b _ ‘O-a s.- ‘mt--.--_.__... v‘-——.'_~‘ 21 o W I! ~j+§7-2;_:;2¢ ':- 1 \111 1L"? I1I1'\:‘\YVI|\E:$"‘ST LA“: . '3‘? :29. ‘VLI’J‘Z" u' J-Z- 4y;- _ 5' :4 k u- -’L'_- @5289: ;- “3' . ‘ ._._ . :4. - 4.;f. 1* .3 ‘3’ ' 'nJIs ..‘ \\ -.- ' i . fag. . I‘1111‘1‘V :“vlt‘c‘a ' H'éV'U 1"“ ~Uh .11 :11“; |1I,' \:.1;¢ .1-- .-1<~.-'w1' «1:§ -1 .1 .1' ‘ I «o .1 1 1 - 1: Unix???“ $33).“ 5:31.15“ 1 ; 1.6.: n+3 1.14? 1.113.111: 1' 1‘. 1x3 Né' {(1.15 ‘1‘ .' ‘ '11 ‘ 1-' 411M 1 '- H--"::‘14‘u':m* "'19..“ 1.1% :V 0 *\I 1114.3 1.3..Huflnz -fP.1‘ >' aim: 11 ' _:1\1...:1 '1 ’1'! 1.1/4‘ 11 ,~1~‘||vtu‘:u 1:311:51. Hm? . .3:'.\'r<:\u>' .\ a? 1 ' 1:11:11'3 3"111'1' 11111111111111“ 1 - {:‘Lkef:1.o‘.;. \1‘;V1'1' l’t3- 51:11.21 ‘13:}: ""1: ‘k;' 'l - 1 I 1 w ~. 21 :L‘ " : "tax '1.‘.. 1‘ :41 :' VL‘W :‘ 13113111123 (GUI; " .“ \ ‘ 1 1 1.11fi‘1." . 1 1 ' n 1'1", -. J‘Aflrfi'glékfi mel av." 1 1- .1 .1 :‘iz‘1‘x‘..‘ q 131%,. 1":1" 1 I 3 (11:1 1: 1.1 1 '1‘1‘3‘1‘(11Q':_11¥§V'° .\1‘ :3‘ ' 3' 11.31%: ‘éa' M 3m: 1:133:51” «51 "‘1111 1:117‘ , 11_ '11é:\.'.,1 311,21. , '\‘~ '1'." '11 1 {5 MK}: 1'. NR l 11 1 1 {$11111 +1: ‘1111111111111’111111111‘11:v%:31§8§5‘&,11. “:13?“ LE ' 131111 ‘r,\ 11 1' 1g "1.? _ . 1 3'1': ‘- I. “111111111 ‘QV‘1$I'I"".‘1 .1111”, 1, “$11 9111‘ “NJ: 1 1‘: 1“" ' " W 1. . (,‘fl 1 3'. 11.1.1.) I‘V“ ' 1.;111J'I1 1:111 ”1:111 11-11 A”: 1‘ 1. y :3 :11: 'I 3-5112; | vfl ‘0- '1 I 1 _111.: 3 ' \ :11 , , 91-. '. 1 ,1 11¢ -| I Up .'. I 1 1+ ply“ ’ T‘ 11:: 'Té.m-i‘1 u-V‘ 31;; 111W”? "U‘hfiffila I". "“ «‘ I ’- 11" 1 1|" ”myltnkflfihl‘rh. ,J!‘ m2, “5;, , fi?’ ‘ This is to certify that the thesis entitled LEADER INTERVENTIONS AND THEIR CONSEQUENCES IN THE INTERPERSONAL PROCESS LABORATORY - A HUMAN RELATIONS TRAINING GROUP presented by Felice Schulman-Marcus has been accepted towards fulfillment of the requirements for Doctor of Philosophy degree in Educational Psychology 465wa Major professor Date November 9, 1978 0-7 639 -.,_~_-/‘- {fl ‘ “an mz‘m7.rzur:v.~. .3_ 1!de Millimw' t ' ‘ A w, A this“ Rib-3 ,: . ,. p.‘ - ll: ‘ 5‘ y‘- ' Ki:‘~‘\igeu A , in partial rultjztgt » _. . ' in tr.» , g 1%; "pf ‘ WIDE. 0?" 31110311151 ' ”is LEADER INTERVENTIONS AND THEIR CONSEQUENCES IN THE INTERPERSONAL PROCESS LABORATORY — A HUMAN RELATIONS TRAINING GROUP By Felice Schulman-Marcus A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Educational Psychology 1978 (Ea/Zfiéifl’fiiz_ ABSTRACT LEADER INTERVENTIONS AND THEIR CONSEQUENCES IN THE INTERPERSONAL PROCESS LABORATORY - A HUMAN RELATIONS TRAINING GROUP By Felice Schulman-Marcus This study examines the characteristics of effective and ineffec- tive leader intervention behavior at certain critical points in a human relations training group, the Interpersonal Process Laboratory. It also examines the utility of several new concepts and techniques used to study leader intervention behavior: the critical incident, the intervention episode, and the stimulated recall technique. Effectiveness is defined in terms of student and observer perceptions and changes in student behavior/attitudes. The Interpersonal Process Laboratory differs from many human re- lations training groups because it is objective based, graded, and required course-work. Students are expected to master eight interpersonal com- munication Skills. The group leader has the responsibility of helping students attain and demonstrate the skills, and evaluates student per- formance on a Pass-Incomplete mastery model. The intervention behavior of four volunteer leaders was studied, each facilitating one group of 14-17 students. Intervention behavior was studied in relation to ten critical incidents. These are events occurring throughout the group life that involve issues important to the successful functioning of the group, e.g., fear of self-disclosure. A new unit of analysis was developed, the intervention episode. Perceived effectiveness was determined by analyzing student self- report data and observer ratings of the intervention's impact. Self- report data on student perceptions of leader behavior and the Laboratory were gathered by administering post-session questionnaires. A stimulated recall technique was sometimes used that asked students to react to a tape recording of a past episode to get direct feedback. An observer re- corded nonverbal behavior and student reactions. The interventions being studied aimed at changing students' detrimental behaviors or attitudes. Data were gathered to see if the interventions were successful and desired changes occurred. Intervention episodes were categorized as Perceived Effective, Perceived Ineffective or Neutral. Further breakdowns focused on the immediate and long-term impact of the intervention, and the effects of two modifying variables studied: group Stage and the critical incident. All sessions were tape recorded. Written transcriptions were made of each intervention episode. Complex quantitative analyses of intervention episodes in various categories were performed using a modi- fied version of the Issacson Scale of Interaction Analysis (1976) and other variables. The quantitative analyses indicated that perceptions of leader effectiveness depended upon adequate amounts of caring and cognition. Failure to exhibit caring behaviors was strongly associated with perceived ineffectiveness. Leaders also provided less cognitive information during ineffective episodes. The group stage and critical incident had little effect on the nature of perceived effectiveness or ineffectiveness. The intervention episode, critical incident, and stimulated recall technique all proved useful methods of studying leader behavior. Further examination of the data based on clinical impressions led to the formulation of a descriptive model of effective leader inter- vention behavior which incorporated the original findings. The model is composed of six categories of behavior called the Six C's of Effective- ness: Caring, Cognition, Clarity, Communication, Consequences, and Closure. Caring deals with the leader's verbal and nonverbal behavior which reflects a genuine desire to understand and assist students. Cog- nitive behaviors help students conceptualize and attach meaning to indi- vidual and group experiences. Clarity refers to the leader's verbal ability to clearly communicate. Communication behaviors increase the personal nature and the amount of communication. Consequences deals with behaviors which ensure students are presented with the negative consequences of their behavior/attitudes. Closure involves behaviors which make sure the central cognitive and affective needs of students are adequately addressed. Behavior in the first four categories leads to perceived effectiveness while all six categories result in actual changes in behavior/attitudes. The model has implications for training group leaders. It also suggests new methods of teaching communication skills. ACKNOHLEDGEMENTS I would like to acknowledge my gratitude to two special people, my chairman, Dr. Joe Byers, and my husband, Randy Marcus. Joe Byers has been a trusted friend for many years now. His untiring efforts and kindness deserve more thank you's than I can put here. Most importantly, he realized that a dissertation is done by a person, not a student, so that he gave me the understanding I needed to undertake and finish this project. My husband Randy provided me with the anotional and intellec- tual support I needed to finish my task despite many obstacles. His patience, unfailing humor, and love did more than get me through; it kept me sane and gave me a reason to wake up in the morning. I am also grateful to my committee members who showed patience and conviction throughout this process. Thanks to Dr. Cyril Warby, who had no reason to give me his time and energy, but did with caring and integrity, and to Dr. John Lopis, who has given me genuine warmth and wisdom at many a moment in my graduate career. He has contributed to my being, not just this project. Lastly, I would like to thank Dr. Lee Shulman, a man I respect, enjoy and admire. He was the first to make learning a truly exciting experience for me and he will always hold a special place in my thoughts. Above all, I am indebted to the group leaders and students who allowed me to come into their groups and study them. They gave much more than they received, but I hope this dissertation will make a con- tribution worthy of their generosity and helpfulness. Thank you Butch, Jim, Dick and John! 11 A I also want to express my appreciation to several people who helped me along the way, Randy Issacson, Michael Radke, Janet Shroyer, and Silvia and Victor Marcus. Finally, I want to thank my parents for the years of love and understanding that made it possible for me to reach this point. They were there before everyone else and have always re- mained a source of comfort and assistance. iii TABLE OF CONTENTS LIST OF TABLES . . . . . . . . . . . . . . LIST OF FIGURES . . . . . . . . . . . . . . Chapter I. INTRODUCTION . . . . . . . . . . . . II. III. Purpose and Rationale of the Study General Approach Further Description of Major Methodological REVIEW OF THE LITERATURE . . . . . . what Should the Leader Do? . . . How and When Should a Leader Intervene? . Models for Studying Intervention Behavior . Resistance and Support for Research on Effective Trainer Interventions . . . . . The Trainer and Group Process . . Quality of the Research . . . . . . Conclusions . . . . . . . . . . . METHODOLOGY . . . . . . . . . . . . Program Description . Characteristics of the Sample and Sampling Techniques Instruments and Observation Techniques used for Data Collection . . . Description of Perceived and Real Impact Measures . Description of the Analysis of Intervention Research Questions . . . . . . . . Statistical Procedures . . . . . . RESULTS . . . . . . . . . . . . . . Descriptive Statistics . . . . . . Changes in the Methodology . . . . Research Questions . Foreword on Presentation of Results Results for Questions on Methodology Results for Questions Pertaining to the Episodes . . . Results for Questions on the Immediate Consequences of Intervention Episodes and Long-Range Behavioral/ 0 Episodes Components Characteristics of Perceived Effective and Ineffective Intervention Attitudinal Changes . . . . . . . . . . . . . . . . . iv vi ix Chapter V. DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . Review of Findings for Research Questions . . ..... Qualitative Data . . . . . . . . . . . . . . . . . Discussion of the Methodology . . . . . . . . Implications for Research on Leader Intervention Behavior . . . . . . . . . . . . . . . . . . . . . . VI. SUMMARY AND CONCLUSIONS . . . . . . . . . . . . . . . The Six C's of Effectiveness . . . . . . . . . . . . The Effects of Modifying Variables . . . . . . Methods for Studying Leader Intervention Behavior . . . Implications of the Results . . . . . . . . APPENDIX A - Student Forms . . . . . . . . . . . . . APPENDIX B - Data Analysis Forms and Materials ...... APPENDIX C - Data Analysis Tables and Sample Transcript . . REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . 160 160 170 192 200 204 205 210 211 212 218 222 233 282 10. 11. 12. 14. 15. 16. 17. 18. LIST OF TABLES Differences between the IPL and T-group Leaders . Relationship between Trainer Type and the Amount of Emphasis on Trainer Functions . . . . . . . . . . . . Leader Type and Outcome . . . Deacriptive Variables for the Four Groups Factor Structure and Loadings for Peer Perception Data, Time 1 and Time 2 . . . . . . . . . . . . Group 1 - Student Types at Time 1 and Time 2 . . . . . . . . Group 2 - Student Types at Time 1 and Time 2 Group 3 Student Types at Time 1 and Time 2 . . . . . . Group 4 - Student Types at Time 1 and Time 2 Frequency of Intervention Episodes by Critical Incident and Group Leader . . . . . . . . . . . . . . . . . . . . . . Availability of PSQ Data for Sessions with Intervention Episodes by Group Leader, in Percentages Observer Ratings of the Impact of Each Intervention Episode on Group Members . . . . . . . . . . . . . . Mean Scores and Results of T-test Comparisons of Parallel Questions on Stimulated Recall and Post—Session Question— naires for Nat's Group . . . . . . . . . . . . . . . . . . Mean Scores and Results of T—test Comparisons of Parallel Questions on Stimulated Recall and Post-Session Question- naires for Hank's Group . . . . . . . . . . . . . . . . . Commonalities Across Intervention Episodes for Questions 1 - 3, by Major Variables . . . . . . . . . . . . . . . . Commonalities Across Intervention Episodes for Questions 4 and 5, by Major Variables . . . . . . . . . . . . . . Commonalities Across Intervention Episodes for Questions 6 and 7, by Major Variables . . . . . . . . . . . . . . . Commonalities Across Intervention Episodes for Questions 8 and 9, by Major Variables . . . . . . . . . . = . . . . . vi 10 58 59 7O 75 76 76 76 76 92 94 96 107 111 120 124 127 19. 21. 22. 23. 24. C1. CZ. C3. 04. CS. C6. C7. C8. Commonalities Across Intervention Episodes for Question 10, by Major Variables . . . . . . . . . . . . . . . . . . . . Commonalities Across Intervention Episodes for Question 11, by Major Variables . . . . . . . . . . . . . . . . . . . Summary of Findings for Questions on Perceived Effective and Ineffective Intervention Episodes . . . . . . . . . Observer thations and Ratings of the Immediate Consequences and Long-Range Impact of Selected Intervention Episodes (bomonalities Across Intervention Episodes for Questions 12 and 13, by Major Variables . . . . . . . . . . . . . . . Commonalities Across Intervention Episodes for Questions 14 and 15, by Major Variables . . . . . . . . . . . . . . . Comparison of Intervention Episodes Perceived as Effective based on Post-session Questionnaire Data Only - Question 1 Comparison of Intervention Episodes Perceived as Effective based on Post-Session Questionnaire Data and Observer Ratings - Question 2 . . . . . Comparison of Intervention Episodes Perceived as Effective Lasting Over Five Minutes based on Post-Session Question- naire Data and Observer Ratings - Question 3 . . . . . . . Comparison of Intervention Episodes Perceived as Effective based on Post-Session Questionnaire Data and Observer Ratings for the Critical Incident "Non-Participation" - Question 4 . . . . . . . . . . . . Comparison of Intervention Episodes Perceived as Effective based on Post- Session Questionnaire Data and Observer Ratings for the Critical Incident "Resistance/Hostility" - Question 5 . . . . . . . . . . . . . . . . . . . . . . . Comparison of Intervention Episodes Perceived as Effective based on Post-Session Questionnaire Data and Observer Ratings for Group Stage 1 — Question 6 . . . . . . . Comparison of Intervention Episodes Perceived as Effective based on Post-Session Questionnaire Data and Observer Ratings for Group Stage 2 - Question 7 . Comparison of Intervention Episodes Perceived as Ineffective based on Post-Session Questionnaire Data only - Question 8 132 140 144 154 158 233 234 236 237 239 240 C10. 011. C12. C13. C14. 015. min-.12. . Comparison of Intervention Episodes Perceived as Ineffective based on Post-Session Questionnaire Data and Observer Ratings - Question 9 . . . . . . . . . . . . . . . . . . . 241 Comparison of Intervention Episodes Perceived as Ineffective based on Post-Session Questionnaire Data and Observer Ratings for the Critical Incident "Non-Participation" - Question 10 . . . . . . . . . . . . . . . . . . . . . . . . 242 Comparison of Intervention Episodes Perceived as Ineffective based on Post-Session Questionnaire Data and Observer Ratings for Group Stage 1 — Question 11 . . . . . . . . . . 243 Comparison of Intervention Episodes with POsitive Immediate Consequences based on Observer Ratings - Question 12 . . . 244 Comparison of Intervention Episodes with Positive Immediate Consequences based on Observer Ratings for the Critical Incident "Non-Participation" - Question 13 . . . . . . . . 245 Comparison of Intervention Episodes Perceived as Effective based on Post-Session Questionnaire Data and Observer Ratings which also lead to Long-Range Behavior/Attitudinal Changes - Question 14 . . . . . . . . . . . . . . . . . . . 246 Comparison of Intervention Episodes Perceived as Effective based on Post- Session Questionnaire Data and Observer Ratings which did not result in Long— Range Behavior/ Attitudinal Changes - Question 15 . . . . . . . . . 247 viii LIST OF FIGURES Overall Conceptualization of the Study and Its Component Parts....................... 29 CHAPTER I INTRODUCTION Over the last twenty years there has been a tremendous growth in the use of human relations training groups in the fields of education, business and the health professions. The group movement exploded onto the American scene in the 19608 with the emphasis on humanistic ideals. It seemed a perfect answer to the calls to develop heightened awareness, interpersonal competence, and self-actualization. Group encounter, sensitivity training, and marathon became common words in the home as well as in psychology courses. Though there was a good deal of fadism in the rising interest in group experiences, behind it lay a lasting and deep-rooted realization that understanding of self and effective inter- personal communication skills are necessary in our complex, stressful society. Education was also swept into the Zeitgeist of the 19603. There was discussion on the need to educate the whole child; to meet his or her affective as well as intellectual needs. Schools had to now do more than impart information; they had to give the child personal and social skills for living, which included coming to understand self and others. This was not a new idea, but for the first time it began to weave itself into the standard fabric of education. There was also the cry to human- ize education, to make the school a place where feelings, creativity and independent strivings could survive. Schools were described as cold, unfeeling, harsh places where intellectual abilities and self-esteem were 1 ‘ smothered instead of fostered. Open classrooms, alternative programs, individualized learning, elective courses, and importantly, affective education were all ideas proposed to make the schools responsive to the needs and desires of students. Teachers had to be competent in the areas of interpersonal com- munication and group dynamics in order to deliver affective education programs and be responsive to students' affective and cognitive needs. They also had to be—aware of self, especially their values, since these colored their interactions with students. This meant career preparation programs were needed to give these skills and abilities to prospective teachers. Educators turned to the group movement as the most visible model of how to teach these skills, i.e., participation in a human rela- tions training group. Consequently the first groups in teacher education were patterned after so-called sensitivity or T-groups which originated from the National Training Laboratory at Bethel, Maine (Bradford, Gibb & Benne, 1964). This is evidenced by the fact that most of the reported literature on groups in education is still listed in the journals under the heading of "sensitivity" groups. The growth of human relations or sensitivity training in schools of education is documented by Issacson (1976). He found over twenty schools of education that reported some type of training in this area. As he noted, the number is probably higher since the figure only reflects reported usage. Along with the rise in human relations training in teached edu- cation programs, there has been increased questioning of the utility of this training and its safety. These are important issues, particularly when students are required to participate and are graded on their per- formance. A failing grade can halt a student's progress through his or her program of study, and continued failure can mean the student is pre- vented from entering the field. Concerns over utility stem from the paucity of research to validate that this training results in improved teacher and, ultimately, pupil performance. Deciding to use this tech- nique in spite of this lack of validation must be balanced against studies that show how participants can suffer lasting emotional problems from these experiences (Hartley, Roback & Abramowitz, 1976; Lieberman, Yalom & Miles, 1973). One of the primary concerns of those who question the use of human relations training groups is the competency of the leaders. Group leaders are usually empowered with a great deal of real and psychological power in relation to participants. A major study of encounter groups found that leadership style can have a profound positive or negative im- pact (Lieberman et al., 1973). Yet very little is actually known about the characteristics of effective leadership behavior in human relations training groups. This is somewhat ironic given that most writers in the field feel the leader is the central factor in bringing about changes in group members (Lieberman, 