@3339 Maw-Lena { ‘ mm ‘H 6% if?! é // W" M £88 ABSTRACT THE EFFECTIVENESS OF TWO TYPES OF GROUP COUNSELING UPON THE SELF-CONCEPT AND OBSERVED CLASSROOM BEHAVIOR OF LOW-MOTIVATED MALE HIGH SCHOOL JUNIORS by Dolores Storey This research was designed to test the relative ef— fectiveness of two types of group counseling upon the self-concept and observed classroom behavior of low- motivated male high school juniors. The design of the study was a post-test—only with control group model. The treatments were: (1) group counseling only, and (2) group plus individual counseling. All group sessions were con- ducted by a conseling team made up of one female and one male counselor. The amount of counselor contact was held constant. The two basic hypotheses of the study were: 1. The self-concept scores of students who received both group and individual counseling in conjunction will be more positive at the completion of treatment than the self-concept scores of students who had either group counseling or no group counseling. 2. The teachers' ratings of behavior scores of students who received combined group and individual counseling will be more positive at the completion of treatment ==_ VT"! _—-‘ .-»w .. Dolores Storey than the behavior rating scores of students who had either group counseling alone or no group counseling. The Minnesota Counseling Inventory was used as a criterion measure of self—concept and an abbreviated eight item form of the Haggerty-Olson-Wickman Behavior Rating Scale was used as a criterion of observed classroom behavior. Originally 96 students were identified as low- motivated on the Michigan M—Scales. Of the 96, 7“ students accepted the invitations extended to them to participate in the study. Stratification categories were determined accord— ing to the free hours that these 74 students had in common. From each group with free hours in common, seven students were randomly selected. Six groups of seven individuals each made up an experimental group. The six groups were randomly assigned to either group or group plus individual counseling treatments. Groups were also randomly assigned within each treatment to the counseling teams. Those stu— dents who were unassigned were designated as the control group and received no treatment. In order for subjects to be included in the final analysis they must have attended 80 per cent of the sessions or 16 for the group counseling treatment and eight sessions for the group—individual treatment. Eighteen students in each of the two treatment groups or 36 out of the original A2, met the minimum requirements for post-treatment analysis. Twenty-eight of the original 32 subjects were in the control Dolores Storey group for post-treatment analysis. Thus the proportionate attrition rate was evenly distributed between the counseled and non-counseled groups. The analysis of variance technique was used to test the two null hypotheses of the study. The level of signi- ficance for rejecting the null hypotheses was set at five per cent. The major findings which emerged from the study were: 1. With exposure to treatment time held constant, group counseling was effective with or without the addition of individual counseling for all variables tested. 2. Group plus individual counseling exceeded group counsel- ing alone in generating more positive perception of the students' ability to cope with reality. 3. Group plus individual counseling exceeded group counsel- ing alone in producing teacher estimates of more gon— forming behavior in the classroom. 4. No differences were found among group counseling plus individual counseling, group counseling alone, and the control group on changes in coping with family relation- ships (FR sub-scale), and in self-perception of conformity_to social standards (C sub-scale). .‘m ‘u‘ -- _ THE EFFECTIVENESS OF TWO TYPES OF GROUP COUNSELING UPON THE SELF-CONCEPT AND OBSERVED CLASSROOM BEHAVIOR OF LOW-MOTIVATED MALE HIGH SCHOOL JUNIORS By Rafi» Dolores Storey A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY College of Education 1967 C;4é*53§ 2/:477'; P ACKNOWLEDGMENTS My sincere appreciation is expressed to the many peOple who have made this study possible. i To Dr. William Farquhar, my major advisor, for his valuable assistance, understanding, and personal support throughout the years of my graduate program. To Dr. John Suehr for his counsel in, about, and during group counseling sessions. To Dr; Norman Abeles for helping me formulate and conceptualize toward a theory of group counseling. To Dr. James Costar for his support on my committee. To Mr. Joseph Mezzano, Miss Charlene Eisenlohr, and Mr. Arthur Howson for serving as perceptive and dynamic counselors in this study. To Mr. William Hanichen, Principal of Mona Shores High School, for his cooperation and interest in this study. To the students and staff of Mona Shores High School for their cooperation and understanding. To my husband Dick for his abiding love, patience, and fortitude during the years of our marriage. To my somewhat bewildered offSpring Deborah, Jeffrey, Denise,_and Ric for their constant love and encouragement. ii 11,- .I-v-z‘ “—5.. TABLE OF CONTENTS Page ACKNOWLEDGMENTS . . . . . . . . . . . . . ii LIST OF TABLES . . . . . . . . . . . . . v Chapter I. THE PROBLEM 1 Statement of the Problem. 3 The Hypotheses 9 Definition of Terms . 9 Unique Aspects of the Study. . . . ll Organization of the Study . . . . . . ll II. REVIEW OF LITERATURE . . . . . . . . . 13 The Effect of Group Counseling in Pro- ducing Behavior and/or Attitude Change . IA The Process of Group Counseling: Quasi-experiments . . . . . . . . 25 Summary . . . . . . . . . . . . 28 III. DESIGN AND METHODOLOGY . . . . . . . . 30 Design. . . . . . . . . . . . 30 The PopulatiOn . . . . . . . . . . 31 Random Assignment . . . . . . . . . 32 Instrumentation. . . . . . . . . . 33 Sample. . . . . . . . . . . . . 37 The Counseling . . . . . . . . . . Al Counseling Setting. . . . . . . . . 43 The Counselors . . . . . . . . . . M3 The Null Hypotheses . . . . . . . . 43 Statistical Treatment. . . . . . . . A5 Summary . . . . . . . . . . . . M5 IV. ANALYSIS OF EXPERIMENTAL RESULTS . . . . . A7 Analysis of Variance of the Sub-Scales of the Minnesota Counseling Inventory . . 47 Analysis of Variance of the Haggerty— Olson- Wickman Scores . . . . . 56 iii Chapter Companion Research Conclusion Summary V. SUMMARY, CONCLUSIONS, AND DISCUSSION. Summary Conclusions Discussion BIBLIOGRAPHY APPENDICES. iv Page 58 60 60 62 61+ 71+ 79 Table 3.1 LIST OF TABLES Summary of the basic design of the experiment Students included in the post-treatment analysis Mean scores for the Otis Test of Mental Ability and GPA means for each group. Analysis of variance of the Otis Test of Mental Ability raw scores of the randomly assigned students Analysis of variance of fall term grade point averages of the randomly assigned students. Analysis of variance of the Family Relation- ship Scale (FR) for the group-individual counseling GI , group counseling (G), and control groups (C) . . . . . Analysis of variance of the Emotional Security Scale (ES) for the group- individual counseling (GI), group coun- seling (G), and control groups (C) Analysis of variance of the Social Rela— tionships Scale (SR) for the group- individual counseling (GI), group coun- seling (G), and control groups (C) Analysis of variance of the Reality Scale R) for the group—individual counseling (GI), group counseling (G), and control groups (C) . . . . . . Analysis of variance of the Conformity Scale C) for the group-individual counseling GI), group counseling (G), and control groups (C) . . . . Analysis of variance of the Mood Scale M) for the group-individual counseling GI), group counseling (G), and control groups (C) . . . V Page 31 38 MO 40 Al 49 SO 51 52 53 5A Table Page 4.7 Analysis of variance of the Leadership Scale (L) for the group—individual counseling (GI), group counseling (G), and control groups (C) . . . . . . . 55 4.8 Analysis of variance of the Haggerty- Olson-Wickman Behavior Rating scores of the group-individual counseling (GI), group counseling (G), and control groups (C) . . . . . . . . . . . 57 vi CHAPTER I THE PROBLEM Group counseling has been called the Pandora's Box of therapy.1 Within it, one can find more notions about its methods and values than almost any other therapeutic tech- nique. A major need is to find means to consolidate the) relationship between facts and theories. Among the diffi- culties encountered iS the fact that a simplified paradigm around which may be built a reliable and meaningful theory of group counseling does not exist at present. Furthermore, existing methodological means are not adequate to enable researchers to quantify the qualitative data of groups. Finally, it is almost impossible to manipulate the group Situation for the purpose of replicating research findings. In Spite of these difficulties, counselors expect through painstaking observations, careful collecting of data, formulation of creative hypotheses, methodical rea— soning, and continued observation and experimentation, to eventually arrive at a tenable basis for separating fact from fantasy in group therapy. So although critics believe that group counseling cannot be justified until significant 1R. J. Corsini, Methods of Group Psychotherapy (New York: McGraw—Hill, 1957). 1 results are obtained from research, proponents hold that if counselors waited for Significant results for everything that is done in the name of counseling, there would be a long wait indeed.2 Most research projects involving group counseling have not established positive results. As Ohlsen3 states, researchers did not obtain significant results because they failed to (a) define the treatment process, (b) define apprOpriate goals for the individuals treated, (c) define criteria, (d) select apprOpriate evaluation measures and, above all (e) allow sufficient time for the experimental treatment. Counselors are well aware of the complex problems that face the modern adolescent, but solutions to these problems are less obvious. The counselor sees the adoles- cents' expressions of hostility, aggression, asocialness, and forced compliance in his behavior. And because this behavior pattern usually inhibits the potential development of the student and disrupts the ordinary functioning of the school routine, the counselor is faced with the problem of how to aid the student within the framework of the school setting. 2Buford Stefflre, Theories of Counseling (New York: Wiley & Sons, 1965). 3Merle Ohlsen, ”Adapting Principles of Group Dynamics for Group Counseling,” The School Counselor, XIII (1966), 159—161. . .ov Statement of the Problem The purpose of this study is to assess the effects of group counseling with low-motivated students upon (1) self— -concept and (2) change in observed classroom behavior. This study seeks to determine if low-motivated students Care more effectively reached with group counseling alone or with group counseling in conjunction with individual counseling. The theory underlying the study attempts to explain how behavior change can be effected through the group process. Because adolescent behavior problems arise in a social Situation and often involve authority figures, it is assumed that these problems may find solution in a group Situation in which authority figures are present and active.LL Hence the use of a counseling team, made up of one male and one female counselor, effects a quasi-family atmOSphere for the group counseling sessions.5 In so doing, a social climate is set up which becomes a reality-testing ground, if we assume that the counselors represent parental figures who are perceived as both inhibiting and facili- tating growth; facilitating in the sense that as in a family, competition will inevitably be evoked among group members. Also group members are ever mindful, in a high 4E. E. Mintz, ”Special Values of Co-therapists in Group Psychotherapy,” International Journal of Group Psy-, chotherapy, XIII (1963), 127-132. 5J. Adler and J. R. Berman, ”Multiple Leadership in Group Treatment of Delinquent Adolescents,” International Journal of Group Psychotherapy, X (I960), 213-225. n“ .7.- school setting, that although these quasi-parental figures may be warm and accepting, they nonetheless remain as symbols of potential authority. In addition to providing a quasi-family structure, the use of two counselors allows each counselor comple- mentary roles and ample opportunity to observe and analyze client-counselor interaction.6 And by interacting with the group in dynamic, direct involvement the counselors can bring a variety of change tools into force. Some of these change tools are seen as:7 l. bringing about peer and adult pressure and support through the use of warmth, acceptance, coercion, and approval—disapproval. 2. fostering of a threat—free, secure climate without fear of humiliation or reprisal 3. permitting client dependence without group or counselor domination A. providing training in Open and honest communi- cation by permitting and accepting all expressions of thoughts 5. giving each group member a warm and meaningful relationship with adults 6G. KonOpka, "Group Work and Therapy," A Decade of GrOu Work, ed. C. E. Hendy (New York: Association Press, 1938),'pp. 39—44. 