are of greater magnitude than the corresponding correlation coefficients between specific 8-0 and total GPA. This trend lends further support for Hypothesis 3. 0n the basis of the data provided in Tables 5, 6, and 7, it is concluded that Hypothesis 3 is tenable, sig- nifying that the self-concepts of ability in specific school subjects of delinquent and non-delinquent boys vary from one subject to the other and from their general self- concepts of ability. 42 .m.h cmozpon mocogm%MHU pom pmop UoHHMpuocounfipm Umpdaogpoo moanmfigm> ucmsvcHHmQ - Icoz pcmsvcaamm mpcmaoammmoo coauwaohpoo mucoUSpm mam: ocmgc unpcfiz pCmSUQHHancoz Hoonom oaansm 62m poosvcHHom UmmaamcoHuSpHumce pom «mo cam pdoocooumaom poonnsm oamaomdm 62m AmH£o< Hahocmo Ugo pdoocoo:maom Hahmcoo coozpom coapmHmLLoo mo mucoHOHmmmoouu.N mnmoa mo. pm uanHMchHm ma mocmhmmmfia* .oocmOHMchHm mo Ho>oH Ho. 63p pm 0 Song pcmgmMMHU hapcmOHMchflm ohm mcofipmHmmhoo HH< 44 mowmsH .mponmcsoo a “.mpcmpmdmmsom ms. mm. in . . 89689 do :38. mm. mm. is. mommsH .mmoammcsoo 5:. 0:. om. mommsH .mpcmpmdomsom Hm. cm. as. mm. momeH .mgmzowme a «.mpomm «.mpcmgmm mo Hmpoe as. 00. mm. as. mommEH .mgonomoe so. ms. ow. mm. mmmeH .mhoom *ms. *Hm. sm. mm. mommsH .mpcmhmm OOHHZ OOHHZ omuz omuz Hopoe ouMum opm>fipm "do cofipdoopmm pawsvcaaoo .mpchSpm :oz acmSUQHHoQ pdmocooumHmm Hmmmcmo spas soapmHommoo mucoUSpm mewgu:nchz was: quSUCHHancoz 6cm pcmzvcaamm apom mom mmeHHHQ¢ LHoQB mo caom o» mhmnpo pascachme m>aoommm moan mmwwEH map can thHHn< no pdmocoo umamm ddmmcou .mpcoUSpm cooZpom coapmnghoo mo mucoHOHmmmoouu.® mqm¢e 45 and the images they perceive their parents hold of their abilities in these subjects were also all significantly and positively correlated. These correlations, for both delinquent and non-delinquent male ninth—grade students, are listed in Table 9. Third, Table 10 presents the coefficients of correlation between the students' academic grade point averages and the images they perceive signifi- cant persons (parents, peers, teachers) to hold of their abilities. Again, it is noted that the obtained corre- lations are positively correlated for both the delinquent and non-delinquent students. The data presented in Tables 8, 9, and 10 and the tests for Hypothesis 4 have demonstrated that this hypothesis is tenable. The tenability of this hypothesis suggests that the expectations of significant others as perceived by delinquent and non—delinquent ninth-grade male students influence the students' self-concepts as learners and consequently influence the students' academic perfor- mance in school. Hypothesis (5).--The expectations of significant others as perceived by delinquent male students differ significantly from the expectations of significant others as perceived by non-delinquent male students. 46 .OHH amnp pmmp wflp pom mo. V9: *m:. *Hm. *os. ecupmosem Hsoamscm *mw. *om. *Hw. mmmgsoo aonm *NN. *33. *mm. #5:. .mocmaom *mm. *mm. *Nm. *wm. mmHUSpm Hmfloom *Nm. *Nz. *mw. *mm. Emaawcm *mw. *HN. *mw. *mm. oaumsnuahd oofinz ooHuz omuz omuz somnnsm asses spasm mpm>fism pCmSvCHHancoz unmadcflaom .mpcmcdpm memgwunpzaz pcosucHHmQucoz 6cm pcmavcaawm pom mpoonnsm ommne CH mmfiuaaand Mamas mo caom on mucogmm Adena m>awogmm hone mmmeH map 6cm mpoonnsm Hoosom oamaowdm Ca zuaafin¢ mo uqmocooumamm .prOUSum Cmmspwm COHUMHmHhoo MO mpcmHoHMMmOOII.m mqmdB 47 .OHQ pars Pwmp mflp «HOM m0. .Vm .x. mmwmeH .whmsomwe é ..mhmmm *0». *ms. *mm. *m:. ._mpemnsm so mpHmOQEOU *om. *mm. ma. *3:. mmwmsH .mgmmm *mm. *m:. *H:. *mz. mommaH .mgonoMmB *mm. *mm. *mm. om. mmwmsH .mpcmgwm ooHuz ooanz omuz omnz so conpamopmm Hapoe madam ouw>amm .mpc665um pawswcfiaoancoz psoSUCHHmQ «mo Spas mcoaumHmhhoo .mpcoosum mam: wnmgwunpcaz unmadcfiawancoz cum pussycaamm pom mmeHHHQ¢ passe mo caom on mumcpo unonMcham m>ao0hmm hone mmmeH map 6cm Admuv mmwmgo>¢ pcfiom mummo .mucmcsum mnp Cmmzumm Coapwamhpoo mo mucmHOfiMM®OOII.OH mdm<8 48 A two—tailed t-test, testing the difference of means, was used to test Hypothesis 5. Table 11 lists the obtained means, standard deviations, "t's," and probability of difference between the delinquent and non-delinquent mean scores for perceived images of ability. The evidence presented in Table 11 indicates that the proposed hypothesis is tenable, that the mean scores differ significantly be- tween perceived expectations of significant others of delin— quent male ninth-grade students and perceived expectations of non-delinquent male ninth-grade students. For each of the significant others (parents, best friends, and teacher) and composite of these, the expectations of significant others as perceived by non-delinquent students were significantly higher than the expectations of significant others as perceived by delinquent students. Other Relevant Results In addition to the hypotheses tested, several questions were raised in Chapter I concerning the relevant variables in this study. The first questions were in regard to who are the significant others to whom delinquent and non—delinquent boys relate and do significant others differ for delinquent and non-delinquent boys. The second question concerned itself with the delinquent samples only. The essence of the question was to'see if there was a dis- crepancy between how delinquents expect significant Others to perceive their abilities and how these significant 49 .ms on ma mpHmOQEOQ mom “mm o» m 809% ma gmnuo unwofimacwam HM5©H>HUQH mom mmhoom go emcee manammom Hmuoe* Ho. Hm.m- sm.m so.mm ow.m ms.:m scene so spamOQEOU mo. :m.m- mm.m mm.ma NH.: mm.mH pmsowme mo. oo.m- mm.m sm.ma mm.m sm.sa eemfism swam mo. mm.m- ss.m Hm.mH ss.m mm.wa mpemssm mocmpmeefio a .m.m use: .o.m new: mnmnpo so passaefiewam spaaansnosm ooHuz oofinz acmSUCaHancoz ucmswcaaom mogoom mmmsH Uo>amogmm new: |I *mmmcpo pamOH umficmam mossy an cam: noncom huaaap¢ mo momeH Um>amommm new: .mpchSpm messeunpeaz was: pcmsucaawoucoz new pcmsucaflmo as mocmpmeeao--.afi mamas 50 others actually perceived their abilities. These questions are considered in the following pages. Identification of significant others.—-Two open-ended questionnaires were used to gain the names of individuals who were significant others. (See Appendix A) In the first ques- tionnaire, students were asked to list the names of the peOple whom they felt were important in their lives; in the second questionnaire, students were instructed to indicate people whom they felt were concerned about how well they did in school. No specifications were given as to the number or type of individuals to be listed. Responses to these questionnaires are categorized and summarized by percentage of responses to each category in Tables 12 and 13. The responses of the delinquent and non-delinquent students to these questionnaires indicate that parents are named more often than any other persons both as ’important in their lives and concerned with their school work. Peers and relatives (both adult and age-level) were also frequently mentioned as being important in the lives of both delinquent and non-delinquent students. Non— delinquents mentioned teachers as being important in their lives significantly more often than did delinquents. Be- cause of the delinquents' common feelings against school and authority this difference is not surprising. Delin- quents mentioned teachers both within and Outside the institution although reference was made more frequently to .Hoccomgod OHEoUMom gonpo Spas UmUSHocle .oHQEMm uneducaamvucoc mnu 0p mHQMOHHQam poz * 51 Hm m 3 SH mHQmHMHmmmHocoz * Hm o m: Hmccomgmm choHQSpHmeH mmnpo % :m mm mm scammcsoo * 3H ma ma mucogmammsom ma mm mm mm measea Hsooq smnpo mm mm om m: mm>HpmHmm Hm>ma-mw< mm mm on em mm>spsflmm passe 0H m m m Hoccomhmm oasmumo¢ gmnuo Sm ma w om mhmnomme mm om om :w mamHUmmso no mucogwm mm mm mm m: xmm gonpfim .mgmmm ma m: rm mm Kmm mpfimOQQo «mgwom om om :H mm xom oemm .mgmmm OOHHZ OOHHZ omuz omnz Hence spasm mps>anm msmnpo L pQMOfimficmHm Hwhmcmc psoSUCHHmmncoz pcosvcaamm