1976). The dearth of information has and still does create difficulties in the screening, hiring and training of competent professionals. Relying on the leader's past experiences in leading groups is insufficient because leaders may have developed their skills in personal growth or therapy groups with populations and con- straints that are different from those faced in academic settings. Even if the leader's experiences are similar, there are limited methods and criteria for actually measuring his or her abilities. There are many aspects of human relations training in career preparation programs that deserve attention. As suggested, one of the fi “f" key areas warranting investigation is determining the characteristics of effective and ineffective leader behavior. The research project reported here is a step toward meeting this need. Purpose and Rationale for the Study The nature of an intervention is of paramount importance to both the understanding and control of group growth and development. If the fact is accepted that there is a large measure of regu- larity and consistency among groups over a period of time and if there are identifiable classes of responses available to the group leader, then an attempt to gather and systematize these incidents and apply appropriate modes of response would seem to be a necessity. (Cohen & Smith, 1971, p. 105) While we acknowledge that there is more than one set of responses to a given critical incident, we nevertheless assert that there are certain preferred alternatives that can be evaluated on the basis of empirical observations of their effects upon the group. (Cohen & Smith, 1971, p. 104) This study is based on the beliefs that leader interventions do have differential effects, and that research can and should proceed to define the characteristics of interventions that are facilitative and inhibitory in nature. It examines the characteristics of effective and ineffective interventions at certain critical points in a specialized human relations training group, the Interpersonal Process Laboratory (IPL). There has been a paucity of research on this particular type group with its unusual characteristics of being objective based, graded, and required coursework for teacher trainees. Such research is needed because of the growth of similar group experiences in career training programs. The required nature of these groups makes it especially im- portant to provide the most effective leaders possible. As already dis- cussed, there is a lack of substantive empirical data on the nature of effective leadership behavior in human relations training groups. The purpose of this study is to examine leader interventions at certain key A points in the IPL group to determine some of the variables associated with positive and negative impact. The literature on groups suffers from a lack of carefully designed studies on the finpact of trainer interventions. Though the importance of trainer or leader comments is constantly discussed, little work has been done to determine which aspects of the trainer's interventions are facilitative (Bradford et al., 1964; Cohen & Smith, 1971; Lakin, 1972; Lippitt, 1971; Psathas & Hardert, 1966). Many writers have advocated this type of research but few studies have been attempted because of the difficult measurement problems involved (Campbell & Dunnette, 1968; Lundgren, 1971; O'Day, 1968; Psathas & Hardert, 1966). The research that has been done has focused on the T-group leader's behavior. The only notable exception is the work on encounter group leaders by Lieber- man, Yalom and Miles (1973). There were thought to be problems in gener- alizing the results of these studies to the IPL leader because there are significant differences between the IPL and the encounter or T-group, differences that are reflected in the leader's role in each. An under- standing of these differences is prerequisite to appreciating the need for this study. The organizational and structural components of the IPL will be described so that they can be compared to those of the average T-group. The Interpersonal Process Laboratory (IPL) is somewhat unique in the general human relations training movement. First, it is objective based, that is, there are a set of eight objectives that students are expected to achieve in order to reach mastery level in the course. These objectives involve interpersonal communication skills and can be briefly summarized as follows: ‘ 1. Interaction Analysis - the ability to assess if a verbal message is primarily affective or cognitive and to respond appropriately 2. Self-description - the ability to describe one's immediate feelings and thoughts as they relate to the on-going process 3. Active Listening - the ability to execute the component parts of this skill such as maintaining focus, paraphrasing, tenta- tive sharing, and so on, in a manner that assists the sender in exploring his/her feelings and ideas 4. Questioning - the ability to ask clarifying and exploratory questions related to the sender's feelings, ideas and non- verbal behavior 5. Feedback - the ability to give responsible positive and nega- tive feedback according to the criteria outlined in the course 6. Observation Skill - the ability to verbalize the perceived meanings and specific attributes of nonverbal behavior rele- vant to interactions in the group 7. Values Awareness - the ability to verbalize and explore one's belief system and the degree of consistency or inconsistency between stated beliefs and actual behaviors demonstrated 8. Process Model - the ability to use the process model (Assess- ment, Goal Setting, Strategies, Evaluation) in relation to one's own interpersonal growth The IPL has two other features that distinguish it from encounter or T-groups: it is a required part of a career training program and par- ticipants are graded on a mastery system. Specifically, it is part of the curriculum for teacher trainees at Michigan State University. v ' v \ , . 0152‘ “L 7"?" Grading on the mastery system means that the student either receives a Pass indicating acceptable competency with the skills, or an Incomplete, which requires the student to remain in the course. A more complete picture of the IPL is necessary to fully understand the nature of the experience. Typically a group consists of 15-18 college sophomores and juniors who sign up for the IPL as a required part of the teacher edu- cation curriculum. This is the initial course in their teacher train- ing program and few have past or concomitant teaching experience. Stu- daus are told they will have to learn and exhibit certain interpersonal communication skills in the IPL in order to receive a passing grade. The group meets five hours each week in a two- and a three-hour time block over a period of ten weeks or one term. A teaching assistant (TA) is assigned as facilitator for each group. It is his responsibility to present the skills and lead the participants through a varied series of discussions aimed at attaining and demonstrating the skills. An effort is made to develop an attitude toward the potential value of the skills in the classroom, the transfer situation of interest. At the same time students are encouraged to incorporate the skills into their general manner of functioning so that they can maintain them until the tflme they become classroom teachers. Acceptable student competence in the skills is determined by the TA with the assistance of informal and formal peer and self-evaluations. Twice during the term the TA gives each student a written evaluation on their progress toward mastery. As the brief description reveals, much of the leader's role is defined by environmental factors and his position as teacher and evaluator. This somewhat sterile representation of the IPL is misleading since the actual process is far from that. In reality it is character- ized by considerable enthusiasm, good feeling and learning as revealed by process observations and end-term evaluations. An overwhelming major- ity of the students (75-80%) report that retrospectively they would choose to take the IPL even if it were not required. Other data indicate a strong desire to participate in an advanced IPL or similar type group and a positive attitude toward the inclusion of such experiences in teacher training programs. Contrast the IPL to the average T-group and the leader's role as described by Campbell and Dunnette (1968): Thus, the T-group learning experience has as its focal point the small, unstructured, face-to-face group, usually consisting of 10-15 people. Typically, no activities or topics for discussion are planned. A trainer is usually present, but he does not accept, in fact he overtly rejects, the leadership role. The participants are to discuss themselves and the way they portray themselves in the group. . . . the focus is on the "here and now" . . . cogni- tive aspects of problems are ancillary to [an] affect-laden orien- tation. Focusing on the here and now is facilitated by the train- er's abdication of the leadership role and his lack of responsive- ness to the status symbols brought to the group by the participants. . . . Frequently, the trainer merely specifies the length of time the group will be meeting and that the major concern is with seek- ing to understand one's own and others' behavior. He then falls silent or otherwise refuses further guidance. (p. 76) Typically, hostility and frustration snerge in this setting as the group members try to determine their purpose and the norms they will be operating under. The behaviors exhibited by the members in this proc- ess become the initial topics for discussion. The trainer serves as a catalyst, reflecting the group dynamics and occasionally posing questions for the manhers to consider. Via this process the group comes to under- stand the nature of the experience and to learn about themselves, their behavior, feelings, perceptions of self and others. Issues discussed usually center on the need for structure, feelings toward authority, ways of decision making and, significantly, feelings toward the leader. As this description reveals, the role of the leader in the IPL and the T-group is notably different. These differences are highlighted in Table 1. The differences outlined in Table l are even more pronounced in the encounter group situation because the objectives are looser and the leader's role is less defined. The encounter group emphasizes experience for experience's sake and sensation awareness. A wider range of behavior is considered acceptable and exploration can lead into areas considered outside the content boundaries of the IPL. In addition, such groups are rarely required or related to on-the-job training or career preparation. On a continuum of similarity the T-group would be much closer to the IPL than the average encounter group. It would be a misrepresentation to portray the T-group and the IPL as totally different in nature. There are basic, important similari- ties such as the emphasis on the here and now, the general nature of the objectives, the focus on interpersonal feedback and communication skills, and the content for analysis being the behaviors generated by group mem- bers as they interact. If these similarities are underplayed it is to emphasize the seeming limitations of generalizing research done on the T-group leader to the IPL facilitator. In fact, it was the differential functioning of the leaders in carrying out the similar intent behind such groups that necessitated this study. The importance of doing research on the IPL leader is accentuated if one considers the growth of such experiences in career training pro- grams. Because of their required nature, participants are often prevented A 1| 2| Ilmuclal I Ills-I ll'o~. -‘III\ I... . III! .8 Fmt'h‘)‘ “ill.“ tairll,.!-v.‘!‘.~ .N. ‘3“ tr t i - uhmh {'5‘ 10 .mouuwamnmuu on so owuoasocx mo zoom on ma wanna use .msoom aoauvaonoumn ..w.o .mwoaovonuoa onu mo mocwmdoo mam casua3 wnfixuoa was name name use» ouamew cu hacwma ma oHou m.uommoH any .vuoaou o>Huuw on one muonEoS mono namom no mo>wuoonno mo use AHmu0m>w£onncosv moswwom manhood m mm: panama maoumnH och .hummmm Hmowmofionumma Snagxma mo meowuamcoo monouo cu wanfimmoa mm o>mumsam>oeoa no on so monowusmo mum mnovmoa macaw-H nosw sH .oHuuHH mocumwa o>mn ou mason on: panama m cu anewuocmm uuo>o mam manna cu mason onu o>m£ no: snow omHm o: .=w>aw sum muuoaou amouwoun no mommuw oz .owsom Hm5u0m has cw woumsam>o so no: m“ mucosa macaw-H och .ouswww hufiuorusm mama umHHaHm has no nonomou m :mnu uoausu conuoa commence a ma uo>aoouom panama osu o>o£ cu m“ Sam ofiH .ousuosuum can :oHumau0mcfl mo um>mw no cash Homowou as» aw ammo wagon Scum manned: uno> noun ou mxuo3 o: .=o«uwmon was mo osuuw> An umsomou m uo: mu novmofi msouwus was .oHOH aaamuovmoa o>fiuoouwvsoe a some» on .m:0wmmom macaw unauusuum hHHouwwHoomm or moon no: .hnufi nowanxo mo>uuoonno unu unwound uo: moon on .aouuumon huwuozusm so nose aw a“: wswuusn anew nuocaoa omu uco>mnn cu mofiuu use each nunmuovmos onu muoofiou mucosa macaw-H 05H .ouwuomum wow meoHufivcoo ovw>oun on was mHHme onu we mucus: can ozone owmwazoax o>fiuaawou mo zoos mamuuoo m uwamcmuu on use: osu moHHnaH many .oocumoH mum mono umau om macaw oau muumu -fiv on was “momma ou mum mucomsum unnu mw>wuuonno mo umm oawaomam a an: <9 05H .cammm uo>o omusoo onu oxen umsfi Ho>oH Hafixm ammuuoo a comma no: on 033 mucooSum .ahou sumo ooHBu mmoumoun ufionu so mucosa xumnowow :ou uuHuB ao>aw was hose on: .mmeo umaawou co>fiw use can mucomsum .vouusouo was umnu mafiaumoa mo unseen can so mucomsum onu movmuw new nouwSHm>o an m« d8 och .momusoo umoa mo coaumuaowuo o>auaswoo hsouan we» muuonou wHomuH QMH man no mfiamcowumawu usooSumauonomou nonhuoouwum ecu muuofiou <9 onu wo>o3o= .Emuwoum soHumoseo Homosou a ma mmmHu a ma ANH man madam uosomou a ma ¢H oak .oHou mafiMSuonuum .o>wuoouwo huo> m momma om .mcownmom emu mo ousuusuum onu mooaoov meanness» new Amomuouoxov moauweauom aw mwcwun .mueomsum emu ou moeuuoohno mam meow omu mo>am on .HommoH onu nu ma any mm>auowfino .s acuos~o>m .n Hosanna .N nqnnwouoofl .a Havana msouwna Hflfiflwfl «EH ooaua~«Aaauoa.oa a ounce muovoofl asoumna new AmH onu noozuun amonowuuuunnu.n Hamdn 1 1 .scwuosuwm macaw o c« wnfisuooa sass: mo moasmcmc oau cc mucoau o>ww mamsconoucomm couwc occwu uno>uouca was .uo>osom .ccmuom nonuosm mm macaw ocu ocfimusc once one mos» Ham on so>Hm ouo omonu wH .msucoufic macmmoa mucosa mo>Hw maoumu noomos macaw-H onH .uo>.:oc cu cowuoa mucunw no mean uom on on: o; omH< .on unumofl uaonu no stA onu you oHAHmsommou ouo muonaoz .waaauooa wow coHuHHHnHmocmc now csou>no mcoa mm: .och opaucouwvso: once can no omsmoon .uoomoH mocwmaa oak .msOwuso>uou=H a9 oH> co>flw on has on no oucowso>bcu u0u oocE caucmcwv m :H counowoum on has ucoucoo monasmou mass moafiu u< .mscuw m cm mewcuooa can wafinooou mo mxwmu onu cu couoaou mamawouma xoonuxou onu scum mumoo nacc cocoa camuuou uo>cu on on: m9 omH cuuxu .o .mowcumoa HonEoE new muwflwnwmucmmou Howuaoumnsm moESmom om .mo>wuoofipc oSu mo unoaawcuum chosen wswxuos ma macaw uoccofl msouwna omu ouso wcwxma cam ucoueoo o>auwewoc wswcuood you ocu wafiusomoum now oHnHueommow mu a9 oak muqawnuanommom .n avenue .EH «3 u: Hana-:83.“ a noacu vanadunooun.n ounmfl [lfl'l II: t- ‘— 12 from entering the profession involved if they are not able to success- fully complete their group experience, success being primarily evaluated by the group leader. This endows the leader with a great deal of power in relation to the sometimes unwilling participant, both psychologically and objectively. There is also resistance to such experiences by outside groups as pointed out by Kaplan and Sadock (1973): What is of concern is the growing tendency of relatively few ultra- ccnservatives to put great pressure on boards of education and school administrators to cut out of their curricula any innovative procedures involving dialogue within groups of students, especially if the dialogue or discussions allows freedom of expression of the students' personal feelings and fantasies. All such group discus- sions are now labeled sensitivity training by ultraccnservatives and then, by . . . implication become dangerous brainwashings, group criticism, self-criticism and invasion of the privacy of the individual. (p. 90) If such programs are to continue to flourish they need competent personnel and adequate procedures so that criticisms from outside groups can be counteracted or acknowledged and dealt with. There were also concerns in the selection and training of group leaders. Though the IPL type group is considerably different from the therapy or T-group, the leaders of IPL groups usually have their back- ground experience in such settings and need help to adjust to their new roles. This is typically done through modeling other experienced TAs and participation in training sessions on relevant issues. Unfortunately these practices had severe limitations because knowledge of the factors involved in effective leadership of the IPL was lacking. Until these factors were known there was little guarantee that the essential elements of leader behavior would be modeled and that training sessions would con- centrate on the most important issues and skills. A 13 There was and is an obvious need for refined research studies on the impact of leader interventions in human relations training groups. This study is valuable because it went beyond being a piece of action research to meet the immediate needs of the IPL; it also explored a method of examining leader interventions that has generalizability to other group- oriented programs. Though the group work literature is replete with state- ments about the importance of variation in leader interventions, few seri- ous studies have been done to study systematically the nature of inter- ventions and their hypothesized impact. This study was designed to remedy partially this situation by providing information on the characteristics and consequences of selected trainer interventions. General Approach Following Banet's (1974) contention that "not every grunt uttered by the psychotherapist is an intervention," it can be argued that not all TA interventions are of equal importance. Though there is merit in look- ing at all interventions, this study examines only select cases of inter- vention at points where they appear of special significance. The type of intervention under consideration is one that aims at modifying, revers- ing or otherwise changing some behavior or attitude. The points are marked by certain "critical incidents" that serve as cues that an intervention is called for. The incidents are seen as symptoms of significant under- lying issues which need to be addressed if the group is to successfully carry out its purpose. A sequential unit of analysis is employed, the intervention episode. Previously researchers looked at an intervention in two ways: emphasizing its function, e.g., to convey theory, show support; or the sentences that composed it, the verbal utterances of the leader (Benet, 1974; Cohen & A _nb- 14 Smith, 1976; Fiebert, 1968; O'Day, 1968). Both have limitations. By concentrating solely on the function of the intervention one loses the particulars that account for the variation among interventions with the same purpose. However, when the sentence is the unit the psychological import of the intervention can be lost. Neither allows the investigator to look at the intervention as a series of verbal remarks or moves that together account for the major impact of the intervention. The inter- vention episode is designed to do this. This sequence unit is defined by all the verbal exchanges that take place from the first TA utterance following detection of a critical incident to the point where the episode is implicitly or overtly resolved. Criteria for determining the terminal boundary are discussed at a later point in this chapter. The TA's verbal remarks during an intervention episode are studied using a modified version of a content analytic system developed by Issacson (1976) for the IPL. With this system the TA's remarks are placed into one of nine categories which roughly correspond to the communication skills taught in the IPL, e.g., active listening, positive responsible feedback. Once a remark is categorized, its emotional tone is also rated as either positive, neutral or negative. The content area of the remark is not ex- amined because this is seen as less important and it is already restricted since only interventions that pertain to certain topics are studied. The impact of an intervention is assessed in several ways as it is felt that a multiple dependent variable approach is best. First, an observer in each group rates the overall flmpact of the intervention epi- sode on the participants directly involved and the group in general. This is useful because the participants' nonverbal behavior can be taken into account. The observer is able to detect member hostility, comfort, ‘ 15 withdrawal and other emotions expressed nonverbally. This information would otherwise be lost since only audio-recordings of sessions are avail- able. This index suffers frcm all of the problems related to live ob- server ratings; therefore data collected in this manner is intended to serve as a secondary source to be used in interpreting other impact measures. Second, students are asked to fill out a post-session question- naire at the end of each session. This form contains questions in four areas: 1) student perceptions of TA behavior, 2) how they think others see the TA's behavior, 3) perceptions of the level of group unity, and 4) overall reactions to the IPL experience. Questions are designed to focus on the types of behavior TAs usually exhibit during intervention episodes. Member responses should therefore be a reflection of the way in which the TA handles an intervention episode during that session. Again there are limitations on the conclusions that can be drawn from these data since students may be reacting to aspects of the session other than the intervention episode. The third measure used involves a stimulated recall technique. Where possible an intervention episode is rerun for the group at the start of the next session. Participants are then asked to answer a series of questions about how they felt when the interaction originally occurred. This technique overcomes the limitations of the post-session question- naire because participants are responding directly to the intervention. The stimulated recall technique has been extensively studied by Regan (1969a, 1969b, 1972) and results indicate that this procedure greatly enhances the participants' ability to reinstate their feelings at the time of the interaction. 