7Wm. Farquhar and Norman Stewart, "Counseling the Low- mOtivated Male: A Working Paper," mimeographed paper, MlChigan State University, April, 1966. 10. ll. 12. forcing insights by reducing and/or blocking defenses to help clients understand how and why they have previously maintained them helping members learn skill in developing relationships through direct observation of the counselors' interactions with them and with each other giving instant feedback to the client about the kind of person he projects . . . (filling him in) on his uniqueness manipulating experiences to provide clients with some rewards and success as well as a new awareness of how he has handled repeated failures in the past teaching the clients flexibility in their be- havior by responding to and rewarding behavior which is acceptable and specific to a given situation helping clients accurately label and understand their own emotions and the emotions of others showing clients how to react selectively to different situations rather than in a generalized pattern of behavior It is assumed that through the use of these change tools the counselors can actively guide, direct, and manip- ulate the group members toward changed behavior. Many Change tools are necessary because although clients may ____.‘,.. arc-~‘J. .2“ 2; ‘H ”wish” to change at one level of cognition, change repre- sents fear of the unknown and is usually frightening and anxiety-provoking. Fear and anxiety may operate initially in the group situation to inhibit growth and open up com- petition for individual support and nurturance from one or both of.the counselors.8 To modify these needs and in— F crease client self-awareness, the counselors theorized that perhaps growth would be facilitated by giving the clients individual counseling as well as group counseling. L In individual counseling the client might feel less threatened and learn to safely relinquish his defenses. HOpefully he would then transfer this new learning to the group experience. In the conceptual approach to change, the continuity of interaction in the total group is a series of ever shifting and alternating emotional balances. The group moves through a succession of emotional phases in which the atmosphere may be described as one involving hostility, wittflrawal, irrelevance, pairing for security, dependency, or combinations of these. These phases and/or combinations Characterize the group Situation and work to support and ifurther the goals of the group. Whatever change occurs in A the individual group members results from the direct impact ‘ Of these phases upon him. Also, impact of change is \— 8J. Mann, ”Some Theoretic Concepts of the Group Process," International Journal of Group Psychotherapy, V (1955), 235—242. reinforced through the use of a dynamic counseling approach:‘ an affective approach in which the adolescent can find re— assurance, acceptance, understanding, and self—awareness. As a result of the group or group-individual experi- ences and interaction clients' behavior change should be reflected in: (a) Open self-acceptance, (b) opening lines of communication in family relationships, and (c) social relationships, (d) developing a greater degree of emotional stability, i.e. ability to differentiate accurately among stimuli, and (e) achieving more acceptance by teacher standards.9 Three implicit assumptions underlie this theory of change: 1. Because attitudes and behavior patterns take many years to develop, it is unrealistic to suppose that behavior can be modified in a short period of time; i.e. most previous coun- seling has been of five to ten weeks duration. This theory assumes a minimum of 16 to 20 weeks of intensive, dynamic counseling in order to induce change. 2. Although non-focused, non—evaluative counseling should motivate the client toward change, research does not present positive findings. x , 9Dorothy Stock and Herbert Thelen, "Emotional Dynam- lCS and Group Culture,” Group Therapy anerroup Function, eds., Rosenbaum and Berger (New York: Basic Books, 1963), pp. 83-86. Hostility, anger, fear, failure, and frus— tration have been too long a real part of the low-motivated client's existence and he cannot patiently endure the wait for insight to arrive. It seems that non-directive counseling can become one more kind of rejection and/or humiliation. The low—motivated client is pre- cisely that: low—motivated. He requires a dynamic relationship of direct involvement with peers and significant adults. 3. The etiology of low-motivation can best ex- plained by a multiplicity of factors. Too often research has made the erroneous assump- tion that underachievers are somehow homo- geneous. Because they are in fact quite the Opposite, counselors should employ a variety of tools to induce change. It is not enough to make the underachiever aware of his poten— tial, he may already be well aware of it and is using the knowledge effectively in his defense system against adults. In summary then, the theory underlying this study affisumes that low-motivated students can be induced toward behavior change through the approach of dynamic group coun- Seling conducted by a counseling team. And, it predicts that students who receive group counseling in conjunction With individual counseling will develop a more positive self-concept, achieve improved family and social relation- ships, a greater degree of emotional stability, and more acceptance from their teachers than those students who eXperience only group counseling. The Hypotheses Within this study two basic research hypotheses are investigated: I. The behavior of students who have experienced both group and individual counseling will be more congruent with teacher expectations than the behavior of students who have experienced only group counseling which in turn will be more congruent with teacher expectations than the behavior of students who did not partici— pate in any form of counseling. II. The self-concept scores of students who have ex- perienced both group and individual counseling will be more positive than those who experienced only group counseling which in turn will be more positive than the scores of students who did not participate in any form of counseling. Definition of Terms For purposes of clarification, the following terms are defined: Vic“. lO Low-motivated student: any male, eleventh grade student who scored on the lower half of his class on an objective measure of academic moti- vation, the M-Scales. Individual counseling: a learning—oriented process conducted in a simple one—to-one environ- ment in which a professionally trained counselor attempts to assist the individual student to better understand himself. Group counseling: a learning-oriented process in which professionally trained counselors meet with a small group of students, usually seven or eight. Together they explore their own feelings in an attempt to better understand themselves and each other. The ultimate goals of both group and individual counseling are identical. Group—individual counseling: an experience in which the students receive both group and indi- vidual counseling, but they receive the coun- seling in alternate weeks. The group meets as a group every other weekailternating with individual counseling. Counseling team: two professionally trained counselors, one male and one female, with ex- perience in both individual and group counseling. -.v.'- ~' . .- are": ‘3':- . I"; a. . amt-morn . .= ., nigh-7A 11 6. Behavioral maladjustment: any displayed be- havior consistently thought of by teachers as unacceptable over a period of at least six months. 7. Dynamic counseling approach: counseling in which the counselor takes an active role in interactions and reinforces expressed feelings about self, attitudes toward teachers, parents, and other authority figures, hostile-angry feelings, and tentatively stated purposes and/or goals. Counselors by active participation encourage free expression of experiences and feelings and only block clearly irrelevant or defense—producing communications. Unique ASpects of the Study This study is unique because counseling teams con— ducted all of the counseling. Each team is made up of a male and a female counselor. The uniqueness of this approach at the secondary level of education is verified IVY published research. Organization of the Study The general plan of the study is to present in the following chapter a review of research which is related to the problem of aiding underachievers through group proce- dures. In Chapter III the design of the study will be 12 described with reference to sampling procedure, method of treatment, the null hypotheses, and the type of analysis. The results of the analysis are reported in Chapter IV. Chapter V will include the summary, conclusions, discussion, and recommendations. CHAPTER II REVIEW OF THE LITERATURE The task of reviewing the literature on group methods with adolescents is both complex and confusing. It is complex because the boundaries of the research problems are often too broad, it is confusing because the research is often vaguely described. Although in recent years coun- selor educators have become proponents of the group process, there is still a dearth of research which clearly defines or demonstrates conclusively the types of situations and/or conditions for which group counseling is appropriate. Most secondary school counselors are still uncertain as to which guidance functions may be best handled by the group process. For the purpose of this review only empirical research studies which directly relate to either junior high or senior high students have been considered. Much of the literature is descriptive and, therefore, not included in this review, but is presented in Appendix A with a separate bibliography. The remainder of the literature falls into two categories: l. Studies investigating the effect of group coun- seling in producing behavior and/or attitude change 13 14 2. Quasi-eXperiements investigating the process of group counseling The Effect of Group Counseling in Producing Behavior and/or Attitude Change One of the first investigations of the effects of group counseling in producing behavior change was reported by Driver.l She attempted to test the usefulness of small discussion groups in helping students achieve self- understanding, understanding of others, and interpersonal skills. Eight groups of senior students with ten in each group mixed heterogeneously by sex were formed. Prior to treatment the students had all indicated that they were dissatisfied with themselves or with their family situa- tions. Six group sessions were held over a period of three weeks. The process of counseling emphasized a permissive atmOSphere and in addition to group discussions, she used the technique of socio-drama based on Moreno's work. Follow-up interviews and questionnaires were admin- istered to group members three months after completion of the study. Returns indicated that the students had ac- quired new facts, gained in self—understanding, and had deveIOped skill in interpersonal relations. Driver con- cluded that the group process was an appropriate method for providing an atmOSphere for personality growth. lHelen Driver, "Multiple Counseling: A Small Group Discussion Method for Personal Growth," Personnel and Guidance Journal, xxx (1952), 173-176. fl..- 15 Caplan2 investigated group counseling in terms of its effectiveness in changing self-concept and improving aca- demic achievement. Thirty-four junior high boys with con- sistent records of conflicts with the school were divided into six groups, three eXperiemental and three control. The experimental groups met for fifty minute sessions for ten weeks. Groups were conducted in a permissive manner with only minimal behavioral restrictions. The counselor was non-directive in manner and provided an atmosphere in i which the boys could openly eXpress their feelings of hos- tility toward the school and authority figures. At the conclusion of the study changes in self- concept were measured by comparing pre and post-experimental administration of Q-sorts. Academic achievement was meas— ured by the traditional honor point system. Findings indicated that the changes in self-concept for the experi- mental groups were highly significant, but no change occurred in the control groups. Caplan concluded that the effect of group counseling upon academic achievement was equivocal. Caplan's work is open to criticism on several points: he did not descrfie his sampling technique; no assessment of group homogeneity was made pre-experimentally; and no 2S. W. Caplan, "The Effect of Group Counseling on Junior High Boys' Concept of Themselves in School,” Journal of Counseling Psychology, IV (1957), l2A-l28. 