mwmpcmogmm Ammonpo vase uHmstHm ngocmov =.mo>Hq gHonB CH peepgoaeHz wcfiom mm mofigomopmo wcazoaaom och no comm scam comsmm wco unwoq pm wcfiswz mucoozpm mo omencoogmmus.ma mumfipsamm Hm>mq-mw< mm um mm ma mm>apmamm pH3©¢ mm mm ma mm Hoccowgmm anmUMoa gonpo Hm a: m: o: mgmcowme >0 mm mm mm mcmacgmsu no mucogmm ma mm :m ma xom mmnpfim .mgomm 5 mm :m NH xom opHmOQdo «mummm N d m, m Now 68mm .mgmom OOHuz OOHHZ omuz Omuz kuoe waspm- mpm>fipm mgmzpo . pascamficwfim OHEmUMQ< psoSUCHHoancoz psoSUCHHmQ owmucmogmm Amgmnpo prOHMHCMHm anmpmo¢v =.Hoonom CH cc 2626 Haoz 30m: Qua: Umcmmocoo wchm mm mofigowmumo mcfizoaaom map mo comm Eomm compom oco ammoq pm wcfismz mpcmpsum mo mprcwogmmun.mH mqmHocfi pom mogoom mo mwcmg mHnHmmOQ Hmuoe* Hoe. ms.m: Ha.mm Ho. ma.ms :m.mm Hoo. mm.H: mm.mm metre mo mpHmOQEOO goo. ms.mg Hm.mH goo. wo.mH mm.mH goo. mm.mH mo.mH toammesoo Hoo. mm.:H mo.oH om. mo.mH mm.mH Hoo. mo.:H mm.eH pcmssammsom goo. m:.mH mm.ma OH. mo.ma om.sfi goo. :m.:H so.mH assumes .eeam new: new: .eeam new: new: .ueflm use: cams- mtmsso pemoaeaemam mo Heapo< vm>aoo mo Hm590¢ Uo>Hoo mo Hm3p0¢ Um>amo .pogm ngmm .QOAm unmm .nogm ngom Aooanv quSUCHHmQ Hmpoe Aomuzv mumpm Aomuzv mps>Hsm *mpcmespm mesto-:peaz was: pcmsUCHHoa ewuafiscoaussfipmeH toe mpmzpo peasaeficwam means so mmfipaaana means so came saHHHp< so mewseH em>amosmm est: ca museumeefio--.:a mamas .mucoUSQm mo coauawohom Heapo< cam mnmcpo unwoamacwam 57 Hooo. moemtmeefle mo spfifiansnotm: Hooo. mocwsmaefie mo ngHHQmQOtmm -do. mocmwmeeae so spHHHQmQOtmm Ho. mocmtmeeae mo apflaanmaosma :wm. ma. H4. m>onm mo opamoasoo :ms. mm. we. m>onm mo oprOQEoo mmH. mo. :m. whoammcsoo mwm. mw. as. . mgoammcsoo mmo. moi: ma. mucogwdwmsom N53. 0:. om. mpcmmwammsom H::. mm. mm. whonomme Hmo. mm. as. mum£ommfi ooHuz omuz omwz oonz omuz omuz HMpOB mpwum opm>Hhm.. "Mo soapmSHm>m . Hapoe mumpm mpm>Hnm "mo cofipmSHm>m Hmzuo< co>HmQMmm pamocooumamm Hmmmcou pdmocoonmaom ngocou spas coaumnggoo Spas codumnggoo mucwozpm womuuunucaz «mam: .ucmSUCfiHwo Umuaam unoHSpHpmcH pom whoammcsoo cam .mpcohmammsom .mmozomme s9 soapmsHm>m Hmdpo¢ one paoocooumamm ngocmc cmmzpom coapwamhgoo mo mucoaoammmoo on» cam mmponm ucsoo new “mucmhwammsom ‘mhmnomme mo COHQMSHM>M ©m>fimogmm map new hpfiafing mo pdoocoouuaom ammocou .muchSQm semapom coaumHmhgoo mo mucmfioammmoonu.ma mum¢e 58 are compared with the correlation coefficients between general self-concept and actual evaluation by teachers, houseparents, and counselors. The significant differences between the correlation coefficients give further evidence of a discrepancy between perceived and actual evaluations of academic abilities. Table 16 gives the correlation coefficients between perceived evaluations of teachers, houseparents, and counselors, and actual evaluations of teachers, house- parents and counselors for the delinquent samples. Although several of these correlations are significantly different from zero, the obtained correlations are significantly lower than the correlation coefficients between perceived evaluations and general self-concept, giving further evidence that there is a discrepancy between how delinquents expect significant others to perceive their abilities and how significant others actually perceive their abilities. Comparative Analysis A major phase of this investigation called for a systematic comparison of the data obtained from the delin- quent and nonvdelinquent samples investigated. The two statistical techniques employed were: one, the t-test, as discussed in Chapter III, to test for the significant difference between means, and two, a "Z” transformation test to test for significant differences between obtained 59 TABLE l6.--Coefficients of Correlation Between Perceived Evaluations of Teachers, Houseparents, and Counselors, and Actual Evaluations by Teachers, Houseparents, and Counselors for Institutionalized Delinquent, Male, Ninth- Grade Students Correlation Between Perceived and Actual Evaluation Private? Statefifi Tdtal Significant Others N=50 , N=50 N=100 Teachers .46* .33* .40* Houseparents .12 .07 .09 Counselors .31* .21 .28* Composite of above .35* .22 .29* W *P<.05 for the test that r=0. 6O correlation coefficients. (See discussion in Chapter III.) The null hypothesis tested states that the two populations are equal. Differenceszetween the Delinquent and Non—Delinquent ggean Scores for the Major Variables.--Table 17 reports a summary of the obtained means, standard deviations, and t— tests between the delinquent and non—delinquent mean scores for the major variables of this study. No statistical comparison was made of mean grade scores. Table 17 shows that except for three variables, the non—delinquent mean scores, as predicted, were all significantly greater than the delinquent mean scores. On one variable, age, the delinquent group, as expected, had the highest mean score. The average age of the delinquent group was 15.62 as compared with 15.19 for the non~delinquent group. On two variables there were no significant differences in the mean scores between the two groups: the Specific Self-Concept in English scale scores and Perceived Image of Parents in English scores. Earlier in this chapter were discussed poSsible reasons for“the comparative low self—concept of ability in English scores for the non-delinquent population. Table 5 had indicated that the mean self-concept of ability in English score was significantly lower than the mean general self—concept of ability score for the non-delinquent group. This difference was not indicated for the delinquent sample. .houmogw ma cams pcosvcafime m£B% .mo. sesofleficwfim ma mocmtmecam* 61 **mo.:. 2:. mH.mH em. mo.mH ewe *mm.mu ms.ma :m.moH mm.oH mH.ooH dH *wm.mu mm.Hm mw.m: :N.ma mm.am xmecH OHEocoomnoHoom *Hm.m- sm.m so.wm om.m me.:m genomes s .eemasm pmmm .mpempmm mo mowmeH ©o>fimohom mo 55m *sm.m- mm.m. oz.mH NH.: mm.mH genomes so mwmeH em>Hmopmm *oo.mu mm.m bm.ma mm.m sw.sa ecmHgm ummm mo mwmsH Uo>Hmogom cocoaom ca *mw.mn 2m. mm.: am. 00.: mucohmm Mo mwmsH em>Hmopmm moHUSpm HmHoom ca *mw.mu Fm. mm.: mm. :m.m mucmgmm mo mmeH Um>amogwm cmfiawsm :H mm. HH.: mm. HH.d mnemAMm mo omeH Uo>fimogom .npmz CH *sa.mu ms. N:.: mm. HN.: mpcmhmm mo omeH ew>Hoomom *mm.m- swim. am.mH ss.m sm.ma mpemtmm do mwmeH em>amotmm *mm.m- mw.o mm.mm mm.m ss.mm moemaom - pamoeoo-eamm *mm.H- mm.w om.sm mm.m mm.mm mmaespm HsHoom - pamoeoo-eamm ms.m mm.sm mm.m mo.sm emaawcm - pamoeoo-eflmm *mm.m. om.m mm.mm ms.o mo.sm ass: - uamocoo-mHmm *mz.mu Hm.: ss.wm mm.: OH.NN Hmhmsmo u pdwocooumamm p .Q.m new: .Q.m new: moanmflgm> ooHuz ooan quSUCHHoQucoz unmadcaamm, mmeBSpm sumac nusaz .mamz usoswzfiflom coz Hoosom oHHnsm use pseudcfiamm nomaamcoHpSpHpmcH on» non moanmfihm> 90nd: Ham Lou mmAOom cmmzpmm mums» p was .wcoaumH>mQ Uhmccmum .mQMmzun. NH mqm¢e 62 Differences Between the Delinquent and Non—Delinquent ,anrelated_pata.--Table 18 shows the comparative coeffi— cients of correlation for the crucial variables for the delinquent and the non-delinquent students. (To note all combinations of correlation coefficients see the correlation matrices of major variables for the various samples in Appendix E.) The information in Table 18 indicates that for slightly over half of the variables correlated the coefficients of correlation for the non-delinquent group were significantly greater than those for the delinquent sample. No coefficients of correlation were significantly greater for the delinquent group. Nearly half of the com- binations of correlated variables were not statistically significant although the correlation coefficients tended to be greater for the non-delinquent group. The relative differences observed indicate a greater degree of relationship between the relevant variables for the non- delinquent population. The coefficients of correlation were significantly greater for the non-delinquent group for the following variables correlated with general self—concept of ability: academic GPA, each specific subject GPA, IQ, socio-economic index and parents' images. Except for the relationship between general self-concept of ability and socio-economic index, the above have been discussed earlier in this chapter. The information in Table 17 indicated that the 63 **mo.: mm. mm. mmwmeH .mpemgmm I amc **m:. m wH.I om. xmeeH anOQOomIoaoom I «mm mm. ON.I mm. mw< I «mo **ms. m :H. mm. dH I OOHHZ OOHHZ uemsvcfiaom unmadeHHoQIcoz *mpcmesum pcosvcaamm cam pcmsweaammIeoz one mom mpmoBIN cam cofiumHognoo no mpcmfioammoou m>HpmngEooII.mH mqm Hmoapfimo Mia. u mu I HNI5.