16 Finally, empirical data are gathered to assess the impact of the interventions on participant behavior. For example, if the intervention involved the need to self-disclose more, do the students involved now exhibit a higher rate of self-disclosure? Wherever possible the tape re- cordings are analyzed to determine if such behavioral changes follow inter- vention episodes. To guard against the confounding effects of extraneous variables, only the first three sessions following the intervention epi- sode are examined for changes. The assumption is that if the intervention did have an impact, the effects should show up quickly and changes that occurred in the next few sessions could more safely be attributed to the intervention itself. In summary, the impact of an intervention is studied in several ways, via perceived impact as determined by the observer and the partici- pants, and empirically in terms of actual behavioral changes that appear to be associated with the intervention. This is a more comprehensive ap- proach to investigating impact than has yet been tried. Two other modifying variables are studied to account for the fac- tors that contribute to the effectiveness of an intervention: the group stage and the type(s) of students involved. Whether an intervention occurs in the early, middle or late parts of the group life may influence how it should be carried out and the reactions of participants. Similarly, what might be seen as an effective intervention with one type of student, for example, a verbal, confident one, may not be perceived as such if the stu- dent is shy and withdrawn. Questionnaire data are gathered at the end of the third and seventh week to determine the perceptions students have of each other. This information is used to see if student reactions to TA behavior vary with the type of students involved in an intervention episode. 7"?— 17 Four TAs participated in this study, each facilitating one group. They are volunteers selected from a larger group of TAs to ensure varia- bility in style and effectiveness. All are reasonably experienced having led groups for at least two years. Variability in style is desirable so that a range of effective interventions can be examined. Further Description of Major Methodological Comppnents Critical Incidents1 In order to compare intervention patterns the situations or issues to which they are directed need to be comparable across groups. Cohen and Smith (1976) investigated such situations for the T-group and devel- oped the concept cf the critical incident. These incidents are said to occur in almost all T-groups with varying modifications and their effec- tive handling by the leader is felt to have a great significance for the productive growth of the group. They define a critical incident as follows: A critical incident is defined as the confrontation of a group leader with one or more members, in which an explicit or implicit opinion, decision, or action is demanded of the leader. It may also be observed conversation, a confrontation among members, an event taking place, or a period of silence. The essential property of a critical incident is that the phenomenon is judged hmpcrtant enough by a group leader to consider, consciously and explicitly, a decision to act in a way assumed to have an impgrtant impact on the group. This implies that the group leader is faced with a number of "choice points" or alternatives in both the content and style of possible responses. (p. 88, underlining mine) As with the T-group, critical incidents occur in the life of an IPL group that reflect the unique demands of that group situation. In 1Not to be confused with Flanagan's (1954) use of the term. Flanagan coined the term "critical incident technique" during World war 11 while studying the critical requirements of activities such as combat leadership and disorientation of pilots. ‘ M. 12“».- ' 18 identifying these incidents certain features of a critical incident were considered. 1. Implications - The proper handling of the incident by the leader is assumed to have substantial impact on the member(s) involved. It may also affect other group members via vicarious learning. If not handled correctly, nonfacilitative behaviors and/or feel- ings might ensue that would hamper the members from having a successful group experience. An effective intervention might also deflect or prevent negative behaviors or feelings from ever developing. 2. Relationship to Group Objectives - The incident involves issues that are significantly related to the objectives of the group. 3. Shared Meaning - The issues implied in the incident have rele- vance for many group members because of their close connection to the objectives and the nature of the group experience. 4. Frequency - The incident occurs frequently across groups. The critical incident as defined refers to the precipitating event that triggers an intervention episode. These incidents act as cues or signals that an intervention is called for. They are critical because they reflect underlying issues that need to be treated in a cor- rective manner by the intervention. Work was done to determine some of these critical incidents and underlying issues for the IPL. A questionnaire was developed (see appendix B) consisting of ten issues and possible corresponding critical incidents. The form was given to seven experienced IPL leaders, all of whom had been conducting IPL groups for several years. First they were asked to decide if the issues and incidents qualified as critical given the criteria outlined above. They were also asked to list any other issues they felt should be included. Finally they were to rank the issues in terms of their frequency of occur- rence across groups including in their ranking any issues they added. Based on these data the ten most frequently occurring critical incident issues were picked for this study.1 of corresponding critical incidents: Issues Involved l. Reluctance to Give Negative Feedback 2. Nonparticipation by Group Mamber(s) - Withdrawn, Shy Mbmber(s) 3. Failure to Own One's Feelings They are listed along with examples Critical Incidents discomfort expressed by members about giving it verbalizing that it means criti- cism and someone might get hurt expressing discomfort about group process right after negative feedback is given or trying to change the topic continually following their nega- tive feedback with a "but" state- ment and positive feedback reluctance of members to give opinions, feelings, especially in a large group feelings by some that they are doing all the work, that others are not contributing verbalizations by some of feeling pressured to speak or pressuring others who do not want to extended periods of silence in the group same members speak all the time while the rest remain fairly silent speaking for everyone, e.g., "This is how we feel" not using the term "I" but using we, one, they, you, people in general, etc. when discussing ideas and feelings blaming others for your feelings, e.g., "You make me angry" 1Henceforth the term critical incident will be used to refer to the critical incident issue, so that, for example, critical incident number 4 would be Resistance/Hostility. 4. Resistance/Hostility 5. Refusing to be Responsible for Changing Self 6. TA put in Authority Position 7. Nonfacilitative Emotional or Nonverbal Behavior 8. Overtalk/Inability to Facilitate Self-Closure 20 b I U1 l questioning the right of the TA to evaluate their skill usage claiming this is an artificial setting which is why they are not performing saying they do it with children which is where it counts feeling constantly evaluated/ana- lyzed attacking verbally the TA's com- ments, opinions comments such as "I've always been that way, I can't change," or "That's just the way I am" other statements to the effect that the person has little control over what they are, that their past or present environment prevents them from changing continually addressing comments to the TA rather than directing them to the entire group saying it is the TA's responsibil- ity to make sure things are con- structive, not theirs being overconcerned about what the TA has to say, his reactions and feedback nonverbal behaviors experienced negatively by group members - e.g., uninvolved, angry, grim cries or appears to be near crying whenever confronted exhibits aggressive actions when disagreed with member rejects or has difficulty making nonverbal contact with others frequent and inappropriate giggling when discussing ideas and feelings person monopolizes conversation, goes from one idea to the next Others become frustrated and con- fused person presents a view and then proceeds to elaborate and exhaust all possibilities for discussion 21 9. Failure to Stay with the 9 - continually trying to discuss Here and Now topics outside the group experi- ence, e.g., roommates, friends, sports continually rehashing past sessions without relating them to the present - continually telling how they use the skills outside the group 10. Expectations for the Group 10 - expressing fears about what might happen in the group - sharing rumors about negative ex- periences others have had in groups continually expressing desires to move the group more in the direc- tion of therapy or encounter This list does not include all the critical incidents that might occur, but it does present many of the more common ones confronting the leader. The Intervention Episode One of the major problems in the literature has been disagreement in defining an intervention. Though interventions have been defined for coding purposes, the rationale for most definitions seems to be based more on practicality than theory, i.e., sentence units; remarks between student comments (O'Day, 1968; Psathas & Hardert, 1966). Usually inter- ventions are defined in terms of their general purpose, e.g., to show support (Fiebert, 1968; Lakin, 1972; Lippitt, 1971). Characterizing the intended purpose of an intervention is useful but this does not clarify the attributes or dimensions along which various interventions with the same general function can vary. Some attenpts to do that have been made (Cohen & Smith, 1976; Mann, 1967; O'Day, 1968) but these authors have not established effective criteria for determining the boundaries of an intervention. For instance, O'Day defined the boundaries in terms of the “uninterrupted verbal intervention or comment made by the trainer A 22 between two student comments." He then goes on to break the intervention into sentence units for coding purposes. This method seems incomplete because there is no psychological or theoretical rationale for defining an intervention in this manner. As stated earlier, this study uses a sequence unit of analysis, the intervention episode. It is a "tactical sequence unit" as discussed by Dunkin and Biddle (1974): "tactical units concern the immediate give and take among classroom members as they concern themselves with but a single topic." This type of unit was chosen because it seems logical to view an intervention as a series of verbal moves that have cumulative impact. This view seems particularly applicable to the type of inter- vention being studied because of its corrective intent. The aim is to alter or modify a detrimental behavior or attitude which differentiates this type of intervention from others with a different intent, e.g., to clarify or direct. Since the intervention episode is a tactical unit it pertains to a single topic, i.e., one of the critical incident issues already pre- sented. It also meets the need for a sequence unit with reliably dis- tinct boundaries. To fully understand its boundaries it is necessary to discuss the concept of an episode as developed by Smith and Meux (1962). Smith and Meux developed the concept of an episode to study classroom interaction. They defined it as "one or more exchanges which comprise a completed verbal transaction between two or more speakers." It is topic-centered and has three phases which serve as its core. Vari- ations can occur within phases, especially the second one. The phases are as follows: 23 1. Opening Phase - verbal entry move in the form of "a remark or set of remarks (questions, assertions, etc.) signaling it will be followed by discussion, and setting the direction of that discussion" (p. 22). 2. Continuing Phase - "is made up of the remarks (verbal moves) which are: a) either replies or answers to questions; b) claims, comments or opinions; c) questions that sustain the entry under discussion; d) anomalous questions" (pp. 22-23). 3. Terminal Phase - Overt or Implicit a) Overt - "remarks which serve to cut off the flow of discus- sion" (p. 24). They could be initiated by the student or teacher. b) Implicit - "the episode is taken as terminated by the occur- rence of remarks which signal the opening of a new episode" (p. 24). The criteria Smith and Meux used to designate the opening and terminal phases are used in determining the boundaries of the interven- tion episode. The entry move in the opening phase begins with the first verbal utterance of the TA as he moves into the ongoing interaction at the point where he senses a critical incident. The terminal phase is redefined in terms of overt and implicit resolutions. Examples of an overt resolution might be a person saying he/she does not want to con- tinue or that the problem has been resolved. An implicit resolution results when the leader or a student prematurely redirects the focus to another student or topic. 24 Issacson's Scale of Interaction Analysis (ISIA) A modified version of this content analytic system is used to analyze the TA's verbal remarks during an intervention episode. This instrument was chosen because it was specifically developed for the IPL and it is proven to be valid and reliable. It focuses on the communi- cation skills being used by the subject(s). Another advantage of the ISIA is that it calls for low-inference judgments on the part of the rater. Interaction analysis requires that verbal interactions be broken down into segments which are then categorized. Issacson calls each of these segments an "event," defining it as "the shortest possible act that a trained observer can identify and record." Each event is given one of 33 possible codes depending on the type of communication skill being used, who is speaking and whether the domain is primarily affec- tive or cognitive. It is a multiple coding category system where each event is coded twice. As Issacson explains it, The codes (in most instances two codes per event) come from two category clusters; the first category cluster (cluster A) identi- fies who is speaking (either teacher, student or silence) and in what domain the speaker is talking (either cognitive or affec- tive). The second cluster (cluster B) identifies what particular communication skill the speaker is using. Every event is coded with one code from cluster A and one code from cluster B. . . . The categories are listed below according to clusters. Cluster A Cluster B - Silence or confusion l - Positive Feedback - Teacher - cognitive a. responsible - Teacher - affective b. irresponsible Active listening Elicits Information Directs or Suggests Offers Information Self-Description Negative Feedback a. responsible b. irresponsible (p. 203) - Student - cognitive - Student - affective wab-Io I n I \lO‘U‘bWN I 25 This system was designed to be used by a trained observer listen- ing to audio-tapes or making live observations. The rater is expected to identify and code events at a speed of one code every five seconds. No record is made of who is speaking except to designate if it is a stu- dent or the leader. The ISIA was modified to meet the demands of this study. Since only TA comments were of interest, it was not appropriate to use the criteria of one code every five seconds since this would also include student comments. It was also not acceptable to code each "event" since coding would be done from written transcriptions and this would usually result in one code per sentence. As discussed earlier, coding each sentence has practical advantages but the TA's overall strategy is dis- guised when the sentence is the unit of analysis. An event was there- fore redefined as "a sentence or group of sentences having a single code." Designating an event in this manner emphasizes the strategies taken by the TA and reduces the problem of having to decide where a sentence begins and ends for coding purposes. This is difficult to determine at times since speech contains many overlapping, disconnected, embedded and broken sentences. Using this definition the decision does not have to be made as often. Changes in the codes were also necessary. Though cluster B was used in its original form, modifications of cluster A were made. It was unnecessary to distinguish between student and leader comments since this was done when making the transcription. However, it still was im- portant to determine the domain as affective or cognitive, to be coded A or C. Thus each event receives a two-part code consisting of a letter, A or C, and a numerical code from cluster B, for example, A6; Cla; C5; A2. 26 The affective quality of each event is coded as mainly positive, neutral or negative. This is a rather subjective judgment based on the tone of the speaker and the way in which the message is delivered. General criteria are established to help in making these decisions with the guideline that when in doubt the event is coded as neutral. The criteria used are based on the work done by O'Day (1968) to study the affective tone of T-group leader intervention. They are as follows: Positive (P) - Expressions of liking, trust, praising - Expressions of satisfaction with member perform- ance, showing support for a member's actions or ideas - Apologizing for prior hostility, denying the intent to hurt to be hostile Neutral (Nu) - No clear affective message comes through Negative (Ne) - Expressions of anger, dislike, strong disapproval - Expressions of indifference, frustration, feelings of giving up on the individual(s) or group - Guilt- indficing behaviors, i. e., shaming, blaming - Attacking members, parrying members' suggestions in an obvious attempt to "win" Impact Measures Very little has been done to determine the impact of leader be- havior in human relations training groups. In large part this is due to the difficult problems of assessing the impact of an entire session, let alone a particular intervention within that session. When participants have been asked to rate the leader it is usually on global variables such as warmth, directiveness, or empathy. The researcher then infers the reasons why the leader received the ratings. This study takes a more focused approach to examining the impact of leader interventions. Several types of outcome measure have been developed that direct the respondent's attention toward specific leader behaviors and interventions. Information is also gathered on a more consistent basis than is typical of most studies. 