THE EFFECT OF LAY GROUP COUNSELUNG ON ADULT PROBATIONE-RS AND PAROLEES Thesis {or the Degree of M. A. MICHIGAN STATE UNIVERSITY James Francis Ghasfin, 5r. 1 9.64 v.09. -.Q ‘ "=”§‘“ LIBRARY Michigan State University (D THE EFFfiOT OF L‘Y ROUP COUNSELLING ON ADULT PROBATIONERS AND PAROLEES James Francis Ghastin, Sr. Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Sociology and Anthropology 1954 ACKNOWLEDGEMENTS The writer has been employed by the Michigan Depart- ment of Corrections since August 1954. He is presently assigned to the Youth Division as a Youth Crime Prevention Specialist and, in this role, is reSponsible for the development of new programs at a local community level. Much of what is done is carried out through the assistance and c00peration of other divisions. Thus, the project presented in this study is an example of a truly c00perative endeavor. Given the collaborative nature of this effort, personal thanks are extended to Don Schmidt, Probation Officer, Ingham County Circuit Court, for his tireless efforts in the testing program. The data collection would have been an impossible task without his assistance. Appreciation is also expressed to William Hauso, Chief Probation Officer, Saginaw Municipal Court and his staff for their assistance; also to Lloyd G. Service, Chief Probation Officer for Ingham County and his staff and the Lansing Parole Office for their cooPeration. Special thanks must go to Robert H. Scott, Assistant Director of the Department of Corrections in charge of the Youth Division for his support and counsel, Without his confidence the endeavor would never have begun. Appreciation is expressed to the Volunteer Bureau of Lansing for their help in recruiting leaders and volunteers to assist with the testing, to the Y.W.C.A. and Y.M.C.A. and to the volunteer group leaders. ii i de‘1 Ccmnica' far their this proj Alsc Eua'ell o: guidance Fin her hus‘: A debt is due to Thomas Danbury and Albert Talbot of the Communications Research Center at Michigan State University for their valuable assistance with the statistical work in this project. Also gratefully acknowledged is the help given by Dr. John Howell of the Department of Sociology and Anthropology. His guidance, counselling and assistance were much appreciated. Finally, many thanks to Jean for Just putting up with her husband. Page mrwm 0.00.00.00.00 ii ”HT VluIOi JJ-UDleeA-JM‘J- 0.00.0.000000000...0. LIST OF TABLES 0.00000000000000000000000000.0...O. V Chapter .. l 1. GEZIH-L‘A-L PROBIIE'I.0...0......0...o.o...0.o Introduction Social Group Work Approach Purpose of Study II. HliOTUT ES AND EXPLORATORY Q3 ES TIONS ...... $04.] 12 Formulation of Hypotheses Theoretical Rationale Design of Project III. PRES“ MITA “IOIq OF DATA 0000......000000000000 28 Statistical Procedures Analysis of Data IV. SEE-LIAN Iii-I'D GOES-I.- 531014.? 3 . . . ...... . . . . . . . . . . 53 Summary of Data Implications of Study Suggestions for Further Research BIBLIOGRAPHY 0.0.0.000000000000000.0000000000.00.0 6O 63 MPECDILJI 0.0.000...00.0.0.0...OOOOOOOOOOOOOOOOOOO. iv O C 'n'a" , ‘ h'ad- n) 0 5,4 Table rage 1. Experimental and Control Group Designations 19 2. Selection Criteria Distribution For Probation Groups ...................... 30 3. Selection Criteria Distribution For Parole Groups O0.00.0.0...OOOOOOOOOOOIOOOOOOOO 31 4. Analysis Categories and Corresponding SJb-Groups 0.00.00...OOOOOOOOOOOOOOOOO0 53 5. Self-Ideal Correlations (raw scores) ...... 34 6. Self-Ideal Correlations by Violator Groups Before&M-ter 00.0.0.0...OOOOOOOC..... 33 7, Delinquency Proneness (raw scores) ........ 30 8. Delinquency Proneness Mean Scores By Violator Groups Before & After ........ 37 9. Rank Distributions of Individual Self-Ideal Correlations and Reductions of Counselees 0000000000000.00000000000000 4O 10. De-Scale Distributions .................... 43 ll. Q-Correlations and Net Change Distributions of Counselees ......................... 4o 12. Self-Ideal Correlations by Groups and Leaders 0.0.00.0...OOOOOOCOOOOOOOOOOOOO 1+8 and m as o—Liee] 113 . 9-39 DI‘ CHAPTER I GEEERAL PRO ELEM Traditionally,the approach in corrections to probation and parole has been on a one-to-one basis. For the most part, individuals are placed on probation or released on parole from an institution under supervision of a probation or parole officer. The individual generally reports regularly (usually monthly) to his supervising officer and individual counselling or casework techniques are employed. In some cases where specialized services are needed, such as employment, financial assistance or psychiatric consul- tation, they are referred to apprOpriate agencies. A major limitation of the one-to-one relationship is the frequency and duration of contacts. Caseloads for probation officers in Michigan vary from 40 to 100 men per officer with the median about 80. For the parole officer, the caseloads vary from 65 to 150 men with the median about 110. In addition to supervision, the officer must also do pre-sentence and pre-parole investigations that are time consuming. He must also maintain routine paper work (i.e. progress reports, statistical reports, etc.). The time remaining for direct supervision is greatly reduced. In addition to seeing the subject in his office monthly, the officer averages about one or two field contacts per month. (1) 0 '. vi 7! - A o s 4‘ O 0 "I 900 I t which 0 Hist D Caselo; ”7e tit/5 Jr a. v...» S Cones: F,‘ ND Wu. Dun -‘. a ‘ vs inter: QvL ~rom A: ‘HA I n {0* ha 1V Qw Both office and field contacts would average about 15-30 minutes. This means that the typical officer would see each man on his caseload about one hour per month or less. The time spent by the parolee or probationer in direct contact with the officer as compared to the time spent with others leaves little doubt as to where the prepon- derance of contacts exists. The caseload realities under which officers operate are dramatized when compared with what professionals feel are model loads. Ideally every officer would do a maximum of 50-100 work units per month.1 By comparison probation and parole officers in Michigan (a state that is a leader in probation and parole practice) will average 130-150 work units per month. However, lack of funds and personnel makes substantial reduction in caseloads unlikely. A central problem facing corrections is the evaluation of different or supplementary approaches. Techniques capable of increasing effectiveness without increasing operating costs are to be sought. The primary concern of this study is to evaluate the merits of one such possible approach. The social group work approach to the problem as used in this context is a form of treatment which involves interaction between more than two persons. It can range from group therapy as utilized in various institutions to group living experiences where the influence of the entire lFifty units recommended by The National Council on Crime and Delinquency. One hundred units represents standards set by The Michigan Department of Corrections. (2) a» 4 41 u .s.» NO _ W . . .. .. ad . . u G. ‘ a a . . . S 2 e .p . m . . VA A. V De — - .U 09 01:“ “N Q“. I I u...» "v I “Q “I‘M. flu. 1* U o s Q J" Au fix“ I-» .. x .. “ rd 4; 3.... ma. ..nu an.“ m” .1“ A.» “a. 1 .ni 0 A... AC Q. Q» .o 1. ab ; a l 5!. 4 «r. 14 . m ._. as L 1V 0. . ,_ ww. MG. Am . L M; r ...., M $14 “A 3 "1 .,.. NC a . A. .3. .n.. O n I~ \. -~ U ( m“ u... an .n .3. :0. an. .5 .0 3 a X t C 3 C .1 q . . . . setting is brought to bear on the individual in a thera- peutic way. It may be viewed as one approach supportive of the work of probation and parole officers.2 This approach as specifically focused upon the law violator and his treatment has been reported in several studies over the past several decades. An early example of this approach was the work of August Aichorn. In Wayward Youth} he reported his experiments with disturbed youths in a detention facility. Later Maxwell Jones wrote of his work with therapy groups in The Therapeutic Community.4 More recently an eXperiment was conducted at a juvenile facility in New Jersey known as Highfields. McCorkle, Elias and Bixby in The Highfields Story described the program of group living and the utilization of group counselling procedures.5 In 1954, Dr. Norman Fenton, Deputy Director of Classification and Treatment for the California Department of Corrections, began a period of informal experimentation with group counselling at Folsom 2Glesla Konopka, Group Work in the Institution, (New York: Whiteside, Inc., 1954). Gertrude Wilson and Gladys Ryland, Social Group work Practice, (Boston: Houghton Mifflin Co., 19497. 3August Aichorn, wayward Youth, (New York: Viking Press, 1956). 4Maxwell Jones, M.D., The Therapeutic Community, (New York: Basic Books, Inc., 1953). 5Lloyd w; McCorkle, Albert Elias and Lovell F. Bixby, The Highfields Story, (New York: Henry Holt and Company, 1958). See also: HT Ashley Weeks, Youthful Offenders at Highfields, (Ann Arbor: University of Michigan Press, 1958). (3) 0 18 W“ .4 In 1‘ 0“ care: \. series c p." Ea tecfl” 4" "fl“ U-b V l“ larole aft the role 0 R U 4 wflu .fi.\ .‘b b; .1 7.... .1 E «5 .43 State Prison in California. This program was eXpanded and is now being used in parole as well as institutionally.6 In the California program, staff personnel are utilized in the role of group leaders. In the parole program, the parole agent himself acts as group leader. In 1960, Bassin, Berlin, and Smith reported their study of the impact of group therapy on probationers.7 A program of group therapy similar to that reported in this thesis was carried out under relatively careful controls. In the study, the investigators, through the use of the Thematic Aperception Test (TAT), found significant im- provement among the probationers who were eXposed to a series of group therapy sessions. These group sessions were conducted by staff members of the BARO (Brooklyn Association for the Rehabilitation of Offenders) clinic. Prior to the Brooklyn study, an experiment was con- ducted in Macomb County, Michigan with a selected group of probationers.8 This group went through twelve weekly sessions of group counselling with lay leaders. The Thematic Aperception Test and Machover Figure drawings were employed. Indications at that time were that the 6Norman Fenton Ph.D., An Introduction to Group Counseling in State Correctional Service, The American Correctional Association, 1956. 7Alexander Bassin Ph.D., Louis Berlin, Alexander Smith Ph.D., Group_Therapy With Adult Probationers: Federal Probation, September, 1960. 8James F. Ghastin, An Experiment in Group Counselling for ngihfulerobationeEE, An unpublished report, Michigan Department of Corrections, September 10, 1959. (4) 1‘“ my finy‘w- - vv‘.. 4 aw i2: 36‘ ‘ I thn‘ «5‘1: at“ 9. P. .0. sh; an~ AMV :1‘ o‘ ‘ h\ ~|s .. .1 n o. as. T9 0 \d group showed significant improvement in two major areas: (1) Concept of the adult male role and (2) Insight. The third area, change in anti-social thinking, was incon- clusive.g While there were some weaknesses in the research design, the indications were hat lay group counselling, as a treatment tool, was effective. Other researchers have studied similar programs. Stuart Adams and Joachim P. Seckel made a study of guided group interaction in a youth training school in California.10 In summarizing their report they stated that the boys showed less scapegoating, reduction in tensions, breakdown of racial cliques, self-enforcement of conventional norms, less discipline problems and more emotional maturity. In addition, there was an improvement in Delinquency Orientation. In 1960 and 1961, Dr. Hans A. Illing supervised the training of parole agents in group counselling methods.11 In this project he found no significant improvement in the counselees. He did suggest that this might be due to the attitude of the individual parole agent since he functioned as the group leader. 9"The Effect of an Organized Program on the Anti-Social Themes Elicited by the Thematic Aperception Test From Youthful Prison Inmates." Ernest L. V. Shelley, Doctoral Dissertation, Michigan State University, 1959. 10Stuart Adams and Joachim P. Seckel, Assessment of Guided Group Interaction in a Youth Training School, Current Projects, National Research and Information Center on Crime and Delinquency, National Council on Crime and Delinquency, New York, Winter 1962-1963. llIbid., V01. 1, Spring 1962, Project no. 54. (5) ..o n"' *1 c :3“: 81-5- " arafeC izlrove: 6' .3 95.838 1!. V‘uL V U- nho‘r‘g r: .dV- v u “V~—L .vr' ”Ov- 8! '“iri To assess the impact of group psychotherapy with juvenile parolees, Robert wald,M.D. conducted a research project in woodland, California.12 His findings indicated improvement in recidivism of those participating. While there were some conflicting results, he concluded that they resulted from inadequate tests. The variety of programs studied in the past tend to support the hypothesis that the group approach is an effective tool in the correctional treatment process. There still remains, however, a need for a careful study of group counselling utilizing volunteer lay group counse- lors. Group counselling began in the Michigan Correctional system in 1954 with the first experimental groups at Camp Brighton, a corrections-conservation camp located in the Brighton Recreation Area. Group leaders were recruited from the surrounding community and came to the camp one evening per week for about two hours. Each leader met with a group of 6-12 inmates. During these sessions, the groups discussed social problems and situations that were of concern to the group members. Following these sessions, the group leaders and staff members at the camp discussed the progress of their groups and the problems that occurred in their leadership roles. These sessions served as in- struction periods in which the leaders were aided in under- standing some of the fundamentals of group dynamics. This 12lbid., Vol. 1, Spring 1962, Project no. 369. (6) 1_¢‘ "“C "”3: -- An..