g .. .w. “mm w. , . r. W as is? wizzéféi... , 3.9.57 v . _ 1 . ..... .. . t: . 3 . . . . .. . , inhuman. @178 3..,_v..v.;y..._.3, ..._!.,...::.....‘.as‘.i. . . cm a V . . . . . . . ‘ .. ‘ :.. r...~.......a..w,h,.. A . .z . fl" I .I. 1‘. IHIIIWIIlllllllllllllllllllllllllllllllllllllllllllllllllll 1 31293 10373 5100 LIBRARY " Michigan State ' University 5. ‘ This is to certify that the thesis entitled AN EVALUATION OF A COMMUNITY BASED DELINQUENCY PREVENTION PROGRAM ON THE BASIS OF GROUP AND INDIVIDUAL EMPLOYMENT presented by Wilbur Montgomery Whitney, Jr. has been accepted towards fulfillment of the requirements for Ph.D. Psychology degree in Date 0-7639 ABSTRACT AN EVALUATION OF A COMMUNITY BASED DELINQUENCY PREVENTION PROGRAM ON THE BASIS OF GROUP AND INDIVIDUAL EMPLOYMENT BY Wilbur Montgomery Whitney. Jr. Juvenile delinquency is a multivariate problem involving numerous social and economic characteristics in our society. There is a deficit of systematic evaluative research in the natural environment of delinquents and a need for the development of community-based programs encom- passing social and economic variables in order to prevent and reduce delinquent behavior. The present study exam- ined the effects of developing a community-based. peer- operated automotive repair business on the reduction of delinquency. The results indicate that participation in the peer-operated business seems to have its major effect in reducing the occurrence of arrest relative to the control Wilbur Montgomery Whitney, Jr. treatment during the initial part of the experiment. The experimental groups superiority is largely due to induc- ing socially acceptable behavior in its participants through an intensive small group dynamics experience and structured vocational training. Various programmatic procedures were discussed. Recommendations for future research and the need for community-based delinquency prevention programs were made. AN EVALUATION OF A COMMUNITY BASED DELINQUENCY PREVENTION PROGRAM ON THE BASIS OF GROUP AND INDIVIDUAL EMPLOYMENT BY Wilbur Montgomery‘Whitney. Jr. A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Psychology 1974 To Vannie. Erika & Michelle ii ACKNOWLEDGMENTS An innovative community longitudinal research project could not be accomplished without the help of many people dedicated to finding solutions to problems of our society. I would first like to thank my beauti- ful parents for without them I wouldn't be writing this acknowledgment. I would also like to thank all my brothers and sisters for without their sweat and blood. the educational opportunity afforded me would not have happened. I would like to express my great appreciation to my committee chairman. Dr. Louis G. Tornatzky. whose friendship. patience. guidance. understanding. and faith- ful support have made this research possible. Working with Lou T. has made my educational experience at Michi- gan State University an appreciative and rewarding exper- ience. My special thanks go to Dr. George W. Fairweather. whose sense of humor. advice and understanding has helped me cope with the problems and frustrations encountered in iii this research. I would also like to thank my other com- mittee members. Dr. Robert Calsyn and Dr. George Logan. And thanks also to Mrs. Marjorie Curtis for her patience and secretarial skills. Special thanks to the entire staff of Youth Development Corporation for providing me a unique oppor- tunity to utilize my skills for the people. Especially. I thank Bill Leavell and Walker Beverly for their contin- uous support in helping me overcome numerous frustrations. as well as all the beautiful young people of Community Automotive Corporation who served as participants for this study. I thank my friends and fellow associates for being interested in my research and providing moral sup- port. eSpecially my friend and companions Garret Payne and Lonnie Anderson. Finally. and most important. my deepest thanks go to my loving wife. Vannie. for her unwavering support. assistance and companionship throughout. iv TABLE OF CONTENTS Page LIST OF TABLES O O 0 O O O O O O O O O O O O O O 0 vii LIST OF APPENDICES o o o o o o o o o o o o o o o o Xii CHAPTER I. INTRODUCTION. . . . . . . . . . . . . . . l The Problem . . . . . . . . . . . . . . . 1 Small Group Dynamics. . . . . . . . . . . 7 Employment and Delinquency. . . . . . . . l4 Purpose . . . . . . . . . . . . . . . . . 19 Experimental Hypothesis . . . . . . . . . 20 II. METHOD AND PROCEDURE. . . . . . . . . . . 21 Experimental Design . . . . . . . . . . . 21 Participants. . . . . . . . . . . . . . . 22 Experimental Subsystem. . . . . . . . . . 23 Control Subsystem . . . . . . . . . . . . 32 Measurement of Dependent Variables. . . . 35 Measurement of Internal Group Processes . 39 III. RESULTS . . . . . . . . . . . . . . . . . 42 Intake Results. . . . . . . . . . . . . . 42 Follow-up Results: Self-Reports. . . . . 45 Follow-up Results:r Police Data . . . . . 52 Follow-up Results: School Data . . . . . 57 Internal Analysis on Basis of Police Data . . . . . . . . . . . . . . 62 Group Process Results . . . . . . . . . . 7S TABLE OF CONTENTS (cont'd. CHAPTER IV. REFERENCES. APPENDIX. DISCUSSION. ) vi Page LIST OF TABLES TABLE Page 1.0. Experimental Design of Study. . . . . . . 21 2.0. Matching Procedure of Participants. . . . 24 3.0. Entry Dates and Follow-Up Periods for Participants. . . . . . . . . . . . . . 37 1.1. Comparison of Participants on Age Upon Entrance into Experiment . . . . . 43 1.2. Comparison of Participants on Educa- tional Status Upon Entrance in the Experiment. . . . . . . . . . . . . . . 43 1.3. Comparison of Participants on Y;D.C. Module Classification Upon Entrance in the Experiment . . . . . . . . . . . 44 1.4. Comparison of Participants on Race Upon Entrance in the Experiment. . . . . . . 44 2.1. Comparison of Participants on Self Reports on Cumulative Arrests for 90 Days of Follow-Up. . . . . . . . 46 2.2. Comparison of Participants on Self Reports on Educational Status for 90 Days of Follow-Up. . . . . . . . 46 2.3. Comparison of Participants on Self Reports on Employment Status for 90 Days of Follow-Up. . . . . . . . 47 vii LIST OF TABLES (cont'd.) TABLE Page 3.1. Comparison of Participants on Self Reports on Cumulative Arrests for 180 Days of Follow-Up . . . . . . . 47 3.2. Comparison of Participants on Self Reports on Educational Status for 180 Days of Follow-Up . . . . . . . 48 3.3. Comparison of Participants on Self Reports on Employment Status for 180 Days of Follow-Up . . . . . . . 48 4.1. Comparison of Participants on Self Reports on Cumulative Arrests for 270 Days of Follow-Up . . . . . . . 50 4.2. Comparison of Participants on Self Reports on Educational Status for 270 Days of Follow-Up . . . . . . . 50 4.3. Comparison of Participants on Self Reports on Employment Status for 270 Days of Follow-Up . . . . . . . 51 5.1. Comparison of Participants on Self Reports on Cumulative Arrest for 360 Days of Follow-Up . . . . . . . 51 5.2. Comparison of Participants on Self Reports on Educational Status for 360 Days of Follow-Up . . . . . . . 52 5.3. Comparison of Participants on Self Reports on Employment Status for 360 Days of Follow-Up . . . . . . . 52 viii LIST OF TABLES (cont'd.) TABLE 6.1. Comparison of Participants on Police Data on Cumulative Arrests for 90 Days of Follow-Up. . . . . . . . . 6.2. Comparison of Participants on Police Data on Cumulative Arrests for 180 Days Of FOllOw-Up o o o o o o o o 6.3. Comparison of Participants on Police Data on Cumulative Arrests for 270 Days Of FOllOw-UP o o o o o o o o 6.4. Comparison of Participants on Police Data on Cumulative Arrests for 360 Days of Follow-Up . . . . . . . . 7.1. Wilcoxon Sum Rank Test on School Absences (1/2 days) from School Data. 7.2. Wilcoxon Sum Rank Test on School Suspensions from School Data. . . . . 7.3. Wilcoxon Sum Rank Test on Mean Grades from School Data . . . . . . . 8.1. Comparison of Participants on Police Data on Arrests. Who Entered the Experiment During the First Six Months. for 90 Days of Follow-Up. . . 8.2. Comparison of Participants on Police Data on Arrest. Who Entered the Experiment During the Second Six Months. for 90 Days of Follow-Up. . . ix Page 54 55 56 57 59 60 61 64 65 LIST OF TABLES (cont'd.) TABLE 8.3. 8.4. 9.4. 10.1. 10.2. Page Comparison of Participants on Police Data on Arrests. Who Entered the Experiment During the First Six Months. for 180 Days of Follow-Up . . . 66 Comparison of Participants on Police Data on Arrests. Who Entered the Experiment During the Second Six Months. for 180 Days of Follow-Up . . . 66 Comparison of Participants on Police Data on Arrests. Who were in Experi- ment More Than Three Months for 90 Days of Follow-Up. . . . . . . . . . 69 Comparison of Participants on Police Data on Arrest. Who were in Experi- ment More Than Three Months for 180 Days of Follow-Up . . . . . . . . . 69 Comparison of Participants on Police Data on Arrest Who were in Experi- ment More Than Three Months. for 270 Days of Follow-Up . . . . . . . . . 70 Comparison of Participants on Police Data on Arrest. Who were in Experiment More Than Three Months. on 360 Days of Follow-Up. . . . . . . . . . . . . . 70 Comparison of Participants in Experiment Less Than Three Months on Police Data Arrests for 90 Days of Follow-Up. . . . 71 Comparison of Participants in Experiment Less Than Three Months on Police Data Arrests for 180 Days of Follow-Up . . . 71 LIST OF TABLES (cont'd.) TABLE 10.3. 10.4. 11.1. 12.1. 12.2. 12.3. 12.4. 13.1. 14.1. Page Comparison of Participants in Experiment Less Than Three Months on Police Data Arrests for 270 Days of Follow—Up . . . 72 Comparison of Participants in Experiment Less Than Three Months on Police Data Arrests for 360 Days of Follow-Up . . . 72 Comparison of Participants on Self Report Explanations for Leaving the Experiment. . . . . . . . . . . . . 74 Matched Pairs on 5/15 Sociometric Rat- ings Done Before Some Participants Self Reported Arrests . . . . . . . . . 77 Matched Pairs on 2/15 Sociometric Rat- ings Done After Some Participants Self Reported Arrests . . . . . . . . . 78 Matched Pairs on 2/15 Sociometric Rat- ings Done Before Police Reported Arrests . . . . . . . . . . . . . . . . 79 Matched Pairs on 5/15 Sociometric Rat- ings Done After Police Reported Arrests O O O O O O O C O O O O O O O O 81 Comparison of Participants Mean Ratings on Morale . . . . . . . . . . . . . . . 83 Comparison of Participants Mean Ratings on the Job Behavior Scale . . . . . . . 84 xi LIST OF APPENDICES APPENDIX Page A. JOB EVALUATION FORM . . . . . . . . . . . 104 B. CONTROL PARTICIPANT'S JOB PLACEMENTS. . . 105 C. PARTICIPANTS INTAKE FORM. . . . . . . . . 106 D. PARTICIPANTS FOLLOW-UP FORM . . . . . . . 108 E. SOCIOMETRIC RATING SCALE. . . . . . . . . 110 F. MORALE RATING SCALE . . . . . . . . . . . 111 G. JOB BEHAVIOR SCALE. . . . . . . . . . . . 113 H. ADMINISTRATIVE AGREEMENT. . . . . . . . . 115 xii CHAPTER I INTRODUCTION The Problem Juvenile delinquency is a multivariate problem and has been related to numerous social and economic characteristics in our society. i.e. the family (Glueck and Glueck. 1968). social class (Empey and Erickson. 1970). education (Coleman. 1966). and socio-economic status (Shaw and McKay. 1970). In general. researchers have been concerned with why some children become delin— quent and how delinquency can be prevented. controlled and treated. While many delinquency programs have been conducted in institutional settings. there is a great need for systematic and evaluative research and the development of programs encompassing social and economic variables in the natural environment of delinquents. There are many factors which seem to contribute to the development of delinquency and most have been thoroughly documented. Many theorists consider the fam- ily the most significant factor in the development of juvenile delinquency (Glueck and Glueck. 1968; Monahan. 1957; Browning. 1960; Gold. 1963; Slocum and Stone 1963; Peterson and Becker. 1965). In general. these studies indicate that delinquents are exposed to broken or dis- organized homes and family environments. It has also been shown that delinquency is related to family struc- ture (Rubin and Hill. 1970) as well as socialization in the family (Shafer and Knudten. 1970). Another variable which has been related to delin- quency is social status. Empey and Erickson (1970) did a rather comprehensive analysis of the relationship of delin- quency to social status. They found that middle class boys in contrast to lower and upper class boys commit the most serious offenses. 0n the other hand. Gold (1966) and Dell (1963) have indicated that one of the factors appearing to facilitate delinquency was a low level of socioeconomic status; while Pine (1965) found no significant relation- ship between social status and delinquency. The contradiction in the aforementioned results indicate one of the prevalent problems in delinquency research; that different conclusions can be drawn depend- ing upon the variable studied. sample used and method of data collection. For example. those youth whose anti- social behavior brings them into court are more likely to be from a lower socioeconomic level. than those released by police to their parents. Therefore. if a researcher is interested in the relation of delinquency to socio- economic status. results will vary depending upon the source of collecting the data. In addition to the characteristics of the family and socioeconomic status a number of other character— istics have been shown to be related to delinquent behav- ior. such as delinquent behavior is directly influenced by urban areas (Spergel. 1966; Gordon. Short. Cartwright and Strodback. 1963). education status of delinquents tends to be lower than non-delinquents (Coleman. 1966; Cloward and Jones. 1963; Elliot. 1962) and delinquency often develops from identification with delinquent peers (Geiss. 1967; Empey. 1966). As shown from the preceding literature review. delinquency has been related to a number of character- istics. However. the exact relationship between delinquency and these characteristics is still relatively unknown and has made the development of prevention. con- trol and treatment programs a difficult task. Wheeler et a1 (1969) has adequately summarized the difficulty of developing delinquency research by stating the following: The field of delinquency touches a wide variety of social institutions. Its causes are still incompletely understood. Indeed. the number of proposed solutions is as great as the number of occupations. profes- sions and organizations that have a stake in delinquency prevention and control. However. in recent years a wide variety of com- munity. state and federal agencies have become involved in delinquency prevention. control and treatment. The major criticisms of these efforts are the lack of experimental evaluation. inappropriate testing. the inability to con- trol important variables and the difficulty in determining their impact on reducing crime and delinquency. Several community-based delinquency prevention pro- grams have been initiated. such as. the Fuld Neighborhood House. Carson Pirie Scott EE Program and the Los Angeles Youth Project (Trojanowicz. 