27 In the past group members usually rated the leader once or twice during the group's life which forced them to make generalized reactions. In this case the participants are asked to make several ratings and an ob- server is constantly present in the group. Also three different outcome measures are used to compensate for the limitations of each. These measures have already been diSCussed earlier in this chapter (see pp.14-l6). Modifying Variables It seemed to be misleading to speak of a single effective inter- vention pattern since it might differ depending on situational variables. Not all variables are of equal weight, though, and it is probable that knowledge of the more significant ones is sufficient to guide the leader in deciding how to intervene. With these key variables in mind, one can discuss the nature of an effective intervention under certain conditions although other environmental variables are still unknown. It is hypothe- sized that group stage and the type of student(s) involved are two of the most crucial modifying variables. Interventions that fail to con- sider these should have less chance of being effective in terms of their perceived and real impact. Most authors in this field agree that groups go through a series of developmental stages where different issues are of primary concern (Bradford et al., 1964; Cohen & Smith, 1971; Fiebert, 1968; Lundgren, 1971). Though they differ on the number of stages and their exact nature, all seem to agree that groups go through some initial stage where problems of goal definition and self-disclosure are prominent, then a period of conflict and frustration where the leader is often attacked, and finally a period where the group seems to solidify and intensify constructive, goal-directed behavior. These stages represent the development as it 28 occurs in the T-group, but the idea of stages of development can be applied to any similar type group experience. It is probably that the leader's behavior is judged according to the dynamics operating during that stage of the group's development. For instance, a strong, assertive interven- tion in the early stages of the group could have negative effects on par- ticipant risk taking. Correspondingly, the same type of intervention during the eighth week of the term might be seen as highly effective and constructive. To deal with this, student reactions to an intervention episode are examined in light of the stage they occur in. Stages have been somewhat arbitrarily defined by dividing the number of sessions into thirds. Group member perceptions of an intervention can also be affected by the types of student involved. Three variables are being used in the definition of student type: activity level, verbalness, and personal security. At the end of each stage, approximately three and seven weeks, students are asked to fill out a brief questionnaire on every other stu- dent in the group. It contains eight 7-point semantic differential scales keyed to the three variables. The number of items and variables is limited since each student has to fill out the form twice on anywhere from twelve to sixteen other group members. Even with eight items, this comes to over 100 responses each administration, a somewhat time-consuming procedure. However, past experience indicates that the variables chosen have a significant impact on student perceptions of the "proper" approach to making an intervention. For example, students feel that the TA should take a tentative, gentle approach with peers seen as quiet and insecure. Figure 1 presents an overall conceptualization of this study and a summary of its component parts. An overview of the remaining chapters is also given to guide the reader. mmooo>wu noowmo oo>woouom so some on Han: soaumcHEuouss m .Aommv ouwmncofiumosv sowmmownumom no women n mass: m we mmmHo mosmmfl sausages was an evade IE“ unusuauum\muoH>mnon onu :H ovumououcfi mace ooOmHmo emu :H oo>~o>cw mauuouwo woonu cw voumououcw mace u woocoswmwcoo oumaooeaa um xooH ome : snowman cowuco>uoucw cm mcasoaaou macammom momma ou w=0wmmom wawooooum mounu we“ nummeou hp newcono Hosanna nwuum\fimuoH>m£mn you seed 1 ow>Ho>eH mamnow>aonH 29 muow>onon Honno> no Has: no Hmnuo>cos no momma mcwumu Honon u no>uomao monsoon: uuonan N meu uaoosuw\¢9 .N mafia Hence .H mommamcm Hmsoaufioo< .m o>fiumwoc uo Hmuusm: .o>aufimom n huwamsv o>auoowu< .m o>auficwoo no o>mu0ommm u sHman .N new: Hawxm :oHuMUHasfifioo wo umhh .H "moons oousu n mammfimc< eowuomuoucH no oamum somommmH mo scamno> ooamwooz .< "wean: uonHmcm on Haws u coausaomou ufiofifimxo Ho uonHme ou o>oE Hmnuo> umufiw Eouw wusooo umau Ham 1 UUOmHNN coauso>uoucu muumm uaosomaoo ousmoaonmaom ououuaqoom on suasunmsu\xamuuo>o uofi>memn Hmnuo>coz no HocOuuoam w>wumuHHwoomnoz cowufimom mufiuonus< cw usm <9 MHom wafiwsmso now oHnwmmommoM on Ou wcwmswom zuwfiaumom\oocmumfimom mwcaaoom m.oco sac cu ousamom muons»: mam .azmuessns I unease: macaw an couummaouuummcoz xoonooom o>wumwoz o>ww cu oucmuusaom d) n o In M N .H mucuofiosH Hoowuwuo mwfiHU OHog MO>HO>BH ovomw N .300th III-n co lull: uomfih lull: mmn SUHSS IIIII m— mnmafiu>wufiH cowucm>umucH uo>uonnb -nunn snowman» nun-n uauvflonu nouauuuu 30 muumm uaonomsou sum was hvSum ecu mo newumuafimaumoocoo Hamuo>o .H .wfim n no .N .H . owuum macaw .m hoaum now ammono onus mocha ucoosum know u oaks uaovsum .< $33 :5 was five: dam Du muuzmcm chanson one Hanson oouofissaum uoavaoo Haas .m~no~«o>u sons I fiasco“ vousHsEHum .e macaw may you macauwuoonuu .o~ zealoaonondm ecu suns mean on 393..» .a im — a- i _—.__._ 31 Chapter 2, Review of the Literature, discusses the research that has been done on human relations training group leaders, with a special focus on the studies that examine effective and ineffective intervention behavior. It reveals the need for further in-depth research in this area, a need this study is designed to meet. Chapter 3, Methodology, provides a further look at the Inter- personal Process Laboratory and describes the complex methodology used in this study. It includes a description of the sample, instrumentation, methods of analysis, the research questions, and statistical procedures. Chapter 4, Results, presents the findings from the quantitative analysis described in Chapter 3. It also includes a description of the modifications made in the methodology and the final set of research questions explored. Chapter 5, Discussion, begins with a summary of the findings from Chapter 4. It then outlines a new descriptive model of effective leader intervention behavior derived from clinical impressions and the results of the quantitative analysis. It also examines the utility of the methods used in this study. Chapter 6, Summary and Conclusions, reviews the major findings and methodology. It also examines the implications of the work for the training of group leaders and teacher trainees. .34, CHAPTER II REVIEW OF THE LITERATURE Theories of personal change in groups emphasize the relationship of the leader to participants. . . . In spite of fundamentally different conclusions about what leader inputs are crucial, all these theories agree on the centrality of the leader to the change process. It is he who sets up the learning experiences, who makes the interpretations or analyzes resistance, who establishes norms, who is the "model," and so forth. The specific content of the leader's actions and responsibilities may vary, but the underlying assumption is that the central factor in changing people is what the leader does or how he expresses himself. (Lieberman, 1976, pp. 232-33) One of the most widely held and unquestioned beliefs in the human relations training movement is that what the leader does or does not do will have an important impact on group process and participant learning. As Lieberman points out, many theories View the leader as central to the change process; that without him little or no learning would occur. Yet there has been little substantive research to support these claims and to examine how the leader's behavior affects the group. The result is that the group leader is given little information on how and when to in- tervene to promote group goals and prevent problems. It is only recently that investigators are beginning to realize the need for more systematic, in-depth studies of leader behavior and are developing the technologies needed to carry out this complicated research. Almost all of the research on human relations training group lead- ers has focused on encounter and T-group leaders. As pointed out in chap- ter 1, there may be problems in generalizing this research to the IPL leader, who has different demands placed on him. One of the secondary 32 33 goals of this study is to determine the extent to which the literature on encounter and T-group leaders is generalizable to the IPL leader. In order to do this, this chapter will present a review of the work done to determine effective and ineffective intervention behavior for encounter and T-group leaders. Unfortunately there have been few studies on the impact of leader intervention behavior. More often the literature pre- sents general guidelines and principles of intervening that leaders are advised to follow, taking as a given that doing so will have productive results. Though authors may not provide empirical proof to back up their suggestions, their validity cannot be automatically discounted since they are usually based on years of experience leading groups. Therefore these suggestions will also be examined to see what information they provide on the characteristics of effective and ineffective intervention behavior. Discussion of the literature on guidelines for intervening will be organized around three questions: what should the leader do, how should it be done, and when should he do it? From discussing guidelines, the chapter will continue with a review of several models for studying intervention behavior. The last part will concentrate on those studies dealing with the impact of leader behavior on group process and partici- pant learning. This should yield a picture of what is known about effec- tive and ineffective behavior, and what directions future research needs to take. What Should the Leader Do? Relying on the subjective truths of their own observations, many writers have outlined what they feel are the roles and activities the leader should perform. Tannenbaum, Weschler and Maszrik (1970) presented (a list of activities that all trainers must engage in to some extent. ‘ 34 They are: H . creating situations conducive to learning - this may involve introducing structured verbal and nonverbal exercises. 2 establishing a model for behavior - a model of what is accep- table, the ways to give feedback, what and when to self- disclose, how to analyze group process. 3. introducing new values - in relation to the authority role, personal growth, the value of analysis, risk taking, and so on. The values will differ depending on the theoretical orienta- tion of the leader and the group goals. 4. facilitating the flow of communication - to step in as a moderator, to share perceptions about what is happening so as to clarify communication processes and feelings, to act as a guide. 5. participation as an expert - to provide theory, knowledge, conceptual handles and a framework to help members extract meaning and generalizations needed for transfer. These basic activities have been addressed and expanded upon by others. Relating to the first activity, creating conditions for learn- ing, the role of the average T-group leader can be looked at from this view. As discussed in chapter 1, the T-group leader refuses to take the leadership role thereby frustrating the expectations of group members (Bradford, Gibb & Benne, 1964). In trying to resolve this problem and their feelings the members exhibit various behaviors toward the leader and each other. These behaviors become the data to be analyzed with re- sultant participant learning. Through his failure to take a traditional role the leader can be said to create the conditions for learning. 0f 35 course the leader can also do this by suggesting various strategies or exercises that members can use to resolve an issue or reach further levels of understanding (Cohen & Smith, 1976; Fiebert, 1968; Lakin, 1972; Lippitt, 1972). In this same vein, Psathas and Hardert (1966) point out that the leader influences group process by affecting its normative structure. They conducted a study to show that trainer interventions are actually norm-sending messages on what should and should not be done. Though the trainer himself is not solely responsible for establishing all the norms, he influences them by reinforcing or punishing norms formed by group mem- bers. He does not simply reflect what is happening, he creates or helps to create it. The problem with this study is that although it establishes that trainers send norm messages in their interventions, it does not exam- ine the effect of these messages on what actually occurs. The authors do present some indirect evidence of impact by showing that many of these norm-sending interventions are seen as "significant" by group members. Concerning the idea of the leader acting as a model, many writers address this role (Bradford, et al., 1964; Culbert, 1968; Gordon, Jr., 1972; Lakin, 1972). This modeling function is especially crucial in the T-group where the participants are given very little structure and natu- rally look toward the leader for guidance. Interestingly enough, this reviewer was unable to locate any study on this phenomenon. Some work has been done to examine the general area of modeling and vicarious learn- ing without special reference to the leader (Lieberman, Yalom, &'Miles, 1973). They found that group members did see modeling and vicarious learning as important to their growth. However, viewing modeling as important did not distinguish between learners and non-learners although it did show some relationships to long-term maintenance of change. 36 As for the impact of the leader on participant values, the research by Lieberman, Yalom and Miles (1973) throws some light on the validity of this perception. Their study of seventeen encounter groups found that "changes in values and attitudes were the most systematically related to differences in leader style" (p. 249). For example, those leaders who emphasized other-oriented values produced changes in members' valuation of things external to themselves. In another instance it was found that leaders who pushed the concept of growth and "experiencing" were success- ful in instilling these attitudes in group members. Changes in values and attitudes also persisted over time so that 6-8 months later they represented the most stable areas of change. The leader's impact cannot be attributed to the encounter nature of these groups because they in- cluded a wide spectrum of types including T-groups, psychoanalytic, and personal growth. As the authors point out, their findings show it is not the group label but the leader's behaviors that are important. Labels were not predictive of leader behavior or participant learning. The leader is also seen as responsible for facilitating the flow of communication. It is his responsibility to highlight and clarify group processes especially when an impasse occurs (Bradford et al., 1964; Cohen 6 Smith, 1976; Kaplan & Sadock, 1973; Lakin, 1972). An offshoot of this role is the leader's function as protector of group members who may be under excessive stress. This frequently occurs when members are given large amounts of negative feedback resulting in a psychological "over- load." It is up to the leader to determine if the person is okay and to halt or modify the incoming feedback. The importance of this function cannot be overestimated. Lieberman, Yalom and Miles (1973) found that many of the psychological casualties attributed their problems to an attack 37 by the group where the leader failed to intervene to stop the assault. Finally the leader acts as an expert, someone who provides cogni- tive input and a framework participants can use to understand the meaning of group processes (Bradford et al., 1964; Cohen & Smith, 1976; Lippitt, 1972). As will be discussed later, this leader activity is significantly related to participant learning. What is more, it has been found that those who maintain their learning do so because they have the necessary cognitive schema to transfer their learning to other situations (Lieber- man et al., 1973). The more the leader can do to provide such schema, the better and more long lasting the learning can be expected to be. Other writers have emphasized the importance of the personal characteristics of the leader, that is, the way he presents himself while carrying out his activities or functions. Carl Rogers is the spokesman for this viewpoint. As he sees it, it is the leader's job to facilitate the growth of the participants, not to control the group process. The leader must exhibit certain characteristics to be successful in facili- tating growth. First, he must be congruent, which implies consistency between words and behavior and the ability to communicate his internal reactions. Next he must be empathic; able to understand the internal states of others and to communicate this understanding. Finally, he must have unconditional positive regard for group members. Positive regard means a warm, accepting attitude and unconditionality implies that the person will be accepted no matter what he does. This nonevaluative stance is held to be a necessary condition for participant growth. T-group theory also encourages the leader to be as nonjudgmental as possible (Bradford et al., 1964). Other trainer characteristics thought to be facilitative of member growth are honesty, genuineness, commitment and self-disclosure (Gordon Jr., 1972). 38 The concern with leader roles and activities is based on the premise that his behavior will have a significant impact on participant learning (Back, 1972). His interventions are held to carry extra weight because of his position as an authority figure and someone who understands what should be happening (Golembiewski & Blumberg, 1970; Lakin, 1972). This is especially true in the T-group because members are initially given minimal guidance on what to discuss and how to proceed. Members naturally seek the leader's guidance and are likely at first to follow whatever sugges- tions or directives he gives (Golembiewski & Blumberg, 1976; Lakin, 1972; O'Day, 1968). Because of these factors, the leader's interpretations and viewpoints are usually accepted by group members. Culbert (1970) addresses this point: Like individual behavior, group behavior is overdetermined. This means that any single event can be accurately understood from a number of perspectives. The trainer‘s comments can, to a large extent, determine the primary interpretation that group members assign to a group event. (p. 34) The leader's interventions are seen as influencing which issues are discussed and the level at which they are dealt with (Banet Jr., 1974; Cohen & Smith, 1976). Appropriate interventions will promote participant learning and move the group toward greater cohesion, growth and independ- ence (Cohen & Smith, 1976; Lippitt, 1972). On the other hand, if the leader intervenes at the wrong level or brings up issues prematurely, the group process can be significantly hmpeded (Banet Jr., 1974; Stock, Brad- ford & Horowitz, 1970). Cohen and Smith (1976) feel that the leader's interventions are especially important at certain times during the group life when "critical incidents" are occurring. For example, how the leader handles an attack on his competency will impact on future expressions of negative feelings toward him and others. 