- - n VV “AU :Ja Venn A... IN w 2011“ ‘n ‘ ~.. I ‘. 2-5 M a - \ A - bul‘ :~ 8- :pi 4. are=~ program is now a continuous and integral part of treatment. The first Corrections Department program of group counselling involving probationers was begun in 1956 in Macomb County. Here experimental groups were utilized in conjunction with a summer vocational training program for probationers. Later in 1959, an experimental project was conducted in Macomb County using group counselling for eighteen selected probationers.13 The procedure here was basically the same as was followed in the program at Camp Brighton. Leaders were recruited from the community and met one evening per week. In this case, they were super- vised by probation staff personnel. This particular conception of group counselling has Spread throughout the Michigan correctional system and is now being used in several institutions. It would be appropriate here to briefly summarize the formal philoso- phy and selected operating features of these programs as ideally conceived by the Michigan Department of Corrections. 1. Group counselling, as used in Michigan Correctional settings, is a process whereby positive attitude and value changes are sought through informal small group discussions. The focus is on personal insights in the realm of social relationships rather than into deep-seated emotional con- flicts. 2. Through small group discussions, members also have an opportunity to learn, develop and maintain: (a) skills of 130o. cit., James F. Ghastin . (7) Av Au $'-« 9-- n‘fi"fi": D U- ,.U V“ H 3..., .9» WC “C 5. ”TL .RU :L «5 n1‘ V. .5: n: self-empression and communication; (b) tolerance through the acceptance of individual differences; and (0) con- structive relationships and understanding of society by interaction with lay group leaders. (This can also be a ”two way street” in that volunteer leaders develop new attitudes towards violators and the correctional process by seeing the former as individuals rather than as stereo- types and the latter as potentially constructive rather than merely repressive.) 3. Group leaders are selected who have the ability to relate to people easily, accept individual differences without undue concern, and who can guide conversations with tolerance and understanding. An individual who is honest with himself as well as others is potentially an excellent group leader. For the most part, group leaders are volunteers coming from colleges and from communities in the area. Some members of the Corrections Department also take part. All are asked to commit themselves to a minimum of twelve weekly meetings barring emergencies. 4. One aspect of the group counselling process is super- vision and training. Both are provided by Corrections Department personnel. Before a leader is assigned a group, he is given an opportunity to observe actual group counsell- ing sessions. After the group session, an evaluation (or supervisory) meeting is conducted. During these meetings techniques and methods are discussed. Supervision is seen as an on-going process that enables leaders to develop and eXpand skills. The meetings provide leaders with an (8) Opportunity to discuss and share problems of mutial concern regarding group methods. 5. The group leader functions as kind of a catalyst - a stimulator of discussion. It is not his job to teach a class, give a lecture, or to provide information about corrections. The leader acts as an enabler whose job is to encourage spontaneous discussions of social problems and situations that are of concern to the group members. The leader is responsible for helping the members discuss a topic or concern rather than presenting or initiating topics that he feels may be important. 6. The leader attempts to provide a situation where each member is able to participate. He must be sensitive to the feelings of others in order to prevent the shy or "put extremely verbal member from becoming conSpicuous, or on the spot". 7. He attempts to focus attention on group problems as well as individual concerns. He attempts to create an atmos- phere that is conducive to free interchange of opinions and attitudes without concern as to their objectivity or merit. 8. The leader, then, is a person who enables the group to discuss personal attitudes and situations that are of particular concern to the members. He provides, through his own objectivity, an opportunity for members to discuss spontaneously,personal attitudes and values without loss of status or prestige with the leader or other members. (9) -- .c-1%fl ‘ U..“*‘ 0 0» .fl.‘ A .r. . V‘- i vario‘ 0""! G'-\J Y 3. — .gv my 0c .1A 1. .5. .3 a; x: Purpose of Study The problem of changing anti-social behavior has been outlined. Some of the various approaches to this problem have been discussed as well as the background and history of the group approach in the Michigan correctional system. As indicated, the Michigan Department of Corrections has had eXperience with various programs of group counselling with probationers since 1956. However, little in the way of systematic evaluation has been carried out with such endeavors, notwithstanding some empirical evaluations of various programs, most of which has been based on obser- vations, interviews, questionnaires, etc. Nonetheless, there appears to be general concensus and some empirical evidence that these programs merit careful evaluation inasmuch as they appear to achieve aSpects of the ob- jectives previously discussed. The purpose of this study is to evaluate a method of group counselling for adult probationers and parolees conducted by lay group leaders. Primary concern will be with what effect, if any, this experience may have upon the self-concept of participants. Because the sglf is not a rigid and essentially fixed pattern but,rather,as Rogers has said, "...an organized fluid but consistent conceptual pattern of the 'I' and 'me' which are admissable into awareness...",it would seem possible that under eXperi— mental conditions, change in self-concept might take place.14 14Carl R. Rogers and Rosalind F. Dymond, Psychotherapy and Personality Change, (Chicago: University of Chicago Press, 195KT p. 55. (10) «.3 .C '3': be U 11'5”. a a n C o. m e :1 .. u v n 9 U I! ‘— Cy h» Q U-. 9. s a Y VK‘. h...‘ C. nfil. .Fu 3 .1 r u a. 1:. ow (to My. hv Nu 0.» LU n .1 C. h». 7 ~ Db - -1. , o. .. "VC- «0 :I* fink Specific attention will be given to any reduction attributable to the counselling eXperience in the diSpari- ty between how the individual actually sees himself and what he would ideally want to be. This will be done on the assumption that the findings of previons studies,which indicate that such a reduction in disparity will result in a better adjusted individual, are findings applicable to the population groups under the conditions of this study. While better adjustment, both social and personal, is a function of the correctional process, it also has a responsibility to reduce law violations on the part of its clients. For this reason, determination of any change in " is another factor with which this "criminal proneness study will be concerned. The question under consideration 'in this regard becomes one of the impact, if any, of the eXperimental conditions upon the subject's tendencies toward laWiviolating behavior. In the following chapter, specific hypotheses re- lating to these central concerns will be formulated and discussed. Also presented will be selected exploratory questions relative to the major issues. Data from the study will be employed not only to test specific hypothe- ses but, also, to develop these exploratory questions. (11) 4 ‘3 v In .11 It" !1‘ ‘m—J CHAPTER II HYPOTHESES AND EXPLORATORY QUESTIONS The preceding chapter outlined the general problem with which this study is concerned as well as one approach to the problem, that of social group work. The purpose of the study was also discussed in relation to the problem. This chapter will deal first with the hypotheses and ex- ploratory questions. Next,the rationale by which the study will attempt to achieve its purpose will be presented. Finally, the chapter will outline the design of the project by which the hypotheses were tested. Hypotheses and Exploratory Questions. This study is con- cerned with the following central hypotheses: H1. Participation of adult parolees and probationers in a program of lay group counselling will result in a reduction in disparity between the self-percepts and ideal self-percepts of counselees. H2. Participation of adult parolees and probationers in a program of lay group counselling will result in a reduction in criminal proneness. H3. Reduction in diSparities between self-percepts and ideal self-percepts of counselees will be correlated with the reduction in criminal proneness. (12) Tn L4 questic: of these In addition, the study will examine a number of questions of an eXploratory nature. The more significant of these are listed below: 1. What is the relationship between self-ideal diSparity and the amount and direction of change? 2. Are there any significant differences in the self and ideal concepts of parolees and probationers as a group? 3. What is the relationship between these two groups ‘ and the amount and direction of change? 4. To what extent does the group leader influence change? Theoretical Rationale The first hypothesis is based primarily upon the work of Carl Rogers and others with the self concept.l Therein the gglf is viewed not as a rigid and essentially fixed pattern but an organized, fluid and consistent pattern of the "I" and "me" that is admissable to awareness together with the values attached to these concepts. The self-concept is not a given but rather the self depends upon the way in which we perceive others as treat- ing us. Self-esteem or self-value depends upon interaction with others. Thus, the self is a social product which can arise only through social interaction. Butler and Haigh hold that a discrepancy between the self-concept and the concept of the desired or valued self reflects a sense of self-dissatisfaction, which in turn lOp. cit., Carl R. Rogers and Rosalind F. Dymond. (13) .1. This it“: to 01?. at. A: r Al .1 Av Dy ‘1 ¢ U C a a. r generates the motivation for coming into counselling.2 This implies that the greater the internal dissatisfaction (disparity), the greater will be the individual's desire to change. If, as Butler and Haigh have suggested, counselling creates conditions under which the individual can reorganize his self-structure so as to make the self and ideal more consistent with each other, then the eXperi- mental conditions of this study might be eXpected to produce the results proposed in the first hypothesis. These assumptions of diSparity and diSparity reductiom also seem consistent with dissonance theory. This latter theoretical approach includes within its concern the problem of the relationship between discrepancy and atti- tude change by way of its assumption of tensions created by engaging in attitude-discrepant activities. It assumes that when tension (dissonance) is created, the person will react so as to reduce this tension and that the greater the attitude-discrepant activities, the greater the tensions, the more likely the person will attempt to reduce these tensions. In the present study, the individual can do this in two ways. One, he may withdraw by refusing to partici- pate or passively resisting. Two, he can participate in which case he will need to reorganize his self-structure. Dissonance theory does not predict when a person will engage in attitude-discrepant activities. It does say that when an individual does engage in such activity that 29p. cit., Carl R. Rogers and Rosalind F. Dymond, pp. 58-59. (l4) L J vi“ ~L a: I1J an“ 381;- ram fi'y‘e r t. '5 .- isi ‘ L de, a 7L .. a ‘1' .5 e C _ ..d h... .‘b 30 A: 01‘ u i d l as “U‘ ~ slq 1" h. C a. x « A4 FIN Hi \I \ .1.