1973). These programs local- ized neighborhood efforts and attempted to work with socially and economically "disadvantaged" area residents. Unfortunately. these programs have not been subjected to appropriate testing and it is difficult to evaluate them objectively. Upon evaluation of these programs it was felt that although most did influence the youngsters in the area and did help reduce rates of delinquency. the evaluations were not rigorous and many of the variables used were not controlled. Therefore. it is difficult to evaluate their success realistically and to definitely state that they did have a major impact on reducing crime and delinquency. In addition to community-based preventive programs. a number of juvenile rehabilitative programs have been initiated in recent years. The Provo Experiment (Empey and Rabow. 1961); the Highfields Project (McCorkle. Elias and Bixby. 1958); and the Fremont Experience (Seckel. 1967) provide a sampling of innovative approaches to the problem of delinquency. However. these three programs are concerned with rehabilitation of delinquents once they have come in contact with the formal criminal justice system. But these very institutions often reinforce the youths negative attitudes toward authority and make it increasingly difficult to work out personal problems. This coupled with the artificial atmosphere of the insti- tution. does not create a situation conducive to personal growth. rehabilitation and increased social functioning (Trojanowicz. 1973). The present study. therefore. is concerned primarily with prevention in a community set- ting. In summary. there is a great need to develop pro- grams which not only prevent youth from being institution- alized. but which also allow for the development of social and personal growth and individual autonomy in their com- munity environment. Further there is a need for programs which can be systematically evaluated through experimental research and which can provide measurable quantitative results of their impact on reducing crime and delinquency. The present study will attempt to meet these needs through the development of an innovative delinquency pre- vention program which: 1. Includes a longitudinal naturalistic experiment. 2. Provides built-in methods for systematic program evaluation. 3. Is based within the community of the delinquent youths. More specifically. the present investigation will provide an experimental study utilizing small group dynamics and employment as a means of preventing juvenile delinquency. The major concern here is developing a program in which delinquents can achieve personal independence and auton- omy through peer support. Small Group Dynamics One of the purposes of the present study is to investigate whether or not the formation of small task- oriented groups will affect the reduction in delinquent behavior. For this reason. groups will be defined and some of the paramount features of small groups and related research will be discussed. There is no single definition of "group" that is generally accepted by all students of small group behav- ior. Most researchers have defined small groups in terms of specific characteristics rather than a single defini- tion. Groups have been defined on the basis of a number of characteristics. such as perception of members (Bales and Slater. 1955). motivation and need satisfaction (Bass. 1960; Cattell. 1951). organization (McDavid and Harari. 1968). and interdependency (Cartwright and Zander. 1968). While many of the previously mentioned character- istics are involved in the formation. perception and con- sequences of group processes. none of these factors is either necessary or sufficient to define "group." Hare (1962) has summarized several characteristics which dif- ferentiate the group from a collection of individuals. These are: 1. The members of the group are in interaction with one another. 2. They share a common goal and set of norms. which give direction and limits to their activity. 3. They develop a set of roles and a network of interpersonal attraction. which serve to differ- entiate them from other groups. The definition of group by Shaw (1971) will be used in the present study. He states that a group is defined as two or more persons who are interacting with one another in such a manner that each person influences and is influenced by each other person. Utilizing the preceding definition of a group. several salient parameters of small groups will be dis- cussed in relation to the development of small. informal. task-oriented groups. Although there are several conceptions of group principles pertinent to the develop- ment of groups in this investigation. empirical data rela— tive to these factors and their effect is sparse. Those principles to be discussed are: 1) group goals; 2) com— munication; 3) cooperation; and 4) group support. In the typical group. there exists at least one goal. which is acceptable to a majority of the group and can be properly identified as the group goal. Zander and Newcomb (1967) have shown that group members who accept the group goals are motivated to enact activities that are expected to aid in the achievement of this goal and they are pleased (experience tension reduction) when there is movement toward the goal and when the goal is achieved. These findings might be relevant to the development of task-oriented groups in the present investigation. Group communication is the second group principle pertinent to this study. The studies of Maier (1950) and Shaw and Blum (1965) show that on difficult tasks group performance is facilitated by the extent to which group members can freely communicate their feelings to satisfac- tion or dissatisfaction. According to these researchers effective group functioning can be facilitated by providing 10 the opportunity for group members to express their feel- ings and opinions in an uninhibited manner. There are several ways in which this opportunity for expression can be provided--through a group leader. who encourages the expression of minority opinion. or by some device which permits an indication of satisfaction or dissatisfaction without focusing attention on the group member. The third group principle to be discussed is the relationship between cooperation and competition and group performance. Cooperation can be defined as a situation in which the goals of the group are homogenous. and competi- tion as a situation in which the goals are heterogenous. Several studies have shown that groups perform more effi- ciently when they are cooperative (Blau. 1954; Hammond and Goldman. 1961; Raven and Eachus. 1963; Deutsch. 1949). This effect is produced largely through the specialization of individual contributions and through helpful actions of each member vis-a-vis other members in the cooperative situation. Motivational factors are also important. Even though the competitive situation may arouse greater moti- vation than the cooperative situation. this increased motivation does not always increase group performance. 11 It has also been shown that group support is important in sustaining group activity (Hollander. 1967). The approval of others often provides a significant rein- forcement function in lending support to individuals' actions. Such support can also play a vital preservative role and often acts as a gauge for the individual to know how he is doing in the group (Kelly. 1952). Group sup- port is closely tied to the motivation to take part in functional group tasks. Thus. whenever individuals come together to achieve some function. they often rely on one another for adequate performance. Wyer (1966) has shown that to the extent that individuals find the achievement of the group's goals rewarding. they will act in conson- ance with its achievement and be supported by the positive response of others. On the basis of the preceding literature review on group goals. communication. cooperation and group support. it was decided to investigate these principles as they relate to task-oriented groups in this study. In addi- tion. since delinquents often manifest their behavior as a part of a group or gang (Thrasher. 1936; Cohen. 1955; Ohlin and Cloward. 1960; Miller. 1958). it is felt that 12 the incorporation of group dynamics would be a natural vehicle through which to view how delinquent behavior may be reduced. Many approaches and methods incorporating small group dynamics have been used in treating delinquents and other types of "social deviant" behavior. Parole and pro- bation agencies have been experimenting with group treat- ment in recent years. However. rather loose and ambiguous terminology has resulted as well as the lack of experimen- tal evidence which shows the effects of these methods. One group treatment method used with delinquents is guided group interaction. This is based upon the assumption that through the group and its processes the delinquent can solve his problems. According to McCorkle (1954). guided group interaction assumes that delinquents will benefit from the freedom to discuss their problems and their own roles and relationships within the group. While guided group interaction has been used primarily for rehabilitation purposes within institutional settings (i.e. Highfield Project). it has not been proven to be effective in detering delinquency in the natural environ- ment of delinquent youths. 13 A number of other group treatment methods have been used with delinquents. such as group counseling (Sharp. 1959; Sarri and Vinter. 1965; Walker. 1959). gggup therapy (Gazda. 1968; Shellow. Ward and Rubenfield. 1958; Allen. 1970). and group psychotherapy (Hesko. 1962; Schulman. 1957). Some of the deficiencies of these meth— ods were noted by Slaikeu (1973). who reviewed twenty- three studies evaluating group treatment of juvenile de- linquents. The major criticism drawn from the review is that. although the evaluation studies report a variety of positive results. as a whole the investigations fall short of meeting the criteria of scientific research. This makes it impossible to determine if these group treatment methods are effective modes for rehabilitation of delin- quents. In a somewhat different approach. Fairweather et a1 (1964) has shown that. compared to traditional treat- ment programs of schizophrenic patients. a small group program demonstrated a heightened social activity in all situations and patients more frequently perceived their fellows as socially desirable. In exploring group pro- cesses of these problem solving task groups. they also 14 discovered that they adequately solved complex interper- sonal problems and they frequently considered their own task group and members as helpful in their recovery. In a later study. Fairweather et al (1969) incor- porated small group dynamics in developing a community treatment program of mental health patients. They found that in the community the activated groups resulted in excellence of group leadership and group performance. Although the findings of Fairweather (1964. 1969) did not deal directly with delinquents. they do indicate the use- fulness of incorporating small group dynamics in a com— munity treatment program. Employment and Delinquency In addition to utilizing small group dynamics. the present study will also provide employment opportunities to delinquent youths. Income and employment are areas so fundamental to our society that they become a natural set— ting for efforts designed to reduce delinquency. The ability to find a suitable job or occupation is one of the 15 basic means our society provides for the legitimate achievement of common success goals. Fairweather et al (1969) derived a series of prin- ciples for the operation of a community treatment program which should be relevant to the present investigation. These principles involve the employment of mental patients in their own business. They noted that as the members developed a greater stake in the social system. they be- came more responsible; pride came with personal indepen- dence and the ownership of a business. Their results further indicate that community treatment programs need to provide as much autonomy to their members as possible. as well as meaningful work (as society defines it). These findings suggest that for both the mental patient and other marginal individuals there is an urgent need to cre- ate new and more participative social statuses and roles. There are a number of investigators who believe economics and employment are closely related to delin- quency. Cloward and Ohlin (1960) hypothesize that: Delinquency represents not the lack of motivation to conform but quite the opposite; the desire to meet social expectations itself becomes the source of delinquent behavior. if the 16 possibility of meeting these expectations are limited or nonexistent. According to Shaw and McKay (1970) there is a precise relationship between economic status and delinquent behav— ior. They believe the greater the economic deprivation the greater the delinquency; the lesser the economic de- privation the less the delinquency. Although there are many other variables which con- tribute to delinquent behavior. economic factors clearly do have some criminogenic importance. The exact etiolog- ical importance and independent effects of economics and employment on the shape and incidence of delinquency are still unclear (Schafer and Knudten. 1970). In other words. merely providing an employment opportunity to a delinquent youth does not mean the automatic disappearance of delin- quent behavior. The relationship between employment and delin- quency is evidenced by the comprehensive study of Fleisher (1966). who has shown that both income and employment can be correlated with delinquency rates. He found that in- creases in the unemployment rates were associated with increased delinquency rates in several large cities over a period of years. Estimates of this study indicate that 17 a ten percent rise in employment might produce a twenty percent drop in delinquency. Singell (1966) has also con- cluded that delinquency is significantly and positively correlated with unemployment. The use of employment as a means of reducing delin— quency is neither new nor innovative. Since 1962. Congress has passed many laws dealing with manpower. unemployment. and related problems of "disadvantaged" and delinquent youth (i.e. Manpower Development and Training Act of 1962). At least thirty—five federal financed programs for youth have been developed. such as Jobs Corps. Neigh- borhood Youths Corps. On the Job Training. etc. However. these programs have encountered a multitude of program- matic and evaluative problems and little is known concern- ing their effectiveness in reducing delinquent behavior. Probably the most serious prdblems of employment programs for delinquent and "disadvantaged" youth is their inability to obtain jobs that provide youth with market- able skills and lead to a higher social status. once they have completed the program. Unskilled youth are often given low-paying. "dead end" jobs characterized by high turnover. not necessarily because they are filled with 18 marginal workers. but because the jobs themselves are mar- ginal (Cohen. 