39 Though many writers claim there is a significant correlation be- tween leader behavior and participant learning, only one study by Lieber- man et al. (1973) has produced acceptable evidence of this. They were able to identify effective and ineffective leadership styles in their study of encounter groups. A detailed discussion of their findings will be included at a later point in this chapter. How and When_§pguld A Leader Intervene? Accepting for the moment that the leader's intervention behavior is important to the group, it is necessary to look at ppgp the leader should intervene and Egg that intervention should be carried out. The leader must know when to be caring, provide meaning, and so on, for even the most seemingly positive action can be taken at times when it would hamper, rather than enhance, learning. Most writers have discussed the issues and problems the leader should respond to. Usually these are albeit equivalent to the list of roles or functions the leader is supposed to play, so that a leader is told to intervene in order to: 1. help the group learn from its processes, to highlight important issues or feelings that may otherwise be missed (Bradford et al., 1964; Fiebert, 1968; Lakin, 1972; Lippitt, 1972) 2. help expose group and individual behavior for analysis QArgyris, 1971; Gordon Jr., 1972; Lakin, 1972; Lippitt, 1972) 3. protect a group member who is being attacked or overloaded (Cohen & Smith, 1976; Golembiewski & Blumberg, 1970, 1976; Lakin, 1972; Lippitt, 1972) 4. give and model feedback and self-disclosure (Gordon Jr., 1972; Lippitt, 1972) 5. convey understanding and acceptance to create psychological safety (Gordon Jr., 1972) 6. invite members to share their feelings and ideas (Fiebert, 1968; Lakin, 1972; Lippitt, 1972) 40 Such general guidelines are useful in that they sensitize the leader to important issues that might call for his intervention. However, their utility is lunited because they do not help the leader in deciding [p23 to make the intervention. After all, there is a difference between intervening and intervening effectively. A leader has to do more than simply recognize that an intervention might be called for; he needs to be aware that there are various ways of making that intervention that may have differential consequences. Then he has to have some basis for choos- ing among these alternative responses so that he can make the most appro- priate, helpful response. Many writers feel that the first step in doing this is to take group goals and situational factors into account, i.e., the group phase and the individuals involved. It seems to be one of those self-evident truths that the leader should take group goals into consideration when making an intervention. The goals will naturally influence which issues are given priority and whether the trainer focuses on group, interpersonal or individual con- cerns (Bradford et al., 1964; Golembiewski & Blumberg, 1976; Lippitt, 1972). Of course, there are other variables to be considered if the leader is to successfully intervene to help the group reach its goals. One of the most crucial variables is thought to be the individual(s) involved. The question of who is involved can be looked at from two angles, overall group composition and the personalities of specific group members. Typically group composition is discussed in terms of volunteers versus nonvolunteers. Mbst human relation training groups are composed of volun- teers. This has several advantages. Such participants are less resistant to change and experimenting with new behaviors. They are also more open to influence from the leader and other group members (Lieberman et al., 41 1973). What is more, because they are looking for a positive experience they are simply more likely to feel they had one. Such qualities make the leader's job easier and he is usually not held as accountable as he would with unwilling recipients. Nonvolunteers not only bring resistance, they also raise ethical questions since they may be forced to participate in training groups because of work-related or academic program require- ments. This is not to say that volunteerism works entirely toward the leader's benefit. Volunteers frequently have more emotional problems than a random sample of individuals and such people can be especially vul- nerable and/or disruptive to the group process (Back, 1972; Lieberman, 1976). This brings in the other side of "who is involved," that is, the personalities of specific group members. There is some evidence to suggest that certain types of indivuals who voluntarily enter encounter groups are more likely to be harmed than others (Hartley, Roback & Abramowitz, 1976). This is especially true when they come into contact with aggressive, confrontive leaders who at- tack them or leaders who allow attacks from group members. Lieberman et a1. (1973) carefully studied the characteristics of those encounter group " A person was labeled members categorized as "psychological casualties. as a casualty when he had an enduring negative psychological reaction that was directly related to his group experience. The authors found that characteristically casualties had: 1. low self esteem 2. unrealistically high expectations about what the experience would yield, e.g., close friendships, solutions to important problems 3. negative attitudes toward people in general 4. feelings of low interpersonal adequacy yet high interpersonal 42 sensitivity. In other words, the misunderstood person who understands others well 5.-a strong need for growth and change It seems that it would be important for the leader to be especially cogni- zant of his interventions toward someone exhibiting a similar profile. These findings also point to the need to assess each member's expecta- tions and personal internal resources so that potential casualties can be identified and handled accordingly. Besides the study by Lieberman, Yalom and Miles, very little has been done to study the types of people that enter human relations training groups and how they are effected. Until now research has concentrated almost exclusively on the average level of improvement or the number of participants who changed. Future research should aim to correct this situation. Another factor to be considered in making interventions is the group phase. It is hard to find a writer who does not agree to this though they have different opinions on how the leader should modify his behavior accordingly (Cohen & Smith, 1976; Dies, 1977; O'Day, 1968). One interesting study by Dies and Cohen (1976) involved the interaction be- tween group phase and leader self-disclosure. They gave subjects a list of statements a leader might make involving his personal history and here-and-now reactions. Subjects were then asked to judge how helpful or harmful they thought it would be for the leader to share each state- ment during the first, eighth and fifteenth session of a weekly group. They found that subjects felt it was harmful for the leader to confront the group or individuals early on, especially if the confrontation in- volved the expression of angry or frustrated feelings. The data were unclear on which self-disclosures were helpful during various phases. 43 Fiebert (1968) deals with the issue of group phase and leader intervention strategies. The emphasis is on strategies because he contends that con- tent and style are less important than the overall intervention strate- gies used. Fiebert conceptualizes the group as having three phases and suggests that the leader adopt different strategies for each phase. At first the leader should act as a catalyst to get things going in the right direction. This means helping group members to understand how to work effectively in the group by clarifying general procedures and norms, i.e., emphasis on the here-and-now; concentration on feelings, not just ideas; self-disclosure and risk taking. At this stage interventions should focus on the group as a whole, the way it is operating, making decisions and so on. During the next phase the leader becomes an orchestrator working to help deepen interpersonal exploration and shar- ing of feelings. The leader can suggest various structured exercises to increase self-other exploration, for example, sensory awareness techniques or the sharing of fantasies. Interventions are now directed toward inter- personal feelings and relationships. Finally, in phase three the leader becomes more of a gggup member, self-disclosing and relinquishing his control to the other participants. The number of interventions the leader makes should also decline as the group progresses. Culbert (1970) also addresses the question of group phase and leader interventions but from a slightly different vantage point. He proposes that once the group establishes a specific set of training goals there are phases it must go through in order to reach its goals. Given this assumption, the leader can use his interventions to influence the group so that it moves productively through these phases. This means taking responsibility to insure that resolution of one phase is completed 44 before going on to the next and to prevent stagnation in any one phase. Interventions are of two types, those required within a phase to increase participant learning and those that shift the group's focus from one phase to the next. Culbert offers some general suggestions on what to do when intervening. For example, he suggests that when making a phase-shifting intervention, the leader should give a rationale for the shift, summariz- ing the process of the group until that point and the need to move on. His contribution to understanding how to intervene is of less importance than his acknowledgement of group phases and the need for the leader to accommodate his interventions to them. Blumberg and Golembiewski (1976) sidestep the whole issue of leader interventions by saying it is difficult to make any general state- ments on how and when to intervene. They feel that the number of factors to be considered make generalizations meaningless. Despite this, they present some ideas others have offered on interventions. For example, they cite Harrison (1970) who feels there are two criteria to be used in deciding how to intervene: l. Interventions should be made at a level no deeper than is needed to produce lasting solutions to the immediate problem. The idea is to avoid interventions which focus on psychodynamic causes unapparent to the person involved since "a person is more likely to feel competent to deal with a problem that he sees and feels at the moment, rather than working abstractly with some other's interpretation of the problem" (p. 88). 2. Interventions need to take into consideration the energy and re- sources of the person involved so that they are not pushed beyond their capacities. Sometimes, it may be too costly to the person, in terms of anxiety and personal upheaval, to make changes and the leader needs to be aware of this and respect it. Excessive pressure to change could be harmful in such situations. Blumberg and Golembiewski also cite Argyris (1971) who emphasizes that interventions should not take away from the group's responsibility 45 for being their own problem solvers. Consequently, the leader should intervene to help develop: 1. valid data on the problem 2. internal commitment to problem resolution 3. a climate of free choice If these are present then the group will be able to resolve the problem themselves without the leader doing it for them. Models for Studyingflntervention Behavior The articles discussed so far offer very nonspecific guidelines on how to intervene, that is, the form and content of the intervention. This is an advantage in that they allow for individuality of approach thereby maintaining the spontaneous flavor of groups and leader flexi- bility. But it is a decided disadvantage in that they do not help the leader to choose among the various intervention responses that would satis- fy such general guidelines. The need to choose is based on the assumption that different types of intervention will‘ggg have the same effects. Up to now the assumption seems to be the opposite, that different interven- tions have similar consequences. Which of these is true and to what ex- tent is still an open question that warrants investigation. If leader interventions are as influential as many propose, then more needs to be done to study the intervention itself and how variations can affect the group process and participant learning. Three groups of researchers have made an attempt to develop the technology needed to do this, i.e., Cohen and Smith (1976), Psathas and Hardert (1966) and O'Day (1968). The most promising work is that of Cohen and Smith. Cohen and Smith feel that not enough guidance is given to the group leader on how and when to intervene in the service of various goals. 46 More needs to be done to help the leader learn how to respond in the immediate give-and-take situations of group life. They disagree with those who feel that effective trainer interventions are essentially a matter of consistency, empathy, or being oneself. Rather they favor "a systematic technological approach to leader intervention with a focus upon specific outcomes and the probabilities of certain desired outcomes of specific interventions" (p. 86). A technological approach involves examining the dimensions along with which an intervention can vary. They write that all interventions can be categorized according to their: 1. Level - The focus is either individual, interpersonal or toward the entire group. Focus depends on the individual re- cipient(s), not the content of the intervention 2. Type - 3 major modes a. Conceptual - "attempts to abstract or conceptualize some significant idea or issue" (p. 91) Provides theory and concepts for understanding group process. It may be planned or a spontaneous reaction to ongoing process b. Experiential - involves the sharing of feelings about the ongoing situation; the leader may share how he is experiencing things and/or ask for student reactions to the here-and-now of group life c. Structural - use of structured activities including verbal and nonverbal exercises. They may be very con- ceptual and task oriented or emphasize the emotional level 3. Intensity -The emotional level of the intervention - can be high, medium or low. It involves "the extent to which the underlying theme of behavior is exposed, interpreted and directly communicated to the individual, group or subgroups to achieve an awareness of the underlying dynamic of the behavior" (p. 102) These three dimensions are said to comprise an Intervention Cube, "a model that can be used to observe, categorize and analyze interven- tions by group leaders, regardless of theoretical and/or practitioner orientation" (p. 87). Using this model, the leader or researcher can 47 begin to examine the effects of various types of styles, i.e., combi- nations of the dimensions. But comparisons require that situations be comparable or nearly so. To deal with this problem Cohen and Smith propose the critical incident model. As discussed in chapter 1, it has been determined that there are certain incidents (issues, problems) common to most groups. The leader's intervention response to them is thought to have considerable significance for group growth and development. Accepting this, it follows that re- search should concentrate on examining the effects of different inter- vention responses to these recurring incidents. Cohen and Smith propose the critical incident model as a way of doing just that. The model "is a way of arranging events in sequence, from those that led up to and im- mediately preceded some critical incident to those that specify the con- sequences of certain interventions" (p. 123). An outline of the critical incident model reveals its component parts. 1. The context of the critical incident a. phase of group - beginning, middle, end and session number b. climate of the group - its mood, e.g., silent, hostile c. brief description of person(s) involved - includes past and current behaviors 2. Behavior and/or conversation immediately preceding the intervention 3. The surface issue(s) and the underlying issue(s) involved, e.g., the surface issue might be asking about how other groups have gone while the underlying issue could involve feelings that this group does not seem to be going well 4. The level, type and intensity of the leader's intervention response 5. The consequences for the group a. the intended directional movement of the leader in.making his response b. the actual behaviors exhibited by the group members, e.g., agreement, silence, hostility 48 This model has several desirable features. First, it incorporates many of the modifying variables held to be important in making interven- tions, i.e., group phase, immediate climate, individuals involved. It also looks at consequences, something sadly lacking in the studies so far reviewed. The critical incident model should serve as the basis for substantive research into the effects of leader interventions. Of course it may turn out that it is more profitable to examine dimensions of an intervention other than the ones discussed. Banet Jr. (1974) proposes a somewhat different scheme for cate- gorizing the type of intervention. He breaks it down into five categories as follows: 1. Meaning-attribution - describes or suggests a definite meaning to the process event. It connects here-and-now events to the past or context by means of theory, association or metaphor 2. Evocation - attempts to elicit specific emotional responses 3. Structural - exercises and all forms of the therapist's nonverbal behavior in relation to the members 4. Experiential - self-disclosure in relation to group events 5. Prescriptive - telling patients to test behavior in another context outside the group and report back on what happened No doubt other researchers will come up with their own category systems, but the comprehensiveness of the model should insure that its basic outline remains intact. It is the overall approach, rather than the individual categories, that seems to have considerable promise. As discussed earlier, Psathas and Hardert (1966) have attempted to develop a methodology for examining the norm message sent by the leader. Norms refer to messages that imply or directly state what members should and should not do. Their study was based on seven T-group meetings over a two-week period. Participants and trainers were given a questionnaire 49 after selected meetings that asked them to write down the trainer inter- vention they considered the most significant. Later these chosen interé ventions were isolated on tape for further study. The data collected in this manner were combined with findings from the literature to develop nonmative dimensions into which leader interventions could reliably be categorized. Space does not allow for a detailed description of these categories but a brief list reveals the scope of the normative dimensions to which the leaders addressed themselves. They are as follows: 1. Feedback - giving and receiving 2. Feelings - expression, acceptance, encouragement of 3. Acceptance Concern - promotion of trust, acceptance of self, others 4. Analyzing Group Interaction or Process - past and present member interaction 5. Coal and Task Concern - definition of, purpose, procedural concerns 6. Behavior Experimentation - encouragement of 7. Leadership Behavior Among Members - power, struggles, sharing leadership 8. Participation - nature of, responsibility, encouragement of 9. Trainer Membership Authority Problems - dependence, problems with 10. Decision Making - procedures for, problems 11. Structure Concern - member feelings, reactions to unstructured atmosphere These normative dimensions are characteristic of the T-group and the list would probably vary with the type of group being considered. Lieberman et al. (1973) found many other norms operating in the encounter groups they studied. The norms found in this study are of less importance than the fact that it attempted to develop a methodology for examining the norm messages being sent by the group leader. This method could be used in conjunction with other measurements to determine the influence 50 of these messages on the actual normative structure of the group. Such a study would seem to be what Lieberman (1976) called for when he said future research should focus on the leader's impact on the group social system. Normative structure would certainly qualify as an important intervening variable between leader input and member learning. O'Day (1968) was also concerned with leader interventions in the T-group. However, his approach was more comprehensive as he developed a content analytic system for analyzing‘gll trainer interventions. Using his system each intervention is classified in terms of three component parts: 1. definitional - the issues being addressed, the content, e.g., de- pendency, participation 2. behavioral - the way in which the content is expressed, the form, e.g., a question, suggestion 3. emotional - the feelings being expressed, e.g., hostility, caring Together these three components comprise a leader's intervention style. Leader intervention style can therefore be examined in various ways, for example, holding the issues constant and examining the behavior. Each component has a complex set of categories under it that are derived from the theoretical and empirical literature on T-groups. Many of the defi- nitional categories reflect the typical concerns of the T-group, i.e., dependency and problems with authority. Because of this, O'Day's system would not be applicable to other types of human relations training groups. One interesting finding from his research was that the four T-group leaders he studied behaved in theoretically inappropriate ways. O'Day (1968) writes: The description of training style derived in this study casts doubts on the typical image of the trainer as completely tolerant, accepting, self-assured and nondirective. . . . trainers frequently 51 function as evaluators, authority figures and experts and also express anxiety, depression and various forms of anger in response to many of the issues which arise in the sensitivity-training group experience. (p. 634) This finding illustrates the important kinds of information