\ RAM tensions are produced and must be reduced. Butler and Haigh, in dealing with the reduction of diSparity, make essentially the same statement when they say that self-dissatisfaction generates the motivation to seek counselling. While there is no clear cut indication as to when they will seek help, the implication is that they are more reSponsive to help when the dissonance is greatest. Reduction in disparity (here roughly equated with dissonance) is correlated with better social adjustment. This implies, at least, that such reduction should lead to a reduction in law violations (unless, of course, the ideal self incorporates law violating attributes). If this were true, then we would expect the second hypothesis to hold true. However, there is little research in this area, particularly with the use of prediction scales. The prediction scale developed by Gough and Peterson is based on a role-taking theory of psychopathy.3 In the deve10pement of the scale, they pointed out that de- linquency and psychopathy are not necessarily synonymous, but they do assume that the total delinquent and criminal population includes a large enough proportion of psycho- pathic types to make feasible the application of role-taking theory. The present study makes the same assumptions with respect to the population under study. For this reason, the second hypothesis is,to a large extent,Speculation. It 3Harrison G. Gough and Donald R. Peterson, "The Identification and Measurement of Predispositional Factors in Crime and Delinquency", Journal of Consultingngychology, XVI (June, 1952), p. 207.: (15) E a 9 A an 7-? i S s E- incre two i“ '3'”: to d :J «0 Dy 9 r A. r +9 x? may also be that this is an independent variable not tied to dissonance reduction. What is the relationship between self-ideal diSparity and the amount and direction of change? For the purpose of this study, dissonance is equated with diSparity. Cognitive dissonance theory suggests that the greater the dissonance (disparity), the greater the likelihood of change. It further stipulates that in many cases dissonance must be increased in therapy before reduction takes place.4 These two factors will have to be taken into account in any in- terpretation of data dealing with amount and direction of change. Are there any significant differences in the self and ideal concepts of parolees and probationers as a group? Parolees reputedly have more difficulty in adjustment and may have a greater diSparity between self and ideal concepts. What is the relationship between these two groups and the amount and direction of change? If greater dissonance results in greater pre-dispositions to change and parolees as a group have greater dissonance; then it should, perhaps, follow that they will show more change. We would be interested in the direction and amount of that change. There is some evidence that parolees as a group are more resistive to change because of lengthy eXperience with treatment. This may, perhaps, tend to retard either nega- 4Qp. cit., Carl R. Rogers and Rosalind F. Dymond, jp. 58. (16) we 0 3.: 211* ally 'n’e wil 41' 3 Ohi‘ + U A.» a Us ,3 7J tive or positive change. To what extent does the group leader influence change? Of all the unknowns, this factor is among the most per- ‘ plexing. In a situation where you are dealing with essenti- ally untrained leaders, many questions immediately arise. We will want to eXplore the relationship between self-percepts of counselees and leaders and the direction of change. We will want to examine the nature of counselees' and leaders' ideal selves and any changes that result. Data on the relationship between leader and client may give some insight into what type of leader is most effective in this kind of role. Design of Studv Selection of Samplg. It was initially decided that two samples would be taken; one from the caseload of the Lansing Parole Office, and one from the caseload of the Ingham County Probation Department. At that time, there were approximately 279 parolees and 460 probationers under active supervision. From these caseloads, 36 parolees and 36 probationers were to be selected. Both probationers and parolees were selected so as to determine what differences, if any, the responses of the two groups would be to the counselling program. The two samples were to be matched with reSpect to the following characteristics: l..égg - 17 through 30. 2. Intelligence - To be maintained within normal limits. To be eliminated were those who were very high or very low. (17) v . O .n a. P.» u “I. «\u‘ v. . I) "1 n u :1. Mb. .I. Pa. O. on; C. » 9w 4:. Ow .14. O O 0 “. v“ Cu Dy .1 1 AC «D n.“ .. .. Q. I u and l u I... G 1 .3 ,C e Q ~ u l P. a l T a +U 2. .n-.. e a r. U. ,3 D C .1 .1 .l a 7.. w u ,0 a .1. .r O 0 e +u 1 a .1. -. O -. mm «.0 1+ «mu 0 o a an. 3 C . u a.-. C «7 ~ 9 «a o. I «a e a u as n .. r . r. e e .71... a a. . m u 0 ..... ..... u u .5; h; wL e C .3 U. n. “I C nv at by C «a n4 ‘u fiv ‘5 U. .Ao - a. a 3. Education - Maintained at a minimum of eighth grade and a maximum of twelfth grade. In the case of probationers, school records were used. For parolees, the recorded A.G.R. (Average Grade Rating) at the time of reception at the Diagnostic Center at Jackson was used. 4. Socio-Economic Status - This factor was determined by the ratings on a modification of the Warner Classifi- cation System.5 5..3gg§ - Sample restricted to white males. 6. Emotional Stability - Based on the supervising officer's evaluation. Eliminated were those with severe emotional problems requiring intensive treatment. Caseloads were carefully screened and a list was compiled of all those who met the foregoing criteria. In the initial screening of parolees, it was found that only a small number of men met the criteria for selection. For this reason the maximum age limit, which had been previ- ously set at 25, was raised to BC. With the increase in the maximum age, a total of 38 parolees was available. From the probation caseload, a total of 46 men was availa- ble. Beginning May 2, 106}, letters were sent to these 84 men.6 The letter informed them that they had been selected for an eXperimental counselling program and that they were 5Scale furnished by Dr. Fred C. Pesetskey, Psycho- logical Clinic, State Prison of Southern Michigan, Jackson, Michigan. 6See appendix E. (18) to appear at the YWCA for testing and instructions. Of these 84 a number were lost for various reasons such as; working nights, errors in selection, under de- tention, etc. Ultimately, 24 probationers and 24 parolees were selected for the study. Determination of Groups. From the final sample of 24 pro- bationers and 24 parolees, assignment to experimental groups was made by use of a table of random numbers. One-half of the probationers was assigned randomly and in equal numbers to two experimental groups. Parolees were assigned in the like manner. The remaining probationers and parolees were placed in the appropriate control groups. Table 1 illustrates the breakdown of these groups.7 TABLE 1 Experimental And Control Group Designations (N = 48) Experimental Control Probation (N = 6) (N = 6) PEel PEG-l Parole (N : 6) (N : 6) PEe2 PEC2 For purposes of statistical analysis, the total number of subjects finally treated was 36; 18 probationers and 18 parolees. For statistical reasons to be discussed later, data pertaining to the actual groups was treated in terms 7It was originally intended that the control and experimental groups were to be numerically equal, comprised of six persons each with a total of 48 persons, but because of various factors this was not possible. (19) of four categories of equal numbers of subjects. These were two experimental categories, one probationer group and one parolee group, each with 9 subjects and two corre- sponding control groups. Leader Selection. In programs of this nature, where volun- teers are involved, the problem of recruitment and selection is substantial because such people are not always readily available. One source most often used is the service clubs and organizations. One inherent weakness in this source is that sometimes it is difficult to find men who can or will commit themselves to the program. Too often it is seen as a "project" type activity and, therefore, difficult to get men to envolve themselves in the program in a way that is essential to an effective leadership role. There is a great deal of difference in the role they assume in a therapeutic group and that which they fill in selling news- papers, putting on fund drives, etc. As important as it may be, the requirements of group counselling are not the same. Another source of leaders has been the local Volunteer Bureau. The eXperience here has been somewhat better. This perhaps is due to the nature of the individual who becomes involved through this kind of organization. There seems to be more personal commitment on the part of these persons. Their involvement, as a volunteer, many times has been a task-oriented experience. Having been exposed to the counselling role, they reSpond very well to a challenging experience. (20) J4 1i :9 "D o. .a 1.3 :rou; u ’3‘ fl‘fi " Motto-" .1“ 1 part is Church groups have, to some extent, involved them- selves in the programs. Two drawbacks are apparent. First, the leaders may tend to identify with a religious group and may bring too much "religious" orientation. For many of the clients, their exposure to religion has been negative. Second, there is a tendency for these group leaders to develop rigidity in some areas of discussion which seriously impairs their effectiveness. This is brought about by moralizing or judgmental behavior on the part of the leader. A related problem is that of time. To do the job with any degree of success requires that the individual commit himself for some period of time on a regular weekly basis. The general practice in Michigan Corrections system has been 12 weeks. This requires the leader to be in attendance for one hour per week with his group and about a half hour to an hour in an evaluation session after the group meeting. Regular attendance to the group meetings is essential to maintaining continuity, interest, and to the maintenance of rapport. Attendance at the evaluation sessions under the supervision of staff personnel is essential to the maintenance of direction in the groups and the development of skills. In the counselling groups of this study, three of the four leaders had previous eXperience.as group leaders in one capacity or another. The fourth leader was recruited through the Junior Chamber of Commerce and had no eXperi- ence in this type of program. (21) 1 a a x J - O . ~h- A \u .. 1" I1 VM— QV Ni r ‘]U Y«§\ P‘J 7‘ . . . 41 E 4. 2. i c w n n. .. a C a .1 . .. .l ._ . . 3 h... T .3 9 a . .1 a a. 9 N .. .. e. e T L. o T a ”u I a u E I ..... v. I .1 Z .- _ .i .5. O O .Q a O U. 3. v“ a _ .1 .. . v" S ,. .. .2. S 3 ...J mu. it a- . r. C .l it U. V: U. 0 o O . . All leaders were given a short orientation session prior to the beginning of their groups. At that time, the nature of the program was explained. Each leader was furnished with a handbook to use as a guide and for instructional purposes.8 Instrumepts - To test the major hypotheses, two instruments were used and were administered to both the leaders and all subjects. They were administered twice; at the outset of the experiment and at the termination of the program. The first instrument is the 100 statement developed by Butler and Haigl .9 This test consists of 10:) cards with statements printed on them.10 These statements are de- signed so as to enable the individual to describe himself. Using the Q-sort technicue, the individual is asked to place the cards into nine piles with the extreme left being the the pile containing the cards that are most like him. The pile at the extreme right contains the cards that are mo :3 t 1.1.nliI-ze him. Two sorts are used; the first sort is to indicate his .concept of himself and the second sort is to indicate his ideal self-concept. The two sorts are then correlated and the resulting correlation coefficient indicates the dis- parity between the self and ideal self-percepts or "level n Jmichigan Department of Corrections, Handoook for Volunteer Group Counselors, Lansing, 1963. 90p. cit., Carl R. Rogers and Rosalind F. Dymond, loSee appendix B. (22) of ad of adjustment". If, as indicated in previous reports, participation in a counselling program produces a shift in self-perception and the shift is positive, it should be indicated in a reduction in disparity between self and ideal percepts in a test-retest situation. The second instrument is the Delinquency Proneness Scale reported by Gough and Peterson, §§_gl.11 This instru- ment was used to test the second major hypothesis and con- sists of 64 statements to which the individual answers 'true or false. The resulting score will give an index of the subject's pred15position or proneness to delinquency. The test is administered before and after the counselling series. It is hypothized that any reduction in anti-social tendencies should be reflected in a reduction in the prone- ness scores on the Gough-Peterson scale. In the final retest phase of the project, an addition- al test was administered. This was the adaptation of the Dymond-Cottrell method of measuring empathic ability reported by Miyamoto and Dornbusch.12 This was a collateral study and, while not directly connected to the hypotheses and exploratory questions of the present study, it is of 11Harrison G. Gough and Donald R. Peterson, "The Identification and Measurement of PrediSpositional Factors in Crime and Delinquency”, gournal of Consulting Psycholo- gy. XVI (June. 1952). 12Frank S. Miyamoto and Sanford M. Dornbusch, A Test of Interactionist Hypotheses of Self Conception, American gournal of Sociology, LXI (March, 1956). ' (23). MC. wt. +u a. . fly I‘.. .1 Dv ~. oale.' U a 1 Ms W... I E nu): m... ¢ .U “A 3.. 