1966). Another programmatic problem is placing delinquents on jobs because of the opinions of employers about delin- quent youths. Recent studies (Moed. 1967; Cohen. 1965) of employer's opinion have revealed general dissatisfaction with the motivation and attitudes of delinquent youth. Many employers complained that these youths are irrespon- sible. shortsighted and unstable. In addition. employers abuse arrest records on a large scale and are unwilling to hire youth who are "hard core" delinquents. These factors can be easily perceived by delinquent youth in search of employment and may partially account for the difficulty of federal employment programs to maintain youth over a long period of time. Employment problems of delinquent youth are fur- ther complicated because it is often assumed that once the youth have been trained. they are prepared to compete in the labor market. meet the qualifications for jobs. and are able to find employment. Often youth are left on their own to find themselves a job (Levitan. 1967). Ferman (1967) has also noted that if young people do obtain a 19 suitable job. they often find themselves with additional pressures and responsibilities in the form of being the only young person employed under the provisions that they must prove themselves capable. perform in an environment perceived as hostile and threatening. and without any supportive services (i.e. counseling). Purpose The present study represents an experimental test of the two issues discussed above. The purpose is to examine the effectiveness of small group dynamics and employment on the prevention and reduction of delinquent behavior. In an effort to test the effect of these vari- ables. the experimenter developed informal task-oriented groups of delinquent youths. who were employed and oper- ated a service-oriented business. The delinquent behavior of these individuals was compared to the delinquent behav- ior of youth employed in individual employment situations. 20 Experimental Hypothesis The Specific hypothesis tested in this study is stated as follows: Participants in the experimental group will show significant improvement as compared to the control group on three dependent measures: a. Occurrence of delinquent offenses b. Educational status c. Length of employment In addition to testing the above hypothesis. the secondary goal of the present study is to explore the relationship between internal group processes and the reduction of delinquent behavior as measured by these three dependent outcomes. CHAPTER II METHOD AND PROCEDURE Experimental Design The conditions and experimental design for the present study are shown below in Table 1.0. The experi- ment consists of.a posttest only-control group design (Cambell and Stanley. 1963). Table l.0.--Experimental Design of Study Conditions Treatments Experimental Subsystem Group N = 31 Employment Control Subsystem Individual N = 30 Employment 21 22 Participants The participants of the present study were sixty- one delinquent youths between the ages of 14 and 19. All participants were clients of Youth Development Corporation (Y;D.C.). which is a Model Cities delinquency prevention program in Lansing. Michigan. All participants volun- teered to take part in the present experiment. Partici- pants were referred to the experimenter from the Outreach Staff of Y.D.C. Delinquency was defined by the modular classifi- cations used by Y.D.C. which were: 1. Youth who have not displayed behavior resulting in official police or school action. but who have been identified as in danger of becoming delinquents. 2. YOuth who have been arrested or suspended but who. as yet. have not come under the official jurisdiction of the court. 3. YOuth who are under the official jurisdiction of the court. but who have not been institution- alized. 4. Ybuth currently in public or private institutions for juvenile offenses. 5. Youth once institutionalized and now re-entering the community. 23 These five classifications were collapsed into a dichoto- mous scale. either classifications l and 2 (low delin- quency) or modules 3. 4 and 5 (high delinquency). Partic— ipants were assessed for classification on the basis of a personal interview with the experimenter and case records from Y.D.C. Participants were matched on the basis of past delinquent behavior (modular classifications) and past experience in automotive mechanics. which was determined as either high or low from the personal interview with the experimenter. The matching procedures are shown below in Table 2.0. After each participant was classified according to the four matching categories. he (she) was then randomly assigned to participate in the automotive repair center (experimental subsystem) or given employ- ment as Y.D.C. work interns (control subsystem). Experimental Subsystem The experimental subsystem consisted of a service oriented business (automotive repair center) operated as a non-profit cooperative and located in the community of 24 mm H 2 NH H 2 OH H 2 ma u z o>HDOEous¢I BOA m>HuOEous<1£mHm 0>HUOEODSHuoeou5¢lnmem mocosvceamnl BOA wocmsvcflamol 30A wocmsqcflamQISmHm mocmsvcflacnlnmflm e >uommum0 m anommumu m muomoumo H wuommumu .mucmmeoeuumm wo ouspmooum mcencumzll.o.m magma 25 Lansing. The subsystem provided delinquent youth with training in automotive mechanics and skills in business management. It also was a service to the community and offered rates to its members on minor repairs. The participants selected to take part in the experimental subsystem were given stipends by Y.D.C. at two dollars per hour for ten hours per week. These sti- pends covered their participation in automotive training and operating the business. The experimental subsystem was developed in three distinct phases: 1) Training in business management skills and the purpose of cooperatives; 2) Training in automotive mechanics; and 3) The application of training in operating the business. However. all participants were not involved in all three phases of the subsystem. Only the initial twenty participants selected participated in the first two phases. while new participants entered the subsystem during the third phase. Phase 1 During the first phase. participants in the exper- imental subsystem were required to attend an extensive 26 training seminar on business management and the purpose of cooperatives. These seminars were held three times per week and lasted approximately six weeks. A consultant with forty years of experience in operating cooperatives was hired by Y.D.C. to conduct these seminars. This training included a weekend retreat to discuss the prog- ress and problems of establishing the automotive coopera- tive. Sessions were devoted to the history of coopera- tives. the current economic system in the United States. community organization and bookkeeping procedures. Participants were formed into three small groups (six to seven members) after the second meeting during this phase. Selection of group members made by the experi- menter on the basis of the two matching variables. This procedure was followed to insure an equal distribution of talent and delinquent behavior. Since Fairweather (1964) has shown that groups of patients worked best in the absence of staff. it was decided that the experimenter would serve as a participant observer. That is. the experimenter indicated to group members that they had the decision-making power on issues affecting the group and the experimenter would intervene 27 only when they could not reach decisions. when asked for advice and when their decisions would not be in the best interest of the project. During this phase each group was given a Specific task related to making the automotive cooperative opera- tional. Each group decided to elect a chairman. who would be reSponsible for assigning tasks to individuals and mak- ing progress reports to the experimenter. Some of the tasks accomplished by groups during this time included finding a business location. obtaining possible members for the Board of Directors. and searching for a qualified mechanic to train and supervise their work. Phase 2 Once participants completed Phase 1. they began to receive additional training in minor automotive repair skills. An agreement was made with Lansing Community Col- lege to run a ten week seminar in basic automotive mechanics. The seminar was geared to the type of skills necessary to operate the repair center and included 28 training in handling of tools and equipment. garage safety. basic engine theory. tune-ups. brakes. etc. Sessions were held fifteen hours per week. which included a five hour session on Saturdays devoted to work- ing on customers“ cars. The seminar was operated as a regular college course. which meant reading assignments. homework and exams. The three small groups developed during the first phase remained intact. However. since training sessions had to include all twenty participants. it was more diffi- cult to develop distinct groups than it would have been if training was done by groups. Yet. a number of procedures were developed to maintain the independence of the groups. First. all participants were informed that the group with the best attendance and work performance (mea- sured by class grades) during training would have their choice of work shifts in the automotive repair center. At this point. each group decided to elect one member to keep attendance and another to make sure members of their group had rides to class. This procedure helped to develop a sense of concern for group members. 29 Secondly. groups were given specific tasks to accomplish. such as determining what equipment would be necessary. obtaining bids on equipment and searching for means of advertisement. Also. during this time a business location was found and participants spent their spare time in the renovation of the garage. i.e. cleaning. painting. etc. These tasks helped to maintain and increase the interest for the participants in the program. Phase 3 Thirteen of the twenty participants completed the course at Lansing Community College and began operating the automotive repair center. During this time the busi- ness was established as a non-profit corporation in the State of Michigan. The first month was used as an adjustment period. At this time equipment was installed and a trained mechanic was hired to supervise the training of participants. Par- ticipants immediately began to apply the skills learned during the first two phases. Automobiles of the Y.D.C. staff were repaired in order to develop working schedules. 3O garage operating procedures and to measure the abilities of each participant. The repair center. known as Community Automotive Corporation was open for business to the general public on March 26. 1973. The project director (experimenter) and the head mechanic assumed the responsibility for the daily business operation. Participants were divided into two working shifts per day--the first from 8 a.m. to 12:30 p.m. and the second from 12:30 p.m. to 5:00 p.m. Work roles. such as assistant manager. secretary. and workers. were assigned to each participant. These roles were assigned by the project director on the basis of individual performance during the first two phases of the program. In addition to the stipends from Y.D-C. partici- pants also received financial reimbursements from the pro- fits of the business. Individual pay rates were deter- mined on the basis of individual work evaluations (Appendix A). which were completed on each job the partic- ipants worked on by the head mechanic. Participants also received high school credit for working in the garage. 31 During this phase all participants met collect- ively for the first month to decide basic operating pro- cedures. such as hours of business. the types of services to be offered. membership fees for the co-op. and how to advertise the business. It was also decided to hold weekly meetings with all the participants as well as weekly meetings with each group. The prdblem arose as to the working hours of par- ticipants still going to school and what would be done with individuals in the "best" group. who decided to work from 8 a.m. to 12:30 p.m. At this point. it was decided to rearrange the groups according to school commitments and work schedules. which reduced the participants into two groups. The participants decided to develop a system of shift rotation. once school was out for the summer. Throughout the remainder of the experiment each group met at least once a week. These meetings were used to discuss operating procedures. work performance. orien— tation of new members. and pay rates. Each group decided to elect its assistant manager. who would be responsible for the work performance of the group. 32 Control Subsystem Participants not selected by random assignment to take part in the experimental subsystem were given employ- ment as Y.D.C. work interns. As part of its regular pro- gram. Y.D.C. provides employment and skill training for young people between the ages of 14 and 20. that have some delinquent problems. The work intern program is Y.D.C.'s private employment component. because private businesses and agencies serve as placements for youth. The work intern supervisor and vocational coun- selor assist youth with any employment problems concern- ing the performance or non-performance. problems of adap- tation. tardiness. etc. This team's primary function is to insure that youth are remaining employed. In addition. they screen applicants. locate jobs. match youth to employers' requirements and provide written agreements between employers and youths. These agreements describe the type of training the youth will obtain. hours of employment and any supplemental salary or hours. With each placement. Y.D.C. attempts to seek a financial supportive match. As work interns participants earn two dollars an hour for working ten yours per week. 33 Some businesses and agencies have been able either to match the ten hours or increase the employee's hourly wage. Twenty employment positions were set aside for control participants and they received priority over other Y.D.C. clients in selecting employment. The con- trol participants' positions ranged from technical and skills training to recreational and community services. Each participant was interviewed by the work intern super- visor to determine his employment interest. educational background and training. and the types of positions Y;D.C. had available. Most of the control participants received place- ments in positions involving technical and skill training. such as radio technicians for local stations. printer apprentices in a Y.D.C. project. sales clerk at local businesses. janitorial workers. training in construction. etc. However. some participants were employed through local service agencies and neighborhood centers. such as Ybung Adult Program. Indian Center. Garvey Institute. Boys Club and YMCA (Appendix B lists these placements). 