* HH.H OH. mm. xoUCH OHEoeoomIoaoom I UH mo. mm.I sm.I . ewe I dH **mm.m mm. mm. oIm mocmaom I amc **©m.m mm. mm. on mmaezpm Hmfioom I «mu mm.a :m. s:. OIm nmaawcm I «ma mm.a mm. 5:. oIm .anE I amo **mm.m we. on. mmmmeH .mtmmm s ..mmonomme ”.mpemAMm .mpHmOQEoo I ¢mu mm.a mm. mm. momeH .mgmom I «ma HN.H me. me. mmwmeH .mtmnosme I «am N me an copmHmmsoo moanwapm> ooanz ooHuz uneducaamm unmachHoaIcoz AetscheoovII.mH mamas 65 socio-economic index, as determined by Duncan‘s 1959 Socio-Economic Index, is significantly higher, as expected, for the non-delinquent group. Variation within the non- delinquent sample in respect to socio-economic index is considerably greater than within the delinquent group, indicating that the non-delinquent group is more hetero- geneous in socio-economic status. Likewise, the mean socio- economic index of delinquent boys from the private institu- tion was higher and variation was greater than for the boys from the state institution. Table 18 indicates that the correlation coefficients between S-C and socio-economic index are .34 for the non-delinquent sample and -.08 for the delinquent group indicating that there is no significant relationship between socio-economic status and self—concept for the delinquent students, A similar relationship, which is significantly greater for the non-delinquent group, is noted between GPA and sociofeconomic index. Correlations between GPA and socio-economic index are .30 for the non- delinquent students and -.18 for delinquent students. Again a slight positive correlation is indicated for the more heterogeneous non-delinquent group, but the relationship between GPA and socio—economic index for the more homo- geneous institutionalized delinquent group is not statisti- cally significant. 66 The coefficients of correlation were significantly greater for the non-delinquent group for the following variables correlated with GPA: general self—concept, IQ, socio-economic index, parents' images, composite of parents', teachers', and peers' images, social studies selfvconcept, and science self-concept. Each of the correlations between the specific subject self-concept of ability scores and corresponding subject grade was signi- ficantly greater for the non~delinquent students. Most of these relationships have been discussed earlier in this chapter. When correlations between GPA and the images that students perceive significant others to hold of their abilities are compared, only the relationships between GPA and perceived parents' images are significantly greater for the non-delinquent students; these correlations are .68 for non-delinquents and .25 for delinquents. There were no significant differences between the two groups when GPA was correlated with the images the students per— ceive teachers or peers to hold of their abilities. CHAPTER V SUMMARY AND CONCLUSIONS Summary of Research Objectives The purpose of this investigation was to make a systematic study of the relationships between self-concept of ability and classroom achievement among ninth-grade public school non-delinquent male students and ninth- grade institutionalized delinquent boys enrolled in an academic program. Further, a systematic comparison was to be made between the delinquent and non-delinquent students in those factors investigated relating to self- concept of ability and classroom achievement. The non-delinquent sample (NulOO) was drawn at random from ninth-grade male students, in one Midwestern metropolitan school system, who met the criteria of having been in the school system at least two years and for whom ninth-grade IQ scores were available. The sample of ninth-grade institutionalized delinquent boys (N=100) was drawn from two sources. Fifty ninth-grade students were selected from each of two institutions designed for the custody and treatment of delinquent boys. One of these was a state institution located in the same metropolitan 67 68 area as the non-delinquent sample; the second was a private institution located within fifty miles of the other samples selected. Summary of Research Findings Basing the major thesis of this investigation on the phenomenological approach to learning, it was postu— lated that self-concept is established early in life in an interpersonal setting and modified by subsequent experience, and that the learner tends to evaluate himself as he per- ceives others to evaluate him, and finally, that a learner‘s self-concept of ability is a functionally limiting or facilitating factor in classroom achievement. It was also postulated that certain deviations characteristic of delinquents have delimiting effects on self-concept develop- ment. The major thesis was tested in the form of five specific hypotheses. All were found to be tenable. The major results of this investigation may be summarized as follows: 1. Non-delinquent, public school, ninth-grade male students have significantly higher mean self- concept of ability scores than ninth-grade insti- tutionalized delinquent male students. Stated otherwise, non-delinquent students have more positive self-concepts of ability than delinquent boys. 69 Self-concept of ability is significantly related to classroom achievement of delinquent and non- delinquent ninth-grade male students. The corre- lation coefficients are .41 for delinquents and .74 for non-delinquents. Self-concept of ability is significantly related to classroom achievement of delinquent and non- delinquent ninth-grade male students when the effect of intelligence is controlled. The coefficients of correlation, with measured in- telligence partialled out, are .39 for delin- quents and .63 for non-delinquents. Self-concept of ability is weighted higher than IQ as a predictor of achievement for both ninth- grade public school male students and ninth-grade institutionalized delinquent boys. The multiple correlation coefficients (.42 for the delinquents and .77 for the non-delinquents) have beta weights of .06 for IQ and .41 for self—concept of ability for delinquents, and .22 for IQ and .62 for self-concept for non-delinquents. IQ alone is not a reliable predictor of classroom achievement for male delinquent ninth-grade students enrolled in an academic program. The correlations between IQ and grade point average without and with the effect of self-concept of 70 ability partialled out are .14 and .04 respec- tively. Neither of the correlation coefficients are significantly different from zero. The hypothesis that the self-concepts of ability in specific school subjects of delinquent and non-delinquent boys vary from one subject to the other and from their general self-concepts of ability is tenable. The non-academic (shop courses and physical education activities) self-concept of ability scales measure different variables than the general selfvconcept of ability scale or the specific academic self-concept of ability scales for the delinquent students. The mean self—concept of ability scores for shop courses and physical education were both significantly higher than any of the mean specific academic self-concept of ability scores or the mean general self-concept of ability score. None of the obtained correlation coefficients between self-concept in shop courses or physical educa- tion activities and self-concepts in specific academic subjects or general self-concept of ability were significantly different from zero. 10. 