1:13:14 interest. The results are included in Appendix A. Procedures All subjects, from both the eXperimental and control groups, assembled initially at the YWCA in Lansing on May 14, 1963. The program was outlined. It was explained that the group meetings would be held weekly and that part of them would participate in the weekly meetings and part of them would not. Instruments were then administered. The first was the Gough-Peterson delinquency proneness scale.13 The men were given the following instructions: "This test is called, 'How I feel About Things'. Please read each statement carefully and if it is true, so far as you are concerned, mark it true. If it is false, then mark it false. There are no right or wrong answers; but rather, we want you to tell us how you feel about things. There is no time limit. If you have a question, raise your hand. Take your time and be as accurate as possible. If there is some doubt in your mind, then answer the question the way you feel most of the time." The second test administered was the 100 statements of Butler and Haigh.l4 The men were given the following instructions: "You have been given 100 cards with statements printed on them. You also have nine markers with pile numbers on them. Arrange them so they run from number zero on the extreme left to number eight on the extreme right.15 You are 13See Appendix C. 1Z‘LSee Appendix B- 15The layout of the markers was drawn on the blackboard in the following manner for illustration: Like Pile Number Unlike Me 0 1 2 3 4 5 6 7 8 Me Number of Cards 4 7 12 17 2O l7 l2 7 4 (24) to use the cards to describe yourself to me. If I were to ask you to tell me what you are like, you would probably use statements such as appear on the cards. You can do this with the cards in the following manner: In pile number zero, place the four cards that are most like you; and in pile number one, place the seven cards that are next most like you and so on to pile number four which is the undecided pile. At the right end in pile number eight, place the four cards that are most unlike you. In pile number seven, place the seven cards that are next most unlike you and so on up to pile number four which is the undecided pile. You may find it helpful to go through the deck first and sort the cards into three piles; "like me", ”unlike me", and undecided. Be as accurate as possible. If you have a question, raise your hand." When the men had completed the first sort, the distri- butions were recorded and each man was given a new deck of cards. They were then given essentially the same in- structions, except that now they were told to sort the cards into piles ranging from "what I would like to be like” to "what I would not like to be like". When this was com- pleted, the distribution was again recorded and the subject was then told whether he was in an eXperimental or control group. If the subject were in an eXperimental group, he was told to report to his particular group at the time of the first meeting and that he would be given further instructions. If he were in a control group, he was told that he would not have to do anything further until the tests were given again to all subjects at the end of the twelve weeks of counselling for the experimental groups. Group Meetingg. The meetings for the experimental groups were begun on Tuesday, Kay 14, 1963 following the completion of the instruments just described. The parole groups met at (25) '.AQ! “M v.- V-.— a, r: : .. . Min 3 a a .. a. _ ._ . 3 me + u C .C a. .. .. .o E is O m. l 3 .u a n. .. o a T. a C. a-.. .C n O I the Lansing YWCA and the probation groups met at the YMCA located two blocks away. These meetings were about one hour per week for a twelve-week period. Each group had one group leader assigned who remained with the same group throughout the project. The last group held its final meeting on October 15, 1963. Attendance. It was decided that attendance would be manda- tory for the parolees. In the case of probationers, manda- tory requirements were difficult to enforce due to the administrative relationship of the Department of Cor- rections and the courts.16 In the case of the probationer, the supervising officers strongly recommended their attendance. This kind of dual attendance requirements is- reflected in the attendance rate. Even with mandatory )\ attendance, parolees attended only '5 sessions out of a ( possible 108, or an attendance rate of 60 per cent. For probationers, the attendance was 55 sessions out of a possi- ble 108, or an attendance rate of 51 per cent. It should be noted that in the present project, the probationers were drawn from the caseload of the Ingham County Probation Department and that the caseloads are so excessive that supervision is very limited. A case may be cited that illustrates the problems 16Parolees are under direct jurisdiction and super- vision of the Department of Corrections. Probationers on the other hand are under the jurisdiction of the sentencing court and, therefore, remain essentially under the control of the local county probation office. (26) entailed in trying to enforce mandatory attendance.17 When initially selected for the project, a parolee refused outright to participate. Several contacts were made and after much discussion he decided to take the initial tests. When the random selection was made, his name was drawn for an emperimental group. Again he refused to participate. Repeated contacts failed to change his mind. Even threats of detention or the possibility of return to an institution made no impression. His only answer was that he would resist attendance all the way to the Parole Board18 if necessary. At the end of the project, he was told to report for re-testing and he refused. After several contacts were made in an effort to get him to report for re-testing, he was told to report or he would be placed in custody for the purpose of testing. He did report finally, but proceeded to answer the questions without reading them as he had indicated he would do. The tests, of course, were discarded as invalid. ‘ This is an extreme example, of course, but illustrates the problems that are encountered. For the most part, resistance is passive. The individual agrees to partici- pate but fails to show up for the meetings. When contacted, he shows up for one or two meetings and then misses one or two. With probationers,the problem is even greater. Hanna, :3. 28. l"319arole authority in Michigan- (27) CHAPTER III 7 JLESEHTATION OF DATA The hypotheses, rationale, design and procedures used in this study have been presented. This chapter will deal with the presentation and analysis of findings as the hypotheses and emploratory questions of this research are considered. Some initial comments on the eXperimental and control groups and the statistical procedures used are indicated.. The remainder of the chapter will be devoted to the hypotheses and exploratory questions. Experimental and Control Groups. At the beginning of the project, four experimental groups of six men each were to be selected with matching control groups. This was to produce a total of 48 men in the project. As indicated earlier, this total was subsequently reduced.1 When the project was completed, the sample consisted of twenty-seven men in the eXperimental group and twenty-four men in the control groups. Partly accounting for this reduction were several men who had been returned to ustody, left the area and,in a couple of cases,refused to participate.2 The other factor in this reduction was the statistical technique 1Supra, p. 19. 21h one instance the re-test had to be administered under duress and the results were so far out of line that (28) used in analysis. Described in detail in the following section, this technique required that all groups be of equal numbers. Meeting this requirement resulted in a pro- cedure by which, for statistical purposes, the two experi- mental probationer groups were collapsed into a category of nine persons and the parole groups were similarly treated. These were paired with equal numbers in the control groups. The selection of those to be included in analysis was made on the basis of a table of random numbers. Tables 2 and 3 present, for the 36 subjects finally used in the study, the background information in terms of which matching was attempted.3 Table 2 indicates that both the eXperimental (The) and control (PNc) groups for probation are very closely matched. The mean age varies by only 0.2, I.Q. scores by 1.7, A.G.R. scores by 0.3 and socio-economic status varies by only 0.5. None of these differences are statistically significant. Table 3 illustrates essentially the same pattern for parolees. Age varies by 0.2, I.Q. scores by 5.1, A.G.R. scores by 1.3 and socio-economic scores vary by 0.2. The they were discarded as invalid. In another case, the subject steadfastly refused to participate in the program after the initial tests were administered. No amount of threats, coercion, etc., would change his mind. At the completion of the program, he submitted to re-test only after he was faced with detention. He then proceeded to answer the questions and do the sorts randomly without reading the questions. The results were discarded as invalid. 3Within the categories "parole" and "probation", actual groups have been combined for purposes of analysis in this and in other tables. (29) N 3 gleam“ .mHQmHHmbm Po: mmhoom .O.H # N.N ¢.OH H.>m N.ON smog h.N H.OH ¢.mo ¢.ON qmmE s NH mm ON HH H O we NN OH n O *N NN OH n O mOH OH O H HH ¢OH OH m m NH mm NH w m NH mHH OH 5 N HH be OH N N OH HO mH w w m ow HN m N NH OHH HN m m OH MO ON d a m 6 mm s m ma mow ma m n O 0m ON m N m 3% EN N H HH HOH HN N N NH HOH HN H 03% mam mSPmpm .asoo .G.H mad pommnsm mSvam .Qsoo .m.H mad PomnnSm .soowIOHoom momma .soomIOHoom madam Hoapsoo prsmsHsmmsm mgfiosw sOHpmponm Mom sOHPSQHHpmHQ mHsmpHpO sOHpomHmm NmfiQB (30) msHPma spasm mwmam>4 n .m.w.4: \o ®.H N.m 0.00H O.MN saws O.N O.N O.MOH O.¢N smog N H.OH NHH ON HH N m.o mHH NN mH H m.O mOH wN OH N m.o mOH ON sH N ¢.HH OMH ON 5 N o.m mo HN MH H >.OH mOH mN m N m.OH mOH NN NH N H.b mHH ON m H N.o OHH MN HH N m.o om dN s N 0.0 NOH MN OH N m.~ MOH ON m m m.® ¢OH mN m N O.N dOH NN N N H.m so NN N N O.m NOH HN H N ¢.m #m on m r: can mam 353m. sum d .a .c . H 91.14 pom Sam mfimpm mama .4 .019 was. p8 and .soooIOHoom .soomuoHoom Hoapsoo HmpsmsHHmQNm mmdonw mHosmm pom QOHpanspmHm mHsmpHHO QOHPomHmm M Mam4e (31) only difference that is statistically significant is the variation in I.Q. score. This is due to the one subject with a score of 130. This individual was included in the study on the basis of the parole officer's recommendation. He stated that the recorded I.Q. was not a valid one and that his capacity was well below that. §tatistigal Procedure. The data presented in this chapter requires some clarification with respect to one statistical technique used in their analysis. The method in question was recommended by Thomas Danbury and Albert Talbot of the Communications Research Center at Michigan State University who acted as statistical advisors for the study. Initially, several procedures were considered. The decision finally was made to use an analysis of variance utilizing a fac- torial design on the grounds that it would be the most efficient way of analyzing the data. However, to use this procedure, it was necessary to adjust the groups so as to end up with equal numbers in both the experimental and control groups. To do this, a table of random numbers was used to eliminate the excess cases. Accordingly,the eXperimental groups, for the purposes of statistical analy- sis, were reduced to nine probationers and nine parolees with matching control groups. This made a total of thirty-six men in the study. The Specific reasons for using this technique were several. It was agreed that the more meaningful information derived from factorial analysis would more than effect the loss of numbers and would not drastically reduce the (32) significance of the results. The value of this method lies in its ability to ana- lyze not only the variations between groups, but, at the same time, an analysis can be made of variations within groups. The method consists of seperating the various observations into groups or parts with each part measuring variability attributable to some Specific source, e. g., to internal variation of the several populations, to vari- ations from one population to another, etc.4 In addition to the foregoing advantages, there is the additional value of a decreased level of significance in comparing multiple variables. For example, in the use of t tests in analysis of paired means, the possibility of rejecting an hypothesis at the .05 level of significance can be critical. In experiments with five means, it has been shown that the chance that some, one or more of ten independent 3 values will exceed t .40. If the hy- is .95 pothesis of all five means equal is rejected when some one of the ten 3 values exceeds t it will be rejected with .95’ a chance which may be many times the level of significance .05.5 For the purpose of statistical analysis, three categories were established. These three categories were 4Wilfred J. Dixson and Frank J. Massey Jr., lgtro- duction To Statistical Analysis, (New York: MoGraw—Hill Book Company, Inc., 1957frChapter 10- 502. cit., Wilfred J. Dixson, Jr., and Frank J. Massey, Jr., p. 139. (32) divided into two sub-groups. Table 4 illustrates the ar- rangement. TABLE Analysis Categories And Corresponding Sub-Groups Categories Sub-Groups __‘_ Al : Parole A : Violators --------------- A2 : Probation Bl : Counselling B : Experimental ------------ B2 : No Counselling C1 : Before C : Time -------------------- C2 : After The three categories, A, B, and C are the major vari- ables and the six sub-groups are the secondary variables within the major variables. The raw data was then sub- mitted to the computor for variance analysis using the fac- torial method.6 HYPOTHESES H. l - Participation of adult parolees and probationers in a_prggram of lay group counsellinggwill result in a reduction in disparities between the self-percepts an self-ideal percepts of counselees. The results of the data analysis from the Q-sorts is shown in Table 5. The F ratio for violator differences (A) 6For a detailed discussion of the design see: B.J. Winer, Statistical Principles in Experimental Design, (New York: McGraweHill Book Co., Inc., l§52L Chapts. 5, 6, and 7. (33) indicates no significant relationship. The same is true of differences between eXperimental and control groups (B). 