16 account was given of the differential effects of three counselors who participated in the study. McCarthy3 examined the effects of non-directive coun- seling on attitude change and academic improvement. Twenty- four boys from seventeen different high schools were iden- tified as under-achievers. These were divided non-randomly into two eXperimental and two control groups of equal number. Pre—experimentally the groups were checked for homogeneity on the variables of aptitude, achievement and personality. The groups met one hour per week for six weeks. At the end of the treatment period the groups were evaluated for attitude change by the use of Q—sorts and a sentence completion test. Academic improvement was evalu— ated by means of grade point averages. McCarthy's findings indicated that the group counseling procedure was ineffec- tive in producing either attitude change or grade improve— ment. However, these conclusions also point up the de- ficiencies of the study: the sample was too small (total subjects numbered twelve), the sampling method was poor; and the six week treatment period too short. 3M. v. McCarthy, "The Effectiveness of a Modified Counseling Procedure in Promoting Learning Among Bright Underachieving Adolescents," Research Project A.S.E.—6AOl (Washington, D. C.: Department of Health Education and Welfare). 7* “Smasmfisks‘éi' " ' -_.- — 17 One of the most sophisticated studies was reported by Baymur and Patterson.L'L They selected thirty—two high school junior for study who were matched on the basis of a large discrepancy between scores on an aptitude test and school grades. Due to scheduling problems the students were not randomly assigned. Four groups were formed in which one group received individual counseling on a weekly basis for twelve weeks, one group received group counseling weekly for nine sessions, the third group received one session motivational group counseling, and the fourth group served as a control. The evaluating criteria were improved personal adjustment as measured by a fifty item Q-sort from Hilden's Pool, increased scores on the Brown-Holtzman Survey of Study Habits and Attitudes, and increased grade point averages. With regard to grade point improvement, results were inconclusive, but when the two counseling groups were com- bined and compared with the one session motivational group and the control group, the improvement by the counseled groups was significantly greater. Two—step analysis yielded similar results on the personal adjustment scores. Therefore, the authors came to no conclusions about the efficaciousness of any of the treatments. The weaknesses of this study are relatively common: small sample size, AF. Baymur and C. H. Patterson, "Three Methods of Assisting Underachieving High School Students," Journal of Counseling Psychology, VII (1960), 83-90. 18 short treatment periods, and dependence upon the profes— sional skill of one counselor. In an attempt to assess the effectiveness of client- centered group counseling in behavior problems of junior high school boys, Arbuckle and Boy5 ramdonly assigned thirty-six students designated as boys with consistent behavior problems to three groups of twelve each. The students were homogeneously matched by age, grade, I.Q., scores on the Standford Achievement Test and teacher behav- ior rating scales. Some of the students in the study were not voluntary, but were selected to participate. Of the three groups, Group A was released from after school deten- tion for counseling, Group B was retained for detention, but received no counseling, and Group C was released from after school detention, but received no counseling. Subsequent to twelve weeks of counseling, progress was evaluated using a comparison technique between actual— self and ideal—self. Group A attained a significant corre- lation between actual-self and ideal-self and all members of this group were rated by their teachers as improved in behavior. The other two groups showed no improvement. None of the three groups had gained increased peer accep- tance. 5D. Arbuckle and Angelo Boy, "Client-centered Therapy in Counseling Students with Behavior Problems," Journal of Counseling Psychology, VIII (1961), 136-139. l9 Arbuckle's and Boy's conclusion that group counseling can be effective even if not sought voluntarily is of value to other counselors. But this study, like many others, neglected to describe or explain what actually occurred during the group counseling, except to say that it was "client-centered." Broede1_e_t__a_1.6 studied twenty-nine freshman high school under-achievers to determine if group counseling was appropriate in aiding these students toward increased per- sonal adjustment and improved grade averages. Two experi- mental and two control groups were designated through the use of random numbers. The experimental groups received sixteen sessions of group counseling while the control groups received no counseling. Upon completion of the ex- perimental counseling, the control groups were also coun- seled for sixteen sessions. Thus this paradigm controlled for both treatment and time effects. An evaluation of grade point increase was used to assess academic improvement as well as scores on the Cali- fornia Achievement Test. Personal Adjustment was measured by'a picture story test and the Mooney Problem Check List. Analyses of the data indicated that group counseling had not improved academic performances, but had significantly aided students in acceptance of self and others. Two 6J. W. Broedel, Merle Ohlsen, and F. Proff, "The Effect of Group Counseling on Adolescent Underachievers," (Mimeographed Paper, College of Education, University of Illinois, 1959). 2O follow-up studies (1960) and (1961) conducted four and eighteen months after the original study revealed no changes in outcome. Designs which include controls for both time and treatment effects are not common in the literature and this study represents a study in a positive direction. The weakness of the study lies in its failure to adequately describe the process of group counseling. 7 Caronis set up a design to determine if group coun- seling was effective in producing behavior change in adolescent boys confined in Boys' Vocational School in Lansing, Michigan. Twenty boys were randomly selected and grouped homogeneously by age, I.Q., clinical classification (by the school psychologist), type of offense, and socio- economic status. Prior to treatment the boys were tested using the Stanford Achievement Test, the Mooney Problem Check List, and the Machover Draw-a—Person Test. Each group contained ten boys; one group served experimentally and the other group as a control. The group met twice weekly for one hour over a fvur month period. Analyses were made I 1 Owing the treatment period, but the results did not E,{f ficantly support Caronis' hypothesis that group coxns ing would aid the boys in 7George Caronis, "Experimental Study in Evaluating the Adjustment of a Group of Disturbed Delinquents Ex- posed to Guided Group Counseling Within a Training School Setting," (unpublished Master's thesis, Michigan State University, East Lansing, 1963). 21 making a more satisfactory adjustment. He concluded that his N was too small, the treatment period too short, and the rather directive method of counseling inapprOpriate. Caronis suggested that if the study were replicated with a longer treatment period, the results might be substan- tially different. In addition Caronis did a fine job of describing each student in detail and also included random excerpts from the counseling sessions. .Frfir In an effort to determine if group counseling could produce changes among under-achievers, Lodato, Sokoloff, and Schwartz8 formed six groups from forty-nine students. Four of the groups were composed of elementary children and two were composed of junior high level students. Criteria for selection included I.Q. at seventy—five mini- mum, achievement one year or more below actual grade level and a history of poor behavior in school. The groups met from three to five times weekly for one year with one of the counselors. In addition to group counseling, the counselors used pantomime, role-playing, psychodrama, pup- petry, and individual counseling. Findings were: (1) positive changes in attitudes toward authority figures were observed in the majority of students as rated by teachers on a behavior and attitude rating scale; (2) increase in self-concept as measured by 8F. Lodato, M. Sokoloff, and S. Schwartz "Group Counseling as a Method of Modifying Attitudes,fi The School Counselor, X (1964), 27—31. ,;.u-. :A-l‘ -;:2»w~ ‘ _ ‘ .w’waz'ficg‘r sv- 22 figure-drawing projectives (Karen Machover served as con- sultant); (3) significantly improved attendance; and (4) increased tolerance and insight and understanding by class— room teachers of these students. Conclusions stated that group counseling is an effective means of modifying nega- tive attitudes of students. This study made no provision for control groups and failed to explain sampling methods. Furthermore, although ‘m’ '9.- there were many variables, no statistical analyses of these were presented. 9 In his doctoral study Catron investigated educational- vocational group counseling and its effects on perception of self and others. Each of thirteen small groups of normal high school students met for fourteen EV group coun- seling sessions with one of thirteen pairs of co-counselors. Each session was for one and a half hours and the total of fourteen sessions extended over five weeks. Changes in perception of self and others were as— sessed for forty-Six matched pairs of subjects (counseled versus noncounseled) by pre— and post—administrations of a modification of the Butler and Haigh (1954) S_I_O_ (Self, Ideal Person, Ordinary Person) Q-sort. Analyses revealed that perception of self changed significantly in the direction of ”good" adjustment, but 9David W. Catron, "Educational-Vocational Group Coun- seling: The Effects on Perception of Self and Others," Journal of Counseling and Psychology, XIII (1966), 202—207. 23 no significant change occurred in the perception of Ideal Person and Ordinary Persons. Congruence between Q-sorts for S, I, and 0 did not differentiate between counseled and non-counseled groups. Although this study is more carefully controlled than many, the following limitations are noted: (1) groups varied in Size from five to twelve members which could seriously affect types of interactions, particularly that of transference effects; (2) the use of all four grades 9 - 12 creates different kinds of problems not even con- sidered in the study; (3) pre- and post-testing produced consistent practice effects which could have been elimi- nated in another design; (4) replication of the study would be impossible because no explanation was given about the process of counseling; (5) the study does not provide for homogeneity of counseling among counselors; and (6) the length of the study is too short. Clementslo studied one hundred-eighty college-bound high school seniors to determine whether small group coun- saUng would affect their anxiety level. Sixty students comprised the experimental group and sixty students each in two control groups. Six small groups of ten members each met for one hour per week for six sessions in spring prior to graduation and again for six sessions in fall loBarton E. Clements, "Transitional Adolescents, Anxiety and Group Counseling," Personnel and Guidance Journal, XLV (1966), 67—71. 24 after enrollment in college. The counseling was structured around preparation for the college environment. Such topics as college admission, registration, curriculum, selection of a major field of study, social activities, extra-curricular activities, financial assistance, and vocational opportunities related to major fields of study were identified and discussed. Direction of the sessions was made the responsibility of the group members. Comparison of anxiety levels were made at the end of each series of sessions using Bills Index of Adjustment and Values and an unpublished Self-concept Inventory by Faust and Daane, 1964. Findings showed that subjects in experimental groups had significantly lower anxiety scores than control group subjects in both spring and fall eval— uations. The conclusion was made that the decrease in anxiety reported by the instruments in spring was sustained by minimal counseling in the fall. The results of this study might have been more mean- ingful if: (1) it had been determined that the subjects were anxious prior to the group experience; (2) the experi- mental and control groups had both responded to the instru— ments twice rather than the experimental twice and the controls only once; and (3) the effect of counselor varia- bility had been controlled. 