71 The hypothesis that the expectations of signifi- cant others as perceived by both delinquent and non-delinquent boys are positively related with the students' self-concepts as learners is supported when parents, teachers and peers are identified as significant others. The relationship between students' general self- concepts of ability and the images they perceive their parents to hold of their abilities is significantly greater among the non-delinquent students than among the delinquent students. The obtained coefficients of correlation are .76 for non-delinquents and .61 for delinquents. No sig- nificant differences were noted between students' general selfvconcepts of ability and the images they perceive teachers and peers to hold of their abilities for delinquent and non-delinquent stu- dents. In respect to teachers' images the corre- lations are .71 for non-delinquents and .69 for delinquents. In respect to peers' images the correlations are .67 for non-delinquents and .73 for delinquents. Classroom achievement is positively correlated with the images students perceive significant others to hold of their abilities. 11. l2. 13. 72 The correlations between students' classroom achievement and the images they perceive their parents, teachers, and peers to hold of their abilities are higher for the non-delinquent students than for the delinquent students. The most significant difference is in respect to parents' images. The obtained coefficients between grade point average and parents' images are .68 for non-delinquents and .25 for delin- quents. The correlations between GPA and teachers' images are .62 for non-delinquents and .45 for delinquents. The correlations between GPA and peers' images are .56 for non-delinquents and .35 for delinquents. The hypothesis that the expectations of signifi— cant others as perceived by delinquent male students differ significantly from the expecta- tions of significant others as perceived by non- delinquent male students is supported when parents, teachers and peers are identified as significant others. The mean perceived image scores are all significantly higher for the non-delinquent students. Both similarities and differences are noted be- tween delinquent and non—delinquent students in their listing of significant others. Parents l4. l5. 16. 73 were named by nearly all students as being "important in their lives" and concerned with "how well they do in school." Peers, relatives, and teachers were frequently mentioned. The delinquent students also mentioned counselors, houseparents and other institutional personnel. Non-delinquents mentioned teachers as being more important in their lives significantly more often than did delinquents. Delinquents mentioned Opposite sex peers significantly more often than same sex peers, whereas non-delinquents mentioned same sex peers significantly more often than opposite sex peers. There is a significant discrepancy between how delinquents expect significant others to per- ceive their abilities and how significant others actually perceive their abilities. The perceived expectations are consistently higher than the actual evaluations by significant teachers, houseparents, and counselors. The mean family socio-economic index for the non-delinquent students is significantly higher than for delinquent students. The relationship between self-concept and socio- economic index is significantly positive for non-delinquent students, whereas this relationship 17. 18. 74 is not significant for delinquent students. The obtained correlations are .34 for non-delinquents and -.08 for delinquents. The relationship between GPA and socio-economic index is significantly positive for non-delinquent students, whereas this relationship is not sig- nificant for delinquent students. The obtained correlations are .30 for non-delinquents and -.18 for delinquents. The relationship between IQ and socio-economic index is slightly positive for both delinquent and non-delinquent samples. The obtained correlations are .25 for non—delinquents and .10 for delinquents. Implications of Research Findings The evidence presented in this investigation has given further support to the following theoretical notations basic to this study: 1. The capacity for intelligent behavior is a function of perceptions. How a student perceives his life experiences, how things seem from his point of view rather than how they seem from the outsider's point of view, are considered important variables in the learning process. 75 3. A person will tend to evaluate himself as he perceives others evaluate him. 4. The learner's self-concept of his learning ability is a functionally limiting or facilitating factor in classroom achievement. 5. Chronic threatening experiences encountered by delinquents hinder and narrow perceptions and consequently delimit the effect of self-concept development and school achievement. It is believed that this study affords concrete applications for the education of junior high and possibly high school students, and may provide special guidelines for administrators, teachers, and related personnel con- cerned directly with the education of institutionalized boys. Although the study proved meaningful for groups as a whole, it is believed, because of the variations indi- cated, that individuals rather than groups should be emphasized. Appropriate courses of action can be outlined when the staff analyzes the self-concept of ability for the individual student and identifies significant others in this student's life. Emphasis could be on building a more positive self-concept through the facilitating effect of significant others in contrast to relying upon the face value of an IQ score and planning programs in relation to such a score. The evidence presented in this investigation suggests that human capacities are not as limited as so 76 often assumed when IQ scores alone are considered. It has been demonstrated that IQ is not a reliable predictor of classroom achievement among delinquent students enrolled in an academic program. The correlation coefficients for the delinquent students between IQ (ranging between 73 and 125) and grade point averages are .14 without self-concept controlled and .07 with the effect of self-concept partialled out. Neither of these correlations are statis- tically significant. In contrast to these low correlations, the coefficients of correlation between self-concept of ability and classroom achievement for the delinquent students are .41 without IQ partialled out and .39 with the effect of IQ controlled. These results would suggest a need for much greater discretion in the use and interpre- tation of standardized intelligence tests. Since it is postulated that self—concept of ability is formulated in an interpersonal setting and subject to modification, and since it is postulated that the learner tends to evaluate himself as he perceives others evaluate him, it follows that it should be feasible to elevate the self-concept of the delinquent student and consequently raise his level of academic achievement by working through appropriate others. The delinquent who has his perceptions narrowed and distorted characteristically reacts against dominating adults and meaningless activity set up by adults. He 77 strikes out against domination with anti-social behavior. He has resisted adult concepts, adult power, order and control. A vicious circle of resistance and counter- resistance has developed. Within the institution, by working through appropriate others, steps can be taken to break this vicious circle. The school can find areas of success and encourage the feelings of realistic self- confidence; the teacher can be consistently interested and supportive; the counselor can give the counselee a sense that his counselor has time for him, has faith in him and will give attention to some of the anxieties