that can be derived from detailed study of trainer interventions. Resistance and_§gpport for Research on Effective Trainer Interventigng There are those who feel that research in the area of effective trainer interventions is essentially fruitless. Like Blumberg and Golembiewski (1976), they contend that the large number of situational factors makes it virtually impossible to generalize about the timing and quality of leader interventions. To them it is better to leave decisions about interventions to the leader's discretion. Frequently these same people make statements about the importance of the leader's contributions to the successful functioning of the group. If such statements are true, then it seems to this reviewer that a failure to engage in research on interventions constitutes dereliction of professional responsibility. Efforts to determine the nature of effective interventions must go on, no matter how difficult the methodological problems. Resistance to research in general is especially high among those leaders who run encounter groups CBack, 1972; Hartley et al., 1976; Massarik, 1972). Such groups are typically more experiential in nature, emphasizing expression of feelings, nonverbal activities and risk taking. Leaders of such groups frequently shun any kind of evaluation, saying that it takes away from the "freeing" nature of their groups. They also argue that formal evaluation is unnecessary because they are the best judges of participant learning and the experience as a whole. The in- validity of this idea was demonstrated by Lieberman et al. (1973) when 52 they found that leaders were actually the worst predictors of positive and negative participant outcomes. They tended to overestimate the amount of benefit students experienced and seriously underestimated the number of psychological casualties. Even after leaders were told they had casual- ties in their groups they could only identify two out of sixteen of them. For every skeptic of the value of research on trainer impact, there are numerous writers who advocate more work in this area. Campbell and Dunnette (1968) were among the first to call for more studies. In their comprehensive review of T-group research they noted that no system- atic studies had been done to relate variation in trainer personality or behavior to outcome. Since then others have pushed for research to exam- ine the trainer's influence on group process (Cooper, 1972; Lakin, 1972; Lundgren, 1971) and ultimately on participant learning (Fiebert, 1968; Harrison, 1970; Lieberman, 1976; O'Day, 1968). As recently as 1976, in a review of the literature on change-induction groups, Lieberman writes that though the concept of leader centrality remains prevalent, there is still little research to support this claim. He feels that the lack of findings is in part due to the tendency of many investigators to select single, generalized leader variables (i.e., directive vs. nondirective) for study. This approach has not produced results and is unlikely to because it fails to take into account the social system of the group. Instead of there being a direct relationship between leader behavior and outcome, the leader influences the system in various ways which in turn impact on the participants. He feels that future research needs to focus on the intervening variables between leader input and.member outcome. Also more should be done to explore leader interventions in depth to de- termine variation in terms of time, frequency and type of intervention. 53 Because of the extreme difficulty of doing this kind of research, Lieber- man ends by saying that the "role of research on leader variables in the group area is clouded" (p. 236). O'Day (1968) and Cohen and Smith (1976) feel that many of the problems in doing such research will be alleviated by the development of methodologies for describing leader behavior in detail. Their work toward this end has already been presented. As O'Day points out: Before it is possible effectively to determine the impact of a trainer on a T-group, it is necessary to have a methodological procedure that permits a reliable and valid description of what it is that the trainer does in a group. The Trainer and Group Process What does the leader do that affects group process? Some of the attempts to answer this question will now be discussed. Many authors have addressed the question of the impact of leader self-disclosure. Encounter group leaders advocate it saying it makes the leader more of a person, someone the participant can relate to. T-group leaders typically are more reserved, especially in the early stages of a group (Bradford et al., 1964). On the other extreme are the psycho- analytically oriented leaders who refrain almost entirely from personal self-disclosure. Unfortunately there is little evidence to support the value of any of these positions. Culbert (1968) studied the effects of trainer self-disclosure on T-groups. Leaders were given two sets of instructions, to be self- disclosing in one group and not in another. It was found that the group members with the more self-disclosing leader viewed relationships with the rest of the group as important to their learning. The group with the less disclosing leaders more often cited the trainer and one other 54 member as influential. Further, Culbert found that participants who experienced the more self-disclosing trainers had a higher degree of self-awareness in the early phases of the group though this advantage was lost over time. He concluded that members model their leader's self-disclosing behavior which leads to increased overall participation and group solidarity. Modeling and participation together seem to affect the rate at which self-awareness is achieved. However, and more impor- tantly, they were not found to influence the total amount of self- awareness because in the end both groups were equal. Bolman (1971) found that leader self-disclosure was unrelated to participant perceptions of learning for self or other group members. This study, combined with Culbert's, seems to cast doubt on the value of leader self-disclosure. But as Dies (1977) points out, both of these studies have a fatal flaw in that they treat self-disclosure as a generalized variable. He argues that it is a mistake to do this, that self-disclosure is modified by group phase, content of the disclosure, leader personality, and so on. The impact of the self-disclosure is contingent on these and other variables and must be studied taking this complex relationship into account. Dies' criticism echoes that of Lieberman (1976) when he writes about the need to avoid research which limits itself to studying general- ized leadership variables. To use a statistical metaphor, it is as if they are saying that investigators should refrain from looking for "main effects" and instead should concentrate on detenmining if there are any meaningful "interactions." As it now stands, the impact of leader self- disclosure is still undetermined. Bolman (1971) studied the effects of other leader variables besides self-disclosure. He constructed questionnaires that examined T-group 55 ‘member feelings about the leader, learning and group climate. Data were collected and factor analyzed. Of several factors only one was found to be important in that it correlated with member liking, identification with the leader, and self-rated learning. The items making up this fac- tor, labeled "congruence-empathy," are as follows: 1. The things he says seem to be highly consistent with what he is feeling 2. He is secure and comfortable in the group 3. He seems to be in close touch with howwmembers of the group are feeling 4. He sees things through the eyes of members of the group 5. He is quite comfortable and relaxed when the attention of the group focuses on him These descriptive statements picture the leader as secure, consistent and empathic, qualities that Rogers (1970) has stated as important for anyone in a helping relationship. On the surface this finding would appear to be significant but unfortunately no studies have been reported that repli- cate it. Even if future research does confirm Bolman's work, there is still the question of determining the relationship of self-reported learn- ing to actual change. The frequent inaccuracy of self-report data is demonstrated by findings from the study of encounter groups by Lieberman et al. (1973): Positive testimony was inflated over our assessments of change. . . . At termination only 57% of those who gave positive testimony actually did receive positive benefit; however, at six~month follow- up, 762 of those giving positive testimony actually benefited. So, just after the group there is a good deal of unfounded enthusiasm, but if several months have elapsed, we can be more confident that participants' views are trustworthy. (p. 427) Another area of interest is the leader's effect on the normative structure of the group. The work of Lieberman et al. (1973) provides some information on this relationship. Unfortunately they were not able 56 to provide any definitive answers because each leader taught only one group. This meant that they could not determine if the leader was pri- marily responsible for creating the group norms or they developed and the leader only accommodated to them. Though the cause-effect relation- ship was unclear, the authors felt they did have some evidence to suggest the direction of influence. Prior to the groups, leaders and controls were asked to decide which of a list of behaviors would be appropriate during the group. These behaviors were supposed to reflect various norms typically found in encounter groups. Using this information the authors found that there was a relationship between leader expectations and the actual norms developed but it was not a one-to-one relationship. Rather the norms that developed seemed to reflect a combination of what the leader expected and the expectations of participants. Groups where both leader and participant norm expectations were met had the highest number of learners. Still, confirmation of the leader's expectations was more closely related to member learning. The authors sunmarize their findings this way: It is not the case that high-yield group leaders were solely effec- tive in imposing their desired set of norms on the group. Rather, they were simultaneously able to create conditions where their own norms expectations and general cultural expectations were realized. (p. 297) Group norms were also found to play an important part in determining learner outcomes. In fact their contribution was at least as important as the leader's. Again this highlights the idea that the social system of the group is crucial to the change process and more needs to be done to explore the leader's effect on it. The most convincing evidence for the influence of the leader comes from the work of Lieberman et a1. (1973). They found that leadership 57 style was clearly associated with participant learning in the seventeen encounter groups studied. This study merits special attention because it represents the first noteworthy research to support the belief that the leader's behavior has an effect on participant outcomes. Because of its importance this research project will now be discussed in greater detail. Determination of leadership style was based on ratings of the leader on four dimensions extracted from numerous evaluation instruments filled out by participants, observers and the leaders themselves. Par- ticipant change (outcome) was also assessed using a variety of measures including peer ratings, self-report scales, outside observers and stand- ardized tests. The four dimensions found to characterize leader behavior were as follows: 1. Emotional Stimulation - eliciting, provoking, challenging members, especially in regard to expression of feelings. Emphasis is on modeling desired behaviors (e.g., self-disclosure of anger, love) and on challenging group members 2. Caring - expression of warmth, affection, concern for group par- ticipants. Leader gives large amounts of support, praise, encourage- ment and protection 3. Meaning-attribution - those behaviors that help the participants to conceptualize, to understand the meaning of processes in the group. It "represents the naming function of leader behavior, wherein the leader gives meaning to experiences that members undergo. It refers to the translation of feelings and behavior into ideas" (p. 238) 4. Executive function - managing the group, setting rules, limits, directing what will be done and how. It includes suggestions on approaches the group might take to work through a problem (e.g., nonverbal exercise, paraphrasing) as well as eliciting and ques- tioning group members on their opinions, feelings, etc. When these dimensions were combined to create leader typologies the authors found a significant relationship between leader style and outcome. Certain styles were associated with high numbers of learners while others were associated with few learners and psychological 58 casualties. Table 2 shows the relationship between trainer type and their ratings on the four dimensions. Table 3 shows the relationship between type and outcome. TABLE 2."Relationship Between Trainer Type and the Amount of Emphasis on Trainer Functions -.-. . , . _ __ ».-___._ _.__ Trainer functions Emotional Meaning Trainer type Stimulation Caring attribution Executive 1. Energizer High High Moderate- Moderate- High High 2. Provider Moderate High High Moderate 3. Social Engineer Low Moderate High Low 4. Impersonal Moderate- Low Moderate Low High 5. Laissez faire Low Low Moderate- Low High 6. Manager LOW’ Mbderate Moderate High SOURCE: A. Blumberg and R. T. Golembiewski, Learning and Change in Groups GMaryland: Penguin Books Ltd., 1976), p. 77. 59 TABLE 3."Leader Type and Outcome weighted High Moderate Un- Negative Drop- Casu- Impact Leader Type N Learner Changer changed Changer out alty Average A Energizers 59 8-14% 12-20% 19-32% 4-071 9-15% 7-12% .17 B Providers 37 8-222 13-3SZ 8-22% 3-08Z 3-08% 1-03% 1.03 C Social Engineers 32 6-19Z 3-09% 14-44Z 6-19Z 1-03% 2-06% .16 D Impersonals 18 0-00% 7-39% 3-17% 2-11% 4-22% 2-11% 00 E Laissez- Faires 25 2-08% 1-04% 13-52% 0-00% 7-28% 2-08Z -.20 F Managers 11 0-00% 0-00% 8-70% 0-00% 2-10% l-lOZ -.45 G Tape Groups 24 3-13% 5.21% 13-54Z 2-08Z 1-04% 0-00% .58 Total 206 SOURCE: M. A. Lieberman, I. D. Yalom and M. B. Miles, Encounter Groups; First Facts (New York: Basic Books Ltd., 1973), p. 245. As the tables reveal, Providers were the most successful leader type in producing positive changes while minimizing the number of partici- pants with negative outcomes. Social Engineers were next in line fol- lowed by Energizers. However, Energizers also had a high number of drop- outs and psychological casualties to offset the number who learned. Overall, the authors felt these three leadership styles could be cate- gorized as beneficial. 0n the other hand, the three remaining styles were far from pro- ductive. The Impersonal, Laissez-Faire and Manager types all had more negative outcomes than they did positive or neutrals. The Impersonal and Laissez-Faire leaders produced no high learners at all. This is in 60 marked contrast with the 22% yielded by the Providers. The Manager was the worst type of leadership style in that it yielded no high or moderate changers and a few negative outcomes. However, it is difficult to make generalizations about this type because only one leader exhibited this leadership style. High psychological casualty rates were associated with Energizers and Impersonals. Both types "were characterized by aggressive stimulation and relatively high charisma" (p. 246). The authors relate this to the finding that casualties were often associated with direct attacks by the leader or with the leader's failure to protect participants from attacks by other group members. Interestingly enough, the taped groups did relatively well, with no casualties and only one dropout. They also had 34% high or moderate learners making them fourth in the number of high learners and third in moderate learners. The authors discuss these findings saying that these leaderless environments were safe because they provided structure through interventions (mainly Mbaning-attribution and Executive functions); yet they avoided the high levels of leader stimulation associated with nega- tive outcomes. Clearly, what the leaders did, their style, was very important in determining if participants had a beneficial or harmful experience. Just how leaders influenced participant learning was less clear. There did not seem to be a simple cause-effect relationship between leader behavior and change. Rather the leader style affected the learning environment, and factors in this environment (other members, norms, goals) were influ- ential in inducing change. As the authors put it: 61 Perhaps the most useful way of thinking about leader style is in teams of the overall impact it has on learning rather than in terms,of specific areas in which people may change. Leader style creates conditions or a setting under which individuals can learn. The particular kinds of change that take place are probably medi- ated by many other conditions such as the individuals' initial level and interests or the particular kinds of learning experiences (mechanisms) that occur to the person in the group. (pp. 258-59) anlity of the Research Before concluding this chapter it seems obligatory to discuss the quality of the research that has been presented. Just how valid and reliable are the findings of the various studies discussed? The answer must be, not very, for all of them suffer from various types of important design problems. The list of problems is long: small sample sizes, lack of random assignment, failure to use control groups, exclusive use of self-report and other subjective measures of change, leaders having only one group thereby confounding group and leader effects, insufficient at- tention to the effects of testing, and so on. Even the major research effort of Lieberman et al. (1973) has several glaring flaws. Uppermost among these is the one group-one leader problem resulting in a confound- ing of the two. It is surprising that the investigators did not foresee and prevent this problem from occurring, given the large scale on which this project was conducted. In addition, it is very likely that there were reactive effects due to the extensive testing and observation of the groups. Despite this, there are many positive aspects about the design (e.g., use of controls, followbup, measures of change) and it represents the best effort to date. It also pointed out the value and necessity of doing large-scale, carefully designed research. The objec- tive here is not to present an extended critique of the research done so far; rather, it is to point out that the study of the group leader is still in its nascent stages. 