2AM Self-Ideal Correlations (raw scores) Source Sun of Souares df Mean Souare F A .25632?00 1 .25632800 2.17529 N.S. B .12335556 1 .12835556 1.08927 N.S. AB .00259200 1 .00259200 .02200 N.S. R (within AB) 3.77076555 32 .11783642 C .15773472 1 .15773472 9.56078 .01 AC .03354050 1 .05354050 2.03299 N.S. BC .05723472 1 .05723472 3.46917 N.S. ABC .01566450 1 .01566450 .9494? N.S. CR (within AB) .52793955 32 .01649311 F ratio Significance 4.15 : .05 7050 = .01 Variable (C) represents the variations in the self-ideal correlations over time. With an F ratio of 9.56078, it is significant well beyond the .01 level. This indicates that the total sample (eXperimental and control groups) made a significant reduction in their self-ideal disparities. Variable (AB) or the interaction between (A) and (B) indi- cates no significance. Variable (AC) or the interaction between violators and time, while it showed no significance, did indicate a tenden- cy to move in the direction of significance. The variable (ABC), which tests the interaction of all three variables (34) (violators-counselling-time), resulted in no significant differences. The critical interaction for testing the first hy- pothesis is (BC). The F ratio of 3.46917 is just short of the required 4.15 mark to make it significant at the .05 level. It is, however, significant beyond the .10 level. To illustrate the probable reason for this, Table 6 shows the breakdown by sub-groups of this interaction. may. Self-Ideal Correlations By Violator Groups - Before & After A1 B2 B1 B2 B1 52 Cl .503 .434 .299 .312 +.O77 +.O23 Net Change+~223 ..051 In the parole group (Al), the net change for the experimental group (B1) was only +.O77 while the net change for the probation eXperimental group (Al-Bl) was +.223. While there was a positive net change for both the parole and probation group, the major change was in the probation group. This large change on the part of the probation group accounted for most of the F ratio for (BC) interaction. However, the change was not large enough to compensate for the much lower parole score. If the total sample had been larger and the change ratio remained the same, the F ratio for (BC) interaction would probably have reached the 4.15 mark. (35) cv- While the hypothesis did not receive unqualified support, the data does not warrant rejection. Even though the results were not Spectacular, the data definitely favors the counselling groups. Additional research is indicated for the purpose of confirming the foregoing conclusion. H. 2 - Earticipation of adult_parolees and probation- ers in a program of lay group counselling_will:result in a reduction in criminal_proneness. The results of the delinquency scale analysis are indicated in Table 7. T AEL E Z Delinquency Proneness (raw scores) Source Sum of Squares df Mean Square F A 544.50000002 l 544.50000002 5.13438 .01 B 304.22222223 l 304.22222223 4.5451 .05 AB .88883589 1 .33335889 .01323 N.S. R (within AB) 2141.38883395 32 66.93403 0 14.22222222 1 14.2222222 1.32500 N.S. AC 2.00000000 1 2.0000000 .1864? N.S. BC 20.05555556 1 20.05555556 1.86986 n.3, 130 .50000000 1 .50000000 .04662 N.S. 03 (within AB) 343.22222222 32 10.72569 F ratio Significance 705:) = .01 Two variables (A) and (B) show differences that are significant beyond the .01 and .05 levels reSpectively. To (36) better illustrate the differences, Table 8 shows the break- down of mean scores by violator groups. TMEES Delinquency Proneness Mean Scores By Violator Groups - Before & After A1 A 2 B1 B2 B1 B2 C2 33.1 30.4 28.3 24.9 -0.7 +1.7 Net Change +0.3 +2.1 The findings are not surprising in light of correction- al practices. Probationers as a group represent a low risk group. They are for the most part individuals who are first offenders or,at least,minor offenders. The courts have felt that the ends of justice and the individual can best be served by placing such persons on probation and allowing them to remain in the community. What the court is doing is, in effect, holding sentence in abeyance con- tingent upon good behavior for some specified period of time. If probationers complete this period of time with- out any difficulty, the case is discharged. Parolees, on the other hand, are,as a group,high risk cases. They represent, as a rule, men whom the courts have judged to be in the need of confinement. In the vast majority of cases, the individual has been a probationer at least once (37) or, perhaps, has served a jail term or two.7 If one were to accept the premise that delinquency proneness is a predictive measure of the individual's chances of becoming delinquent, then it would follow that parolees, by the nature of their status, would be more delinquent prone. The data illustrated in Table 3 shows the mean scores of the parole and probation groups broken down into eXperi- mental and control sub-groups. The F ratio for the variable (B) indicates a signifi- cant difference beyond the .05 level (Table 7). In looking at the data from Table 8, it can be seen that the vari— ations between experimental and control groups were high. This was particularly true in the case of probationers. In both groups the differences were greatest at the beginning of the program. There seems to be no apparent reason for this other than sample fluctuation. For variable (C) about all that can be said is that the eXperimental groups (parolees and probationers) held fairly steady while the control groups tended to lose ground slightly. In both cases, however, the change was not sufficient to be significant. Based on these data, the second hypothesis is to be rejected. H. 3 - Reduction in disparities between self:percepts and self-idealgpercepts of counselees will be correlated 7Based on statistics gathered by the Michigan Department of Corrections. Courtesy of Robert Glass, Statistician. (38) with the reduction in criminal proneness of counselees. Since the third hypothesis is tied to the second, then it is to be rejected along with the second. While there might be some tendencies suggested by the data to support the second hypothesis or some ideas as to the source of possible error, the data does not warrant any conclusions relative to the third hypothesis. Horeover, it is quite possible to hypothesize that under Specified conditions reduction in proneness may be independent of self-ideal relationships. Exploratogy Questions In addition to the three major hypotheses, this study had as a secondary purpose the examination of several eXploratory questions. Such questions serve to help illumi: nate the nature of the problem and to suggest possibilities L for future research. *3 O elati nshi _b_§gg§n self-ideal lisparitv —-— — _———. - I- ‘755, J—‘Y what is pee and thqmggggnt_and direction ofl_qhanee? Dissonance theory suggests that the greater the d ssonance the greater he likelihood of change.8 In this context, dissonance is thought of as self-ideal diSparity. The question then becomes one of determining the degree to which the data of this study seems to support this princi- ple. Referring to Table 6 (Self-ideal correlations by violator groups - before and after), the results tend to support the dissonance principle. The parole eXperimental 8Supra, pp. 15-16. (30) _I group had a mean score of .503 on the initial tests while the probation experimental group had a mean score of .209; a difference of 104 units. Subsequently, on the re-testing, the parole group had a mean score of .580 for a net change of +.077 and probationers had a mean score at re-test of .522 for a net change of +223 units. The control groups, while not so pronounced, nevertheless, tended to follow the same pattern. Applying the same principle to the individual group members, it would be eXpected that there would be a corre- lation between the initial test scores and amount of dis- parity reduction. Table 9 presents the data for this correlation. E3122“ Rank Distributions of Individual Self-Ideal Correlations and Reductions of Counselees Parole Group Probation Group Pre-test Net Pre-test Net Rank Scores Change Rank derank Scores Change Rank dl l .229 .260 8 -l 1 -.l46 .362 8 -7 2 .357 .153 7 -5 2 .020 .243 7 '5 3 .385 -.078 2 l 3 .062 .375 9 -6 4 .463 .115 5 -l 4 .067 .167 3 l 5 .502 .144 6 -l 5 .333 .188 4 l 6 .518 .287 O -3 6 .401 .195 5 1 7 .682 .044 3 4 7 .563 .140 2 5 8 .634 .050 4 4 S .643 .237 6 2 Q .708 -.270 l 8 9 .776 .038 l 8 qu rS:-.517 N29 rS=-.800 (significance >.07<.05) (significance 7.01) (40) To compute the correlations, pre-test scores for the experimental group were placed in rank order and the corresponding net changes listed in the next column and ranked, and the rank - correlation coefficients (rs) computed.9 For the parole group, the coefficient of -.517 was just short of the .05 level of significance (critical value .589). For the probation group, the coefficient of -.800 is well beyond the .01 level of significance (critical value .750). Both the individual and group analysis lend to support the expectation that the greater the self—ideal disparity, the greater the change. At least the results in this parti- cular instance are at a high level of confidence. It should be noted here that the generally accepted view is that,in many cases, dissonance must be produced or increased before positive change can take place. Because there are no inter- mediate tests, it would be difficult, if not impossible, to determine if there were negative changes prior to positive changes that would have affected the net change in the) scores. It would be interesting to more carefully plot changes in time with other eXperimental groups to determine if the dissonance principle is Operative here. Are there any signifipant differences in the self and ideal concepts of parolees and probationers as a_grqu? 90p. cit., Wilfred J. Dixon and Frank J. Massey, Jr., pp. 29 ’2950 (41) At the beginning of the study, it was suggested that, as a group, parolees would have a greater disparity in their self and ideal correlations than probationers. This was based on the assumption that parolees,by reason of their having had more difficulty in adjustment,would be so characterized. Referring to Table 6, re find that the results indi- cate that the opposite is true. At pre-test, the parole experimental group had a mean correlation of .503 while probation had a mean correlation of .239. The parole group had a mean correlation of .434 and the probation group had a mean correlation of .312. This suggests that parolees as a group have less of a self-ideal disparity. Using the procedures of Dymond, scores were computed for the ideal sorts of both the probation and parole at pre-test.lo This is done by scoring one point for every negative item on the Q-sort that the subject places on the "unlike me" side of his piles and one point for every positive item that he places on the "like me". The total score is then computed. The results are presented in Table 10. The results indicate no significant differences between the probation and parole groups. Means range from 45.66 (PNe) to 48.00 (PEc). Parolees as a group had a mean score of 47.06 and probation had a mean score of 46.17. When these scores are viewed in light of the mean correlations, they do not support the expectation that b 100D, cit., Carl R. Rogers and Rosalind F. Dymond, pp. 76-78 (42) rm-“ “Hams.-.“ .—K . . _ . . . _ m I ;. parolees have a greater degree of maladjustment (disparitr than do probationers. Rather, it would suggest that parolees' self-ideal elements are in closer approximation. This would Stggest that concern with therapy in SlCh cases would entail a concern not sinply with the matter of dis- parity, but also with the actual content of these self-elements. Putting it another way, it may be that parolees are better adjusted criminals. TfidLE lg De-Scale Distributions Barole Probation gage (3:0) {Sam—:9) i.-- Pm use) ___::';~__<:*:‘> 4Q 46 53 44 29 42 24 43 46 25 49 37 51 55 52 45 59 50 47 43 53 50 55 51 54 50 50 52 3 40 44 51 53 41 57 43 46.ll 45.00 Kean 45.66 46.66 Parole Mean : 47.06 Erobation Mean : 46.17 In this connection two points which may partially account for the parolee's self-ideal relationship should be mentioned. First, parolees, by reason of longer eXposure to the correction system, more adjustment difficulty, etc., have (43) ) I i “HAW—ik-_m l -. ‘4 » ' - r .1" . . . . l I. v 1. come to accept their lot and see little chance for change in their status in life. In view of this, they learn to adjust to the "inevitable". To some extent this might be viewed as "marginal adjustment” suggested by Merton and Sherif.ll While they percieve themselves as excluded from society, they nevertheless maintain society’s values and, n L at the same time, they remain members of an out-group. 4 Second, the parolee may be using what Butler and Haigh : describe as defensive sorting.12 That is, he may be so f '1 motivated by defense needs, that he pictures himself as 3 pretty much like his ideal self when in actuality, at a ’1 ‘si- (: deeper level, he does not resemble this ideal. This po bility would appear to carry some weight when one sees the parolee as an individual who has had extensive eXposure to the correctional process and has to learn to protect himself from the battery of testing to which he is subjected. Having accomplished this, he becomes extremely resistive to change. Whatever the case may be, certainly these points would be worth xploring further in the interest of better under- standing the parolee and the impact of the correctional process. It might well prove fruitful in shedding light on more effective treatment programs. 11R.K. Merton and A.S. Kitt, "Contributions to the Theory of Reference Group Behavior.", Readings in Social Psycholog‘, (New York: Henry Holt & 00., 1952). 