that plague him as well as free his capabilities for realistic and more positive perceptions; houseparents, by exercising "power with" rather than "power over," will demonstrate faith in their boy and provide him with activities which will offer him responsibilities and enlarge his social abilities and perceptions. These approaches could possibly produce changes in the delinquent's self-concept of ability and consequently change his level of classroom achievement. Implications for Further Research This investigation has given support to the pro- positions that the learner tends to evaluate himself as he perceives others evaluate him, that the learner's self- concept of his learning ability is a functionally limiting or facilitating factor in classroom achievement, and that 78 non-delinquent, male, ninth-grade students have more positive self—concepts of ability than institutionalized delinquent, male, ninth-grade students. A most likely sequel to this study is to investigate the possibility of changing students' self—concepts of ability and thereby raise their level of performance. Brookover and his associates at Michigan State University are presently conducting such an investigation in a public school setting. A worthwhile counterpart to this investi- gation would be a similar approach at an institution for delinquent boys, concentrating on working through appropriate others at the institution. Other questions which might be investigated more thoroughly are: To what extent does academic achievement reinforce the learners' self—concept of ability? What is the nature of the self-concept of ability and consequent relation to achievement for socially and emotionally mala- justed learners in a public school setting? Can results obtained in this study be generalized to delinquent girls? How do "others" become more meaningful in the student's life? What would be the effects of enrolling institution- alized delinquent students only in those areas where they have positive self—concepts of ability and extending enrollment only as positive self-concepts are generalized to new areas of study? What is the relationship between self-concept of ability and institutional and 79 post—institutional adjustment? What effects do socio— economic factors have on self-concept development? What role does self—concept of ability have in changing socio- economic status? BIBLIOGRAPHY 8O BIBLIOGRAPHY l. Amos, William Earl. "A Study of Self-Concept: Delinquent Boys' Accuracy in Selected Self- Evaluations," Dissertation Abstract, 1961, 20:3179. 2. Association for Supervision and Curriculum Develop- ment. Learning More About Learning. Washington, D. C.: ASCD, 1959. 3. . Perceiving, Behaving, Becoming, 1962 Year- book. Washington, D. C.: ASCD, 1962. 4. Balester, Raymond J. "The Self-Concept and Juvenile Delinquency," Dissertation Abstract, 1956, 16:1169-1170. 5. Bodwin, Raymond Franklin. "The Relationship Between Immature Self-Concept and Certain Educational Disabilities." Unpublished Doctor's Thesis, Michigan State University, East Lansing, Michigan, 1957. 6. Brookover, W. B. "A Social Psychological Conception of Classroom Learning," School and Society, 1959: 87:84-87- 7. . "Improving Academic Achievement Through Students' Self-Concept Enhancement.” Mimeographed application, Office of Education, Department of Health, Education and Welfare, March 15, 1962. 8. . Research in process. East Lansing: Bureau of Educational Research, Michigan Statelkuversity. 9. . Thomas, S., and Patterson, A. Self-Concept ' of Ability and School Achievement. Final report of cooperative research project No. 845, Office of Research & Publications, College of Education, Michigan State University, 1962. lO. Bruck, Max and Bodwin, Raymond. "The Relationship Between Self-Concept and the Presence and Absence of Scholastic Underachievement," Journal of Clinical Psychology, 1962, 18:181-182. 81 82 11. Burke, Richard Leonard. "Ratings of Self and Others as a Function of Expectations and Evaluations,” Dissertation Abstract, 1962, 23(5):1814. l2. Combs, Arthur W., and Syngg, Donald. Individual Behavior, Revised Edition. New York: Harper and Brothers, 1959. l3. . "Intelligence from the Perceptual Point of View," Journal of Abnormal and Social Psychology, 1952: 47:662-73- 14. . Taylor, Charles. ”The Effect of the Perception of Mild Degrees of Threat on Performance," Journal 'gf Abnormal and Social Psychology, 1952, 47:420- 24. - 15. Cooley, Charles Horton. Human Nature and the Social Order. New York: Charles Scribner's Sons,fil902. 16. Davidson, H. H., and Lang, G. "Children's Perception of Their Teachers' Feelings Toward Them Related to Self-perception, School Achievement and Behavior," Journal of Experimental Education, 1960, 29:107-118.’ 17. Deitche, John Howard. "The Performance of Delinquent and Non-Delinquent Boys on the Tennessee Depart- ment of Mental Health Self-Concept Scale,” Dissertation Abstract, 1959, 20:1437-1438. l8. Dinitz, Simon, Reckless, W. C., and Kay, Barbara Ann. "A Self-Gradient Among Potential Delinquents," Journal of Criminal Law, CriminologyL and Police Science,rl958, 49:230-233. 19. . Kay, Barbara Ann, and Reckless, Walter C. ' ”Group Gradients in Delinquency Potential and Achievement Scores of Sixth Graders,” American Journal of Orthopsychiatry, 1958, 28:598-605. «20. Helper, Malcolm M. ”Learning Theory and the Self- Concept," Journal of Abnormal and Social Psychology, 1955, 51:184-194. 21. . "Parental Evaluation of Children and Children's Self-Evaluation," Journal of Abnormal andeocial Psychology, 1958, 56:190-194. 22. Kipnis, Dorothy M. ”Changes in Self-Concepts in Relation to Perceptions of Others,’' Journal of ‘Personality, 1961, 29:449-465. 23. 25. 26. 27. 28. 29. 30. 313 32. 33- \5 83 Kvaraceus, William. "Delinquency," in Chester W. Harris, Editor, EncyclOpedia of Educational Research. New York: The Macmillan Company, 1960 1 pp ° 365'370 - Lively, Edwin L., Dinitz, Simon, and Reckless, Walter, C. "Self-Concept as a Predictor of Juvenile Delinquency," American Journal of Orthopsychiatry, 1962, 32:159-168. 77 Manis, M. "Social Interaction and the Self-Concept," Journal of Abnormal and Social Psychology, 1955, 51:352-370. McNemar, Quinn. Psyphological Statistics, Second Edition. New York: John Wiley & Sons, Inc., 1955- ’ Mead, George Herbert. Mind, Self, and Sociepy. Chicago: University of Chicago Press, 1934. Morse, Richard Johnson. "Self-Concept of Ability, Significant Others and School Achievement of Eighth-Grade Students: A Comparative Investigation of Negro and Caucasian Students," Unpublished Master's Thesis, Michigan State University, East Lansing, Michigan, 1963. Miyamoto, Frank S. and Dornbusch, Sanford M. "A Test of Interactionist Hypothesis of Self-Conception,” American Journal of Sociology, 1956, 61:399-403. Payne, David A. "The Concurrent and Predictive Validity of an Objective Measure of Academic Self-Concept,” Educational and Psychological Measurement, 1962, 22:773-780. Farquhar, William W. "The Dimensions of an Objective Measure of Academic Self-Concept,” Journal of Educational Psychology, 1962, 53:187-192. Reckless, W. C., Dinitz, S., and Murray, Ellen, "Self- Concept as an Insulator Against Delinquency,” American Sociological Review, 1956, 21:744-746. Dinitz, S., and Kay, Barbara, "The Self- Component in Potential Delinquency and Potential Nondelinquency," American Sociological Review, 1957, 22 566-570. 34. 35- 36. 37. 38. 39- 40. 41. 42. 43. 