25 The Process of Group Counseling: Quasi-experiments In the literature most of the studies focus on the differences in process between types of counseling, but a few have attempted to either define the group counseling process or tocbscribe Similarities within the group process. Defining and describing the process of group work is not novel in the areas of group psychotherapy or group dynamics, but is relatively recent in the literature as the process relates to the educational setting. The papers reviewed in this section have used young adolescents and are con- cerned with investigating the process within group coun- seling. However, rather than generalizable research these are quasi-experiments about which only limited conclusions can be made. In a process study Davisll used individual counseling and group counseling to determine relative effectiveness in producing behavior change. Thirty juniors in high school were selected for study from the group of juniors receiving the lowest citizenship grades. These were ' divided into three groups of ten. Group A received only non—directive group counseling fifty minutes weekly for twenty sessions. Group B received individual counseling with time matched to that of Group A. Group C serves as llDonald Davis, "The Role of the Group," Personnel and Guidance Journal, XXXVI (1958), 135—142. W 26 a control and received no counseling. All groups were homogeneous according to age, grade, sex, citizenship grades, I.Q., number of siblings, number of schools at- tended and number of years of education attained by the parents. Davis reported that at the conclusion of counseling Group A (group counseled) had improved in behavior over Group B or Broup C as evaluated by their teachers' citizen— [4 ship grades. Stockey12 selected three groups of adolescent boys and divided them into groups of fourteen to sixteen sub-. jects. One group received individual counseling, one group received group counseling, and one group served as a con— trol. Pre- and post-counseling measurements were admin— istered to evaluate change in adjustment by attitude toward self, by teacher evaluations of student change, and by GPA. Stockey concluded that there were no marked differ- ences between the processes of group or individual coun- seling in producing attitude change. 120. W. Stockey, "A Comparison of the Effectiveness of Group Counseling, Individual Counseling, and Employ- ment Among Adolescent Boys with Adjustment Problems," (unpublished Doctoral thesis, University of Michigan, Ann Arbor, 1961). "we I... . .l E... i|ulliiz uIIinllll'lln' 27 Similarly Froelichl3 and BaileylLL working independ- ently with different high school pOpulations, but using the same criterion of improvement in accuracy of self—knowledge, found no difference in the effectiveness of individual and multiple counseling. BilovskylS and his associates challenged the concept that the individualized face—to—face counselor-client relationship is always more desirable as "good counseling" than a depersonalized group method. They selected two hundred and one twelfth grade boys for group counseling and two hundred and one for individual counseling. All students were from the same high school. They divided the first group into small groups which met for weekly sessions. The other group received individual counseling. The problem was to determine if there were differences in realism of vocational choice of students who participated in individual counseling or in group counseling. Subsequent to the treatment period a team of four non—participating counselors rated each subject independ- ently on a three point scale as to the realism of his 13Clifford Forelich, "Multiple Counseling," (unpub- lished manuscript, University of California, Berkeley, 1955). 1“Bruce Bailey, "A Comparison of Multiple and Indi- vidual Counseling in Terms of Self-knowledge," (unpublished manuscript, University of California, Berkeley, 1955). 15David Bilovsky et_§l,, "Individual and Group Counseling," Personnel and Guidance Journal, XXXI (1953), 363—368. . . . . v.. E to r a . - h ' n\~ Us ‘ .F.‘ r A 1 . ._ _ rd ad 2,. R u ”L. : H s :. 4,4 A. . . C: (rm ..- r. a: .2» .. r . r ._ . »~u . * C» r“ a. . Y . «s~ . . o a I. - Ant s -~u . v . a l . . «a . er» e. by Ty r .. Q a .i» ~C r .. 2. i e . ~ _~‘ .W a n a. 28 choice as unreal, doubtful, or realistic. Raters felt that there were no marked differences between those students in individual or group counseling with regard to their voca- tional choices. Summary The review of the literature relevant to group coun- seling with adolescents was centered on studies which inves- tigated: (1) the group counseling as a vehicle for behavior and/or attitude change; and (2) quasi—experiments investigating the process of group counseling. The major findings of the review are summarized in the following para- graphs. Investigations of the effectiveness of group coun— seling in producing behavior and/or attitude change tend to show that group counseling is effective. However, studies do not at present support the theory that group counseling is effective in increasing academic achievement. Comparison studies between individual and group coun— seling are inconclusive. They seem to be about evenly divided in that some counselors have found marked differ- ences in favor of group counseling, while about as many others have found that there are no differences in effect between group and individual counseling. The literature does not yield any studies which investigate different methods of group counseling. 1L... .__ A —~'.:r- .. 7 mm.— 29 Few studies have been conducted which adequately define or describe the process within group counseling. There is a need for research on group counseling methods and for research which might be termed "action research" in which the actual process of group counseling is pre- sented in lucid detail. Research is also scanty about which situations, conditions, or for which clients the group process may be considered appropriate. In the research on group counseling with adolescents many limitations in design and methodology are frequently encountered which prevent consistent findings. Common limitations are: l. Improper or vague sampling methods Inadequate sample size Failure to provide for control groups «ll—UL) I'D Failure to reduce variance among subjects in number of sessions attended and time alloted 5. Inadequate definition of treatment variables 6. Lack of follow—up evaluations 7. Failure to allow for sufficient time of treat- ment for change to occur 8. Inadequate descriptions of the content and process of the treatment 9. Too much reliance upon the skill and person- ality of one counselor. CHAPTER III DESIGN AND METHODOLOGY This research was designed to test the relative ef- fectiveness of two types of group counseling upon low- motivated male high school students. The amount of coun- selor contact was held constant. The criteria of effective- ness were measured by increased positive self-concept and observations of acceptable classroom behavior. The three essentials of modern design are incorporated in the experi— ment: randomization, replication, and control. Design The design of the study was a pom}test-only control group model. Campbell and Stanley1 emphasize that such a design has no fixed weaknesses in the control of sources of invalidity. Kerlinger concurs.2 Students were randomly assigned to the various treat- ment groups. The six counseling groups were also randomly divided between the two pairs of counselors and the two 1D. T. Campbell and J. C. Stanley, Handbook of Research on Teaching (Chicago: Rand McNally & Co., 1963). 2Fred N. Kerlinger, Foundations of Behavioral Research (New York: Holt, Rinehart & Winston, Inc., 1964). 3O .‘y-fi . . ‘ kV-Aa.~— Ala o Jfigoa 7‘ 'IA u-.t’ \ K ‘Y:‘ v- 5“?“ saw}. N . C v 31 types of counseling: group and group-individual. The ex- perimental subjects exposure time to counseling was held constant. Those students assigned to group counseling only held group sessions each week. Students assigned to group— individual had group sessions alternating with individual counseling or group sessions every other week. By dupli- cating counseling methods replication was obtained. The control group received no counseling. To determine change in students in the groups, students in treatment groups were compared to control subjects. The design is summa- rized in the following table. TABLE 3.1--Summary of the basic design of the experiment. Grou Group NO' 9 Individual . Counseling Counseling Counseling Method Method Control Counselors A & B 2 groups 2 groups Counselors C & D 1 group 1 group Total 3 groups 3 groups control group The Population A low—motivated male, as operationally defined in this study, was a junior attending high school during the academic year 1965-66 at Mona Shores High School, Muskegon, 32 lkichigan, who ranked in the lower half of his class on an objective measure of academic motivation, the M-Scales3 administered early in the fall term to the entire male pOpulation of the junior class. Ninety-six students were classified as low-motivated according to the above criterion. These were invited to participate in group counseling. A11 invitations were ex- tended by the four individuals who were the counselors in the experiment. Seventy-four students accepted the invi- tation. Random Assignment Stratification categories were determined according to the free hours that the subjects had in common. From each group with free hours in common, seven were randomly selected. These sevenvere placed in experimental groups. Six groups of seven individuals each made up an experi- mental group. The Six groups were then assigned to either group or group-individual counseling by flipping a coin. The same method was used to assign groups within each treat- ment to the counseling teams. Those students who were unassigned were designated as the control group and were informed that they could not participate in counseling during the term of the study due 3This experimental instrument is more fully described in the following sections. 33 t£> the large number of responses and the limited staff available. They were provided with booklets to assist them in improving their study skills. All subjects, both control and treatment, were assured of an interview at a later date for the purpose of test interpretation. Instrumentation Independent variable measure.-—The M-Scales used as a selection instrument for this research relate task char- acteristics, self—concept, and personality traits to aca- demic motivation}1L Farquhar, et_al. develOped four scales for inclusion in the instrument.5 1. The Word Rating List was developed to measure self-concept 2. The Human Trait Inventory was constructed from items which differentiated between discrepant achievers. 3. The Generalized Situational Choice Inventory was developed to assess academic achievement task characteristics. 4. The Preferred Job Characteristics Scale was developed to determine high or low occupational motivation. uWilliam W. Farquhar, Motivation Factors Related to Apademic Achievement, Cooperative Research Project 846. (East Lansing, Michigan: Michigan State University, Office of Research and Publication, 1963). . 5w. w. Farquhar, D. A. Payne, and M. D. Thorpe, The figcthan State M-Scales (U.S. Office of Education, 1961). 34 With regard to the validity and reliability of the M-Scales Farquhar6 states: Reliability of the M—Scales. Using a sample of 240, a Hoyt's analysis of variance reliability estimate of .94 for 139 male cross—validated items was ob- tained. A female sample of 240 yielded a comparable .93 reliability estimate on 136 cross—validated items. For the most part, the reliability estimates for the sub—scales and various defined group of , achievers attain a satisfactory level (r = .68 to .92 for males and .60 to .93 for females). Validity of the M-Scales. Based upon a sample of 254 males and 261 females the validity estimates of the total M-Scales against grades was .56 and .40 respectively. The cross-validation estimates were .49 and .48 for males and females. The correlation of the sub- scales with the grade point criterion (GPA) follows the same pattern with the female correlations lower in magnitude than the males. The range was .27 to .42 for females and .32 to .51 for males. 5!! Dependent variable measures.--Two instruments were used as criteria in this experiment: (1) Measures of self- concept were obtained by the Minnesota Counseling Inventory (MCI) which identifies areas in which students may be ad- justing well or poorly. These are the scores on the Family Relationships (FR), Social Relationships (SR), and Emo- tional Stability (ES) scales. The remaining four scores of the MCI provide information more directly related to methods students employ in making adjustments. These scores are those of the Conformity (C), Adjustment to Reality (R), Mood (M), and Leadership (L) scales. 6William W. Farquhar, Motivation Factors Related to Academic Achievement, Cooperative Research Project 846 (East Lansing, Michigan: Michigan State University, Office of Research and Publication, 1963). 35 The concepts underlying the development of the MCI are based on the needs of the adolescent and the educator in the school setting. More Specifically, the purposes of the MCI are:7 1. To sensitize teachers and counselors to relevant personality characteristics differentiating students. 2. To identify students in need of therapeutic attention. 3. To assist in understanding students as they attempt to achieve more mature self-understanding and integration between themselves and their environment. 4. To provide a means for determining the effects of educational experiments upon relevant per— sonality characteristics. Two types of reliability data are reported for the scales of the MCI: coefficients of correlation between scores on odd-even numbered items, corrected by the Spearman—Brown formula; and reliability as estimated in the test-retest studies done at Austin and North High Schools. For boys the average reliability coefficient was .67; for girls it was .64. (2) Behavior rating--Eight selected items from the Haggerty—Olson-Wickman Behavior Rating Schedule (see 7Ralph Berdie and Wilbur Layton, Minnesota Coun- ggfljggg Inventory Manual (New York: Psychological Corp., 57. 