1200. cit., Carl R. Rogers and Rosalind F. Dymond, pp. 58-59. (44) What is the relationship between prgbationers and parolees and the amount and direction of change? An initial expectation in this study was that parolees would have a greater degree of self-ideal disparity than would the probation group. However, as the data presented in the previous question indicates, the parole group had less diSparity than probationers. In Spite of this, there are some interesting observations of the change that took place in both groups. Table 11 shows the initial test and re-test scores and the net change. As indicated in the data probationers had a mean score at the beginning of .299 and a re-test mean of .522 for a mean net change of .223. The net changes range from .088 to .375. All change was in a posi- tive direction. Parolees on the other hand had a mean score at the beginning of .503 and a mean score at re-test of .580. As a total group (eXperimental and control), parole began with .468 and ended with .519 and probationers began with .306 and ended with .443. This represents net gains of .051 and .137 reapectively. Scores for parolees ranged from -.279 to .237. Two parolees had a net change in a negative direction. While the promise regarding difference in parolees and probationers did not hold up, the expectation regarding the amount and direction of change held true. Probationers had the greatest amount of disparity in their self and idea sorts and subsequently showed the most net change. It was initially suggested that parolees might be more resistive to change. The data suggests that this is (45) probably true. Th e amount of total net showed a net change Q-Correlations parole group not only had the smallest gain but even had two subjects who in a negative direction. TAELE 11 and Net Change Distributions of Counselees Parole Probation Net Net Pre lost Jhanse Pre Eost Change .682 .726 .044 .020 .263 .243 .229 .439 .260 .062 .437 .375 .502 .646 .144 .067 .234 .167 .684 .734 .050 .536 .685 .149 .463 .573 .115 .333 .521 .183 .385 .307 -.078 .775 .864 .088 .708 .429 -.279 .643 .580 .237 .518 .305 .287 -.146 .216 .362 .357 .510 .153 .401 .596 .195 .503 .580 +.O77 Mean .299 .522 +.223 Further study in this area is needed before any valid conclusions can be drawn. It would be useful to know if a longer period of counselling would make any significant differences. It might be that the two parolees with nega- tive changes would reverse their direction of change over a longer period of counselling. duce a larger net change. This would, in turn, pro- Such might be eXpected on the basis of the previous discussed premise that in some cases increased dissonance must be produced before a positive change can be achieved. (46) To what extent does tn group leader influence change? ('0 Some prior research has suggested that ultimate adjust- ment where "lay leaders" are involved is related to group leader orientation.13 The power and influence of the leader in small group settings has in the past been the focus of a number of studies. For example, Kaier and Solem measured the relative differences of task oriented groups with and 14 Both groups showed improvement in their without leaders. achievements after iscussions, but the groups with a leader showed the most improvement. Hare pointed out 'm—m.u_vq n.- n-v. A.‘AI-H essentially the same idea when he discussed leadership training.15 He indicated that an important consideration in leadership is consistency, with the group clearly under- standing the function of the group and of the leader. In this present study, leaders' orientations in the group setting was not a variable as such. However, certain relevant data was collected and are presented herein as suggestive of hypotheses to be tested rather than as demon- strated conclusions. Table 12 lists the net changes by counselling groups along with the scores of the leaders. The data sujsests only a trend indicating that the higher the self-ideal correlation of the leader, he greater the change in the counselee. The pattern of net chance for 130p.*citL, Stuart Adams and Joachim P. Seckel. 14A. Paul Hare, Handbook of Small Group Research, (Kew York: The Free iress of Glercoe, lfi625jp.47. 1 533,13, pp . '50“s-321. (47) those leaders whose results are considered valid is A>B>C with the pattern of mean net change for counselees in the respective groups being Al) 31>C1' r3 bets 1 Qt ./ ' A w n... Self-Ideal Correlations By Groups And Leaders Parole Group I (N = 3) Group II (N : 6) Net Net Subject Cl 02 Change Subject C1 02 Change Leader .322 367 .045 Leader .255 .422 .167 PEe 5 .682 .726 .044 PEe 10 .684 .734 .050 PEe 7 .229 .439 .26o PEe 11 .463 .578 .115 PEe 8 .502 .646 .144 Phe 12 .385 .307 -.O78 M .149 Pne 13 .703 .429 -.270 PEeiU+ .518 QKNS .287 PEe 15 .357 .510 .153 M .160 Probation Group I (N : 5) Group II (N : 4) Net Net Subject Cl 02 Change Subject Cl 02 Change Leader .500 .716 .216 Leader* .765 .700 -.065 PNe l .020 .263 .243 PNe 3 .067 .234 .167 PNe 2 .062 .437 .375 The 7 .776 .864 .088 Pre 4 .536 .685 .149 The 8 .643 .880 .237 PNe 5 .333 .521 .188 PNe 10 .401 .505 .195 PNe 9 -.146 .216 .362 M .172 M .263 *Scores for this leader not considered valid for reasons to be presented s ( 4; 3 U ) Another factor was the backgrounds of the leaders. One leader had no experience at all prior to the project and received a minimal amount of training before the pro- gram began.1'6 One leader had prior formal training (M.A. degree in counselling) and works as a school counselor. The other two, while they had no prior formal training, have had extensive experience in similar programs. If the leaders are ranked in order of their training and experi- ence with their corresponding group achievements, there is a rank-correlation coefficient that approaches significance at the .05 level. If the scores of the leader Whose tests are considered invalid are discarded, then the correlation is even higher. .But what of the impact of the counselling process on the leader? If the leader is in fact a member of the group and not a separate entity, then the process should make an impact on him. Referring again to Table 12, it is noted that all but one leader showed a net change in a positive direction on the Q-sort correlations. One leader had a net change in a negative direction. Of the three leaders who made a positive change, the one making the least amount of change (.045) was the man with the highest Q-sort corre- lation (.822). The remaining two men began with .255 and .500 correlations and made net gains of .167 and .216 respectively. The tendency is for the leader to shift his 16This is the same leader who showed an increase in diSparity on the Q-sort. (49) correlations on the same basis as the other members; that is, the greater the dissonance, the greater the change. If the leaders were paired by violator groups (parole and probation), the net change relationship would be perfect. The data presented would tend to confirm the premise that group counselling is a "two-way street". That is, that the leader develops new attitudes toward violators and the correctional process thereby seeing himself as a better, more productive, individual. This is accomplished throug‘ % a sharing of ideas, values, and concerns with others.17 7'} The fourth group leader, who showed an increase in self-ideal disparity, presents some rather interesting observations of group impact. Leader number four was the youngest (age 21) and lacked any previous eXperience or training. He was recruited from a service organization through a former group leader. Initially he displayed a great deal of enthusiasm for the project. He was given the same instructional material as was given to the other leaders. This was followed by two conferences to discuss the techniques and problems of group counselling. At the time of the initial test, he did not show up at the appointed time, but rather at a testing session the following evening. It was noted at the time that he seemed a little aloof and disdainful of the testing instruments, particularly of taking them with a group of parolees and probationers. Upon completion of the first Q-sort (self-sort), the group was told to re-sort the cards to indicate what they \ 17§ppra. 92- 7-9- (50) would like to be like (ideal-sort). At this point, the leader said that this would not be necessary, that he was happy as h was, and that his ideal would be what he was at the moment. He was ouite firm in his feeling so the matter was deferred for discussion with test specialis s. Two or D three days later he was iniormed that it would be necessary - ~ : if . V... «VI? to do the ideal-sort. He insisted on doing it at home so the test material was given to him and he returned it later 5 completef. # Darin? the project, this leader emperienced some diffi— g; culty with attendance in his group. It was quite apparent that he felt the group members owed it to him to be there since he was giving of his tihe and energy freely. On one occasion, he fail d to show up at the meeting with his (I) group. When he was contacted, he said he had forgotten the meeting and had fallen asleep. At the close of the program, the re-test schedule was set up. The leader was contacted on two occasions, but failed to make the appointments. He agreed to take the tests at home and the test materials were taken to him. 3 After a lapse of about three weehs durinO which two contacts were made, he stated that he had lost the test materials after completing the tests. Another set of test materials was taken to him and after several days, the tests were returned. Upon examining the score sheets, it was discovered that he had failed to do the ideal-sort. When he was con- tacted, he stated that his ideal had not changed since the last time and that the scores from that test could be used. (51) 4.1,. Since so much time had elapsed between the completion of the program and the re-test, it was decided that the *alidi- ty of the tests were questionable and the matter closed. The events and facts concerning this one particular leader are covered in detail because they illus rate some of the difficulties that arise in the selection, training f? and involvement of group leaders; particularly, the "lay a leader”. In this case, the problems encountered seem to indicate.that the leader perceived the program initially s kind of a new and different eXperience, perhaps as a 9: chance to make some kind of spectacular change in social deviants. After exposure, however, h found it very threatening. His apprehe sion became more acute when he was faced with tests that might reveal himself as a person. This is a defensive kind of reaction much in the same manner as previously described as "defensive sorting".l Like any therapeutic relationship, group counselling re— ouires the individual to give of himself and for the leader who must be honest with himself, the process can be hard on the leader and requires some degree of stability and flexi- bility. (52) CHAPTER IV SCIEIARY AND '30:?"LL7SIOI‘33 This study has been concerned with three major hypothe— r} i, L (I: ses and several exploratory questions. Analysis of the data 3 lead to the tentative support of the initial hypothesis and g the rejection of the remaining two. In addition, the E3 evidence has led to the identification of a number of proba- ble factors operating in the experimental situation. Many of these constitutegfigci for'further research. The findinss and their relevance to the hypotheses and eXploratory questions are summarized below. ‘M 7,- major flyootheses H, H. l - Participation 0 adult parolees and probationers in a prosram of lay group counselling will result in a reduction in disparity between the self-percents and ideal m self-percepts of counselee . The data, although not supporting the hypothesis at the level desired, nevertheless, did indicate strongly that the counselling emperience was effective in reducing self-ideal disparity. Had the total sample been larger, the significance level likely would have reached the desired level. On this basis the first hypothesis was tentatively accepted. H. 2 - Participation of aiult probationers and parolees in a program of lay group counsellingpwill result in a (5'5) reduction in criminal proneness. The data failed to support the hypothesis that group counselling would reduce the counselee's delinquency prone- ness. he data suggested that the counselees eXperienced no change to speak of,while control groups seemed to become a little more delinquent prone, although the differences I? Were not significant. However, analysis did point to some important factors relative to differences between groups. he most obvious of which was that parolees as a group were significantly more delinquent prone than probationers. E; H. 3 - Recuction in disparities between selffipercepts and ideal self-percepts of counselees will be correlated with the reductibn in criminalpproneness of counselees. This hypothesis is dependent upon the preceding hy- pothesis. Because the latter was rejected, it was necessary to reject this one. Exploratorypguestions What is the relationship between self-ideal disparity and the amount and direction of change? Data from both group and individual analysis tended to support the premise that the greater the disparity, the greater the change. Probationers as a group had lower corre- lations on the Q—sort than did parolees. At the re-test point, the probationers had made substantially greater increase. Are there any sisnificant differences in the self and ideal concepts of parolees and probationers as a_:roup? At the beginning, it was thought that parolees would (54) have lower self-ideal correlations than would probationers. The data did not support this premise; rather, it indicated that parolees tended to have higher correlations than did probationers. Several factors were suggested to account for this. One such factor was that of marginal adjustment (i.e., the parolee), while alienated from society, nevertheless may accept many of its values. At the same time, he may see himself as pretty close to achieving his ideal-self. Another factor was that of "defensive sorting". Having been exposed to the correctional system and its diagnostic procedures many times, he has soon learned to manipulate tests. In this way he never reveals his tggp self. Finally, parolees,by virtue of having been exposed to the correctional process more often,are not faced with the same emotional impact of their deviant behavior. What is the relationship between probationers and the amount and direction of change? The impact of he counselling process varies, as the data has indicated, with the violator groups. Parolees,as a group,had less diSparity to begin with and subsequently had less reduction. Probationers, on the other hand, began with much more disparity and subsecuently showed sub- stantially more reduction. The results suggested that parolees as a group are less aporehensive and are more es stive to change. The data, however, does not irdicate H why this is the case. A \J 1 U] V The evidence suggests that per counselling might have affected the direction of change for those who had an increase in disparity. This is implied in the theory that many times disparity must be produced before positive change can take place. To what extent does thg group Lgader influence change? 