84 Reeder, Leo G., Donohue, George A., and Biblar , Arturo. "Conceptions of Self and Others,‘ American Journal of Sociology," 1960, 66:153-159. Reeder, Thelma Adams, "A Study of Some Relationships Between Level of Self-Concept, Academic Achieve- ment and Classroom Adjustment," Dissertation Abstract, 1955, 15:2472. Roth, R. M. "The Role of Self-Concept in Achievement," Journal of Experimental Education, 1959, 27: 265“281. Sears, Pauline S. "Levels of Aspirations in Academi- cally Successful and Unsuccessful Children,” Journal of Abnormal and Social Psychology, 1940 35:498-5367 Selden, Edward Harvey. ”A Study of Self-Structure and Level of ASPiration in Delinquent and Non-Delin- quent Boys,| Dissertation Abstract, 1961, 21:2394. Staines, J. W. ”Self-Picture as'a Factor in the Class- room," British Journal of Educational Psychology, 1956, 28:97-111. Stevens, P. H. "An Investigation of the Relationship Between Certain Aspects of Self-Concept Be- havior and Students' Academic Achievement?" Dissertation Abstract, 1956, 16:2531-2532. Sullivan, H. S. Conceptions of Modern Psychiatry. Washington, D. C.: William Alanson White Psychiatric Foundation, 1947. ) . Videbeck, Richard, ”Self-Concept and the Reaction of Others,” Sociometry, 1960, 23:351-359. Wylie, Ruth C. The Self-Concept. Iincoln: University of Nebraska Press, 1961. APPENDIX A QUESTIONNAIRES TO IDENTIFY SIGNIFICANT OTHERS 1. General Significant Others 2. Academic Significant Others 85 86 There are many people who are important in our lives. In the space below, list the Names of the people who you feel are important in your life. Please indicate who each person is. NAMES. WHO IS THIS PERSON? If you finish before the time limit, please sit quietly. Do not turn the page. 87 There are many people who are concerned about how well young people do in school. In the space below, list the NAMES of the people you feel are concerned about how well you do in school. Please indicate who each person is. RABIES. WHO IS THIS PERSON? If you finish before the time limit, please sit quietly. Do not turn the page. APPENDIX B SELF-CONCEPT OF ABILITY SCALES 1. Self-Concept of Ability Scale--Genera1 2. Self-Concept of Ability Scale--Specific Subjects 88 89 Circle the letter in front of the statement which best answers each question. 1. How do you rate yourself in school ability compared with your close friends? (DQOO‘W am the best am above average am average am below average am poorest HHHHH 2. How do you rate yourself in school ability compared with those (DO-00‘3” in your class at school? am among the best am above average am average am below average am among the poorest HFHFHfiki 3. Where do you think you would rank in your class in high school? (DQOO‘QJ among the best above average average below average “among the poorest 4. Do you think you have the ability to complete college? (DD-00“!) yes, definitely yes, probably not sure either way probably not no ~~ 5. Where do you think you would rank in your class in college? (DQOC‘QJ among the best above average average below average among the poorest Go on to the next page 9O 6. In order to become a doctor, lawyer, or university professor, work beyond four years of college is necessary. How likely do you think it is that you could complete such advanced work? (DQOC‘W very likely somewhat likely not sure either way unlikely most unlikely 7. Forget for a moment how others grade your work. your own opinion how good do you think your work is? My work My work My work My work My work (DQOC‘W 8. What kind of grades getting? (990693 GO on to the next page is is is is is do mostly A‘s mostly B‘s mostly C's mostly D's mostly E's excellent good average below average much below average you think you are capable In of 91 Now we would like you to again answer some of the same questions, but this time about four different subjects plus shOp courses and physical education activities which you are now taking or have taken in the past. Put an "X" in the box under the heading which best answers the question. Answer for all four subjectsy shop courses and physical educatign. (You will have one "X" on each line.) 1. How do you rate your ability in the following school subjects compared with your close friends? I am I am I am therelow I am above I am poorest average average average the best Mathematics English (Reading) Social Studies Science Shop Courses Physical Education Activities 2. How do you rate your ability in the following school subjects compared with those in your class at school? I am I am I am I am among the below I am above among poorest average average average the best Mathematics English (Reading) Social Studies Science - Shop Courses Physical Educatio Activities Go on to the next page 92 3. Where do you think you would rank in your high school graduating class in the following subjects? among the below I above poorest average average average the best Mathematics English (Reading) Social Studies Science Shop Courses Physical Educatio Activities 4. Do you think you have the ability to do college work in the following subjects? probably not sure yes, yes, . No not either way probably definitely Mathematics English (Reading) Social Studies Science Shop Courses Physical Education Activities Go on to the nextppage 93 5. Where do you think you would rank in your college class in the following subjects? among the below above among poorest average average average the best _. Mathematics English (Reading) Social Studies Science ShOp Courses Physical Education Activities 6. How likely do you think it is that you could complete advanced work beyond college in the following subjects? most not sure somewhat very unlikely unlikely either way likely likely Mathematics English (Reading) Social Studies Science Shop Courses Physical Educatior Activities Go on to the next page 94 7. Forget for a moment how others grade your work. In your own opinion how good do you think your work is in the following school subjects? my work is much my work my work my work‘my work below is below is is is average average average good excellent Mathematics English (Reading) Social Studies Science Shop Courses Physical Education Activities 8. What kind of grades do you think you are capable of getting in the following subjects? Mostly mostly mostly ~most1y ~mostly E's D's C's B's A's Mathematics English (Reading) Social Studies Science Shop Courses Physical Education Activities NOTE: The questionnaire for the public school students did I not include non-academic (shop and physical education) COUPSGS . Go on to the next page APPENDIX C PERCEIVED EXPECTATION OF SIGNIFICANT OTHERS SCALES Ul-F—‘UOIU Parents Best Friend Teacher Houseparents Counselor 95 96 Please answer the following questions as you think your Parents would answer them. If you are not living with your parents answer for the family with whom you are living. you are presently in an institution, answer for the family with whom you are living while not in the institution. Circle the letter in front of the statement that best answers each qpestion. 1. How do you think your PARENTS would rate your school ability compared with other students your age? Among the best Above average Average Below average Among the poorest (DO-IOU!» 2. Where do you think your PARENTS would say you would rank in your class in high school? Among the best Above average Average . Below average Among the poorest (DQOO‘DJ 3. Do you think that your PARENTS would say you have the ability to complete college? . Yes, definitely Yes, probably Not sure either way Probably not Definitely not w Ombcrm 4. In order to become a doctor, lawyer, or university professor, work beyond four years of college is necessary. How likely do you think your PARENTS would say it is that you could complete such advanced work? Very likely Somewhat likely Not sure either way Somewhat unlikely Very unlikely (DQOO‘SD Go on to the next page 97 5. What kind of grades do you think your PARENTS would say you are capable of getting in general? a. Mostly A's b. Mostly B's c. Mostly C's d. Mostly D's e. Mostly E's 6. How far do you think your PARENTS expect you to go in school? a. They expect me to quit as soon as I can b. They expect me to go to high school for a while They expect me to graduate from high school They expect me to go to secretarial or trade school They expect me to go to college for a while They expect me to graduate from