36 1\D‘pendix B) were evaluated by the students‘ teachers to obtain measures of behavior change. This scale consists of eight items which are related on a five point descrip— tive continuum ranging from poor to acceptable behavior. The authors report reliability of total scores at .86 for elementary Children. Using an abbreviated scale with senior high students the reliability of a single rating is .92 obtained from the correlation between halves of the scales with a prediction for the total. Analysis of results of the use of the scales reveals a tendency to emphasize behavior of an aggressive type and to miss emo- tionally disturbed, non—aggressive types. To estimate the internal consistency of the abbre- viated eight item scale the Kuder—Richardson formula 20 was applied to the ratings made for this study. A relia- bility coefficient of .62 was obtained. This reliability coefficient is actually the mean of all split-half coefficients resulting from different splittings of a testfl8 The ordinary split-half coefficient is based on a planned split designed to yield equivalent sets of items. Hence unless the test items are highly homogeneous, the Kuder- Richardson coefficient will be lower than the split-half reliability.9 8L. J. Cronbach, "Coefficient Alpha and the Internal Structure of Tests," Psychometrika, XVI (1951), 297-334. 9Ann Anastasi, "Test Reliability," Psychological Testing (New York: Macmillan Company, 1961). o.-. 37 A case can be made for reliability of the above mag~ nitude. When a scale is designed to measure change, a low reliability coefficient would be the proper expectation,lO provided they are not so low that no consistency exists at all. The obtained coefficient supports this concept. No hypotheses were made about the students‘ G.P.A. within this research. But G.P.A. data was compiled as part of Mezzano's research.ll Grade point averages for the third and fourth marking periods were calculated and used to provide an assessment of treatment effectiveness during both the experimental period and again ten weeks following the experiment. Sample Before describing the characteristics of the 64 students used in the analysis, it is appropriate to account for students who were initially included in the groups, but not included in the analysis. It was decided that in order for students to be included in the final analysis they must have attended 80 per cent of the group sessions. Thus the minimum was 16 sessions for the group counseling 10R. L. Thorndike and Elisabeth Hagen, Measurement and Evaluation in Education (New York: Wiley & Sons, 1961). 11Joseph Mezzano, ”Group Counseling With Low- Motivated Male High School Students: Comparative Effects of Two Uses of Counselor Time,” (unpublished doctoral thesis, Michigan State University, East Lansing, 1966). 38 ‘treatment or eight sessions for the group-individual coun- seling treatment. This minimum was assumed to be adequate for exposure to treatment and was chosen as a baseline in order to allow for the possibility of missed sessions due to illness and other events that could not be helped by the student. Because individual sessions were on a flex- ible schedule, attendance was not a problem. By referring to Table 3.2 it will be noted that three students did not meet this requirement: two from the group only treatment and one from the group-individual treatment. TABLE 3.2-—Students included in the post—treatment analysis. Less Than Sgiggl Minimum Final N Attendance Group-individual (original N = 21) 2 1 18 Group only (Original N = 21) 1 2 18 Control (original N = 32) 2 2(refused 28 testing) Eleven students receiving group counseling from coun- selors A and B and seven students receiving group counsel- ing from counselors C and D met the minimum requirements for post-treatment analysis. Also eleven students re- ceiving group—individual counseling from counselors A and 39 E>Eind seven students receiving group-individual counseling from counselors C and D met the minimum requirement for post—treatment analysis. Thus each of the two experimental counseling groups contributed 18 students to the post- treatment analysis. There were 32 students assigned to the control group at the beginning of the study. Of these students, two re- fused to complete the testing and two had moved from the city. Therefore, there were 28 students in the control group for the post—treatment analysis. Since subjects were randomly assigned to counselors and treatments it was assumed that they were homogeneous in selection criteria. To lend support to this assumption, raw scores obtained on Otis Test of Mental Ability were averaged for each group and compared by means of an analy- sis of variance.12 Similarly, the mean GPA (previous term) of each group was tested for differences between groups. Data in Tables 3.3 through 3.5 support the assumption of homogeneity of groups. Inspection of the data in Table 3.3 reveals Slight differences among the three treatment groups on the Otis Test of Mental Abilities. The results of the analysis of variance of the OTMA scores are summarized in Table 3.4. The null hypothesis 12D. T. Campbell and J. C. Stanley, Handbook of Research on Teaching (Chicago: Rand McNally & Co., 1963). 40 of no differences cannot be rejected and it was concluded that there were no Significant differences on academic aptitude. TABLE 3.3--Mean scores for the Otis Test of Mental Ability and GPA means for each group. Group-Individual Group Control OTMA 104.89 107.67 102.89 GPA 4.55 4.95 4.86 TABLE 3.4--Ana1ysis of variance of the Otis Test of Mental Ability raw scores of the randomly assigned students.* Source of Variation Group—Ind. d.f. m.s. F Between treatment 285.80 2 142.90 .58 Within treatment 15,008.64 61 246.04 Total 15,294.44 63 *N = 64 A second consideration in determining the pre— eaxperimental equivalence of the groups is the fall term grwade point averages. The mean fall term grade point atmerages for each group reported in Table 3.3 reveals only slgight differences among groups. The results of the analysis of variance of GPA are :Sununarized in Table 3.5. The null hypothesis was not nr_‘mu l .1 n 41 rejected, and it was concluded that there were no signifi- cant differences among the three groups on grade point average. Apparently the differences that do appear in Table 3.3 are of a magnitude that could be expected by chance variation. TABLE 3.5—-Analysis of variance of fall term grade point averages of the randomly assigned students.* Source of variation S.S. d.f. m.s. F Between treatment 1.70 2 .85 .311 Within treatment 147.13 61 2.41 Total 148.83 63 *N = 64 The Counseling The approach used in all of the counseling sessions is one in which the counselors provide types of leads and reinforce those reSponses which are concerned with feelings and experiences about self, school, teachers, parents, future goals, and expressions of anger—hostility. Members (3f the groups were encouraged to freely discuss their ex- Eaeriences and feelings on the above topics. Also an atrtempt was made to encourage students to use their own vexrnacular. For example:* cl. 1 . . . so what d'ya mean about Saturday night? *Indicates a direct quote from a group session tape. 42 cl. 2 . . . not a damn thing! I got screwed! Yeah, Man. c1. 3 new girl? cl. 2 . . . could say that. Cl. 1 . . . so what happened? cl. 2 . . . I went by her house . . . supposed to pick her up. She said 9:15. So I come to pull her and I Says well I'll be a son of a bitch, she ain't there. Yeah. Saw her at the Annex and I walked up there and I was real pissed off and I says so what's the big idea? The counselors as active participants of the groups attempted to relate discussion to scholastic skills (or lack of them) in order to hasten group movement and increase awareness of the ways in which behavior patterns are inter- related. Counselors also blocked clearly irrelevant and defense-producing communications. As an example of the latter: cl. 1 . . . did he drop out or get kicked out? cl. 2 . . . Oh hell, he always says he's getting kicked out, but he never does. He's a big bullshitter. (talking irrele- vantly . . . straight gossip) cnslr . . . I keep getting the message that you guys are knocking on him to stay off your own feelings about it . . . what gives? Because replication was an essential element to the (design, it was necessary that the two types of group coun- sseling be conducted in like manner by both teams of coun- sselors. To insure that the teams were operating in like rnanner, the four couselors met each week after group ses- ssions to compare notes and experiences. 43 Counseling Setting All six of the groups met in a small conference room in the high school. Individuals were not assigned seats. They sat in a circle facing each other. Sometimes they placed a small table in the middle, but more often they did not. All of the individual counseling sessions took place in the office of each counselor. 3!— The Counselors All four of the counselors involved in this research possess a Master's degree in counseling and guidance and all have worked in public schools and have previously par- ticipated in both group and individual counseling. The counselors worked in pairs, one male and one female in both group and individual contacts. At the time of the experi- ment three of the counselors were members of the staff at Mona Shores High School and the fourth was employed as an instructor at Michigan State University. Three of the four counselors were currently working on advanced degrees. The Null Hypotheses The basic research hypotheses of this study were ‘broadly stated previously in Chapter I. A more Specific :formulation of these hypotheses as they relate-to the G > C Statistical Treatment The analysis of variance is the appropriate technique to treat the null hypotheses of this research. This tech- nique differentiates variation among a number of means according to different treatments.13 The statistic used is F. The level of significance for rejecting the null hypotheses was set at five per cent. Summary This eXperiment is designed to test the differences in effect of two types of group counseling, group and group— individual on the self-concept and behavior of low-motivated 13D. T. Campbell and J. C. Stanley, Handbook of _Research on Teaching (Chicago: Rand McNally & Co., 1963). 46 male high school students. The study is unique because both the group and individual counseling was conducted by a team of one male and one female. There were six treat- ment groups: two group-individual (GI), two group (G) assigned to one counseling team, and one each of group- individual (GI) and group (G) assigned to the other coun- seling team. Also there was a control group which received no counseling. The design contained the three essentials of modern design: randomization, replication and control. The analysis of variance was used to test the null hypotheses of this study. The five per cent level of confidence was chosen for accepting or rejecting the null hypotheses. . .e' "vs-Pm , ,_ _, ,, _ .--.~-.«=.\ CHAPTER IV ANALYSIS OF EXPERIMENTAL RESULTS In Chapter IV the analysis of experimental results are reported. Both of the null hypotheses were tested by the analysis of variance technique which was reported in Chapter III, Analysis of Variance of the Sub-Scales of the Minnesota Counseling Inventory The first hypothesis, which predicted the effects of treatment, was tested by the analysis of variance of the sub-scales of the Minnesota Counseling Inventory. Hypothesis One Hypotheses one is restated in null form: 1. There are no differences in the Minnesota Counseling Inventory scores at completion of the experiment in the group-individual (GI), group experience (G), and students who did not participate in group counseling (c). Symbolically: HO: “GI = "G = ”C Legend: G1 = group-individual G = group C 2 control 47 48 Alternate Hypothesis: At the completion of the experiment the mean scores on the Minnesota Counseling Inventory of the students in group- individual (GI), will be lower than the mean scores of the students in group experience (G), which in turn will be lower than the mean scores of those students who did not participate in group counseling (C). Symbolically: HO: “GI < “G < ”C Legend: GI G C group-individual group control There was a total of 18 students in the group- individual treatment, 18 students in the group treatment, and 28 in the control group. The first hypothesis was tested by comparing each of the seven Minnesota Counseling Inventory sub-scales scores of the two treatment groups and the control group using the one-way analysis of vari- ance technique. The results of the analyses are presented in the IRDllowing tables 4.1 through 4.7: A . 