62 Conclusions This chapter began with the aim of exploring the literature on human relations training group leaders to see what information was avail- able to the group leader on the characteristics of effective and ineffec- tive intervention behavior. A second aim was to determine the directions future research in this area should take. The review found that the only noteworthy and validated informa- tion on effective leader behavior comes from the work of Lieberman, Yalom and Miles (1973). Their research on encounter group leaders showed a strong relationship between participant learning and a certain style of leader behavior. Participants were more likely to learn when leaders gave high amounts of cognition and caring. Though extremely useful as a first step in defining effective leader behavior, the study did not produce guidelines for determining when to give cognition and caring, or even exactly how. Aside from the findings in this study, all the group leader has to rely on are unsubstantiated claims by various authors that using a particular technique or approach to intervening will prove effec- tive. Though of some worth, these discussions of general principles for intervening are inadequate when the leader is forced to make continuous decisions about his behavior in the here-and-now of group life. As for future research directions, it is clear that in-depth studies are needed on the characteristics of effective and ineffective leader be- havior. Research that focuses on generalized leader variables should be abandoned. All this entails developing methods to isolate and study leader behavior and its impact. This dissertation is designed to do just that. It develops and explores the utility of various methods of study- ing leader behavior and impact while trying to gather information on 63 effective and ineffective intervention behavior. There are several features which make it a unique contribution to the literature. Instead of looking at the general effects of leadership style, it focuses on the immediate and long-range consequences of selected interventions on group members. In other words, it tries to determine the characteristics of interventions that actually lead to changes in student behavior and atti- tudes, as well as looking for the factors that contribute to perceptions of effectiveness. In addition, it builds on the work of Cohen and Smith (1976) in developing a methodology for studying leader behavior by exams ining it in relation to critical incidents, while reformulating the con- cept of an intervention as a series of verbal moves within a specific time period called an episode. Finally, it attempts to study the effects of two key variables which are purported to influence the nature of ef- fective leader behavior, namely, group stage and member personalities. Though on a small scale, this project is designed to overcome some of the limitations of prior research by trying to deal with the complexi- ties of leader behavior and its effect on group members. The findings should constitute a meaningful addition to the knowledge available to group leaders on how to intervene effectively. It will also be possible to see if the guidelines and research on effective leader behavior for encounter and T-group leaders apply to the IPL leader. CHAPTER III METHODOLOGY Review of the complex methodology used in this study will be organized around six key areas as follows: 1. Characteristics of the Sample and Sampling Techniques 2. Description of Instruments and Observation Techniques used for Data Collection - The Perceived Impact MEasures and Peer Perceptions 3. Description of Analysis of Perceived and Real Impact Measures 4. Description of the Analysis of Intervention Episodes 5. Research Questions 6. Statistical Procedures Before discussing the methodology, it is appropriate to give the reader a more complete picture of the IPL and the larger course of which it is a part. Doing so will reveal many of the special problems TAs face in conducting IPL groups, problems that other leaders of human relations training groups do not have. Research to determine effective TA behavior becomes all the more important in light of these obstacles. Program Description The IPL is one of two parts of a larger course, Education 200. These parts are identified as the Personal Demands of Teaching and the Task Demands. They are succinctly defined in the course outline given to students at the start of each term. 64 65 Because one seeks and wrestles with questions and answers regard— ing self-growth differently from the way he deals with the growth of others, the course is divided into two related but parallel sets of experiences--the personal and task demands of teaching. The personal demands of teaching are taught through Interpersonal Process Labs (IPL) . . . . Here the learning experiences involve direct interaction with people and ideas. . . . The focus is placed upon self-learning, self-development, and personal growth. . the task demands of teaching are taught through reading materials, individualized carrel activities, tutorial sessions, etc. Instruction is self-paced and students read the course handbook on their own. Optional individualized carrel programs are available to highlight the more important concepts in the text. When ready, the stu- dent comes to a special testing room where unit exams are given. Testing is based on a mastery model and the student must achieve an 80% level to receive 8 Pass. When a student fails to reach mastery on the first try, s/he can be reinstructed by a tutor on the concepts missed and then take a parallel test on these concepts. There are four units in the handbook, three of which cover the Tasks of Teaching. These tasks include Assess- ment, Goal Setting and Objectives, Behavior Modification Techniques, Modeling and Evaluation. In addition, Unit 1 examines the general con- cepts of Teaching and Learning and what should be taught to children. Though the course is divided into two parts, TAs are encouraged to incorporate the handbook concepts into the IPL. They are also re- quired to provide instruction on these concepts to prepare students for the exam. As already discussed, the IPL is distinctive in the human rela- tions training movement. Three factors account for this. First, it is required; all students must complete the experience to receive their teaching certificate. This lack of volunteerism affects the group 66 dynamics and the leader must work to overcome the resistance and scepti- cism of many participants who doubt the relevance of the IPL to their preparation for teaching. Next, the IPL has an evaluation component. Contrary to the admo- nitions of T-group practitioners to create an evaluation-free atmOSphere, the IPL leader formally and consistently evaluates group members. Stu- dents are actually graded although letter grades have been replaced by either a Pass or Incomplete. A student with an Incomplete is required to repeat the IPL the following term by participating in a regular group or as a member of special groups established for Incompletes. The mastery model grading system has several advantages. The student's grade point average is unaffected if s/he needs more time to complete the course. Also there is no need to repay for the course. Be- sides these practical advantages there are the attitudinal benefits accrued by helping students see that it is acceptable for people to progress at their own rates and that a system can accommodate to meet these individual differences. However, for many,receiving an Incomplete is psychologically tantamount to failing even if it is not objectively equivalent. As a result, the sense of being evaluated and having to demonstrate the skills "to pass" is continually present and the TA.must effectively deal with this if the group is to be a successful learning experience. Finally the IPL is more goal directed than most training programs. Its purpose is to impart certain interpersonal communication skills that students need to be effective teachers. These skills are defined in be- havioral objectives given to students along with detailed analyses of the constituent parts of the skills. For example, students are taught the behaviors associated with the skill of Active Listening, such as 67 paraphrasing, maintaining eye contact, and sharing perceptions tenta- tively. ‘Written materials describe and illustrate these concepts and the IPL is used as the vehicle for practicing them and assessing their impact. The TA has primary reaponsibility for choosing the activities of the IPL. He can use any techniques or strategies that will assist the group to reach competency level on the objectives. Consequently, though there is a common set of goals and objectives, the way in which they are taught can vary considerably across laboratory sessions. There are some boundaries since the IPL may not become a "sensitivity or ther- apy" group, nor is it allowed to imitate the didactic characteristics of a traditional classroom experience. This inbetween position is often difficult to maintain and at any particular moment the demands of the immediate situation may force the IPL to assume one approach or the other. Though there is room for diversity, certain practices encourage bounds on the techniques used. To begin with, each TA is given a manual suggesting exercises to be used in the service of the different objec- tives. These exercises have been drawn from the work done in the field of human relations training and are similar to those used in many en- counter and T-groups. Next, newly hired leaders are required to team teach one term with an experienced instructor. This introduces them to the basic IPL structure and provides guidelines on how to run their own groups. Third, all new TAs are observed by administrative staff to ensure that activities are appropriate to the IPL experience. Finally, a large part of weekly staff meetings is devoted to sharing conmon problems and useful strategies. All these elements work to increase the uniformity of the experience across the thirty to forty sections of the course offered 68 each term. This is not to say that all groups are run alike. On the contrary, TAs are given considerable latitude in determining group process and in implementing their unique style. This decision is based on the belief that diversity is not only healthy but inevitable when dealing with the individual differences of so many TAs and groups. Other aspects of the course need to be considered. Since it is given in a college environment, IPL meetings are typically scheduled in classrooms with movable desks and chairs. Carpeting and comfortable furniture are not available and the rooms are frequently rather sterile in appearance. Because of this groups may move to more appealing surround- ings such as lounges or apartments. Mere often this is not done because of inaccessibility and time restrictions since students usually have other classes before and after the IPL meetings. One positive feature is the length of the IPL experience. Each group meets five hours per week in a two- and a three-hour session. This means over a ten-week term the group meets for approximately fifty hours, much longer than most human relations training groups. But if there is more time, it is needed because the nature of the IPL makes it difficult to accomplish its goals in a short time. In many senses the IPL is difficult to describe-—there is much that approximates sensitivity training yet it is far from being that; there is considerable freedom and emotion, yet pains are taken to main- tain limits on these; students are urged to examine themselves candidly, yet they are being graded at the same time; it is definitely a class, and yet it is not a class in the traditional sense. The TA has the responsi- bility to balance these features and achieve a positive learning experience. He must guide and monitor the group to ensure that learning occurs and that 69 neither cognition nor affect suffers at the expense of the.other. The nature of the IPL experience makes this an especially difficult task. Because of this it is important to understand the factors that contribute to successful TA performance. Characteristics of the Sample and Sampling Techniques The Group Leaders Four group leaders (TAs) were studied, each leading one group. All were male, unmarried or divorced and physically attractive. Their ages ranged from 26-34 years. Three were white, the other black. Each had considerable experience leading IPL groups and was viewed as highly competent by colleagues. The Students Sixty-three students comprising four groups were involved in the study. Initially the number of participants in each group ranged from 14-17. However, two of the groups lost students during the term so that three finished with 15 members and one with 14. One group, Nat's, was distinctly different from the others because all the members had already taken the IPL and received an Incomplete. In the other groups only one or two students fell into this category. Nat's group was included because the group originally intended for study was cancelled and another compa- rable group was unavailable. It was felt that having a fourth group was valuable enough to compensate for its somewhat unusual nature. Several demagraphic variables were collected and analyzed to see if differences existed between the groups. No meaningful differences were found between the average age of participants, their level of edu- cation, overall GPA, previous work experience or expectations for the course. 70 In general, most students were young (20 years or under), white, sopho- mores or juniors, single, and came from cities with populations less than 100,000. The average gradepoint varied somewhat with the majority being between 2.5 and 3.5 on a 4-point scale. Expectations were mixed since many reported hearing negative comments about the IPL. Also, some of the students in Nat's group said their first IPL experience had been unpleas- ant. Despite this most said they were adopting a wait-and-see attitude or were looking forward to the experience. Interestingly, approximately 30-40% reported having some sort of group experience prior to or concur- rent with the IPL, for example, group therapy, T-group, growth group. Other descriptive variables have been summarized in Table 4. TABLE 4.--Descriptive Variables for the Four Groups Descriptive 1 groups and TA: 4 variables Nat Ken Hank Mark Total Number of Students 17 15 l4 17 Number and Percentage of ‘Male Students 6 (35%) 3 (20%) 7 (50%) 2 (12%) Number and Percentage of Female Students 11 (65%) 12 (80%) 7 (50%) 15 (88%) Number of Minority Group Students 1 - 3 2 Number and Percentage of Students with an In- complete 17 (100%) 2 (13%) l (7%) 2 (12%) Number of Students Drop- ping Course during the Term 2 - - 2 71 Sampl ing Since this was a required course there were certain constraints on the experimental manipulation of subjects. Student and leader schedul- ing demands made it impossible to randomly assign students to the four groups. However, a natural randomization occurred because the student's choice of a group is usually based only on finding one that will accom- modate the rest of his or her schedule. In addition, students do not know who will be teaching a particular section when they register so selective bias is minimized. As it turns out, the groups were very similar on all variables examined except the male-female ratio. It is impossible to determine if these differences affected group process. There were active and passive males in each group and male-female issues did not arise. Because of this it is unlikely that the sexual composition was an impor- tant factor in these groups. More control was possible in selecting the group leaders (TAs) for the study. Though all were volunteers, they were invited to partici- pate on a selective basis from the larger group of TAs (approximately 20). They were chosen to maximize homogeneity in the areas of physical attrac- tiveness, experience, sex, and background. They were also asked because they represented a diversity of styles, some being extroverted and active, others introverted and passive. It was hoped this would lead to variation in their intervention patterns. 72 Instruments and Observation Techgiques Used for Data Collection Perceived impact Measures Observer Ratings The researcher acted as observer, being present during each ses- sion to record important nonverbal behaviors at the time of an interven- tion episode. She also rated the overall impact of the intervention episode on the participants directly involved and/or the class as a whole. Impact was rated as either clearly positive (+), clearly negative (-), indetermi- nate (I), or mixed (M), meaning some combination of the other categories. A simple form was developed for the observer to use in recording obser- vations and judgments (see appendix B). The observer also made note of any unusual occurrences during each session that might affect student reactions. Post-Session Qgestionnaire (PSQ) This 21-item form was developed to be given out to students at the end of a session. It contained statements to which the students indicated their level of agreement using a 5-point Likert scale ranging from (1) Strongly Agree, to (5) Strongly Disagree. Questions covered four areas of interest: (1) perceptions of TA behavior, both positive and negative, (2) how the student thinks,g£§§g§ see the TA, (3) percep- tions of group solidarity and openness, (4) overall reactions to the IPL experience. Questions were both positively and negatively phrased to prevent response sets and were randomly ordered within the four areas. To guard against reactive effects students were asked to fill the form out for sessions with and without intervention episodes. 73 Though originally it was planned that students would complete the form after each session, this did not always occur for a variety of rea- sons. Sometimes the form was inappropriate given the nature of the ac- tivities during a session, for example, students independently went on a "trust walk," TA gave content lecture only, or students worked in small groups without recordable input from the TA. On other occasions the TA would hold the class overtime and some students would have to leave because they were late for appointments. In these cases the student(s) leaving would be asked to take the form home to complete. Unfortunately, few stu- dents returned these forms and the low return rate made analysis impos- sible. These two explanations account for the majority of missing data. Two other sessions were missed because a replacement observer forgot to distribute the forms. Stimulated Recall Technique (SRT) As discussed in chapter 1, this technique was used to overcome some of the problems of the PSQ which asked participants to react to an entire session. The idea of the SRT was to isolate an intervention epi- sode on tape and rerun the tape segment for the group at the start of the next meeting. Students were then asked to complete a l3-item.form with questions on how they felt when the interaction originally occurred. Certain of the questions were identical or equivalent to those on the PSQ and the same Likert scale was used. This allowed for direct compari- sons between student reactions when the episode originally occurred (on the PSQ) and those given later during the stimulated recall (on the SRT). Unfortunately, only two stimulated recalls were done, one with Hank's group, the other with Nat's. Another was attempted with Ken's group but the poor quality of the tape recordings made it impossible for 74 students to hear the episode. There are several reasons for the small number of recall sessions. One, student permission had to be obtained before a segment involving that student could be replayed to the class. In two instances this was denied. Instructors were also reluctant to re-open some issues, especially when the episode had been lengthy or in- volved sensitive issues. They felt it would be detrimental for the group to go back to these incidents. Finally some episodes were either too short or too lengthy to replay. Peer Perceptions As discussed, it was of interest to determine if group member perceptions of an intervention episode would be affected by the type of student(s) involved. Three dimensions were used in the definition of student type--Activity Level, Verbalness, and Personal Security. At the end of each stage (approximately three and seven weeks), students were asked to fill out a brief questionnaire on every other student in the group. Henceforth these administrations will be known as Time 1 and Time 2. The form contained eight 7-point semantic differential scales developed to reflect the three dimensions of interest. The eight scales were: 1. Confident. . . . . . . . .Lacking in Confidence 2. Introverted . . . . . . .Extroverted 3. Secure . . . . . . . . . .Insecure 4. Outgoing . . . . . . . . .Shy 5. Quiet . . . . . . . . . .Talkative 6. Leader . . . . . . . . . .Follower 7. Aggressive . . . . . . . .Unaggressive 8. Pa881ve. O O O I I O O O oACtive The data were factor analyzed for Time 1 and Time 2. Factor analysis revealed two factors, not three as expected. The Activity Level and Verbalness dimensions collapsed into one factor to be called Activity Level. However, the Personal Security dimension, Factor 2, remained essentially intact. The factor structure at Time 1 and 2 was slightly different as seen in Table 5. TABLE 5.--Factor Structure and Loadings for Peer Perception Data, Time 1 and Time 2 Scales and Loadings Time of Administration Factor 1 Factor 2 a 2 - extroverted (.73) l - confident (.88) 1 4 - outgoing (.73) 3 - secure (.89) (3 weeks) 6 - leader (.74) 7 - aggressive (.86) 8 - active (.82) 1 - confident (.77) 3 - secure (.85) 2 - extroverted (.91) 7 - aggressive (.71) 2 4 - outgoing (.89) (7 weeks) 5 - talkative (.85) 6 - leader (.84) 8 - active (.86) 8Scale 5 did not load significantly on either factor Factor scores were computed for all students for both Time 1 and 2. The range of factor scores was then divided into thirds. A factor score was assigned a l, 0, or 2 depending on which third of the range it fell into. A.l meant that the person was high on the factor, e.g., high activity level, a 0 meant s/he was neutral, and a 2 indicated a low status, e.g., low activity level. Thus each individual received a two- part code for Time 1 and 2, e.g., 01, 22, 12. The code reflects their ratings on the two factors of Activity Level and Personal Security. A full list of students and their codes, by group, is given in Tables 6-9. 76 TABLE 6.--Student Types at Time 1 TABLE 8.--Student Types at Time 1 and Time 2 - Nat's Group and Time 2 - Hank's Group Student Student Name Time 1 Time 2 Name Time 1 Time 2 Don 01 01 Rick 11 12 Jan 21 01 Lila 00 01 L 12 22 at 01 01 sign 00 21 fihciey 11 ll Pam, 02 01 James 01 22 Mary 12 02 Shari 20 22 Lisa 20 " Peter 22 22 Jack 00 10 Kate 00 11 Carrie 22 22 Mack 11 10 Tim. 12 02 Dana 01 10 Joan 22 22 Joe 10 20 Sharrie 22 22 Sarah 00 20 Susan -- 20 Sandra 01 00 Byron -- ll Doreen 21 00 Kurt -- -- TABLE 7.