34 Each employer was responsible for training youth in the specific skills necessary to function in their business. In addition. employers were responsible for determining each intern's working hours. establishing the rules and regulations they would have to follow. and sub- mitting attendance records to the Y.D.C. office every week. Monthly meetings were held with each employer and the work intern supervisor to assess the status of work interns. Y.D.C. was responsible for providing supportive services to work interns in the form of counseling. recre- ational activities and relevant cultural experiences. WOrk interns were notified of these activities and most of the youth participated in such activities as attending sporting events. movies. lectures. etc. In addition. each work intern was assigned an outreach worker. whose basic responsibilities included individual counseling. making youth aware of existing agencies for specific problems (i.e. legal aid) and. in general. being a big brother or sister. Outreach workers were required to contact their clients at least twice a month to insure that they were not encountering any difficulties on the job. in school or 35 at home. A confidential file was kept on the progress of each client and contained follow-up information. These files on control participants were then made available to the experimenter for data collection purposes and to assess the status of control participants. Measurement of Dependent Variables Three dependent variables were measured in the present study: 1) Occurrence of juvenile offenses; 2) Educational status; and 3) Length of employment. Data on these three variables were obtained from three sources-- participants' self reports. the Lansing Police Department and the Lansing School District. In addition. background data were obtained at intake. Intake Data Prior to randomization of participants to either experimental or control subsystems. data were collected on demographic characteristics of participants. The charac- teristics of age. race. educational status. modular 36 classifications. parental income. previous work experience. and previous arrest records were recorded during a personal interview with the experimenter (see Appendix C). Self Report Data Data on the experimental and control participants status on arrests. education and employment were recorded on the self report follow—up form (Appendix C). These data were collected at ninety day intervals. The number and dates of follow-up periods for each participant was dependent upon the entry points of participants into the subsystem. Table 3.0 indicates the number and dates of follow-up periods as they relate to the entry points of “participants. Each of the three dependent measures (arrests. educational. and employment status) were recorded as dichotomous data. i.e. arrested or not. in school or out. employed or unemployed. The status of the experimental participants was collected through personal interviews and telephone conversations by the experimenter. The data on the control participants' status were collected by their respective outreach worker and given to the experimenter. 37 Table 3.0.--Entry Dates and Follow-up Periods for Partici- pants. Participants Follow-up Periods Entry Date 90 Days 180 Days 270 Days 360 Days 11/22/72 2/15 5/15_. 8/15 11/15 11/23 to 2/14 5/15 8/15 11/15 --- 2/16 to 5/14 8/15 11/15 --- --— 5/16 to 8/14 11/15 --- --- --- In addition to the data collected via individual follow—up forms a daily diary was kept by the experimenter. The daily diary contained information concerning any crit- ical events occurring in the experimental subsystem. Police Data Data were collected on the number of arrests for all experimental and control participants. These statis- tics were obtained directly from the files of the Lansing Police Department. The Y.D.C. data specialist. who had access to police files. was responsible for collecting these statistics. 38 The arrests of participants were recorded accord- ing to the follow-up periods. All information was kept confidential by removing individuals names once the sta- tistics were collected. School Data --. Arrangements were made with the administration office of the Lansing School District to obtain statistics on the absences. suspensions and grades for all partici- pants attending school. The names and schools of partici— pants were given to the secretary. who was responsible for collecting the data. The experimenter could not Obtain permission to personally inspect the files. Statistics on the absences. suspensions and grades were to correspond to the follow-up periods of this study; however. due to the schools' recording systems it was not possible to get the data to correspond to the follow-up periods of the present study. Therefore. the only data used were statistics from the beginning of school (9/18/73) to 11/15/73. 39 Measurement of Internal Group Processes In addition to the data collected on the three dependent measures. data were also collected on three internal processes in the experimental subsystem: 1) Sociometric choice ratings; 2) Morale ratings; and 3) Job behavior ratings. The instruments used for data collections of these ratings can be found in Appendices E to G. Sociometric Ratings A sociometric rating scale (Fairweather. 1964) was used as an index to determine how experimental par- ticipants felt about their group members. Each group mem- ber was asked to rate individuals in their group on the sociometric rating scale. Since participants dropped out or had to change groups. ratings were completed on the current group members at the time the ratings were com- pleted. The names of group members were entered on the rating form by the experimenter. The scale was adminis— tered at three follow-up periods--2/15. 5/15 and 8/15. 4O Morale Ratings The morale scale. developed by Fairweather (1964). was used as an index of participants disposition to act toward the goals of the group. Each participant was asked to complete this scale at two follow-up periods--5/15 and 8/15. Ratings were completed by participants in the experimental subsystem during these time periods. Job Behavior Ratings The job behavior scale. developed by Fairweather (1964). was used as a means of accessing how participants were performing on the job. The head mechanic. who super- vised and trained the participants. was responsible for completing this scale. The scale was completed on all participants employed at the garage at the time of its administration. The scale was administered at two follow- up periods--5/15 and 8/15. Scoring of Process Measures The responses of the three process measures were scored as follows: Each response to an item was given a numerical value corresponding to the scale value for the 41 response category. For items in the sociometric scale each reSponse was given a numerical value from 1 (do not like at all) to 7 (like very much). The means of each participant's rating were then calculated. For items on the morale scale each response was given a numerical value from 1 (most favorable response) to 4 (least favorable response). The means were then cal- culated for each participant. The job behavior scale consisted of 26 items. Each item had either a positive or negative response. Each negative response counted as 1. These values were totaled and the mean score for each participant was calculated. CHAPTER III RESULTS Intake Results All subjects were randomly assigned to conditions and matched on two variables--past delinquency behavior and automotive mechanic experience; however. it was decided to investigate possible differences between the experi- mental and control groups on the matched variable (past delinquency) and other relevant variables-~age. educational status and race--which might have influenced the dependent measures. The E_test (Hays. 1963) was used to determine any significant differences between conditions on age. while the chi square test (Hays. 1963) was used to deter- mine any significant differences between conditions on the other three variables. The results are displayed in Tables 1.1. 1.2. 1.3. and 1.4. The comparisons between conditions. as presented in these tables. indicates that there were no significant 42 43 Table l.1.--Comparison of Participants on Age upon Entrance into Experiment. Experimental Control Mean Age Group Group Mean N Mean N 17 6.2 6.0 t = 0.20 (4 df) (one-tailed) Table l.2.--Comparison of Participants on Educational Status upon Entrance in the Experiment. . Experimental Control Educational Status Group Group N % N % In School 19 (61) 22 (73) Out of School 12 (39) 8 (27) 2 X = 1.00 (1 df) (one-tailed) 44 Table l.3.--Comparison of Participants on Y;D.C. Module Classification upon Entrance in the Experi- ment. Experimental Control Y;D.C. Module Group Group N % N % One 9 (29) 13 (43) Two 10 (32) 6 (20) Three & Four 12 (39) ll (37) 2 X = 1.76 (2 df) (one-tailed) Table 1.4.--Comparison of Participants on Race upon Entrance in the Experiment. Experiment Control Race Group Group N % N % Black 23 (74) 22 (73) Non-Black 8 (26) 8 (27) 2 X = 0.005 (1 df) (one-tailed) 45 differences on the basis of age. educational status. Y.D.C. modular classification and race. Therefore. the process of randomization was successful in equating the two conditions on these four relevant variables. Foplow-Up Results: Self-Reports Self reported follow-up data on arrests and edu- cational and employment status was collected on subjects in both the experimental and control conditions every ninety days. There were no participant attrition for self-report data. The chi square test (Hays. 1963) was computed to test for any significant differences between conditions on the dependent variables. Tables 2.1. 2.2. and 2.3 reveal that there were no statistically significant differences after ninety days of follow-up on self reports of arrests. educational sta- tus and employment status. Tables 3.1. 3.2. and 3.3 indicate the self reported status of participants on arrests. education and employment after one-hundred and eighty days of follow-up. These comparisons between conditions on these variables 46 Table 2.1.--Comparison of Participants on Self Reports on Cumulative Arrests for 90 Days of Follow-Up. Experimental Control Arrest Status Group Group N % N' % Arrested 5 (l6) 3 (10) No Arrest 26 (84) 27 (90) = 0.05 (1 df) (one-tailed) Table 2.2.--Comparison of Participants on Self Reports on Educational Status for 90 Days of Follow-Up. Experimental Control Educational Status Group Group N’ % N’ % In School 22 (76) 21 (70) Out of School 7 (24) 9 (30) 2 X = 0.08 (1 df) (one-tailed) 47 Table 2.3.--Comparison of Participants on Self Reports on Employment Status for 90 Days of Follow-Up. Experimental Control Employment Status Group Group N %. N % Employed 21 (68) 26 (87) Unemployed 10 (32) 4 (13) 2 X = 3.08 (1 df) (One-tailed) Table 3.1.--Comparison of Participants on Self Reports on Cumulative Arrests for 180 Days of Follow-Up. Experimental Control Arrest Status Group Group N % N % Arrested 6 (22) 3 (11) No Arrest 21 (78) 24 (84) 2 X = 1.20 (1 df) (one-tailed) 48 Table 3.2.--Comparison of Participants on Self Reports on Educational Status for 180 Days of Follow-Up. Experimental Control Educational Status Group Group N % N % In School 12 (57) 19 (70) Out of School 9 (43) 8 (30) 2 X = 0.89 (1 df) (one-tailed) Table 3.3.--Comparison of Participants on Self Reports on Employment Status for 180 Days of Follow-Up. Experimental Control Employment Status Group Group N % N % Employed 15 (60) 15 (56) Unemployed 10 (40) 12 (44) 2 X = 0.20 (1 df) (one-tailed) 49 show no significant differences. as evidenced in Tables 3.1. 3.2 and 3.3. After two—hundred and seventy days of follow-up there were no significant differences of self report com- parisons between conditions on arrests. education and employment. as shown in Tables 4.1. 4.2 and 4.3. No significant differences were found when com- paring conditions on the basis of arrests. educational and employment status after three-hundred and sixty days of follow-up. These results are presented in Tables 5.1. 5.2 and 5.3. Table 5.3 shows a non-significant trend (p < .10) in employment status and reveals that more experimental participants tended to remain employed after one year of follow-up. as compared to control participants. Due to the daily contact of the experimenter with experimental participants and lack of contact with control participants. these self reported data are probably not completely informative. The experimenter depended on the outreach staff of Y;D.C. to collect follow-up data on con- trol participants. which might have affected these data. Another variable affecting the self reported arrests is the fact that many youth consider being stopped or 50 Table 4.1.--Comparison of Participants on Self Reports on Cumulative Arrests for 270 Days of Follow-Up. Experimental Control Arrest Status Group Group N % N % Arrested 6 (25) 4 (17) No Arrests 18 (75) 19 (83) X2 = 0.15 (1 df) (one-tailed) Table 4.2.--Comparison of Participants on Self Reports on Educational Status for 270 Days of Follow-Up. Experimental Control Educational Status Group Group N % N % In School 9 (50) 15 (79) Out of School 9 (50) 4 (21) 2 X = 3.397 (1 df) (one-tailed) 51 Table 4.3.--Comparison of Participants on Self Reports on Employment Status for 270 Days of Follow-up. Experimental Control Employment Status Group Group N % N % Employed 12 (55) 10 (45) Unemployed 10 (45) 12 (55) X2 = 0.36 (1 df) (one-tailed) Table 5.1.--Comparison of Participants on Self Reports on Cumulative Arrest for 360 Days of Follow-Up. Experimental Control Arrest Status Group Group N ‘% N % Arrested 7 (35) 4 (20) No Arrest 13 (65) 16 (80) 2 X = 1.32 (1 df) (one-tailed) 52 Table 5.2.--Comparison of Participants on Self Reports on Educational Status for 360 Days of Follow-Up. Experimental Control Educational Status Group Group N % N % In School 9 (56) 13 (76) Out of School 7 (44) 4 (24) x2 = 1.51 (1 df) (one-tailed) Table 5.3.