3 1,3 - Iqu Due to the limited numbers, no data pertaining to group leaders could be submitted to statistical testing. However, 3 there were several trends in the data. Indications were 1 that change in the leader took place in much tne same way £5 that it did the counselees. Like the counselees, the leaders tended to reduce disparity in proportion to the amount of initial disparity. There were some in ications that effectiveness of the leader tended to increase with training and esperience and that for the ineXperienced leader, the counselling eXperi- ence can be quite threatening. Implications of This Study. At the outset, it was pointed out that the purpose of this project was to evaluate lay group counselling with respect to its effectiveness as a treatment tool. As indicated, the data appears to support its effectiveness. vTherefore, it appears appropriate, giving proper cognizance to the limitations of this study, to consider expanding and strengthening the present use of group counselling techniques. Notwithstanding a number of obscure factors, it appears that the group counselling process aids in the establishment of a more integrated image for co“.selees. A U1 O\ V The data also implies that adequate training and supervision of the lay leader is essential to good results. The s udy thus relates to the matter of the recruitment, selection and training of group leaders. Perhaps of even greater consequence is that this research provides the base for many avenues of further F} inquiry. One measure of any research lies in its ability .. to point to outstanding questions which call for empirical research. While the relevancy of lay group counselling seems supported in this study, many factors operative in £5 the esperimental setting still remain all too obscure. For this reason, posing questions for future research are an important part of his study. finggestionsqu; :urther Research. One major research question relates to the duration effect on the scores. That is, would a longer period of counselling produce even more change? Even more important would be the question of whether a longer series would have significantly affected the parole group which appears to have resisted change. Increasing the number of sessions substantially and the administering of tests at various intervals might shed some light in this area. Further research seems indicated into the differences between parolees and probationers. It may well be that research into the area covered in this study with insti- tutional inmates would be useful in seeing more clearly what, if any, differences do exist and what the nature of these differences are. It may be that more specialized (57) techniques need to be developed for what appears to be a highly resistive individual. Such information could have value in determining the kinds of institutional programs hat need to be developed or altered. A great deal of research is needed in better under- standing the role of the group leader. At best, the j '. ' '0; selection process at the present time is hit and miss, trial and error. More study is needed in the area of group leader orientation to establish the relationship between orientation and leader effectiveness. As pointed out g3 earlier, previous research has indicated that the two (orientation and effectiveness) are related. Since society in general and the Department of Corrections in particular has a primary interest in the counselees' ultimate adjustment, then some form of follow-up study is indicated. It would be well to know what the ultimate adjustment patterns are for the individuals. This would give some indication as to the permanency of the improvement. It may also furnish a predictive device for determining prognosis.if ultimate adjustments are corre- lated with present scores. Such inquiry might demonstrate the precise relationship between self-ideal image and law-viOlating behavior. ‘The differences in the delinquency proneness between parolees and probationers indicate the need for additional research in this area. It would be helpful to know if the delinquency proneness declines over a longer period of time. Another factor to be eXplored would be the relationship (58) between age and delinquency. It might be useful to know if delinquency diminishes with age. Yet, another problem is the determination of the relationship between counselling and the different dimensions measured by the De-scale. It would, perhaps, be useful to determine staff atti- tudes towards this counselling technique as well as other pa approaches to the treatment of social deviancy. Much of what is accomplished is dependent upon the attitude of the supervisors toward the program. Apathy, indifference, or passive resistance seriously impairs the effectiveness of hi this or any other program. While this study indicates that group counselling makes a contribution to the correctional process and as a treatment tool, more needs to be known about the process of group counselling before its full potentials are realized. (5?) SELECTED BIBLIOGRAPHY Adams, Sturart, Ph.D. and Seckel, Joachian, M.A. ”Assissment of Guided Group Interaction in a Boy's Training School: A Preliminary Report.“ California Youth Authority. Pre- sented to 39th Annaul.Meeting of the American Ortho- psychiatric Association, Los Angeles, March 21-24, 1962. Aichornu, August, Waygard Youth. New York: Viking Press, 1936. Bassin, Alexander, Ph.D., Berlin Louis, Smith, Alexander. P3 "Group Therapy With.Adu1t Probationers," Federal Pro- -u bation, September 1960. Brehm, Jack W. and Cohen, A. R. Explogations in Cognitive Qissgnance. New York: Wiley and Sons Inc., 19 2. Cohen, Albert. Delinquent Boys: The Culture of the Gang. b3 Glencoe, 111.: The Free Press, 1955. Current Projects In The Prevention, Control, And Treatment Of Crime and Delinquency, National Research and Infor- mation Center 0 C ins and Delin uenc : National Council On Crime and Delin uenc , SprIng 1962 and Winter 1962-1963. New York: 1962 and 19 3- Dinitz, Simon, Reckless Walter C., and Kay, Barbra. "A SeltGradient Among Potential Delinquents",8go§;nal 9f Criminal Law and Criminology, XLIX (195 Dixson, Wilfred J. and Massey, Frank J.ePr. lptgoductiog To Statis ca Ana sis. New York: MoGraw-Hill Book Company Inc., 19%7. Driedger, Leo. "The Self Concept". A paper submitted to Sogiology 863 Michigan State University. November 13, 19 . Driver, Helen. Counsellina and Learning Through Small Group Discussion. son 8.: onona ca ons, Fenton, Norman Ph.D. An Introduction to Group Counselling in State Correctional Service. New York: The American Corrections Association, 1958. Festinger, Leon. A Thgozz 9f Cognitive Dissonance. Evanston 111.: Row, Peterson and Company, 1957. Ghastin, James F. "An Experiment in Group Counselling For Youthful Probationers", An.unpublished Report, Septem- ber 10, 1959. . (60) Gough, Harrison G. and Peterson, Donald R. "The Identifi- cation and Measurement of PrediSpositional Factors in Crime and Delinquency", Journal of Consultin Ps - ghology, IKV No. 3 (June 19525. Michigan Department of Corrections, Handbook for Volunteer Group Counselors, Lansing, Michigan, 1965. Hare, A. Paul, Bargatta, E. F., and Bales, R. F. Small Groups. New York: A. A. Knopf, 1955. Homans, George C. Social Behavior; %ts Elementary Forms. New York: Harcour , Brace an ompany, 9 . James, William. Painciples pf Psychology. Vol. 1, New York: Henry Holt and ompany, 190 . Jones, Maxwell, M.D. The The a wutic Communit . New York: Basic Books, Inc., I953. Konopka, Giesla, Group Work in thy Institutihn. Whiteside, Inc. New York, 195 . Kvaraceus, William C. et a1. Delinquent BehaviorI Culture and the Individual. National Education Assodation of the United States, 1959. McCorkle, Lloyd W., Elias Albert, and Bixby F. Lovell. The Highfields Story. Henry Holt and Company, 1958. McCorkle, Lloyd W. "Group Therapy in the Treatment of Offend- ers," ngeral Probation, December 1952. McGee, Richard A. Five Years of G on Counsellin . Report to the Governor's CounciI, CaIIgornEA, June 29, 1959. Meade, George H. Mind, Self and Society. Chicago: University of Chicago: University of Chicago Press, 1954. Merton, R. K. and Kitt A.S. “Contributions to the Theory of Reference Group Behavior," Continuities in Sodal Research, ed. R. K. Merton and P. F. Lazarsfeld, Glencoe, 111.: The Free Press, 1950. " " Miyamoto, Frank S. and Dornbusch, SanfordM."A Test of the Interactionist Hypotheses of Self-Conception, " Ameyican Jgurnal of Sociology, LXI (1956). Olmsted, Donald H. Social Groups, Roles, and Leadership: An Introduction to the Concepts. (East Lansing: Board of Trustees, Michigan State University). Reckless, Walter C., Dinitz, Simon, Murray, Ellen, "The 'Good" Boy In a High Delinquency Area". Journal of Criminal Law and Qgiminology, ILVIII (1957). (61) Rogers, Carl R. and Dymond, Rosalind F. Psychotherapy and Personality change. (Chicago: The University of Chicago Press, 1954). Shibutani T. Soc e nd Personalit ° nte actionist Approacp to social sychglpgy. New Jersey: Prentice- Hall, Inc., 19 l . Weeks H. Ashley, Youthful Offenders at Highfields. (Ann Arbor: Unitersity of M chigan ress, 95 . Wflson, Gertrude and Ryland, Gladys. Social Grou Work Ppactice. Hougton-Mifflin, Boston, 1959}. Winer, B. J. Statisical Princi les in Ex erimental D Si n. New York: Mcgraw-Hill Book Company Inc., 19625. Wylie, Ruth C. The Salf Concepp. Lincoln, Nebraska; The Univesity of Nebraska Press, 1961. (62) E ‘3 . ‘- APPENDIX A As a collateral to present study an additional instru- ment was administered at the completion of the project to the counselling groups. While it does not have a direct bearing on the study, it does have some relevance. The purpose was to determine what effect, if any, the reSponses of others had on the self ratings of the individual parti- ma cipants. In the study of Miyamoto and Dornbusch, they attempted to test the interactionist hypothesis of self-conception. They approached the problem from the vieWpoint of how others &5 in the process of interaction affect the self conception of the individual. In doing this, three factors were con- sidered:1 (l) The perceived reSponses of others (2) The actual responses of others (3) The effect of what Meade referred to as the ”generalized other". In their study, they administered rating sheets of four items to ten different groups. The forty results were tab- ulated and applied to the following five hypotheses: l. The mean of the actual responses of others to the subject will be higher for those persons with a high self- rating than for those with a low self-rating. 2. The mean of the perceived reSponses of others will be higher for those persons with a high self-rating than lOp.Cit., S. Frank Miyamoto and Sanford M. Dornbusch. (64) for those with a low self-rating. 3. Self-conception tends to be closer to the mean perceived reSponses of others to the subject than to the mean actual reSponse of others. 4. Those persons sho have high self-ratings on a characteristic will have a higher mean perception of the generalized other than will those with low self-rating. 5. Accordingly, self-conception should corresPond more closely with the generalized other than with the mean of the perceived responses of others.2‘ They found the results supported all five hypotheses and lent empirical support to the symbolic interactionist view of self-conception. In the Miyamoto and Dornbusch study, the groups were not randomly selected from any known universe. Statistical tests for significance were not applied and the data examined for gross differences and trends. If all groups supported the hypothesis, it was regarded as having perfect support. If only half were found to support the hypothesis then it was considered to have only chance success. In the present study, groups were randomly selected. Again no statistical tests of significance were applied and only gross difference was examined. For our purposes, we are interested primarily in whether there seems to be some support of this theory of self-conception in our groups. Four groups were involved and the rating sheets consisted of six items, thus it yielded twenty-four results. If more than half the groups supported the hypothesis 20p. Cit., 8. Frank Miyamoto and Sanford M. Dornbusch fr§a\ -Yi-JN '4 '7. ! “SMILE was considered a trend in a positive direction. The results follow: TABLE A Hypothesis Charactpristic Supported Not Supported Tied self-confidence 00.00000 2 1 T LikeableneSS 00000000000 1 3 0 Intelligence ........... 2 1 1 Physical Appearance .... 