49 'TABII34.l——Ana1ysis of variance of the Family Relationship Scale (FR) for the group-individual counseling (GI), group counseling (G), and control groups (C) __Y__f Means GI = 15.28 G = 16.78 C = 16.50 Source of Hypothesis Variation 5'8 df m'S‘ F Tested is: Among Treatments 30.88 2 15.44 .37 not rejected Within Treatments 2564.60 61 42.04 Total 2595.48 63 Necessary: F .05 1 3.15 to reject Hol, GI = G = C An F value of 3.15 is necessary for 2.61 degrees of freedom before significant differences between groups can be concluded at the .05 level of significance. InSpection of the table revealed an F value of less than 1.00 which was not significant. The differences among the three experimental groups were likely to have occurred by chance. Therefore, the null hypothesis that there were no differences in Family Relationship scores between students who participated in group counseling and those who did not participate in group counseling was not rejected. TABIII4.2--Ana1ysis of variance of the Emotional Stability Scale (ES) for the group—individual counseling (GI), group counseling (G), and control groups (C) Means GI = 16.77 G = 16.55 C = 21.50 Source of Hypothesis Variation 8'8 df m.s. F Tested is: Among Treatments 353.23 2 .176.11 4.55 rejected Within Treatments 2358.25 61 38.66 Total 2710.48 63 Necessary: F .05 i 3.15 to reject Ho GI = G = C 1.3 An F value of 3.15 is necessary for 2.61 degrees of freedom before significant differences between groups can be concluded at the .05 level of confidence. The table revealed a significant F value of 4.55 The differences among the three experimental groups were not likely to have occurred by chance. Differences as large as this would occur by chance less than tive times in one hundred. Therefore, the hypothesis of no differences in emotional stability among the students in group—individual counseling, group counseling, and the students who did not participate in group counseling was rejected. The F ratio when test by post-hoc comparisons of the scheffe test revealed that the group treatment means ex- ceeded the group—individual which in turn exceeded the con- trol group. Therefore the alternate hypothesis had to be rejected. 51 TABLE 4.3-—Analysis of variance of the Social Relationships Scale (SR) for the group individual counseling (GI), group counseling (G), and control groups (C). Means G1 = 22.77 G = 16.77 C = 27.0 Source of Hypothesis Variation 8'8 df m.s. F Tested is: Among Treatments 31676.40 2 1583.82 24.51 rejected Within Treatments 3941.21 61 64.61 Total 35617.61 63 Necessary: F .05~; 3.15 to reject Ho G1 = G = C 13 An F value of 3.15 is necessary for 2.61 degrees of freedom before significant differences between groups can be concluded at the .05 level of significance. The analysis revealed an F value of 24.51 which was significant. Differences as large as this would occur by chance less than five times in one hundred. Therefore, the null hypothesis of no differences on the Social Relationship scale among the students in group—individual, group coun— seling, and the students who did not participate in group counseling was rejected. Post-hoc comparisons using Scheffe revealed that the group treatment mean exceeded the group—individual mean. This finding was not in accord with the alternate hypothe— sis. 52 TABLE 4.4--Ana1ysis of variance of the Reality Scale (R) for the group-individual counseling (GI), roup counseling_(G), and control groups (0%. Means G1 = 20.0 G = 22.7 C = 24.0 Source of Hypothesis Variation 8'8 df m°S’ F Tested is: Among Treatments 1741.80 2 270.90 4.99 rejected Within Treatments 3309.64 61 54.26 Total 5051.44 63 Necessary: F .05 1 3.15 to reject Ho G1 = G = C 13 An F value of 3.15 is necessary for 2.61 degrees of freedom before significant differences between groups can be concluded at the .05 level of significance. A significant F value of 4.99 was revealed by the analysis. Differences as large as this would occur by chance less than five times in one hundred. Therefore, the null hypothesis of no differences on the Reality scale among the students in group-individual counseling, group coun- seling, and the students who did not participate in group counseling was rejected. In accord with the alternate hypothesis the mean of group—individual treatment (G1) was found to exceed the mean of group treatment (G). " ‘ ‘ ' ’- " ' '. . . ‘4 ' "' — . .‘ _‘-. r ‘ .— ~. h. . : . .. . 3&1. . ..- .5, d.f.l'mkqh ~_ .~_,._,.._ a, I: a. 1 -\ _.. .p' 3" 5'1-“ ‘42? 3‘ ‘ "'1“ s.- q __ . . - — V: W. J‘-1‘ L.’ -_ ’ '..- ()A r S 53 TABLE 4.5--Analysis of Variance of the Conformity Scale (C) for the group-individual counseling (GI), group counseling (G), and control groups (C). Means GI = 16.2 G = 16.2 C = 16.4 Source of Hypothesis Variation 8'8 df m.s. F Tested is: Among Treatment .91 2 .45 1.0 not rejected Within Treatments 1536.20 61 25.18 Total 1537 ll 63 Necessary: F .053_ 3.15 to reject H01; GI = G = C An F value of 3.15 is necessary for 2.61 degrees of freedom before significant differences between groups can be concluded at the .05 level of significance. InSpection of the table revealed an F value of less than 1.00 which was not significant. The differences among the three groups are likely to have occurred by chance. Therefore, the null hypothesis that there were no differ- ences in Conformity scores between students who partici- pated in group counseling and those who did not partici- pate in group counseling was not rejected. 54 TABLE 4.6--Ana1ysis of variance of the Mood Scale (M) for the group-individual counseling (GI), group counseling (G), and control groups (C). Means G1 = 17.0 G = 14.0 C = 19.2 Source of Hypothesis Variation S'S’ df m.s. F Tested is: Among Treatments 323.02 2 161.56 10.93 rejected 1 IE Within Treatments 900.98 61 14.77 Total 1224.00 63 :~- Necessary: F .05 1 3.15 to reject Hol; GI = G = C An F value of 3.15 is necessary for 2.61 degrees of freedom before significant differences between groups can be concluded at the .05 level of confidence. The analysis revealed an F value of 10.93 which was significant. Differences as large as this would occur by chance less than five times in one hundred. Therefore, the null hypothesis of no differences on the Mood Scale among the students in group—individual counseling, group coun- seling, and the students who did not participate in group counseling was rejected. Post-hoc comparisons with the Scheffe method revealed that the mean of the group treatment exceeded the mean of the group-individual treatment. Because this finding was not in accord with the alternate hypothesis no other con- clusion is warranted. 55 TABLE 4.7—-Analysis of variance of the Leadership Scale (L) for the group—individual counseling (GI), group counseling (G), and control groups (C). Means G1 = 16.5 G = 12.1 C — 22.8 Source of Hypothesis Variation S'S' df m.s. F Tested is: Among Treatments 200.61 2 100.3 8.3 rejected Within Treatments 729.99 61 11.96 Total 930.60 63 Necessary: F .05 l 3.15 to reject Ho G1 = G = C 13 An F value of 3.15 is necessary for 2.61 degrees of freedom before significant differences among groups can be concluded at the .05 level of significance. A significant F value of 8.3 was found by analysis. Differences as large as this would occur by chance less than five times in one hundred. Therefore, the null hypothesis of no differences on the Leadership Scale among the students in group-individual counseling, group counseling, and the students who did not participate in group counseling was rejected. Following the F test for significance, the Scheffe test was used to compare means of the treatment groups. Results indicated that the mean of the group treatment (G) exceeded the mean of the group—individual treatment (GI). This finding was contradictory to the alternate hypothesis. 56 Analysis of Variance of the Haggerty- Olson—Wickman Scores Hypothesis two was tested by an analysis of variance of the scores on the Haggerty-Olson-Wickman Behavior Rating Scale. Hypothesis Two Hypothesis two is restated in null form: 2. There are no differences in the Haggerty-Olson- (. Wickman Behavior Rating mean scores at completion of the experiment in the group—individual (GI), group experience (G), and students who did not participate in counseling (C). Symbolically: HO: ”GI “G “C Legend: GI G C group-individual group control Alternate Hypothesis: At the completion of the eXperiment the mean scores on the Haggerty-Olson- Wickman Behavior Rating Scale of students in group—individual (GI) will be greater than the mean scores of students in the group experience (G) which in turn will be greater than the mean scores of students who did not participate in group counseling (C). Symbolically: H2a: ”GI > MG > C Legend: G1 = group-individual G 2 group C 2 control _..\.7.-...... 57 The results of the analysis of variance are sum- marized in Table 4.8. TABLE 4.8--Analysis of variance of the Haggerty-Olson- Wickman Behavior Rating scores of the group-individual counseling (GI), group counseling (G), and control groups (C). Means G1 = 6.40 G = 4.91 C = 4.51 “I Source of Hypothesis Variation 8'8“ df m.s. F Tested is: Among Treatments 44.44 2 22.22 7.10- rejected Within Treatments 191.10 61 3.13 Total 235-54 63 Necessary: F .05 l 3.15 to reject Ho G1 = G = C 1: An F value of 3.15 is necessary for 2.61 degrees of freedom before significant differences among groups can be concluded at the .05 level of significance. InSpection of the table revealed an F value of 7.10 which was Significant. Differences as large as this would occur by chance less than five times in one hundred. Therefore, the null hypothesis of no differences on the Haggerty-Olson-Wickman Behavior Rating Scale among students in group-individual counseling, group counseling, and the students who did not participate in group counseling was rejected. Furthermore, the Scheffe test of mean differences .| II . Ill '1'l.‘ .lu Ill ' was significant for all three groups and in the hypothe- sized direction. Companion Research Conclusion The major conclusion from the companion research of Mezzanol ftfllows: jDuring the treatment period and ten weeks following the completion of the experimental period, grade point averages were computed for the three groups as part of Mezzano's research. Findings indicated that there were no differences in academic improvement during the treatment period. However, ten weeks after the experimental period Mezzano found that the grade point averages of the students who received group counseling were higher than the students who received group plus individual counseling which in turn were higher than the grade point averages of the con- trol group. Summary The analysis of variance technique was used to test the two null hypotheses that were stated in Chapter III. The first hypothesis tested each of the seven sub—scales of the Minnesota Counseling Inventory. Results were sig- nificant for five scales: Emotional Stability (ES), Social Relationships (SR), Reality (R), Mood (M), and Leadership lJoseph Mezzano, Group Counseling with Low-Motivated Higp School Students: Comparative Effects pf Two Uses pf Counselor Time (unpublished doctoral thesis, Michigan State University, East Lansing, Michigan, 1966). I' . 59 (L). The remaining two sub-scales Family Relationships (FR), and Conformity (C) did not yield significant findings. Thus the first null hypothesis was rejected in five major parts and was not rejected in two. However, on four of the five sub-scales (Emotional Stability, Social Relationships, Mood, and Leadership) the means of the group counseling treatment exceeded the means of the group plus individual treatment. This finding was not in accord with the alter- nate hypothesis. Only on one sub-scale (Reality) were the findings in agreement with the alternate hypothesis. Scheffe's method for post-hoc comparisons of means was used to further test the sub-scales of Emotional Sta- bility, Social Relationships, Mood, and Leadership of null hypothesis one.2 Subjects who had experienced the group counseling treatment had lower means on these sub-scales than the students who had participated in group-individual counseling which were lower than the mean scores of the control group. The difference between these counseled groups was significant, but not in accord with the alter— nate hypothesis. The null of hypothesis two, designed to evaluate the differences among the groups on the Haggerty-Olson- Wickman Behavior Rating Scale, was rejected; the alternate was accepted. 2H. Scheffe, The Analysis pf Variance (New York: Wiley & Sons, 1959). '— CHAPTER V SUMMARY, CONCLUSIONS, AND DISCUSSION Summary Within this experiment the relative effectiveness of two types of group counseling for low—motivated male high school juniors was investigated. The treatments were: (1) group counseling only, and (2) group plus individual coun- seling. Changes in self—concept and teachers' ratings of observed classroom behavior constituted the criteria measures. The two basic hypotheses of the study were: 1. The self-concept scores of students who received both group and individual counseling in conjunction will be more positive at the completion of treat- ment than the self—concept scores of students who had either group counseling or no group counseling. 2. The teachers' ratings of behavior scores of students who received group and individual coun- seling in conjunction will be more positive at the completion of treatment than the behavior rating scores of students who had either group counseling or no group counseling. 