--Comparison of Participants on Self Reports on Employment Status for 360 Days of Follow-Up. Experimental Control Employment Status Group Group N ‘% N % Employed 14 (74) 9 (45) Unemployed 5 (26) 11 (55) 2 X = 3.309 (1 df) p < .10 (one-tailed) 53 questioned by police as an arrest. Also. many clients of Y.D.C. have expressed that they wouldn't get on a Y.D.C. program unless they displayed delinquent behavior. There— fore. participants might have exaggerated their arrest status. In summary. the comparisons between conditions on the self reported follow-up data did not indicate any statistically significant differences to support the experimental hypothesis of this study which is that experimental participants would be significantly differ— ent on three dependent measures as compared to control participants. Follow-Up Results: Police Data Since the self reported follow-up data on arrests may be less than accurate. it was decided to investigate arrest records of participants from a more reliable source. Arrests records were obtained from the files of the Lansing Police Department and tabulated for all par- ticipants according to the four specified follow-up periods. There were no participant attrition for police 54 data. The chi-square test (Hays. 1963) was used to cal- culate any significant differences between conditions on arrests. Table 6.1 indicates that there was a significant difference (p < .02) between conditions on arrests after ninety days of follow-up. This difference is in the pre- dicted direction. According to the police data. there were no arrests of experimental participants. while five or seventeen percent of the control participants showed being arrested. These results substantiate the inaccuracy of self-reported arrest data and support the experimental hypothesis of the present study. Table 6.1.--Comparison of Participants on Police Data on Cumulative Arrests for 90 Days of Follow-Up. Experimental Control Arrest Status Group Group N % N % Arrested 0 (0) 5 (17) No Arrest 31 (100) 25 (83) X2 = 5.62** (1 df) **p ‘<.02 (one-tailed) 55 After one-hundred and eighty days of follow-up. a significant difference (p < .05) was found between con- ditions on arrest status. as evidenced in Table 6.2. Table 6.2 further reveals that only one participant (4%) in the experimental subsystem was arrested during this period as compared to six or twenty two percent of the control participants. Table 6.2.-—Comparison of Participants on Police Data on Cumulative Arrests for 180 Days of Follow-Up. Experimental Control Arrest Status Group Group N % N % Arrested l (4) 6 (22) No Arrest 26 (96) 21 (78) 2 X = 4.08* (1 df) *p ‘<.05 (one-tailed) Table 6.3 shows the comparisons between conditions on arrests after two-hundred and seventy days of follow-up. As evidenced in this table there were no significant dif— ferences between conditions. 56 Table 6.3.--Comparison of Participants on Police Data on Cumulative Arrests for 270 Days of Follow-Up. Experimental Control Arrest Status Group Group N % N % Arrested 3 (12) 7 (30) No Arrest 21 (88) 16 (70) 2 X = 2.56 (1 df) (one-tailed) Table 6.4 indicates that there were no signifi- cant differences between conditions on arrests after three-hundred and sixty days of follow-up. The results of the analysis of police reported arrests provides significant evidence that the experiment had an effect on reducing the arrest rates. This evidence was used as an indicator of occurrence of delinquent offenses. for participants in the experimental subsystem. Tables 6.1 and 6.2 reveal that there were more arrest of control participants and the differences were significant during the ninety and one-hundred and eighty day follow-up periods. These results. which support the experimental 57 Table 6.4.--Comparison of Participants on Police Data on Cumulative Arrests for 360 Days of Follow-Up. Experimental Control Arrest Status Group Group N % N % Arrested 6 (30) 7 (35) No Arrest 14 (70) 13 (65) X2 = 0.04 (1 df) (one-tailed) hypothesis of the present study. also point to the Speci- ficity of treatment outcomes. When juxtaposed against the self-report data described above. there is a clear lack of correspondence. This is congruent with previous work (Fairweather et a1. 1969) that has found a lack of agreement between verbal and behavioral outcomes. Follow-Up Results: School Data Data were obtained from the Lansing School Dis- trict on absences. suspensions and grades for all 58 participants enrolled in school from 9/18/73 to 11/15/73. The statistics collected only reflect those participants whose records were available during this time period. The Wilcoxon Sum Rank Test (Wilcoxon and Wilcox. 1964) was used to obtain any significant differences between conditions on absences. suspensions and grades. The comparison between conditions on the number of absences of experimental and control participants is contained in Table 7.1. No significant differences were found from these comparisons. The table further indicates that the mean number of absences for experimental partici- pants was 13.9 as compared to the mean of 18.0 for control participants. Absences were recorded for half days. Table 7.2 reveals that there were no significant differences between conditions on the basis of school sus- pensions. The table also indicates that the mean suspen- sion rate for both conditions is less than one and only three participants were suspended during this time period. On the basis of mean grades no significant differ- ences were found between conditions. as evidenced in Table 7.3. Table 7.3 also indicates that the mean grades for both conditions was less than 2.00 and the mean grades 59 Table 7.1.--Wilcoxon Sum Rank Test on School Absences (1/2 days) from School Data. Experimental Control __group Group $3222.? Rank $222311? Rank 17 13 16 12 26 16.5 32 18 10 10 23 14 25 15 2 3 26 16.5 3 4.5 0 1.5 6 8 4 6 12 11 0 1.5 4.5 -- -- 9 9 -- -- i = 13.4 i = 18.0 60 Table 7.2.--Wi1coxon Sum Rank Test on School Suspensions from School Data. Experimental Control ___§roup Group 55222332; Rank 53:22:12; Rank 0 8.5 O 8.5 0 8.5 1 17.5 0 8.5 0 8.5 0 8.5 0 8.5 0 8.5 0 8.5 0 8.5 O 8.5 2 19.0 0 8.5 0 8.5 O 8.5 1 17.5 0 8.5 0 8.5 O 8.5 i = .30 SE = .11 T = 85.5 61 Table 7.3.--Wi1coxon Sum Rank Test on Mean Grades from School Data. Experimental Control Group Group Mean Mean Grades Rank Grades Rank 1.60 9.0 ' .80 4.5 0.75 2.5 .20 1.0 1.50 8.0 .75 2.5 0.80 4.5 2.60 13.0 1.30 6.5 2.20 11.0 2.50 12.0 1.30 6.5 3.00 14.5 3.10 16.0 3.00 14.5 2.00 10.0 i = 1.81 x = 1.62 T = 64.5 62 for experimental participants was .20 points higher than control participants. The comparisons between conditions on the basis of mean absences. suspensions and grades yielded no sig- nificant evidence to support the experimental hypothesis of this study. The difficulties in obtaining statistics on all participants school records did hinder the accurate analysis of these data. Internal Analysis on Basis of Police Data Since significant differences were found on com- parisons between conditions on the basis of police re- ported arrests. it was decided to investigate the specific relationship between experimental and control subsystems and the individuals arrested. Therefore. an internal analysis of police reported arrests was completed to better determine what aspects of the program produced the experimental effect. Furthermore. since more experimental participants were arrested after 270 and 360 days of follow-up than during 90 and 180 days of followeup (Tables 6.1. 6.2. 6.3 63 and 6.4). it was felt that at least two possible inter- vening variables might have effected these results; when participants entered the subsystem and how long partici- pants remained in the subsystem. Therefore. the internal analysis was completed and based upon two parameters: a) The entry point of the participant in the experiment. and b) The amount of time the participants remained in the experiment. The chi-square test (Hays. 1963) was used to compute any significant differences between con- ditions on these two parameters. a. For the first internal analysis (based upon the entry point of the participants) a median split was performed on the participants. Participants were grouped on the basis of whether their entry points occurred dur- ing the first or second six months of the experiment. Since all the participants included in the 270 and 360 follow-up periods entered the experiment during the first six months; comparisons between conditions were only made for 90 and 180 days of follow-up. Table 8.1 indicates a significant difference (p .<.05) on a comparison of arrests between conditions after 90 days of follow-up for participants whose entry 64 Table 8.1.--Comparison of Participants on Police Data on Arrests. Who Entered the Experiment During the First Six Months. for 90 Days of Follow-Up. Experimental Control Arrest Status Group Group N’ % N % Arrested 0 (0) 4 (15) No Arrest 24 (100) 23 (85) X2 = 3.86* (1 df) *p < .05 (one-tailed) points occurred during the figgp six months. These re- sults are in the predicted direction. After 90 days of follow—up. Table 8.2 indicates no significant difference on arrests between conditions for participants Whose entry points occurred during the second six months. The table (8.2) further shows that only fourteen participants had entry points during the second six months. Since the expected cell frequency was less than five. it was necessary to use Yates correction (Hays. 1963) in the calculation of the chi-square test. 65 Table 8.2.--Comparison of Participants on Police Data on Arrest. Who Entered the Experiment During the Second Six Months. for 90 Days of Follow-Up. Experimental Control Arrest Status Group Group N % N % Arrested O (O) 1 (14) No Arrest 7 (100) 6 (86) 2 X = 0.00 (1 df) (one-tailed) Table 8.3 shows a significant difference (p'< .05) on arrests between conditions for participants whose entry points occurred during the first six months. These results are after 180 days of follow-up and are in the predicted direction. Table 8.3 also reveals that only one experi- mental participant was arrested as compared to six control .participants. As evidenced in Table 8.4. there were no signifi- cant differences between conditions on arrests for partici- pants whose entry points occurred during the second six months. Due to the small sample size (three experimental 66 Table 8.3.--Comparison of Participants on Police Data on Arrests. Who Entered the Experiment During the First Six Months. for 180 Days of Follow-Up. Experimental Control Arrest Status Group Group N % N % Arrested 1 (4) 6 (26) No Arrest 23 (96) 17 (74) 2 X = 4.45* (1 df) *p < .05 (one-tailed) Table 8.4.--Comparison of Participants on Police Data on Arrests. Who Entered the Experiment During the Second Six Months. for 180 Days of Follow-Up. Experimental Control Arrest Status Group Group N %’ N % Arrested 0 (O) 0 (0) No Arrest 3 (100) 4 (100) 2 X = 0.00 (1 df) (one-tailed) 67 and four control participants) Yates correction was used in the calculation of the chi—square. In summary. the results of the first internal analysis. based upon police data. indicated that experi- mental participants were less likely to display delinquent behavior (in terms of getting arrested) than control par- ticipants. when their entry points occurred during the first six months of the experiment. In other words. par- ticipation in the experimental subsystem in its initial stage (first six months) had a significant effect in . decreasing the occurrence of delinquent behavior (measured in terms of police arrests). b. The second internal analysis of police arrests. based upon the amount of time participants remained in the experiment. was completed for each follow—up period. Each participant was categorized in terms of either remaining in the experiment for ppgg than three months or lggg than three months. No significant differences were found between con- ditions for participants. who remained in the experiment for £252 than three months. on the basis of arrests after 90. 180. 270 and 360 days of follow-up. These results 68 are evidenced in Table 9.1. 9.2. 9.3 and 9.4. Therefore. no evidence was found to show that remaining in the experiment for more than three months had any significant effect upon decreasing the arrest rates of experimental and control participants. A significant difference (p < .01) was found by comparing participants in both conditions who remained in the experiment l§§§_than three months after 90 days of follow-up. These results are evidenced in Table 10.1 and are in the predicted direction. After 180 days of follow-up on arrests. the dif- ference between conditions was significant (p < .05) for participants who remained in the experiment lgpg than three months. Table 10.2 shows these results. which are in the predicted direction. Table 10.3 shows no significant differences be- tween conditions on arrests after 270 days of follow-up for participants who remained in the experiment ngg than three months. Table 10.4 shows no significant difference between conditions in arrests after 360 days of follow-up. These results are for participants who remained in the experi- ment less than three months. 69 Table 9.1.--Comparison of Participants on Police Data on Arrests. Who Were in Experiment More than Three Months for 90 Days of Follow-Up. Experimental Control Arrest Status Group Group N % N % Arrested O (0) 0 (0) No Arrest 12 (100) 14 (100) X2 = 0.00 (1 df) (one-tailed) Table 9.2.--Comparison of Participants on Police Data on Arrest. Who Were in Experiment More Than Three Months for 180 Days of Follow-Up. Experimental Control Arrest Status Group Group N %1 N % .Arrested 1 (9) 2 (14) No Arrest 10 (90) 12 (86) X2 = 0.12 (1 df) (one-tailed) 70 Table 9.3.--Comparison of Participants on Police Data on Arrest. Who Were in Experiment More Than Three Months. for 270 Days of Follow-Up. Experimental Control Arrest Status Group Group N % N % Arrested l (10) 2 (17) No Arrest 9 (90) 10 (83) 2 X = 0.20 (1 df) (one-tailed) Table 9.4.--Comparison of Participants on Police Data on Arrest. Who Were in Experiment More Than Three Months. on 360 Days of Follow-Up. Experimental Control Arrest Status Group Group N % N % Arrested l (9) 2 (17) No Arrest 10 (91) 10 (83) X2 = 0.29 (1 df) (one-tailed) 71 Table 10.1.--Comparison of Participants in Experiment Less Than Three Months on Police Data Arrests for 90 Days of Follow-Up. Experimental Control Arrest Status Group Group N ‘ % N % Arrested O (O) 5 (29) No Arrest 19 (100) ll (71) 2 X = 6.91*** (1 df) ***p < .01 (one-tailed) Table 10.2.--Comparison of Participants in Experiment Less Than Three Months on Police Data Arrests for 180 Days of Follow-Up. Experimental Control Arrest Status Group Group N' % N % Arrested 0 (O) 4 (29) No Arrest 16 (100) 10 (71) 2 X = 5.25* (1 df) *p < .05 (one-tailed) 72 Table lO.3.--Comparison of Participants in Experiment Less Than Three Months on Police Data Arrests for 270 Days of Follow-Up. Experimental Control Arrest Status Group Group N % N % Arrested 2 (15) 5 (50) No Arrest 11 (85) 5 (50) X2 = 3.19 (1 df) (one-tailed) Table 10.4.