college They expect me to do graduate work beyond college CLO (NF-hm 7. In general, would your PARENTS say you are doing as well in school as you are capable of doing? Yes, definitely Yes, probably Not sure either way Probably not . Definitely not (DQOO‘SD 8. What grade do you think your PARENTS would say you are capable of getting in Mathematics? (DQIOU‘QJ LTJUOWI‘P 9. What grade do you think your PARENTS would say you are capable of getting in English (Reading)? mQIoo‘SD meow» Go on to the next page 98 10. What grade do you think your PARENTS would Say you are capable of getting in Social Studies? (DQOO‘SD L'IJUOUiID 11. What grade do you think your Parents would say you are capable of getting in Science? (DQOU‘DJ WUOUHD 12. What grade do you think your PARENTS would say you are capable of getting in Shop Courses? (DCLo;rm tIJUOUiID 13. What grade do you think your PARENTS would say you are capable of getting in Physical Education Activities? (DQIOO‘SD WUOUJP Go on to the next page 99 Think about your closest friend at school. What is this friend‘s name? What grade is this friend in? Now answer the following questions as you think this FRIEND would answer them. Circle the letter in front of the statement that best answers each question. 1. How do you think this FRIEND would rate your school ability compared with other students your age? among the best above average average below average among the poorest 090693 Where do you think this FRIEND would say you would rank in your class in high school? among the best above average average below average . among the poorest (DOIOU‘SD Do you think that this FRIEND would say you have the ability to complete college? yes, definitely yes, probably not sure either way . probably not definitely not (DQOO‘SD In order to become a doctor, lawyer, or university professor, work beyond four years of college is necessary. How likely do you think this FRIEND would say it is that you could complete such advanced word? very likely somewhat likely not sure either way somewhat unlikely very likely (DQOU'QJ Go on to the next page 100 5. What kind of grades do you think this FRIEND would say you are capable of getting in general? a. mostly A's b. mostly B's c. mostly C's d. mostly D's e. mostly E's 6. How far do you think this FRIEND expects you to go in school? a. He (she) expects I can b. He (she) expects for a while c. He (she) expects school d. He (she) expects or trade school e. He (she) expects a while f. He (she) expects college g. He (she) expects beyond college Go on to the next page me me me me me me me to quit as soon as to go to high school to graduate from high to go to secretarial to go to college for to graduate from to do graduate work 101 Think about your favorite teacher--the one you like best; the one you feel is most concerned about your school work. What is this teacher's name? What subject (s) do you have this teacher for? Now answer the following questions as you think this TEACHER would answer them. Circle the letter in front of the statement which best answers each question. 1. How do you think this TEACHER would rate your school ability compared with other students your age? among the best above average average below average among the poorest (DQOU'QJ 2. Where do you think this TEACHER would say you would rank in your class in high school? among the best above average average below average among the poorest (DQIOO‘SD 3. Do you think that this TEACHER would say yOu have the ability to complete college? yes, definitely yes, probably not sure either way probably not definitely not @9069) Go on to the next page -\] Go 102 In order to become a doctor, lawyer, or university professor, work beyond four years of college is necessary. How likely do you think this TEACHER would say it is that you could complete such advanced work? very likely somewhat likely not sure either way somewhat unlikely very unlikely (DQOC‘DJ What kind of grades do you think this TEACHER would say you are capable of getting in general? mostly A's mostly B's mostly C‘s mostly D's Mostly E's (DQ-OO‘SD How far do you think this TEACHER expects you to go in school? a. He (she) expects me to quit as soon as I can b. He (she) expects me to go to high school for a while c. He (she) expects me graduate from high school d. He (she) expects me to go to secretarial or trade school e. He (she) expects me to go to college for a while f. He (she) expects me to graduate from I college In general, would this TEACHER say you are doing as well as you are capable of doing? yes, definitely yes, probably not sure either way probably not definitely not (DQOO‘SD on to the next page 103 Please answer the following questions as you think your present cottage or houseparents would answer them. Name of houseparent(s) referred to Circle the letter in front of the statement which best answers each question. 1. How do you think this HOUSEPARENT(S) would rate your school ability compared with other students your age? among the best above average average below average among the poorest (DQ106513 Where do you think this HOUSEPARENT(S) would say you would rank in your class in high school? . among the best . above average average below average . among the poorest (DD-0693 Do you think that this HOUSEPARENT(S) would say you have the ability to complete college? yes, definitely yes, probably not sure either way probably not . definitely not (DQOC‘SD In order to become a doctor, lawyer, or university pro- fessor, work beyond four years of college is necessary. How likely do you think this HOUSEPARENT(S) would say if is that you could complete such advanced wOrk? very likely somewhat likely not sure either way somewhat unlikely very unlikely (DQOO‘DJ Go on to the next page 104 5. What kind of grades do you think this HOUSEPARENT(S) would say you are capable of getting in general? a. mostly A's b. mostly B's c. mostly C's d. mostly D's e. mostly E's 6. How far do you think this HOUSEPARENT(S) expects you to go in school? a. He (she) expects can b. He (she) expects for a while c. He (she) expects school d. He (she) expects trade school e. He (she) expects while f. He (she) expects college g. He (she) expects beyond college me me me me me me me to quit as soon as I to go to high school to graduate from high to go to secretarial or to go to college for a to graduate from to do graduate work 7. In general, would this HOUSEPARENT(S) say you are doing as well as you are capable of dbing? a. yes, definitely b. yes, probably c. not sure either way d. probably not e. definitely not Go on to the next page 105 Please answer the following questions as you think your present counselor would answer them. Name of present counselor Circle the letter in front of the statement which best answers each question. 