60 . ‘ ./ ‘ , . . ‘ 61 All group sessions were conducted by a counseling team made up of one female and one male counselor. 0f the 96 students originally identified as low- motivated on the Michigan M-Scales, 74 accepted the invita- tions extended to them to participate in the study. Stra- tification categories were determined according to the free hours that these 74 students had in common. From each group with free hours in common, seven students were ran- domly selected. These seven were placed in experimental groups. Six groups of seven individuals each made up an eXperimental group. The six groups were randomly assigned to either group or group-individual counseling by flipping a coin. The same method was used to assign groups within each treatment to the counseling teams. Those students who were unassigned were designated as the control group and received no treatment. Change in group—counseled subjects was measured by comparing them to the control subjects on two criteria: self-concept scores and behavior ratings. In order for subjects to be included in the final analysis they must have attended 80 per cent of the ses- sions or 16 sessions for the group counseling treatment and eight sessions of the group—individual treatment. Attendance at individual sessions did not prove to be a problem because subjects were scheduled according to hours convenient to them. Eighteen students in each of the two treatment groups or 36 out of the original 42, 62 ‘met the minimum requirements for post-treatment analysis. Twenty-eight of the original 32 subjects were in the con— trol group for post-treatment analysis. Thus the propor— tionate attrition rate was evenly distributed between the counseled and non-counseled groups. Three major findings emerged from the study: 1. Group plus individual counseling helped male stu— dents' ability to cope with reality. This finding substantiated the prediction of alternate hypothesis one. Of the other six sub-scales, used as criterion for change in self-concept, four (Social Relationships, Emotional Stability, Leadership, and Mood) indicated Significant findings, but not in the predicted direc- tion. Two scales (Family Relationships and Conformity) did not yield significant results. 2. Group—individual counseling produced significant and positive change in teachers' ratings of the observed classroom behavior of students. 3. With counselor time held constant, group counseling was effective with or without the addition of individ— ual counseling. Conclusions An examination of the data revealed: 1. Group counseling plus individual counseling ex— ceeded group counseling alone in generating more positive perception of the students' ability to cope with reality. w 63 2. Group counseling plus individual counseling exceeded group counseling alone in producing teacher estimates of more conforming behavior in the classroom. 3. No differences were found among group coun— seling plus individual counseling, group counseling alone, and the control group on ‘ 3 changes in coping with family relationships (FR Scale), and in self-perception Of conformity to social standards (O Scale). 4. On four scales (Emotional Stability, Social Relationships, Mood, and Leadership) differ— ences in the effect of group counseling were observed, but the group counseling alone treatment exceeded the group counseling plus individual treatment which was contrary to the theoretical predictions of the study. This study should be replicated to pull out the ele- ments in it that operated to make the group process success— ful. And in order to cut the attrition rate, the treat- ment period Should begin as soon as school Opens in fall. The treatment period should be continued throughout the school year because low—motivated students need the bene— fits of peer and adult support for longer periods of time. Finally, a follow-up study should be done three months and six months following the treatment period to see if the Changes in self—concept and behavior were sustained. 64 Discussion The use of group counseling in this study was based upon the theory that low-motivated students require a com- prehensive approach. The comprehensive approach is indicated because it appears from research on the low- motivated that these students are not homogeneous. They _1 share the syndrome of low-motivation, but the etiology of that syndrome and the attendant defenses are usually quite diSsimilar. Many different sources appear to generate low- motivation. Some of the moSt salient appear to be: Hostility. Kirkl viewed hostility as an important dimension and further research has supported her observa- tion.2 Intolerance of Delayed Rewards. Delayed rewards 3 hold no appeal or value for the low-motivated student. lBarbara Kirk, "Test versus Academic Performance in Malfunctioning Students," Journal pf Consulting Psychology, XVI (1952). 2Wm. Farquhar and Ronald Taylor, "Personality Moti- vation and Achievement: Theoretical Constructs and Empir— ical Factors,” Journal of Counseling Psychology, XII (1965), 186-191. _— 3Wm. Farquhar, Motivation Factors Related 39 Academic Achievement, U.S. Office of Education and Welfare, COOpera- tive Research Project #846, ER 9 (East Lansing, Michigan: Office of Research and Publications, College of Education, Michigan State University, 1963). 65 Negative Self-Concept. The low motivated student sees himself as worthless and has little respect for him- self.“ Depression. Evidence indicates that depression in the low-motivated evolves from a self-derogation which in 5 turn produces a feeling of inadequacy. Low Academic Involvement. Low-motivated students reject the school's goals and tend to passively endure the 6 educational setting or act—out against it. Persistent Low Achievement. Research indicates that for most low-motivated students the syndrome is of long standing and usually dates to the early elementary grades.7 Because the pattern and dimensions of low-motivation are complex and are intricately woven into the students' reality, the vehicle for the development of insight and change must be able to accommodate this complexity. Group counseling is a vehicle for change and a reality testing 24David Payne and Wm. Farquhar, "The Dimensions of an Objective Measure of Academic Self-concept," Journal pf Educational Psychology, L111 (1962), 187-192. 5Morris Rosenberg, Society and the Adolescent Self- Image (Princeton, New Jersey: Princeton University Press, 1965). 6Wm. Farquhar, Motivation Factors Related 39 Academic Achievement, U.S. Office of Education and Welfare, Coopera— tive Research Project #846, ER 9 (East Lansing, Michigan: Office of Research and Publications, College of Education, Michigan State University, 1963). 7 ‘Morris Rosenberg, Society and the Adolescent Self— Image (Princeton, New Jersey: Princeton University Press, 1965). u 5' - v H 66 ground for low-motivated students. In theory group coun- seling Should enable students to: (l) relinquish their defenses; (2) find acceptance, support, and reassurance; (3) find out that they can help others; (4) become aware of their own personality dynamics; and (5) relate new in— sights to the solution of current problems. Furthermore, it was theorized that low-motivated students would gain insight more rapidly, if the advantages of group counseling and individual counseling were combined. Thus, holding .1. treatment time constant, the decision was made to give three groups only group counseling, and three groups group and individual counseling. There are many who assert that co—counselors of dif- ferent sexes offer a more complete and realistic setting in which to learn new life patterns.8’9 It was assumed then that the co-counselor method would produce optimal results because the personalities and orientations of the counselors were flexible enough to permit considerable variance along the active-passive, directive-nondirective continuae. The presence of co-counselors should also serve to lessen the initial threat for the students and give 8A. Solomon, F. J. Loeffler, and G. H. Frank, "An Analysis of Co—therapist Interaction in Group Psychotherapy," International Journal pf Group Psychotherapy, III (1953), 171—180. 9W. H. Lundin and B. M. Aronov, ”The Use of Co— therapists in Group Psychotherapy," Journal pf Consulting Psychology, XVI (1952), 76-80. them the advantage of participant-learning, while at the same.time offering some protection.lo Description pf the subjects. All of the subjects whether in treatment or control groups had been identified as low—motivated. However, as stated previously, they varied considerably in their personality dynamics and in their overt behavior. While it is impossible to give a ”1 description that would clearly characterize each student, clinically there did appear to be several attitudes that were common to all of them; attitudes which only varied in irtensity from student to student. Self-concept. It was the group leaders' impression that over-riding all attitudes was the subjects' low self- concept. That is, these students Shared a basic underlying feeling of worthlessness, which discouraged them from free expression (verbally or physically). This sense of worth— lessness led them to the belief that they were inadequate to master confronting situations and that in general ”some- thing” was fundamentally wrong with them. These feelings 1r turn generated depression which seemed to manifest arxiety in varying behavior: trouble going to sleep; trouble getting up; headaches; coughs; fingernail biting; rightmares; upset stomachs; hands sweating. In school 10F. J. Loeffler and H. M. Weinstein, "The Go- Therapist Method: Special Problems and Advantages,” Group Psychotherapy, VI (1954), 189-192. Uh} 68 their behavior was either passive—apathetic or passive- defiant. Other factors. Subjectively, the group leaders iden— tified six factors of low self-esteem which may have con- tributed to the subjects' anxiety: instability of self- concept, distrust of others, vulnerability, anticipation Of failure, feelings of isolation, and inability to differ- entiate appropriately among stimuli. l. Instability of self-concept; or a tendency to have ever shifting and unstable pictures of themselves. In short, while these boys held negative views of themselves, they were neither absolutely nor consistently negative and thus their self-image constantly shifted. 2. Distrust of others: their assumptions of trust were based upon how they perceived them— selves treated by others. Their 1ow faith in others tended to show in some form of contempt accompanied by either overt or passive hos- tility. 3. Vulnerability: although they strenuously defended against showing vulnerability, these boys felt criticism and blame keenly and disproportionately. 4. Feelings of isolation: because these boys were vulnerable, they were also relatively awkward with others . . . particularly with ' " "_..-'9. Fir... 69 authority figures. This self-conscious awkwardness Showed in their hostile, but often subdued behavior which in turn produced self-imposed isolation. 5. Anticipation of failure: although the low-motivated students said that they wanted what most students say that they want-~success--their low self-concept made them dubious of their own abilities and set up an anticipation of failure which was likely to produce failure. This degenera- tive cycle compounded anxiety. 6. Inability to differentiate appropriately among stimuli: the low-motivated students seemed to have a truncated repertoire of behavioral responses. So they often reacted with identical responses to totally different stimuli. And, the intensity of their re— sponses did not vary appropriately to the given situation. On the basis of the above observations it appeared reasonabletp assume that not only is low—motivation psy— chologically distressing in and of itself, but it is associated with an entire train of emotions which leads to an inevitable state of generalized anxiety. As a result of generalized anxiety, low—motivated students block and fail to develop solutions to their academic and emotional 7O prntltms Clinically Bruner's statements about "learning- . . 