--Comparison of Participants in Experiment Less Than Three Months on Police Data Arrests for 360 Days of Follow-Up. Experimental Control Arrest Status Group Group N' % N’ % Arrested 5 (45) 5 (56) No Arrest 6 (55) 4 (44) X2 = 0.20 (1 df) (one-tailed) 73 The second internal analysis. based upon the amount of time participants remained in the experiment. revealed no significant differences between conditions for participants who remained in the experiment mpg; than three months. The analysis did show that most of the experimental and control participants who were arrested were in the experiment lpgg than three months. Since most of the participants who were arrested left the experiment. and since an experimental effect was found in the internal analysis of police arrests. it was decided to further explore these differences. A compari- son was completed between conditions on the basis of the reasons participants gave for leaving the experiment. Participants were categorized according to their reasons for leaving the experiment as either leaving for full-time employment or-other reasons. such as moving or termina- tion. The chi-square test (Hays. 1963) was used to com- pute any statistical difference between conditions. Table 11.1 indicates that there was a significant difference (p < .05) when comparing conditions on the basis of reason for leaving the experiment. The table (11.1) shows that a greater number of experimental par- ticipants (68%) left the experiment because they obtained 74 Table 11.1.--Comparison of Participants on Self Report Explanation for Leaving the Experiment. Experimental Control Reason for Leaving Group _§pppp_ N % N % Full-time Employment 13 (68) 5 (33) Other Reasons 6 (32) 11 (67) X2 = 4.74* (1 df) *p < .05 (one-tailed) full-time employment. as compared to the control partici- pants (33%) who left for the same reasons. These results suggest that the experimental effect on arrests of partici- pants was partly due to the fact that experimental partici— pants remained employed after leaving the experiment. In summary. the experimental treatment seems to have its effect on those participants Who were part of the initial sample and remained in the experiment more than three months. This also seems to be related to obtaining future employment after leaving the experim mental subsystem. 75 Group Process Results Sociometric Rating In an effort.to determine the effect of socio- metric ratings of participants on the experimental effect of the present study (occurrence of juvenile offenses). it was decided to investigate the relationship between sociometric ratings of experimental participants arrested and those not arrested. The mean sociometric ratings of each participant not arrested were calculated. These ratings were then matched against each participants rat- ing of their group members arrested. The E test for matched pairs (Walker and Lev. 1953) was used to analyze these comparisons. The purpose of this analysis was to determine the amount of group acceptance or rejection of arrested par- ticipants by their peers. Therefore. each participant acted as their own control and matched pairs were made on the basis of the groups in the present study. Comparisons were first analyzed according to sociometric ratings. which occurred before the group participants were arrested. Two separate analyses were calculated for participants 76 arrested on the basis of police data and self reported data. Table 12.1 indicates a significant difference (p < .10) between matched pairs on mean sociometric rat- ings done before some participants self-reported arrests. The difference shown in Table 12.1 shows a trend toward group rejection of arrested participants. before they were arrested. Therefore. participants who got arrested were not getting along with their group peers prior to their arrests. Table 12.2 indicates a significant difference (p < .025) between matched pairs on sociometric ratings done pfpgg some participants self reported arrests. The results in Table 12.2 reveal that arrested participants were rejected by their peers §§£§£_being arrested. There— fore. not only were arrested participants rejected by their peers prior to being arrested. but they were also rejected by their peers after being arrested. Table 12.3 indicates a significant difference (p-< .05) between matched pairs on sociometric ratings done before police data indices. These findings are con- sistent with the results shown in Table 12.1 and provide 77 Table 12.1.--Matched Pairs on 5/15 Sociometric Ratings ’ Done Before Some Participants Self Reported Arrests. Mean Ratings of All Participants Except Those Arrested Mean Ratings of Those Arrested . 3.82 4.50 6.00 6 00 5.50 3.50 6.50 4.00 4.00 3 50 7.00 7.00 5.80 4.50 32 = 5.52 3': = 4.71 _t_ = 1.825* (6 df) *p < .10 (one-tailed) 78 Table 12.2.-~Matched Pairs on 2/15 Sociometric Ratings Done After Some Participants Self Reported Arrests. Mean Ratings of All Participants Except Those Arrested Mean Ratings of Those Arrested 5.00 4.00 5.53 4.50 5.83 5.50 5.43 3.00 4.33 3.50 4.00 4.00 3': = 5.02 i = 4.08 E = 2.755** (5 df) **p < .025 (one-tailed) 79 Table 12.3.--Matched Pairs on 2/15 Sociometric Ratings Done Before Police Reported Arrests. Mean Ratings of A11 Participants Except Those Arrested Mean Ratings of Those Arrested - 4.50 5.00 7.00 7.00 5.83 6.00 3.66 3.00 6.00 4.00 6.00 5.00 5.00 2.00 x = 5.43 f = 4.52 _t_ = 2.21* (6 df) *p < .05 (one-tailed) 80 sufficient evidence to indicate that peer rejection might be a useful indicator of recidivism of delinquent behavior. Table 12.4 shows no significant difference between matched pairs on sociometric ratings ggppg police data indices. These results are inconsistent with the previous findings shown in Table 12.2; however. the data was based on the police reported arrest of one experimental partici- pant. In summary. the results of the analysis of socio- metric ratings provide significant evidence that group processes are useful in evaluating recidivism of delin- quent offenses. The fact that significant differences were obtained between ratings of participants arrested and those not arrested. both prior and after arrest occurred. suggests that sociometric ratings are beneficial in predicting juvenile delinquent behavior. Morale Ratings In an effort to determine the effect of morale on the significant dependent variable (occurrence of juvenile offenses). it was decided to investigate the relationship between ratings on morale between participants arrested Table lZ.4.--Matdhed Pairs on 5/15 Sociometric Ratings 81 Done After Police Reported Arrests. Mean Ratings of A11 Participants Except Those Arrested Mean Ratings of Those Arrested 5.00 6.00 6.50 7.00 5.80 6.00 6.00 i = 6.04 3; = 1.31 (6 df) 6.00 4.00 5.00 5.00 6.00 5.42 (one-tailed) 82 and those not arrested. Since morale ratings were com- plated at two follow-up periods (5/15 and 8/15) and since most experimental participants were also arrested during these periods. it was decided to combine the morale rat- ings of participants. The mean morale ratings of partici- pants. excluding those participants arrested. were com- puted and compared to the mean morale ratings of partici— pants arrested. The 5 test (Hays. 1963) was used to analyze these comparisons. Table 13.1 shows no significant difference when comparing the mean morale ratings of experimental partic- ipants not arrested with the mean ratings of participants arrested. Therefore. morale ratings were not useful in determining any differences on recidivism of juvenile offenses. Job Behavior Ratings In an effort to determine the relationship be- tween behavior on the jOb and the significant dependent variable (occurrence of juvenile offenses). it was decided to analyze the mean ratings. on the job behavior scale. of experimental participants not arrested as compared to the 83 Table l3.1.--Comparison of Participants Mean Ratings on Morale. Ratings of All Participants Ratings of Arrested for 5/15 and 8/15 Follow-Up Participants on 5/15 Except Those Arrested and 8/15 Follow-Up x = 2.73 i = 2.92 1:. = 0.663 (16.4 df) (one-tailed) mean ratings of those participants arrested. Since job I behavior ratings were completed at two follow-up periods (5/15 and 8/15) and since most experimental participants .were arrested during these periods. it was decided to combine these ratings. The mean job behavior ratings of participants not arrested were calculated excluding the ratings of participants arrested and compared to the rat- ings of participants arrested. The £_test (Hays. 1963) was used to compute this analysis. Table 13.1 reveals no significant difference when comparing the mean job behavior ratings of participants not arrested with the mean ratings of participants arrested. 84 Table l4.l.--Comparison of Participants Mean Ratings on the Job Behavior Scale. Ratings of All Participants Ratings of Arrested for 5/15 and 8/15 Follow-Up Participants for 5/15 Except Those Arrested and 8/15 Follow-Up i=0.77 X=0.76 5 = 0.32 (16.4 df) (one-tailed) Therefore. behavior on the job had no significant effect upon the participants getting arrested. CHAPTER IV DISCUSSION In the introduction it was pointed out that juvenile delinquency is a multivariate problem and there is a deficit of systematic evaluative research in the natural environment of delinquents. In this study. the development of a peer-operated. service-oriented business operated by delinquent youths. with emphasis on group dynamics and self-management concepts. provided an oppor- tunity to explore the effect of a community treatment pro- -ngam on the reduction of delinquent behavior. The results -of this experiment revealed evidence to support the hypoth- esis that. on the basis of police data. participation in athe experimental subsystem significantly reduced the occur- rence of delinquent offenses as compared to control partic- ipants. However. conflicting results were found on the - basis of self-reported data. No evidence was found to indicate any differences between conditions on occurrence of offenses. educational status. and length of employment. 85 86 There are several possible reasons for the failure of the self-reported data collection procedure to produce the desired experimental effect. First. the lack of per- sonal contact with control participants as compared to -the daily contact with experimental participants and the experimenter yielded differences in the degree of reli- ability of self-reported information. In other words. the experimenter was relying on the honesty and integrity of control participants' reports to their respective Y.D.C. outreach worker as well as telephonic conversations with the experimenter. whom they hardly knew. The experi- mental participants. on the other hand. generally reported directly to the experimenter and often consulted with the experimenter concerning their school status. contact with police. and future employment possibilities. Secondly. in order to get preferential or top priority program place- .ment as a Y}D.C. client. youth must display severe delin- .quent behavior. i.e. the higher the Y.D.C. modular class- ification. the higher the prdbability of program place- ~ment. Therefore. in order to remain in a program. many clients would often exaggerate their delinquent behavior in terms of contact with police and problems in school. 87 A recent Y.D.C. evaluation (Anderson and Whitney. 1973) pointed out these findings as well as the fact that the outreach staff was negligent in contacting clients and verifying self-reported information Obtained from their clients. Another important consideration as to the lack of significant differences between conditions on self- reported data was the fact that both experimental and control participants did receive employment. and this fac- tor might have contributed to the improvement of both con- ditions on the dependent measures. The inclusion of the employment variable in both conditions may partly explain the lack of significant differences between conditions. In order to determine the effects of being employed as opposed to the type of employment. the present experiment -might have improved if another control group was incor- .porated into the experimental design. which did not receive employment. Because of the aforementioned deficiencies of the self-reported data collection procedure. verification of the dependent measures was obtained from sources indepen- dent of the participants and the Y.D.C. organization--the 88 police department and the school system. The results indicate that on the basis of data obtained from the schools on participants' absences. suspensions. and grades. no differences occurred between conditions. as evidenced in Tables 7.1. 7.2. and 7.3. Several factors did contrib- ute to the lack of success in obtaining the desired exper- imental effect on school behavior. First. there was a prdblem in obtaining data from the school administration to correspond to the specific time frame of the present study. Secondly. many participants who were no longer clients of Y.D.C. or who left the experiment transferred from one school to another periodically. Because of this. it was almost impossible for the school system to collect «data on these individuals without contacting each individ- ual school to inquire about the participant's enrollment and having them forward the necessary statistics. Since the school administration was not willing to do the neces- sary work and permission could not be obtained for the experimenter to have access to the files of participants. -many of the statistics on participant's school behavior could not be obtained. 