1. How do you think your COUNSELOR would rate your school ability compared with other students your age? among the best . above average average below average among the poorest (09406513 2. Where do you think your COUNSELOR would say you would rank in your class in high school? a. among the best b. above average 0. average d. below average e. among the poorest 3. Do you think that your COUNSELOR would say you have the ability to complete college? yes, definitely yes, probably not sure either way probably not . definitely not (DQIOUW:D 4. In order to become a doctor, lawyer, or university pro- fessor, work beyond four years of college is necessary. How likely do you think your COUNSELOR would say it is that you could complete such advanced work? very likely somewhat likely not sure either way somewhat unlikely . very unlikely (DQOU'SD Go on to the next page 5. What kind of grades do you think your COUNSELOR would 106 say you are capable of getting in general? (DQOU‘SD . mostly A's mostly B's . mostly C's mostly D's mostly E's 6. How far do you think your COUNSELOR expects you to go in school? a. b. O (D f. g. He (she) expects me can He (she) expects me for a while He (she) expects me high school He (she) expects me or trade school He (she) expects me a while He (she) expects me college He (she) expects me beyond college to to to to to to to quit as soon as I go to high school graduate from go to secretarial go to college for graduate from do graduate work 7. In general, would your COUNSELOR say you are doing as well as you are capable of doing? a b c d e . yes, definitely . yes, probably not sure either way . probably not . definitely not APPENDIX D EVALUATION OF SIGNIFICANT OTHERS SCALE 107 108 ‘ to‘ ‘ t‘ 3 r0 0f 1‘ - a ‘ -nt w. o 0‘ answers eaph guestipn in rating the fpllowing gtudept: How would you rate this student's school ability compared with other students his age? among the best above average average below average among the poorest (DQOU‘W Where do you think this student would rank in his class in high school? among the best above average average below average among the poorest (DD-.0693 O O O Q Do you think this student would have the ability to complete college? yes, definitely yes, probably not sure either way probably not . definitely not (DQOU‘W In order to become a doctor, lawyer, or university pro- fessor, work beyond four years of college is necessary. How likely do you think this student could complete such advanced work? a. very likely b. somewhat likely c. not sure either way d. somewhat unlikely e. very unlikely Continued on next page 109 What kind of grades do you think this student is capable of getting in general? a. mostly A's b. mostly B's c. mostly C's d. mostly D's e. mostly E's How far do you think this student will go in school? He He He He . He He He UQHJGQJOO‘QJ will will will will will will will quit as soon as he can go to high school for a while graduate from high school go to secretarial or trade school go to college for a while graduate from college do graduate work beyond college In general, would you say this student is doing as well as he is capable of doing? (DQOU‘W Yes: yes, definitely probably not sure either way probably not definitely not APPENDIX E CORRELATION MATRICES OF MAJOR VARIABLES JI'UUID State Training School Delinquent Sample Private Institution Delinquent Sample Total Delinquent Sample Non-Delinquent Sample 110 111 oe I oe on do I on on ad onI I on an oAI nnI on I nn on ~« oaI «u nn I on nn ~o o~ o~ no- on I an on ouI ooI oo n« no ««I I on nn n«I odI oo o« oo o«I on I nn «n n« no on u« o~ «oI o« no I «n an ndI no nHI noI «AI go o« o~ noI I an vouuwao naqaauooc on o«I o« «oI «a euI n~ «« n« mg on I on n« noI «a noI «oI an «« noI no oo o«I oe I o« o« noI noI on oo ~« odI oe an on no- Nd ed I o« n« oo ~oI «oI noI «a ogI no on ««I oo HaI nuI ad I n« o« no noI n« noI no ms «0 «o no oo o« m« do eoI I o« n« noI noI nn oo on do ma nd ma oo «« o« oo oo an I n« a« «oI n~I mg mg «o oo n« o« -I «n na ou no «« «a «« I . o« 2 so- 8- 8-3 2- 3- 2 on 2- 2 an 3 2 8 2 8 B I 2 I «Raw 5...... . So so... .3 «« no odI o« oo an no «oI ooI no «a on no noI do oo on nu ~« I «« “on zv a u «A o a a m u a ya u m ~« ooI naI na mu noI noI «« on odI ~« «a an no u« oo o« on no a« I ~« imman mono: no xHuH¢l nowhcaumuoo o« ««I noI o~ co no sn o« o~ ca «oI on o~ no do an no oo. ooI on no I o« o~ oo oHI an no mg no so do- «oI «n ouI noI oo o~ no mg on oo mu an no I an cg ~o «oI u« no «M nn o~ o~ «o no «d o« no ad «e on on o« «« o« noI nu I ad mg noI naI no «a noI oo n« as «AI o« on om no on «a on «o no o~ do noI on nn I “a on co n~I co ~oI ooI «o oo on «AI so an an «a on ma o« no «o n« no no on e« on I oH ma noI noI ««I onI “«I ~«I noI no na no oo naI on noI noI noI oaI no mo uaI oo o«I AAI noI «oI I no «a ooI dd do «a uoI noI «o on o« on n« «oI on noI no o« a“ nn o~ n« no no on o« Nd on I o~ nu noI doI ooI noI «oI no eaI no ««I o« o« ouI oo on o« an o« on an n« so no «« n« c« oo nn I nu Nd n~ -I n« ad o« ~o noI «A an no no no «e no on n« A« on on «« no on o« o« n« co on «n I ~H ad -I oo «o «oI co noI noI an «oI an on noI on «aI oo no co- mg «a do oHI euI nu so noI on an on «e I Ad oa -I uoI an- co- oaI o«I noI oo o~I n« no n«I no n« noI nAI an on «oI no on- do no oH o~ oo on nn nn nn I on a noI oo noI no nuI euI «« nn ooI ~« .dn co on an «o no no nu on an eoI o« oo «n no «n no no «o o« «o I a n nu n« naI ««I ««I noI oo noI nn o«I ooI no «a co noI no oo MoI o« oo noI aoI noI «oI no ~n do «oI noI «o sdI ooI I o . n «oI nu mo n« uoI «oI n« o« «n «« «« noI n« no oH u« on on «a «o na n~ on on Nd «a on n« oo «« no n« oo I n o ndI o«I noI ««I doI noI goI no ~«I oo a» on so no on an no do n~ nn n~ n« so oo on oo. so on so no no o« «o no I o n NA noI «n on n« no on on on onI noI «n on no no ma on sn «o on on n« n~ on on nuI n« ooI «n naI aNI on «a do n« I n c nfiI qu noI o~ ooI ~o on q« m«I no nn an «o noI do- noI on o« ch n« odI nu od nn o« naI ««I ndI n«I n« onI noI noI daI on no I o n udI noI n~I noI naI «uI no «« o«I an o~ aoI nu on no ooI oo on noI «n n«I on a« so on ma o« an n« on on no on- on n« «o o«- I n « oaI naI oo n~ oo. «o ha nn noI doI o« on n« so o~ nu no on no no No. oe on no oo o« an «n on «« «« nn on m« on oc nn no I « — voI nAI noI ooI oo od ««I n~I ndI noI «o «~ noI noI «a o« oo «a we no oo nmI ooI “A n~ oo no on noI «o noI so no naI ma doI «o qu do I or on on an on nn on nn «n ~n on m« o« u« o« n« o« n« «« ~« o« ma on an og nu ea nu «A dd on n o n o n o n « a mewmwwwwwwwwwmnmmmouwannaunnunuunnmwmwnw I u 0 u. 1. U. 8 O 3 3 O O O 3 n... n... I 1 1 1 1 1 1 J 11 1. 1. 1. 1. 1. .l 1. D a .. w .I I i I I 1 z .. I .. wcnoaoVN a v” » vm a n a»s»s»s»au a n ... ... .... .... .... ... ... m. w. h 1. V M h h h h h h h h WI.III0.ITO.dI.O.I 1. ....u.....u.....3vIo..I 7pm.}... . I P I I I I I I I I IIdIIIIIA II A T... A AAAOAIADAMAJA A Q m W m m. mu m 1. W r.. m 1.3I333TOIIJIIII IIIdIIITIIIQII w u w w n m n V m D V M W... S m. M m 0033 IIJIPUMPTI PPIP PwPIPIP.PP a a I a P r.. u 1 a 3 1I w .a 3 .A 3 I my 1. I I I I .fl a I I u Irv n .4 I I A o .... I 3 1. .deAIa NMKJMMIM MI... Mu.” “Md d d dd m u n I m D. D. I d I .1 1. u. II3III A I 1. n AS 3 3 .4 .4 3 3 3 I I 3 I I m 1. u I 1. . IIJI... dI II 1 I I IJI I.4I I II . . . . . . . .... 3 .4 u u I I 9 a 1. I datoua I... 1. a IIWII... 1W... 1. 1.... . . . . . . . I TmI3IJIIJI “.9 IHIPIxmannmnanmwmmnuwssmmna .4 n .4 o a 1. I S 1 III...» 1. a II d I nIII 1. I a o I I n u 3 S 1 P O 3 D I aI A9A .u... 13 .4 3 .433 .4 3I3 3 .43 (A o 1. a 3 u u .4 .4 3 V 9 m I 1 n 1 D. .(muI I T 31.. 1.1.3.3 I .90.} 1. 1:... I d I 7' 1. u. I D. 1r 3 O A V. J I D. I I 0 0.1.9. 14% IO 0 O 030 O O O 0 00 11 u I 1: . 1 D o S a 1 I 1. I n P ..I P In" I .I. du u u qu u u u u no I o 1. I I x u o o I 1 .l P a I u I .MMIM o .4 o I m. I u. 1. O J I .- I 5.0 3 no 10 O 0 Duo 0 O O O 00 1. s \I ... I I S 3 .4IIIU.I .4. 0.4. 31...: .4. .9 .4. .4. .4. 3.4. I 3 . . r m a a Tutu". r u s n . m m I... . a... mosaunuuuuuuu.» I . I V P I3 I 0 SM In n m I J 1 ... J 1 Ju n n I I I\ D I .I .4. 3 I D. 3 I m I I I m D 3 J I r. m. a. a I u. u u n u u 3 I I W I .d I Md N I H n .4 3 3 .4 .43 1. S » Ionnmwp.-... ......m n 1. u. I 3.. II I I 1. 1. 1. T I. .l . I u I n u u u u I u m. u I (1 1.4 I 3|: D. I\ I C 1.1.22 “1 MM I MM NM Me I NM MM OA MAI I MM MM MMI MMI MOI I MM MM MNI OA NMI MM I QM MM OMI AO MMI MM QM I MM NM QO MM NAI MOI AM MN I NM AM NO MOI NAI MO ON AA MO: I AM OM MNI MOI MNI OM OM MM MA MA I OM MN NNI MA NOI MN MN OM MN NN MM I MN MN MAI oAI MAI NM oo MN Mo 00 an AM I MN .oouuAso «AeeAuooa NN AMI OO MAI MA NM NM MN MAI MM AM NA I NN MN MOI MN: MNI NM «A «N MNI MOI AN MA MN MA I MN MN AMI OMI OAI AM AM MM MOI OAI OM AM MM OM NM I MN ON ONI NA ONI AM Me MM MA MA MM MN MO MO MN MM I ON MN ONI NO MNI MN AM MM MOI MA NM AM MM MA MM NM NM I MN NN MOI MNI NOI MM MN OM MAI NNI AM MN MM NM NM MM AN NN I NN u ME .~ om- No «A- on on on no no no o~ as .oo on No so no an I a. .on 2. c. we .oo..c...o eoaosoaooea 9...... ON MMI MAI NO MM MM AM MOI OO MM MM NM MA NM MN NA MO NM MN I ON immAM MOMMS MO KAMH<2 ZOAH