11 blocks” seems verified: There is a sharp distinction that must be made between behavior that copes with the require— ments of a problem and behavior that is designed to defend against entry into the problem. . . . CCping respects the requirements of problems we encounter while still respecting our integrity. Defending is a strategy whose objective is avoidtugor escaping from problems for which we relieVe there is no solution that does not Violate our integrity of functioning. Grou) Counsel ng. Perhaps the dynamic counseling ap— 4L preach accounted for the significant findings on both of The hypotheses. The fact that group counseling was effec- tive with or without the inclusion of individual counseling may be a function of the personality dynamics of the low- motivate« subjects. The effects of peer support, reassur- arce, and understanding may have been internalized more readily than the support and reinforcement from the coun- selors. Also subjects who received only group counseling had no Opportunity to ventilate their feelings with a coun— selor anywhere except in the group. They were thereby breed to use the group sessions advantageously. As indicated by the results on the MCI, those stu- 1 ML w;.S who received either treatment of group counseling borefited in Specific ways while students assigned to the :ortrcl group did not. Some of the Specific areas of im— p I (,1 V PHIL-761' C. we I‘ e I .M _J .1 ' F Jerome Bruner, Toward a Theory of Instruction, (Camurid e, Massachusetts: HaYVard University Press, 1966). k/C Social relationships . . . the boys appeared to enjoy talking to others and were interested in what others had to say. In general they gained some social skill and were thus able to conduct themselves appropriately in school situations. Emotioral stability . . . the boys tended to become less self conscious or lacking in self- confidence. Consequently new situations did not seem to generate anxiety. Reality . . . students began to actively deal with threatening situations in order to master them rather than withdraw or avoid them. Often they even became openly competitive with class- mates. Mood . . . they develOped a more appropriate affect and morale. When depressed, discouraged, or angry, they could recover more readily. And because they had gained some measure of self- confidence, they began to be more Optimistic and engaged in some long—range plans. Leadership . . . in addition the students began to show initiative and acceptance of responsi— bility. Their newly acquired self-concept allowed them to use new skills in working rela- tively well with others. .72 That both group counseled groups did not show gains on the Family Relationship and Conformity Scales of the MCI may be explained by the students' perceptions of and inter- actions with their families. Most of the students had experienced serious conflicts with their parents for many years and their parents even now were either slow in recog— _ . . l2 rg change In their sons or denied change had occurred. These students Spent as little time at home as possible and often expressed the feeling that their parents did not wish to understand them and ”couldn't care less" if they were home or not. If it is assumed that inter—family communica— ,—+ ions had been strained or non-existent for a long time, ‘,l;. 0. follows that re-opening these communication lines would l, D extremely difficult and might well take considerably more time than the limits of a study of this nature per— mitted. The eXperimental subjects tended to perceive them- selves as isolated both at home and in school. So they maintained defensive behavior as evidenced by minor class- room disturbances, unexcused absences, belligerence, mouthiness, irritability, and impulsiveness; thus perpetu— ating isolation. It almost seemed to be behavior which said ”O.K. . . . you think I'm a bad guy . . . I'll show you just how bad a guy I can be!" ."_\ 14 n.7- ocseph Mezzano, Group Counseling with Low—Motivated MaLe High School Students: Comparative Effects of Two *” Cf Counselor Time (unpublished doctoral thesis, Mich- ian State UniversIty, East Lansing, Michigan, 1966). M'lndl _-_-l 711-." ' _.! ...-— . .- .- 1 73 AS group counseling progressed the students gained in self—confidence and altered their self-concept which in turn enabled them to give up some asocial behavior as sup— ported by their teachers' ratings of them on the behavior scale. But apparently the observed conformity in classroom behavior did not necessarily reflect the students' under- lying attitudes. It is assumed that they gave the teachers rm} what the teachers demanded: conformance. But this new I pattern of behavior was not internalized and accepted by the students as a real part of them, but used as an effec- tive coping technique. BIBLIOGRAPHY BIBLIOGRAPHY Adler, J. and Berman, I. R. "Multiple Leadership in Group Treatment of Delinquent Adolescents," International Journal 9: Group Psychotherapy, X (1960), 213-215. Arbuckle, D. and Boy, Angelo. ”Client-centered Therapy in Counseling Students with Behavior Problems," Journal of Counseling Psychology, VIII (1961), 136-139. Anastasi, Anne. ”Test Reliability," Psychological Testing. New York: Macmillan Company, 1961. Bailey, Bruce. ”A Comparison of Multiple and Individual Counseling in Terms of Self-knowledge." Unpublished manuscript, University of California, Berkeley, 1955. Baymur, F. and Patterson, C. H. "Three Methods of Assist- ing Underachieving High School Students," Journal Ipf Counseling Psychology, V11 (1960), 83-90. Berdie, Ralph and Layton, Wilbur. Minnesota Counseling Inventory Manual. New York: The Psychological Corporation, 1957. Bilovsky, David 33 21: "Individual and Group Counseling,” Personnel and Guidance Journal, XXXI (1953), 363— 368. Broedel, J., p£_§l, "The Effects of Group Counseling on Gifted Underachieving Adolescents," Journal pf Counseling Psychology, VII (1960), 163—170. Bruner, Jerome. Toward p_Theory pf Instruction. Cambridge, Massachusetts: Harvard University Press, 1966. Campbell, D. T. and Stanley, J. C. Handbook_pf Research pp Teaching. Chicago: Rand McNally and Company, 1963. Caplan, S. A. ”The Effects of Group Counseling on Junior High School Boys' Concepts of Themselves in School," Journal pf Counseling Psychology, IV (1957), 124-128. Caronis, George. "Experimental Study in Evaluating the Adjustment of Disturbed Delinquents Exposed to Guided 75 .\1 O\ Group Counseling Within a Training School Setting." Unpublished master's thesis, Michigan State Univer- sity, East Lansin, Michigan, 1963. Catron, D. W. ”Educational-Vocational Group Counseling: The Effects on Perception of Self and Others," Journa1_pf Counseling Psychology, XIII (1966), 67—71. Clements, Barton E. "Transitional Adolescents, Anxiety and Group Counseling,” Personnel and Guidance Journal, XLV (1966), 67-71. Corsini, R. J. Methods pf Group Psychotherapy. New York: McGraw-Hill, 1957. 5 Cronbach, L. J. "Coefficient Alpha and the Internal Struc- ture of Tests,” Psychometrika, XVI (1951), 297-334. Davis, Donald. ”The Role of the Group," Personnel and Guidance Journal, XXXVI (1958), 135-142. Driver, Helen. "Multiple Counseling: A Small Group Dis- cussion Method for Personal Growth," Personnel and Guidance Journal, XXX (1952_, 173-176. Farquhar, W. W. Motivation Factors Related to Academic Achievement. Cooperative Research Project 846 (January, 1963). East Lansing: Office of Research and Publications, College of Education, Michigan State~University. , Payne, D. A., and Thorpe, M. D. The Michigan State M-Scales, U.S. Office of Education, 1961. and Stewart, N. ”Counseling the Low Motivated Male: A Working Paper." Mimeographed paper, Michigan State University, April, 1966. and Taylor, Ronald. "Personality Motivation and Achievement: Theoretical Constructs and Empirical Factors,” Journal 9: Counseling Psychology, XII (1965), 186-191 Froelich, Clifford. ”Multiple Counseling." Unpublished manuscript, University of California, Berkeley, 1955. Kerlinger, Fred N. Foundations pf Behavioral Research. New York: Holt, Rinehart and Winston, Inc., 1964. Kirk, Barbara. "Test versus Academic Performance in Mal- functioning Students," Journa1_pf Counsulting Psychology, XVI (1952). , I ,-| I "i . - 77 KonOpka, G. ”Group Work and Therapy," A Decade pf Group Work. Ed. C. E. Hendy. New York: Association Press, 1958; PP- 39-“4- Lodato, F., Sokoloff, M., and Schwartz, S. "Group Coun- seling as a Method of Modifying Attitudes," The School Counsaor, X (1964), 27—31. Loeffler, F. J. and Weinstein, H. M. "The Co-Therapist Method: Special Problems and Advantages," Group Psychotherapy, VI (1954), 189-192. Lundin, w. H. and Arnov, B. M. "The Use of Co-Therapists in Group Psychotherapy," Journal of Consulting Psychology, XVI (1952), 76-80. '"’ McCarthy, M. V. "The Effectiveness of a Modified Counsel- ing Procedure in Promoting Learning Among Bright Underachieving Adolescents," Research Project A.S.E.— 6401. Washington D. C., Department of Health, Edu- cation and Welfare. Mann, J. "Some Theoretic Concepts of the Group Process," International Journal pf Group Psychotherapy, V (1955), 235—242. Mezzano, Jos. "Group Counseling with Low-Motivated Male High School Students: Comparative Effects of Two Uses of Counselor Time." Unpublished doctoral thesis, Michigan State University, East Lansing, Michigan, 1966. Mintz, E. E. ”Special Values of Co—Therapists in Group Psychotherapy,” International Journal pf Group Psychotherapy, XIII (1963), 123—132. Ohlsen, Merle. "Adapting Principles of Group Dynamics for Group Counseling,” The School Counselor, XIII (1966), 159-161. Payne, David and Farquhar, W. W. "The Dimensions of an ' Objective Measure of Academic Self-concept," Journal p£_Educational Psychology, L111 (1962), 187-192. Rosenberg, Morris. Society and the Adolescent Self-Image. Princeton, New Jersey: Princeton University Press, 1965. Scheffe, H. The Analysis_pf Variance. New York: Wiley and Sons, 1959. 78 Solomon, A., Loeffler, F. J., and Frank, G. H. "An Analysis of Co-Therapist Interaction in Group Psychotherapy," International Journal pf Group Psychothergpy, III (1953), 171-180. Stefflre, Buford. Theories pf Counseling. New York: Wiley and Sons, 1965. Stock, D. and Thelen, H. ”Emotional Dynamics and Group Culture,” Group Therapy and Group Function. Ed. by Rosenbaum and Berger. New York: Basic Books, 1963, pp. 83—86. Counseling, and Employment Among Adolescent Boys with Adjustment Problems." Unpublished doctoral . thesis, University of Michigan, Ann Arbor, Michigan, 5 1961. Stockey, C. W. ”A Comparison of the Effectiveness of Group W I 2‘ 1 Thorndike, R. L. and Hagen, E. Measurement and Evaluation ip Education. New York: Wiley and Sons, 1961. APPENDIX A SPECIAL BIBLIOGRAPHY FOR CHAPTER II BIBLIOGRAPHY Ackerman, Nathan. "Group Psychotherapy with a Mixed Group of Adolescents," International Journal Psychotherapy, V (1955), 249-260. Boone , W. C. and Foley, W. J. "The Transition Stage in ) Group Counseling in Terms of Congruence Theory,” Journal pf Counseling Psychology, X (1963), 136-138. Cohn 3., Ohlsen, M., and Proff, F. "Roles Played by Adolescents in an Unproductive Counseling Group," Personnel and Guidance Journal, XXXVIII (1960), 724— 73l- Cohn, B., and Sniffen, A. Mead. "A Report of Group Coun- seling,” Personnel and Guidance Journal, XL (1962), 133—138. Crawford, R. E. ”Group Therapy for Seven Junior High Boys,” Clearing House, XXVIII (1953), 235-238. Pcrd, Donald. ”Group and Individual Counseling in Modi— fying Behavior," Personnel and Guidance Journal, XL (1962), 770-773. Harris, W. "A Beginning Counselor's Experience with Group Counselirg,” The School Counselor, XIII (1965), 46—50 Haskell, Martin, ”Role—training and Job Placement of Ado— lescent Delinquents,” Group Psychotherapy, XII (1959); 250-257. Kean, Charles. "Some Role—playing Experiments with High School Students,” Group Psychotherapy, VI (1954), 256-265. McManus, J. E. ”Group Techniques in a Child Welfare Agency," Group Psychotherapy, XV (1962), 63—68. Nccle, F., Ohlsen, M., and Proff, F. "A Method for the Quantification of Psychotherapeutic Interaction in Counseling Groups,” Journal_p£ Counseling Psychology, XII: (1961), 54—61. 8O _ 81 Ohlsen, M., and Oelke, Merritt. "An Evaluation of Topics in Group Counseling," Journal pf Clinical Psychology, XVIII (1962), 317-322. Parrish, Marguerite. "Group Techniques with Teen-age Emotionally Disturbed Girls," Group Psychotherapy, XIV (1961), 20-25. Rappooort, David. ”Small Group Counseling,” The School Counselor, XIV (1966), 116—118. Robinson, Jane. "Venture in Group Therapy on Senior High , School Level, Journal of National Association of ‘ Deans of Women, W11 (11953), 25- 28. Woal, Theodore. "A Project in Group Counseling in a Junior High School," Personnel and Guidance Journal, XLII (1964), 610-613. Yonge, K. A. and O'Connor, N. "Measurable Effects of Group Psychotherapy with Delinquents," Journal pf Mental Science, (1954), 944-952. APPENDIX B HAGGERTY-OLSON-WICKMAN BEHAVIOR RATING SCALE (abbreviated form) |JII|||1|II|I|I1II mammmoooc O>Hpnommw Gog: mvamph moanpz oap>pom spawn zpopppom o>pppomm< flQ30 was moaom maampocoo mppmmmw no>oz meomEpz ppommm ms meow no muonpo Op CH O>Hw on moon .w moonpoe momppm zppmmooo: moospoe so OHQHXOHM 3oz pmooow mm zawcpaaes mEOpmso 3oz pmHEHOMCOOGOC zpo> . Op gonad wepomcoo pmmoom Op 30pm ncaoppspm so: we oapfixoap 303 .w Hams hpo> hpfl pcmpameoo pcopoopo hpppOdem upozpzw mpmooo< «Happoommom xappmcpnppo p0 Hmoppflno pcmfimom wthHOSpsm pmooow on meow 30m .m wcpcpzpo>o condone pmommpCfl mp pmopopcp mapmmm new thmOHpso oopm6poch coopoonooc: monasmcoo mpmmpoch Hmsm: whmammem hampwm «pcomoppHoGH mmwcpgp CH pmopopcp camp on moon so pnoaompfloce on mH .m 3o RC mpp>ppow m>ppom wcoam weave phone nose msosm pommm happmcecpo “newnmspoq cow hump wo>flpom no hump zapwpcoe on mH .3 page» woman hsmcpopo wcpoooam 30am mawcflooooxm vooCflEumppw< Spas mxcHQB nzmpmwzam mHoEospxm wmcpxcpsp CH gonzo no 30am oz mH .m moOppom wGOH Umpoameoo Hep nonpocm pom coapcoppw mo0o on pmss hpopmsvoow xmmp pm mmox Op wspsp moo 6H0: Op oap< CH ooppomp< mccopp< Op pasopmppm Soap mQESh woocflmpmsm soapcoppw we: mH .m oceans oopomppmpm pHOmEHL ppmpb new :pcowosg moeooop CH UOQHOan o>paw mpcoom 0x638 one: zaamsmb zapcmstMM zaawscppcoo eoxmzw one; no oopomppmpm on mH .H zo ma