3 l 0 Leadership ............. O 4 O Chance of Violating .... _;6_ _Q __J_._ Total ..... ll 10 3 Hypothesis 1. - While there is not overwhelming support for this hypothesis, there appears to be some trend in that direction. With the elimination of two items the support would be very strong. At least it would appear that the actual reSponse of others might be a factor in determining one's self-image. It might be Speculated that the results in terms of this item might be tied to amount of attendance. That is to say that if the group had been together longer that responses to the items may have been better. TABLE B ~ Hypothesis Chapacteristic Supporped Not Supported Tied Self-confidence 0.000... O O Likeableness ........... 3 O 1 Intelligence coco-cocoa. 3 l 0 Physical Appearance .... 3 0 1 Leadership ............. 4 O O Chance of Violating .... 4 O 0 Total ..... 2T .1' -§ Hypothesis 2. - The support for the second hypothesis appears to be very strong. This would suggest that in a group situation the way in which we perceive others as reSponding to us is ve ry important to the way in which we see ourselves. We reSpond to ourselves as we think others reSpond to us. This is important in the function of the (66) w group. One of its aims is to establish better communi- cation with others and to share ideas. While poor per- ception may work to the benefit of the individual, it may also work to his disadvantage. TABLE C Hypothesis Chapactepistic Su orted Not Supported Tied Self-confidence ........ 4 O O Likeableness ........... 4 0 0 Intelligence ........... 3 O 1 Physical Appearance .... 4 O 0 Leadership ............. 4 O O Chance of Violation .... _4_ _Q_ _Q_ Total ..... 23 O l Hypothesis 3. - Twenty-three out of twenty-four comparisons supported the hypothesis. This is consistent with social psychological theory that perceived behavior of others toward us of greater importance to the self-image of the individual than the actual response. TABLE D Hypothesis Characteristic Supported Not Supported Tied Self-confidence ........ l 2 1 Likeableness 0.000000... 2 2 0 Intelligence ........... 4 O 0 Physical Appearance .... 3 O 1 LeaderShip cacaooooooooo 3 O l Chance of Violation .... .4; _g .4; Total ..... 15 7 2 Hypothesis 4. - Like hypothesis number one there seems to be only a trend in a positive direction, but no overwhelming support. TABLE E Hypothesis Chapacteristic Supported Not Supported Tied Salf-Confldence 0.00000. 2 l l Likeableness 00000000900 1 2 1 Intelligence ........... 3 1 0 Physical Appearance .... l 3 0 Leadership ............. 2 l l Chance of Violation .... _Q, _g 2 Total 9 o "5 A O\ N - v I . I U I O I I O O o O C I O O l. - c o C Hypothesis 5. - Data for this hypothesis has dropped below the supported level. Because the distribution is scattered the results do not yield any useful information on the affect of the "generalized other" on the individual. In total the data from the rating sheets strongly support hypotheses 2 and 3 and tend to support hypotheses l and 4. As in the Miyamoto and Dornbusch study, the results are viewed with caution and used to strengthen suggested relationships within the group. That is to say that reference groups and the self-concept are tied to- gether. And, Specifically, that the counselling groups can be seen as functioning in this way. Therefore, for our purposes, the first three hypotheses are critical and the results are in the necessary direction. (63) APPENDIX B 1. 2. 3. 4. 5. 6. 7. 9. 10. 11. 12. 27. 28. feel uncomfortable while talking with someone. put on a false front. am a competitive person. make strong demands on myself. often kick myself for the things I do. often feel humiliated. doubt my sexual powers. am much like the opposite sex. have a warm emotional relationship with others. am an aloof reserved person. an reSponsible for my troubles. am a reSponsible person. have a feeling of hopelessness. live largely by other people's values and standards. can accept most social values and standards. have few values and standards of my own. HHHHHHHHHHHHHHHHH have a hard time controlling my sexual desires. It's difficult to control my aggression. Self control is no problem to me. I am often down in the dumps. I am really self-centered. I usually like people. I express my emotions freely. Usually in a mob of people I feel a little bit alone. I want to give up trying to cope with the world. I can live comfortably with the people around me. My hardest battles are with myself. I tend to be on my guard with people who are somewhat more friendly than I eXpected. ' (7o) 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. I am optimistic. I am just sort of stubborn. I am critical of people. I usually feel driven. I am liked by most people who know me. I have an underlying feeling that I'm not contribut- ing enough to life. I am sexually attractive. I feel helpless. I can usually make up my mind and stick to it. my decisions are not my own. often feel guilty. am a hostile person. am contented. am disorganized. feel apathetic. am poised. just have to drive myself to get things done. often feel resentful. HHHHHHHHH am impulsive. It's important for me to know how I seem to others. I don't trust my emotions. It is pretty tough to be me. I am a rational person. I have the feeling I'm just Int facing things. I am tolerant. I try not to think about my problems. I have an attractive personality (71) 56. 57. 58. 59. 60. 61 62. 63. 64. 65. 66. 67. 68. 69. 70. 71. 72. 73. 74. 75. 76. 77. 78. 79. 80. 81. am shy. need somebody else to push me through on things. feel inferior. am no one. Nothing:really seems to be me. am afraid of what other people think about me. am ambitious. deSpise myself. shrink from facing a crisis or difficulty. just don't reSpect myself. am7dominant person. take a positive attitude toward myself. am assertive. am afraid of a full-fledged disagreement with a per- on. I I I I I I I I have initiative. I I I I I I s I can't seem to make up my mind one way or another. I am confused. I am satisfied with myself. I am a failure. I am likeable. My personality is attractive to the opposite sex. I am afraid of sex. I have a horror of failing in anything I want to accomplish. I feel relaxed and nothing really bothers me. I am a hard worker. I feel emotionally mature. I am naturally nervous. (72) 82. 83. 84. 85. 86. 87. 88. 89. 90. 91. 92. 93. 94. 95. 96. 97. 98. 99. 100. I really am disturbed. All you have to do is just insist with me and I give in. I feel insecure within myself. i have to protect myself with excuses, with ratibnaliz- ng. I am a submissive person. am intelligent. feel superior. feel hopeless. am self-reliant. often feel aggressive. am inhibited. am different from others. am unreliable. understand myself. am a good mixer. feel adequate. am worthless. dislike my own sexuality. HHHHHHHHHHHHHH am not accomplishing. (73) APPENDIX C '1. 2. 10. 11. 12. 13. 14. 15. 16. 17. 18. I get nervous when I have to ask someone for a job. T-F i sometimes feel that I made the wrong choice in my ’ ’ occupation. ~ T-F I would never play cards (poker) with a stranger. T-F I would have been more successful if people had given me a fair chance. T-F I think Lincoln was greater than Washington. T-F Life usually hands me a pretty raw deal. T-F A person is better off if he doesn't trust anyone. T-F My family has objected to the kind of work I do, or plan to do. T-F Sometimes I used to feel that I would like to leave home. T-F If the pay was right, I would like to travel with a circus or carnival. T-F I would do almost anything on a dare. T-F As a youngster in school I used to give the teachers lots of trouble. T-F My parents were too strict with me when I was a child. T-F Even when I have gotten into trouble I was usually trying to do the right thing. T-F my parents never really understood me. T-F My home life as a child was less peaceful than those of most other people. T-F I think I am stricter about right and wrong than most people T-F I often feel that I am not getting anywhere in life. T-F (75) 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31 32. 33. 34. 35. 36. Even the idea of giving a talk in public makes me afraid. T-F It is very important to me to have enough friends and social life. T-F I never worry about my looks. T-F I hardly ever got excited ox'thrilled. T-F I have very strong likes and dislikes. T-F My parents have often disapproved of my friends. T-F My home life was always happy. T-F I often act on the spur of the moment without stopping to think. ’ T-F I seem to do things that I regret more often than other people do. T-F I would rather go without something than ask for a favor. T-F I have more than my share of things to worry about. T-F I go out of my way to meet trouble rather than try to escape it. T-F When I meet a stranger I often think that he is better than I am. T-F It is pretty easy for people to win arguments with me. T-F BeforeI do something I try to consider how my friends will react to it. T-F I have never been in trouble with the law. T-F In school I was sometimes sent to the principal for cutting up. T-F I keep out of trouble at all costs. T-F (76) 37. 38. 39- 40. 41. 42. 43. 42+. 45. 46. 470 48. 49. 50. 51. 52. 53. 54. I often think about how I look and what impression I am making upon others. I find it easy to "drop" or "break with" a friend. I spend a good deal of time planning and thinking about my career. I enjoy work as much as play. It is hard for me to act natural when I am with new people. When something goes wrong I usually blame myself rather than the other fellow. I have often gone against my parents' wishes. I have never done any heavy drinking. I have been in trouble one or more times because of my sex behavior. When I work at something I like to read and study about it. My table manners are not quite as good at home as when I am out in company. Most of the time I feel happy. I know who is re3ponsible for most of my troubles. I get pretty discouraged with the law when a smart lawyer gets a criminal free. I have used alcohol excessively. When I was going to school I played hooky quite often. PeOple often talk about me behind my back. I often feel as though I have done something wrong or wicked. (77) T-F T-F T-F T-F T-F T-F T-F T-F T-F T-F T-F T-F T-F T-F 55- 56. 57. 58. 59. 60. 61. 62. 63. 64. I don‘t think I'm quite as happy as others seem to be. I used to steal sometimes When I was a younster. I am somewhat afraid of the dark. I never cared much for school. I sometimes wanted run away from home. The members of my family were always very close to each other. With things going as they are, it's pretty hard to keep up hOpe of amounting to something. My parents have generally let me make my own decisions. I was often punished unfairly as a child. My home life was always very pleasant. (78) T-F T-F T-F T-F T-F T-F T-F T-F APPENDIX D INSTRUCTI NS At the meetings during the last several weeks you have been a member of a small discussion group. During this time you have gotten to know the men in this particular group and they you. We would like to ask your opinions based on this eXperience. Please be frank. All materials will be treated as strictly confidential and will be hand- led only by the Michigan Department of Corrections and the Youth Division. Thank you. Name Date (80) Instructions: In Comparsion with other members in your particular group, HOW WOULD YOU RATE ? (Make an "X" along each scale to indicate your answer) e f-Confidence O O _Q__ 0 ! O O 0 Below About Above Average Average Average ikableness - A A 4 L _Q 1. Below IAbout ' Ifiove Average Average Average Intelligance 0 49 O O. I O O 0 Be ow About Above Average Average Average Physical Appearance : k .L_ O x 3* L O 0 Below About Above Average Average Average As A Leader _ l1 Q 0 l A— O O BeIow About Above Average Average Average Cha ces of Becomin A Proba io Parole Violator . . . l - - A Below IIbout Above Average Avera e Average (81 U. ***The IoIIowing pages ask youto imagine how you think others would rate you*** Instructions: In comparison with other members in your particular group, HOW DO YOU THINK THAT WOULD RATE YOU? (Make an "X" along each scale to indicate your answer) Self-Confidence .1 Q Q Q 1 Q .1. L 0 Be ow About Abbve Average Average Average Intelligence __l L 1 I. 1 Q 4 L. Q Q k fiw j 0 Below About Above Average Average Average Likableness Q l 9 Q _Q Q k _9_ Q 0 4 __l_ Below About Above Average Average Average Physical Appearance w; 0 O I t O __O_ Q 0 Q 0 Below About Abbve Average Average Average As A Leade #- O O O o t L_ _1_ e g L _l_n Below. About Above Average Average Average Chances of Become A Probatioszarole Violator M ‘ ‘ ”“9“ " ‘ ‘ I 'W' 8 Averfiég) Average Instructions: Aside from your paticular group and associates, HOW DO YOU THINK PEOPLE IN GENERAL WOULD RATE YOU? (Make an "X" along each scale to indicate your answer) Self-Confidence I: Q __1 L Q ! 0 _Q_ _Q_ Q 0 C 0 Below About Above Average Average Average Likableness O 0 e o g I 9 0 e o e e BeIow ‘About' Above3_ Average Average Average Intelli e ce L Q j .L g _1 l. I L O 4 Below IAbout Above Average Average Average Physical Appearance Q Q Q t l. .- A L Be ow About Above Average Average Average A As A Leader L Q Li. 0 _Q I 9 9 O 0 Below About Above Average Average Average Chances of Becoming A ProbationZParole Violator O O Q I A A 4 A Below About Above 4 Average Average Average (83) '- ‘ A O" . .0 '1 wr- APPENDIX E State of Michigan DEPARTMENT or CORRECTIONS May 2, 1963 The Department of Corrections is about to begin an exper- imental counseling program. Participants will be selected at random from a pre-selected group. You are one of those who are in the total group. It will be necessary for you to appear at the YWCA auditorium on at 7:30 PM for preliminary testing and instructions. If, for some reason, you are unable to be there at that time, you are instructed to contact your officer immediately. James Ghastin Youth Division (85) no ROOM USE Oi‘vSLY