89 The results of the data from the police department provided the experimenter with significant evidence as to the differences between conditions on occurrence of delin- -quent offenses. As displayed in Table 6.1 (p < .02) and 6.2 (p < .05). the results indicate that on the basis of police arrests there were significantly less experimental participants arrested as compared to control participants after 90 and 180 days of follow-up. However. the police data also indicated an increase in the arrest of experi- mental participants during the 270 and 360 follow-up ~periods (see Tables 6.3 and 6.4). thus indicating the -presence of some intervening variables which may have re- duced the experimental effect during these periods. On the basis of the internal analysis of police- reported arrests a more accurate picture of the variables .affecting the experimental findings was Obtained. First. less arrests occurred for experimental participants as compared to control participants. if they entered the experiment during its initial stages (first six months). as evidenced in Tables 8.1 and 8.3. However. as described in Tables 8.2 and 8.4. there were no significant differ- ences between conditions for those participants who 90 entered the sample during the second six months. Sec- -ondly. significantly less experimental participants as compared to control participants were arrested. if they remained in the experiment l§§§_than three months. as evi- denced in Tables 10.1 and 10.2. However. as evidenced by Tables 9.1 and 9.2 these differences disappeared for those subjects who were in the experiment for more than three months. And finally. these findings are related to the fact that significantly more experimental participants. who left the experiment. obtained full-time employment. as evidenced in Table 11.1. As a summary statement. the experimental treatment .seems to be having its major effect relative to the con- trol treatment during the initial part of the experiment and for those subjects who remained less than three months. Examination of the appropriate Tables indicates that this .experimentalsuperiorityis largely due to the control gppsystem's performance decrement in inducing socially, acceptable behavior in its participants. This may be due in no small degree to the experimental subsystem's ability to marshall peer group influence particularly for those participants in the initial sample who were involved in 91 an intensive group dynamics experience and in structured vocational training. During the first six months experi- mental group participants received formal training. while in contrast. during the second six months training was primarily on-the-job (OJT). Formal training included structured sessions on the purpose of cooperatives and a -seminar in automotive mechanics from Lansing Community College. On the other hand. OJT included repairing cars in the garage and learning by doing under some super- vision. The formal training sessions were conducted by a specialist in the fields of co-operative enterprises and automotive mechanics. These sessions were held indepen- dent of the regular Y.D.C. program. which allowed the instructors to initiate rules and regulations different from the normal Y.D.C. procedures. Participants were expected to attend class regularly. complete daily assign- ments and direct their energies toward the establishment of their business. These activities were in sharp contrast to the regular Y.D.C. work intern activities. which usually con- sisted of youth merely putting in time at an agency or 92 business and getting a check every Friday. Therefore. many participants developed the feeling that they were a select group of youth. who either displayed special skills or potential talents necessary to initiate the auto co-op project. The formal training seemed to give participants a special status in the Y.D.C. program and also seemed to develop feelings of achievement and pride. Training dur- ing the initial phases of the project also seemed to pro- vide participants with some security and assurance of having a jOb in the future. During the latter phases of the experiment train- ing was primarily on-the-job. Participants were expected to have gained knowledge and skills from the formal train- ing. which could be immediately transferred to the OJT situation. Although those subjects who entered the sample after the formal training period did not benefit from this experience. there was a general concern among partici- .pants that they were not adequately prepared to assume the responsibilities of Operating a "real" business. Also. the trainer hired was not formal. but more "street- oriented" and expected participants to have more skills than they acquired. At this point participants were not 93 assured of their success and tended to become skeptical of the business succeeding. OJT actually meant learning by doing. often under stressful conditions from the trainer. and working on "real" customers' cars. Individ- ual roles and expectations were less structured. confi- dence in the trainer was low. and individuals were often performing under stress during the OJT as compared to the formal training. Conceivably. the OJT period of the experiment did not instill as much confidence and achieve- ment in participants as did the formal training. These differences in the type of training could help account for the experimental effect found in the present study. Another difference between the initial and latter phases of the experiment. which may partly explain the major findings of this study. was the differences in the emphasis on small group dynamics. During the formal training period. there was a strong emphasis on peer group dynamics. while during the OJT period this was less so. Because of the structure and organization of the formal training the experimenter was free to manipulate and develop small group dynamics. More time was spent on developing specific group tasks. involving groups in the 94 decision-making process. and meeting with groups as autonomous units with independent leadership. However. during the OJT period the development of small group dynamics tended to dissipate. This was partly due to the increased responsibilities of the experimenter as well as the training structure.- At this time more emphasis was placed upon operating the business as opposed to group tasks. group decisions and group autonomy. The nature of the business limited the development of small —groups. for example. it became necessary to transfer indi- viduals from one group to another. because of school com- mitments and the original three groups had to be condensed into two groups for more efficient business operation. Also. groups became more dependent upon members of the other group in order to complete repairs on customers' cars. The results of the analysis of the sociometric data clearly indicate the need and importance of small group dynamics and peer support in operating a community treatment program. The results indicate that there was a definite relationship between the group's sociometric ratings and the rejection of participants arrested. The 95 results show that arrested participants not only were rejected by their peers before being arrested (see Tables 12.1 and 12.3). but that arrested participants were also rejected pfippngeing arrested (see Table 12.2). These findings indicate that peer-influence and group support can be very important in maintaining non- .delinquent behavior. It seems that by utilizing socio- metric ratings one may predict the reoccurrence of delin- . quent behavior. intervene at this point. and provide spe- cial treatment or attention to the rejected individuals. Since the experimental effect occurred for partic- ipants who were involved in the experiment during its initial develOpment. and since the type of training and emphasis on small group dynamics were different for the initial phase and the latter phases of the experiment. recommendations can be made for future research of delin- quency prevention on these bases. Accordingly. a more comprehensive experimental design. which can determine the effects of formal training and small group dynamics on the dependent measures. might be developed. Such a design might include the following conditions: 1) Partic- ipants receive formal training with a heavy emphasis on 96 small group dynamics; 2) Participants receive formal training and no group dynamics; 3) Participants receive on-the-job training with a heavy emphasis on small group .dynamics; and 4) Participants receive on—the-job training and no group dynamics. The inclusion of the aforemen- tioned conditions would allow the experimenter to better determine why the experimental effects of the present study occurred as well as providing useful information in creating other community-based delinquency prevention programs. In summary. the present experiment had a signifi- cant effect on the reduction of delinquent offenses for. initial participants who remained in the experiment less than three months. The experiment also provided partici- pants with confidence and desire to search and Obtain gainful employment. The results further suggested that sociometric ratings are good predictors of delinquent behavior and that formal training and small group dynamics may be beneficial in maintaining a community-based delin- quency prevention program. REFERENCES REFERENCES Allen. J. E. "The Silent Observer: A New Approach to Group Therapy." Crime and Delinquency. 16. 1970. Anderson. A. and Whitney. W. "An Evaluation of Youth Development Corp.." Unpublished. 1973. Bales. R. F. and Slater. P. E. "Role Differentiation in small Decision-Making Groups" in T. Parsons et al. (eds.) Family. Socialization. and Inter- action. Glencoe. Illinois: Free Press. 1955. Bass. B. M. Leadership. Psychology and Organizational Behavior. New York: Harper and Row. 1960. Blau. P. M. "Co-Operation and Competition in a Bureauc- racy." American Journa; of Sociology. 59. 1954. Browning. C. J. "Differential Impact of Family Disorgani- zation in Male Adolescents." Social Problems. 8. 1960. Cambell. D. T. and Stanley. J. C. Experimental and Quasi- Experimental Designs for Research. Chicago: Rand McNally & Co.. 1963. Cartwright. D. and Zander. A. Group Dynamics: Research and Theory (third edition). New YOrk: Harper and ROW! 1968 c Cattell. R. B. "New Concepts for Measuring Leadership. in Terms of Group Syntality." Human Relations. 4. 1951. pp. 161-184. 97 98 Cloward. R. A. and Jones. R. "Social Class. Education. Attitude. and Participation." in Passow (ed.) Education of Depressed Areas. 1963. Cohen. A. K. Deviance and Control. New Jersey: Prentice- Hall. 1960. Cohen. A. K. Delinquent Boys: The Culture of the Gang. Glencoe. Illinois: Free Press. 1955. Cohen. E. E. "Getting Hired. Getting Trained." U. S. Department of Health. Education and Welfare Publication. Office of Juvenile Delinquency and Youth Development. 1965. Coleman. J. S. "Equality in Education Opportunity." Department of Health. Education. and Welfare. Washington. D. C.. 1966. Dell. G. A. "Social Factors and School Influence in Juvenile Delinquents: An Analysis of Police Cases in the Belfast Juvenile Court." British Journal of Educational Psychology. 33. 1963. pp. 312—333. Deutsch. M, D. "Theory of Co-Operation and Competition." Human Relations. 2. 1949. pp. 129-152. Elliott. D. S. "Delinquency and Perceived Opportunity." Sociological Inquiry. 1962. pp. 216-222. Empey. L. T. "Delinquency Theory and Recent Research." Journal of Research in Crime and Delinquency. 3. 1966. pp. 28-41. Empey. L. T. and Erikson. E. H. "Delinquency Theory" in H. Voss (ed.) Society. Delinquency and Delinquent Behavior. Boston: ~Litt1e Brown. 1970. Empey. L. T. and Rabow. J. "The Provo Experiment in Delinquency Rehabilitation." American Sociological Review. 26. 1961. pp. 679-694. 99 Fairweather. G. W. Social Psychology in Treatinquental Illness. New York: John Wiley and Son. 1964. Fairweather. G. W.. Sanders. D. H.. Cressler. D. L. and Beck. D. S. Community Life for the Mentally ILLL An Alternative to Institutional Care. Chicago: Aldine Publishing Co.. 1969. Ferman. L. A. "Disadvantaged Youth: The Problem of Placement. Job Creation and Job Development." Manpower Administration. U. S. Department of Labor Publications. 1967. Fleisher. B. The Economics of Delinquency. Chicago: Quadrangle Books. 1966. Gazda. G. M. (ed.). Basic Approaches to Group Psycho— therapy and Group Counseling. Springfield. Illinois: Charles C. Thomas. 1968. Geiss. G. "Juvenile Gangs" in President's Committee on Juvenile Delinquency. U.S. Government Printing Office. Washington. D.C.. 1967. Gleuck. S. and Gleuck. Eleanor. Delinquents and Non- delinquents in Perspective. Cambridge. Massa- chusetts: Harvard University Press. 1968. Gold. M. "Undetected Delinquent Behavior." Journal of Research in Crime and Delinquency. 3. 1966. pp. 27-46. Gold. M. Status Forces in Delinquent Boys. Ann Arbor: University of Michigan Press. 1963. Gordon. R. A.. Short. J. F.. Cartwright. D. S. and Strodback. F. L. "Values and Gang Delinquency: A Study of Street Corner Groups." American Journal of Sociology. 69. 1963. pp. 109-128. Hammond. L. K. and Goldman. M. "Competition and Noncom- petition and Its Relationship to Individual and Group Productivity." Sociometry. 24. 1961. pp. 46-60. 100 Hare. A. P. Handbook of Small Group Research. New York: The Free Press. 1962. Hayes. W. L. Statistics for Psychologists. New York: Holt. Rinehart and Winston. 1963. Hersko. M. "Group Psychotherapy with Delinquent Adoles- cent Girls." American Journal of Orthopsychiatry. 32. 1962. Hollander. E. P. Leaders. Groups. and Influence. New York: Oxford University Press. 1967. Kelly. H. H. "Attitudes and Judgements as Influenced by Reference Groups: Two Functions of Reference Groups." in G. Swanson. T. M. Newcomb and E. L. Harley (eds.) Readings in Social_Psychology (2nd edition). New YOrk: Holt. 1952. pp. 410- 420. Levitan. S. A. "Anti-Poverty Work and Training Efforts: Goals and Reality." The Institute of Labor and Industrial Relations and National Management Policy Task Force. Washington. D.C.. 1967. McDavid. J. W. and Harari. H. Social Psychology: Individuals. Groups and Sociepy. New YOrk: Harper and Row. 1968. McCorkle. L. W. "Guided Group Interaction in a Correc- tional Setting." International Journal of Group Psychotherapy. 4. 1954. McCorkle. L. W.. Elias. A. and Bixby. F. L. The High- fields Story. New Ybrk: Holt. 1958. Maier. N. R. F. "The Quality of Group Decisions as Influenced by the Discussion Leader." Human Relations. 3. 1950. pp. 155-174. Miller. W. B. "Lower Class Culture as a Generating Milieu of Gang Delinquency." Journal of Social Issues 14. 1958. 101 Moed. M. "Increasing the Employability of YOuth: The Role of Work Training." Task Force Report on Juvenile Delinquency and Youth Crime. U.S. Government Printing Office. Washington. D.C.. 1967. Monahan. T. P. "Family Status and the Delinquent Child." Social Forces. 35. 1957. Peterson. D. R. and Becker. W. C. "Family Interaction and Delinquency." in Quay. H. D. (ed.) Juvenile pglinquency: Research and Theory. Princeton. New Jersey: D. Van Nostrand. 1965. pp. 36-39. Ohlin. L. E. and Cloward. R. A. Delinquency and Oppor- tunity. Glencoe. Illinois: Free Press. 1960. Raven. B. H. and Eachus. H. T. "Cooperation and Competi— tion in Mean-Interdependent Triads." Journal of Abnormal Social Psychology. 67. 1963. pp. 307-316. Rubin. L. N. and Hill. Shirley. "Assessing the Contrib- utions of Family Structure. Class and Peer Groups in Juvenile Delinquency." in Glasser. D. (ed.) Crime in the City. New YOrk: Harper and Row. 1970. Salaikeu. K. A. "Group Treatment of Juvenile and Adult Offender." Journal of Research in Crime and Delinquency. 10. 1973. Sarri. R. C. and Vinter. R. D. "Group Treatment Strategies in Juvenile Correctional Programs." Crime and Delinquency. 11. 1965. Schulman. I. "Modification of Group Therapy with Anti- Social Adolescence." International Journal of Group Psychotherapy. 7. 1957. Seckel. J. P. "The Freemont Experiment: Assessment of Residential Treatment at a Ybuth Authority Reception Center." Department of YOuth Authority. Sacramento. California. 1967. 102 Shafer. S. and Knudten. R. G. "The Family and Home" in Juvenile Delinquency: An Introduction. Shafer (ed.). New York: Random House. 1970. Sharp. E. P. "Group Counseling in a Short-Term Institu- tion." Federal Probation. 23. 1959. Shaw. M. E. Group Dynamics: The Psyphology of Small Group Behavior. New York: McGraw-Hill. 1971. Shaw. M. E. and Blum. J. M. ”Group Performance as a Function of Task Difficulty and the Group's Awareness of Member Satisfaction." Journal of Applied Psychology. 4. 1965. pp. 151-154. Shaw. C. R. and McKay. H. D. "Cultural Transmission." in Wolfgang. M. (ed.) The Sociology of Crime and Delinquency (2nd edition). New York: John Wylie and Son. 1970. pp. 225-232. Shellow. R. 8.. Ward. J. L.. and Rubenfield. S. "Group Therapy and the Institutionalized Delinquent." International Journal of Group Psychotherapy. 8. 1958. Singell. L. "The Economic Opportunity in Juvenile Delin- quency." in Fleisher. B. (ed.) The Economics of Delinquents. 1966. Slocum. W. and Stone. Carol. "Family. Culture Patterns and Delinquent-Type Behavior." Marriage and Family Living. 25. 1963. Spergel. I. Street Gang WOrk: Theory and Practice. New York: Doubleday. 1966. Thrasher. F. M. The Gang (1313 Chicago Gangs). Chicago: University of Chicago Press. 1936. Trojanowicz. R. C. Juvenile Delinquency: Concepts and Controls. New Jersey: Prentice-Hall. 1973. 103 Walker. G. J. "Group Counseling in Juvenile Probation." Federal Probation. 22. 1959. Walker. Helen and Lev. J. Statistical Inferences. New York: Holt. Rinehart and Winston. 1953. p. 151. Wheeler. S.. Cottrell. L. and Ramasco. Ann. Juvenile Delinquenpy: Its Prevention and Control. Task Force Report. U.S. Government Printing Office. Washington. D.C.l 1969. Wilcoxon. F. and Wilcox. Roberta. Some Rapid. Approxi- mate. Statistical Procedures. Lederle Laboratories. New York. New York. 1964. Wyer. R. 8. "Effects of Incentives to Perform Well. Group Attraction and Group Acceptance to Conform- ity in a Judgement Trial." Journal of Personality and Social Ppychology. 4. 1966. pp. 21-26. Zander. A. and Newcomb. T. "Group Levels of Aspiration in United Fund Campaigns." Journal of Personality and Social Psychology. 6. 1967. pp. 157-162. APPENDIX A JOB EVALUATION FORM JOB EVALUATION FORM MECHANIC RESPONSIBLE DATE TYPE OF JOB EVALUATOR (CHECK EACH QUESTION) 1. AMBITION TO WORK AT JOB WAS: GOOD AVERAGE BELOW AVERAGE 2. INDIVIDUAL IS REALLY TRYING TO LEARN: YES NO 3. JOB WAS COMPLETED IN A REASONABLE AMOUNT OF TIME: YES NO 4. QUALITY OF WORK WAS: GOOD AVERAGE BELOW AVERAGE ADDITIONAL COMMENTS: 104 APPENDIX B CONTROL PARTICIPANT'S JOB PLACEMENTS CONTROL PARTICIPANT'S JOB PLACEMENTS Number of P . . lace of Employment Type of Employment Part1c1pants Garvey Institute Black Cultural 4 Awareness Cristo Rey Community Chicano Cultural 4 Center Awareness Northside Drug Center Drug Counselors 3 Aide Eastside DrOp-In Recreational 3 Center Supervisor WJIM Radio Radio Technician 2 Trainee Y.D.C. Printing Printer 5 Training Program Apprentices Boys Club Recreational 2 Supervisor Industrial Laundry Laundry 1 Indian Center Indian Cultural ’ ' 2 Awareness Legal Aide Legal Intern 3 Free Spirit Sales Clerk 1 Total = 30 105 APPENDIX C PARTICIPANTS INTAKE FORM Interviewer name: PARTICIPANTS INTAKE FORM Address: Telephone: Sex: Male Female Age: (circle 14 15 16 17 18 over Ethnic Origin: Marital Status: Family Income: Nearest Relative: Relationship Referred From: Black Mex-Amer. Indian White fir Other (Specify) Married Single Separated Less than $3,000 $3,000-4,999 $5,000-6,999 $7,000 or more Address Date: (Month)(Day)(Year) Birth Date: (Month)(Day)(Year) Dependents- Outreach Staff (Specify) Agency (Specify) Police f Court Other (Specify)_ Educational Status: Highest Grade Completed: In School, Full-time Where? In School, Part-time Where? 7 8 9 10 11 12 13 14 + Out of School Reason Number of Times Suspended or Expelled: (truant, suspended, drop-out) 10GB 0 1 2 3 4 5 + 107 Correctional History: Has not been Arrested Arrested, but not Adjudicated Arrested and Adjudicated Arrested, on Probation Arrested, on Parole Arrested, Institutionalized Institutionalized, Out Reason for Arrest: _,_ Automotive Mechanic Experience: (List) 1. 2.‘ 3. Employment History: (List jobs for past two years, where and immediate supervisor) 1. How Long? 2. fl How Long? 3. 'How Long? What do you plan to do for employment in the next two years? How do you feel about owning and operating an auto repair shop in Lansing? In addition to having a job would you be willing to spend one (1) night per week to improve your community? Yes No Comments by interviewer on individual's prospects: Module: 1 2. 3 4 5 Auto Experience: None 1 Low High APPENDIX D PARTICIPANTS FOLLOW-UP FORM PARTICIPANTS FOLLOW-UP FORM NAME: DATE: (Month) (Day) (Year) OUTREACH WORIGR: MODULE: 0 1,2 3 4 5 Number of Weeks Client has been in Y.D.C. School Status: Continued in School--Full-Time " Name of School Continued in SchOol-eParthime Name of School Truant (reason) Suspended (reason) Dropped Out (date and reason) Graduated (date and school) Other (specify) Employment Status: unemployed Employed (full-time and place) Employed (part-time and place) Y.D.C. work-Intern (place) Correctional Record: Not been arrested Apprehended, not arrested ‘ Arrested, but not adjudicated Arrested and adjudicated 'Arrested, on parole' Arrested, on probation Arrested, court ward Other (Specify) i If arrested, explain details of arrest and results of arrest: .108 109 RECOMMENDED SUPPORTIVE SERVICES : Psychological Counseling Psychological Testing Medical Treatment Social Counseling-Group Social Counseling-Individual Academic Training vocational Training Other (Specify) Referred to following source for services Explain results of services (if possible): Referred to other program: Police Court (on-going) School Family Social Agency (specify) Other (specify) Explain results of referral (brief) APPENDIX E SOCIOMETRIC RATING SCALE Directions: Names SOCIOMETRIC RATING SCALE Below there are seven scales. one for each member of the group. Please rate each mem- ber. excluding yourself. on the scale as to how much you like or dislike them. (CIRCLE YOUR ANSWER) (1) Do not like at all (2) Mildly dislike (3) Dislike just a little (4) Neither like or dislike (5) Like just a little (6) Mildly like (7) Like very much (1) Do not like at all (2) Mildly dislike (3) Dislike just a little (4) Neither like or dislike (5) Like just a little (6) Mildly like (7) Like very much (1) Do not like at all (2) Mildly dislike (3) Dislike just a little (4) Neither like or dislike (5) Like just a little (6) Mildly like (7) Like very much (1) Do not like at all (2) Mildly dislike (3) Dislike just a little (4) Neither like or dislike (5) Like just a little (6) Mildly like (7) Like very much (1) Do not like at all (2) Mildly dislike (3) Dislike just a little (4) Neither like or dislike (5) Like just a little (6) Mildly like (7) Like very much (1) Do not like at all (2) Mildly dislike (3) Dislike just a little (4) Neither like or dislike (5) Like just a little (6) Mildly like (7) Like very much (1) Do not like at all (2) Mildly dislike (3) Dislike just a little (4) Neither like or dislike (5) Like just a little (6) Mildly like (7) Like very much 110 APPENDIX F MORALE RATING SCALE Name _ Shift Date MORALE RATING SCALE Below are liSted several statements with which you may agree or dis- agree. There are five possible ways you might react to each state- ment. Please put a check mark in front of the one response which most adequately expresses your feelings. 1. I would just as soon have my present job as any other job that I know about. __§trongly agree__Agree__pndecided__pisagree_+Strongly disagree I am perfectly happy_with my present position in the group. __§trongly agree__Agree__pndecided__pisagree__§trongly disagree I have a lot of training or skills that could be used better by the group. __Strongly.agree__Agree__pndecided__pisagree__Strongly disagree Even if it were not for matters of training and experience, I would rather be in my present position with this group than in any other position with any other group. __§trongly agree_;Agree__pndecided__pisagree__§trongly disagree Is your group manager good at figuring out easthays to do things when the group has work? __yery good;_Pretty good__Sometimes__psually no__Not good Does your group manager usually help to settle arguments among the men under him? __Almost always__pften;_§ometimes__Rarely;_Never How often does your group manager help the members out in per- sonal matters? __Almost always__pften__Sometimes__Rarely;_Never When the men or women in your group do something wrong, is your group manager able to handle the situation? __Almost always__pften‘_sometimes__Rarely__Never All of the members in my group work as hard as they can. __Almost.always__pften__Sometimes;_Rarely__Never 111 112 10. All of the members in my group cooperate with each other. __Almost always__pften__Sometimes__Rarely;_Never 11. So far as overall performance is concerned, I would rank my group as: __pne of the best __pne of the next best __Somewhere between the best and worst __As a poor group __As the poorest group APPENDIX G JOB BEHAVIOR SCALE WOrker's Name Rater's Name 1. 2. 10. 11. 12. 13. The worker The worker The worker The worker The worker The worker The worker The worker The worker The worker of time. The worker The worker The worker The worker JOB BEHAVIOR SCALE Group Date makes few errors. makes many errors. seldom needs prodding to get job done. usually needs prodding to get job done. rarely needs help. often needs help. follows directions on the job. does not follow directions on the job. works for extended periods of time. fails to work on a task for any reasonable period abides by rules and regulations. violates job rules and regulations. usually profits from constructive criticism. usually does not profit from constructive criticism. The worker The worker The worker seldom becomes upset by failure. usually becomes upset by failure. usually does not complain about tasks that are given him. The worker The worker The worker The worker The worker' The worker The worker The worker The worker usually complains about tasks that are given him. usually applies self to task. usually does not apply self to task. works with average accuracy. 3 work is often inaccurate. works constantly on task. rarely works on the task. reports regularly to work. often fails to show up for work. 113 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 114 The worker usually profits by his mistakes. The worker repeats his mistakes over and over again. The worker does not do or say anything which is disturbing to others. The worker sometimes says or does things that disturb others. The worker is courteous and considerate, showing a normal concern for the feelings of others. The worker is discourteous and often shows no concern for the feelings of others. The worker is reasonably friendly and agreeable. The worker is usually unfriendly and disagreeable. The worker often speaks to others. The worker rarely speaks to others. The worker is usually happy. The worker is usually sad. The worker takes pride in the quality of his work: i.e.. shows work or talks about work to others and/or super- visor. The worker does not take pride in the quality of his work. The worker seldom finds fault with others and/or their work. The worker often finds fault with others and/or their work. The worker pays attention to the work and activities of others. The worker does not pay attention to the work or activities of others--is in a world of his own. The worker makes some worthwhile suggestions about tasks. The worker makes no worthwhile suggestions about tasks. The worker usually accepts constructive suggestions from the supervisor. The worker seldom accepts constructive suggestions from the supervisor. The worker is not usually disturbed by constructive criticism. The worker is usually disturbed by constructive criticism. The worker seldom finds fault with his work. The worker often finds fault with his work. APPENDIX H ADMINISTRATIVE AGREEMENT ADMINISTRATIVE AGREEMENT The following agreement has been drawn up between all interested parties with regard to the automotive co- operative research project being conducted by Youth Devel- opment Corporation (Y.D.C.). The project is being carried out in an effort to determine the effect of delinquents working in group employment (automotive co-op) and indi- vidual employment (work-interns) on their rehabilitation and ability to become effective organizers in their com- munity. In order that the responsibilities of all indi- viduals involved in the project are not misunderstood. the following responsibilities of each are hereby agreed to: ON THE PART OF THE ADMINISTRATION OF Y.D.C.: 1. Finance the project according to the specified budget and provide stipends for the project par- ticipants according to the work-intern criteria. 2. The utilization of Y.D.C. clients as participants of the project. 3. The random assignment of project participants either to the automotive cooperative or to work- intern positions. 4. That all data concerning project participants shall be made available to the project director. 115 ON THE 116 That the outreach staff of Y.D.C. will participate by referring prOSpective project participants and assist the project director in collecting follow- up-data on the project participants. That the project director can administer ques- tionnaires and interviews to the participants of the project upon their approval. Individuals not selected to participant in the CO-Op will be given employment commensurate with their education and ability. If employment is not obtained in 30 days. the project director will then be notified. PART OF THE RESEARCH PROJECT DIRECTOR: Assume complete responsibility for the daily operation of the automotive cooperative project. Keep the confidentiality of all data concerning the project participants. Make available all reports on research evaluation and monitoring of project to Y.D.C. Control project expenditures as specified in the budget. Assign the project participants into small groups . for community organizational purposes. Be responsible for assigning the duties and responsibilities to members participating in the operation of the cooperative. Will use the information collected from this project only to meet doctoral dissertation requirements. 117 These agreements shall be in effect during the eleven months the project is expected to run. beginning the weekuof October 30. 1972 and ending the week of Octdber l. 1973. subject to any changes or extensions by the granting agency--H.E.W. Ybuth Development Corporation Administration Project Research Director Date "I7'11?@11111'1’11'1111T