--Student Types at Time 1 TABLE 9.--Student Types at Time 1 and Time 2 - Ken's Group and Time 2 - Mark's Group Student Student Name Time 1 Time 2 Name Time 1 Time 2 Kitty 10 10 Sally 12 10 Jane 20 20 Ali 00 -- Betty 20 20 Liz 11 ll Lana 20 02 Sherry 00 01 Caron 22 02 Tess 12 00 Meg 22 02 Lana 01 01 Joshua 21 20 Ruth 22 22 Jerry 12 10 wa11y 22 22 Barry 22 22 Janie 01 11 Daisy 12 12 Suzie 22 -- Bess 10 ll Rhoda 00 01 Annie 11 ll Rozie ll 11 Peg ll 10 Paula ll 12 Laura 02 20 Dawn 20 22 Dora 00 02 Noreen 01 01 Jill 22 22 Al 10 10 Note: 11 - High Activity Level and High Personal Security 12 - High Activity Level and Low Personal Security 21 - Low.Activity Level and High Personal Security 22 - Low.Activity Level and Low Personal Security 77 Four student types were chosen for this study: 11 - High Activity Level and High Personal Security 12 - High Activity Level and Low Personal Security 21 - Low.Activity Level and High Personal Security 22 - Low Activity Level and Low Personal Security It was possible for a student to belong to a "type" at Time 1 and not at Time 2 and vice versa. Student type was noted so that it could be used in the analysis of data from the impact measures. Description of Perceived and Real_;mpact Measures The impact measures were designed to collect information on stu- dent reactions to the TA, the group, and the IPL in general. The aim was to sort out those intervention episodes seen as effective and inef- fective so that their characteristics could be determined. The PSQ supplied the main data source for this categorization. Post-Session Qpestionnaire (PSQ) It was originally proposed to factor analyze the data from each PSQ administration to determine if students had similar positive or nega- tive views. This method was later rejected because of small sample sizes and a fairly high level of consistency in student reactions found in the raw data. Instead an alternative approach was taken. Further discus- sion of the structure of the PSQ is needed to understand the analysis done. As discussed earlier, the PSQ covers four areas: (1) perceptions of TA behavior, both positive and negative, (2) how the student thinks others see the TA, (3) perceptions of group solidarity and openness, (4) overall reactions to the IPL experience. These can be viewed as four scales and will henceforth be referred to as View'TA, Other TA, 78 View Group and View IPL. The items for each scale are as follows: Scale 1 -- View TA 1. The TA seemed accepting of viewpoints different from his own 2. At times the TA acted overly critical and disapproving 3. The TA made a real effort to help others gain insight into their own behavior and/or feelings 4. At times the TA pushed too hard, beyond a point where it was pro- ductive 5. The TA did a great deal to help build trust in the group today 6. At times the TA seemed too manipulative in getting others to react the way he wanted 7. The TA made a sincere attempt to understand why people reacted the way they did 8. The TA's actions did.ppp help build trust in the group today Scale 2 -- Other TA 9. Some people seemed to feel overly controlled and manipulated by the TA 10. Most members seem to trust the positive intent of the TA 11. Some members seemed to agree with the TA just for the sake of it 12. Most members seemed receptive to feedback from the TA 13. Most members seemed to feel that the TA.made an honest attempt to understand and appreciate differences in opinion Scale 3 -- View Group 14. Most members seemed to be guarded and hiding their feelings 15. Many members seemed uninvolved in the group 16. There is a strong sense of unity in the group 17. Many people seem to identify with one another's needs and/or problems Scale 4 -- View IPL 18. I feel a low amount of trust towards this group 19. I feel I can rely on the TA to be understanding and constructive towards me 20. I am hesitant about expressing my feelings in class because I don't think the TA will give them fair consideration 21. I feel positive towards the IPL experience so far Students were asked to agree or disagree with each statement using a 5-point Likert scale ranging from (1) Strongly Agree to (5) Strong- 1, Disagree. Statements were purposely value laden to assess if students were reacting favorably or unfavorably. Scale scores were computed for each administration of the PSQ. This meant reversing some of the item scores so that a (5) always represented the most favorable rating and a (l) the least favorable. 79 It was initially reasoned that a scale score of 3.00 would repre- sent the theoretically neutral point, that is, students were neither clearly positive nor negative, or the positive ratings cancelled out the negatives. Given this assumption, a two-tailedIE test could be used to see if the actual mean scale score differed significantly from 3.00. If it did then the score could be categorized as positive (+), negative (-), or neutral (N). Preliminary analysis proved 3.00 to be an inappropriate neutral point. Student reactions were so negatively skewed that using 3.00 al- most every score would be categorized as positive. Such categorizations would be contrary to many observer ratings and student comments recorded in the transcriptions. The negative skewness is probably due to the tendency to give the TA and the IPL experience the benefit of the doubt, that is, to give fairly high ratings even in cases where the respondent is unsure or even feels slightly negative. As a result the distribution of responses moves up the scale and so does the "psychological" neutral point. Given this, a new neutral point, 3.50, was chosen. Analysis proved this to be more appropriate and categorizations were more in line with what would be expected from observer ratings and student comments. Determination of BehaviorallAttitudinal Changes The interventions being studied all had a corrective aim, that is, to change some behavior and/or attitude that was inhibiting the person(s)' full participation in the group. To be totally effective the intervention episode should lead to behavioral and/or attitudinal changes in the desired direction. It was important to identify intervention episodes with real, as well as perceived, impact so that their character- istic features could be examined. To do this a group of intervention 80 episodes was studied to determine their impact on the participants involved. Only a select group of intervention episodes could be examined because of financial and data-processing limitations. Selection was based on several criteria: 1. There had to be three or more intervention episodes dealing with the same critical incident issue, e.g., Reluctance to Give Negative Feedback 2. Two or more TAs had to be represented 3. PSQ data had to be available for that session. In three instances this cribaion was bypassed because there was clear evidence of the impact and the episode was judged as important for study Altogether twenty-four episodes representing eight critical incident issues met the criteria. In order to assess if change had occurred a baseline performance level was established using the three sessions directly preceding the intervention episode. Only the behaviors and/or attitudes of concern in the intervention episode were monitored. Performance measures were then taken on the three sessions following the episode to see if changes occurred. In addition, an assessment of the immediate impact of the epi- sode at the time it occurred was also made. In both instances the impact will be described in paragraph form. The reasons for limiting the analy- sis to three sessions pre- and post-episode have already been discussed in chapter 1. 81 Description of the Analysis of Intervention Episodes Recording of Sessions Audio recordings were made of all sessions using two standing microphones placed at strategic places in the room. The researcher (observer) sat in one corner to run the equipment, gather forms, and answer questions, as well as collect nonverbal information at crucial times. Each group used a different meeting room. They were all regular classrooms used at other tines by various deparhments. This made it im- possible to use a permanent overhead microphone or to leave equipment since security could not be guaranteed. The acoustics varied in quality from good to poor. In one room interference from overhead lighting pro- duced tapes of such poor quality that students were unable to understand them when a stimulated recall technique was attempted. Future attempts to use this technique with this group had to be abandoned. Determination of an Intervention Episode The researcher (transcriber) had the responsibility of determin- ing when an intervention episode occurred and its boundaries. This was done using the guidelines discussed in chapter 1 ( p. 23). Briefly, the episode was said to begin at the point where the TA became sensitive to the critical incident and made a move to address the problem. Conse- quently, each transcription begins with a comment by the TA. The termi- nal boundary of the episode could be either implicit or overt. To be implicit, the leader had to redirect the focus to another student and/or topic or a student did this. An overt resolution meant that at least one of the parties involved verbally acknowledged the resolution of the con- cern or asked to stop the interaction. 82 On occasion one intervention episode was nested within another so that they could not be separated without distorting the nature of the interaction. When this happened the episodes were not separated but a line was drawn across the transcription to indicate that another issue was being addressed. When two episodes occurred on the same transcrip- tion they were analyzed together. Episodes were separated for transcrip- tion and analysis purposes wherever possible. Transcriptions Written transcripts were made of all intervention episodes. An example of a portion of transcript is included to illustrate the stand- ardized format used (see following pages). Coding As described in chapter 1, a modified version of the Issacson Scale of Interaction Analysis was used to code the transcripts. Both TA and student comments were coded although only data relevant to the TAs were analyzed. Student comments were coded because future research may examine the interaction and sequence patterns between TAs and stu- dents. Given this possibility it seemed best to code all comments at the same time. The author was responsible for coding all transcriptions. This was felt to be justified by reasonably high inter- and intra-rater relia- bility figures. Inter-rater reliability was determined by having the author and another graduate student code a sample of transcripts and comparing the results. It was found to be .78. Intra-rater reliability was determined by having the author re-code several transcripts after a three-month interim.period. The extent of agreement was found to be .84. 83 N<, \.ooo one: ow omen oo one as snow 9H \umwaum one mum o3 hm3 one ous“ um oumuwoucw 9H.o3 «mono mo uoo um» 63 mono menu new .85 no umuwm um um no :uHB eumsxam on On wowow mo uuom ow.:om pom mHHfixm oau wcaowuumuo ou.o3 omomooa umsfi ow m9 .xowsu maoooo one? wcauoocoa omen mmz H .wcwumoooa 8.9 mmmow H \.mwoun N< pom mouoocoo onu wcwhmm .hm3 moo umofi one xomoeoom wafi>qm 8.9 does mafia «mod lHNMQI-GONGDGS w< \ asses use; lgOOOOOOOOOOOOOOOOOO000.00....0.0.000...OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOIO0.0.0.0.000... omen mafia woaxamu announces .ewmax3m macaw one when ooA Am muaafiumom\ooooumwomm n efi ”unoowonH adequauo An message m an mu eoammmm an apex A< N e m x» m NIH mcaflaoo xoono u muaanou <9 mo :aooo a muooaaoo <9 a muooaaoo ucoooun mo wcwooo n mucoaaoo unmosum u women: mean u Aoawu oau mwo xuma Ou own: one nocwav mouoowa.cfi oEH9 u H cabfioo NMQUNON "m3oHHom mm oonqoumoomum ma unawomoouu ago no woooamaou oe9 <9 can he coauco>uoucw as» ou wowed umofi wcficoooma nm3 owes mo nouuofiuumoo < Am oo>~o>=fi Amvuooowucw Hmo«u«uo 0:9 An unnamed“ one you new» House on9 Au Hones: :onmon os9 Am <9 on9 n< «nowuomoouu some mo mom can nmouom summons mcaaoaaow op9g u Huomcou9 mo coauuom m we umwamm 84 T: No no ND ND o .......................n....................... .3... \ Ao>HuHmoov mop AHHmou H .umsu woaumsm woo» uuooeom H .o: umnu unwoouo :0» uses amoum hHHmmu m.ua xcwnu H \.ua memos now common a saw no» ma and H \.uomoa ones u.an umnu use omuoooxo m.uH HHoz \.Homw menu 30: zooomEOm onxm< \NuH onoo now common m one so» \uoumzxsm mHomm ammo and: 00 04 co no no mo 04 o< a .UMQ 53wan man “50‘“ o o o o o o o O o o o o o unwaoau omHm H \.oum3x3m mo mowx Hoom gosh H .m.um:u Hoom mo coax umofi H one .mQaHuoEOn .nwoosu oaHo sou HHo uoc .moeauoaom «00H \.oOu onau mama moo osmooau H «Homm \.m: mo oouoooxo m.um£3 mo once mmB ammo wowsosu H .cooa H .nmow \Aumea wagon \.um£u wonmowo oHoaouu on>mn hHHmou E.H one com um mo anon on we cu comma nooumum on was umnu one: now H ouowoo omomoon m.uH mmoow H \.umnu uaonm Hoom com oo 30; on .mHnu one mHau one anu oxHH .3oox a» .umsu oxHH meu uo>oa H .Hoow mesa 3o: moonoEOm xom H cuss oxHH \.wanoooo3 omen mos H van ow nuHa oum333m onHoom zHHoow E.H onamo .hmxo «who: aH Aoooowucoov uoawomomw9 mo coHuuom n no an an on mm «m. mm Nm an on ma m~ AN em nu em MN Nu Ha o~ an we AH in. so no so no NH OOOOOOOOOOOOOOOOOOOOOInOOOOOOOOOIOOOOOOOOO...N H EH00 oHoamm 85 m.nqmmm couuna mH man: was :OHummuo>aoo oau mwounouou omH moon .Hsmm «0 non» ma umooom ou can: oxen 6:9 .:0Hummuo>boo on» oucH heads no: Hmcan one was muaoooou ou mono mHoo monsoon memo m.ooH uoHuomamuu oHoamm moo oH .woaxmoom mH unmeaum same one away cabana coo wooden may awesome man: ucmummeo m mmmHon .mmeu unwound onu wooH so: umuuoa o: mono hHoo couuna mH was: oe9 .coHuMmuo>coo moo nuouco on\m we can: he oonHuoooH mH uoooaum 50mm .muaoooum macaw anowcHumHo cu once on: wowsoHHom os9 .uowuomdmuu some «0 ammo umuHH on» no oou on» us ao>Hw one c3oox homouHm ow mHnu macaw uxou sou aH <9 mnu no man: man hmwooom ou use: on mH ouoou momma an9 .ooHn one: umoH men so muaoaaoo unooaum HHm can onm one: unwwu one no wcHon munoaaoo <9 HHm sow: wuoHuomomuu one no umEHOH onu cH some mH nowuocHumHo ucooaum\<9 «59 n wcwxooom oH on: no noHuuUHocH .mumo wnHmmHE_mn wouuoumHo uos nH ooHucm>houaH ouwuao sou mo auwooH Hmouum ecu umnu om aOHuuom wnammwa man no auwooH onu mouoaaxouooo oawH mouuoo 6:9 .uH opauowcouu cu nwoooo HHos mcHouooou moo use; cu oHonmoo uos was uH umnu momma muoo mo oan < n A......v moon Ho mooHH .moHuHaonmHo coHuoHuooomuu onuoomouo mm HHo3 no ucooHooH onu mo nuwooH HmuOu sou uuoumHo oHoos mHsu omomooo axon on» awsuHB mooou one oomHo ou oHnHomoo uoa mus uH .moHHooo uH noHea ou uxou onu ou axon oaaHoo sou :H wanmoooo oooo onu .0u summon oooo m uxou mo mooauuoo :oHna 3o£m on own: mu “\v ammHm < .suooom mo newsman moo oHnuHa nucoaououm oHomooo ucoummmHo woos no o3u on HHH3 ouoeu one wcwamu on HHHs oncoEOm amumo u a\v monmmHm .m .N .H unBoHHom on uoHuumcmuu man no mowaumow unauo nHmo>ou noHuuoo oHoamn m£9g Aooaowuaoov uoHuumcmu9 mo aoHuuom m H0 onamm To prevent bias, all coding was done prior to the examination of student reaction data. Some of the criteria for coding have already been discussed in chapter 1. For a fuller description see appendix B and Issacson (1976). Summarization of Codes Each transcript was coded using the modified version of the Issacson Scale of Interaction Analysis. Once this was done the TA codes had to be summarized. To do this each code was conceptualized as a "verbal move" by the TA. It was then possible to determine the percent- age of moves (codes) in a category, e.g., C3, A2, by dividing by the total number of TA.moves (total number of codes). This was done for all B cluster categories resulting in nine percentages for each intervention episode.1 The affective quality of the move was ignored in this calcu- lation so that, for example, moves A2 (Affective Active Listening) and C2 (Cagnitive Active Listening) were placed in the same category. Other percentages were calculated to examine different aspects of the TA's behavior. 1. Percentages to look at the Domain of each move. It should be recalled that the domain could be affective or cognitive depending on its focus. Affective meant the focus was on feelings, cognitive that it dealt with thoughts or anything not related to feelings. Two percentages were calculated: a) The number of Affective moves b) The number of Cognitive moves The total number of TA moves The total number of TA moves 2. Percentages to look at the Affective nglity of each move (see chapter 1, 1see chapter 1, p.24 for description of B cluster categories. 87 p. 26). Two percentages were calculated in this area: a) The number of Positive moves b) The number of Negative moves The total number of TA.moves The total number of TA moves . Percentage of TA/Student Participation. This was done by comparing the number of lines of text spoken by the TA to the total number of lines of text in the intervention episode. The number of lines was used be- cause it is much easier to calculate than time or words, yet it still provides a good picture of the proportion of TA to student talk. These percentages and the actual frequencies were placed on a 5" x 8" index card along with other descriptive data for each interven- tion episode (see appendix B). Information on Summary Index Card for Each Intervention Episode 1. 2. 10. of Critical Incident Number (1 - 10) Session Number TA Number (1 - 4) Group Stage (1 - 3) The Number of Students Directly Involved (for future data analyses) Total Time (to the nearest half minute) Type of Student(s) Involved (ll, 12, 21, 22 and the number of each type) Frequencies for all Code Categories broken down by Domain and Affective Quality Percentages for all Code Categories plus the five special categories discussed (maximum.of 14) Resolution Type (Implicit or Overt) These cards now became the source for analyses examining the nature effective and ineffective intervention episodes. 88 Research Quespions This project was designed as a fact-finding, hypothesis-generating piece of research on the characteristic features of effective leader interventions in the IPL setting. A secondary aim was to examine the features of interventions that have negative effects on the group. It was also desired to obtain information on the utility of the untested methods and the unit of analysis, especially the stimulated recall tech- nique and the intervention episode unit of analysis. These aims are re- flected in the original research questions of interest. As will be dis- cussed in chapter 4, some of these questions were later dropped and others were substituted. 1. LOOKING AT THE UTILITY OF METHODS AND THE UNIT OF ANALYSIS a. Is the intervention episode a useful sequence unit of analysis? b. Are there benefits to using stimulated recall techniques as opposed to post-session questionnaires? 2. LOOKING AT THE CHARACTERISTIC FEATURES OF EFFECTIVE INTERVENTION PATTERNS a. What is the nature of those interventions perceived as effective by students which also lead to behavioral/attitudinal changes? (modifying variables are not controlled) Dimensions of the intervention episode to be examined for this and all other questions: 1. Total Time 2. Percentages for ISIA categories 3. Percentages for Domains - Affective & Cognitive 4. Percentages for Affective Quality - Positive & Negative 5. Percentage of TA/Student Talk 6. Nature of an Acceptable Resolution - Overt & Implicit b. What is the nature of those interventions perceived as effective by students but which do'ppp lead to behavioral/attitudinal changes? (modifying variables are not controlled) 89 c. Controlling for the critical incident issue only, what is the nature of those interventions perceived as effective by students which also lead to behavioral/attitudinal changes? d. Controlling for the critical incident issue only, what is the nature of those interventions perceived as effective by students which do not lead to behavioral/attitudinal changes? 3. LOOKING AT THE EFFECTS OF MODIFYING VARIABLES a. Controlling for the type of student only, what is the nature of those interventions perceived as effective in each of the three group stages? b. Controlling for group stage only, what is the nature of those interventions perceived as effective with different types of students? c. What is the nature of those interventions that lead to behavioral/ attitudinal changes with different types of students? 4. LOOKING AT INTERVENTIONS WITH NEGATIVE CONSEQUENCES a. What is the nature of those interventions that are perceived as ineffective by students? Perceived effectiveness was based on the analysis of the PSQ data. Though there were four scales in the PSQ, only two were of inter- est in determining perceived effectiveness, View TA and View Other. Both these scales focused on student perceptions of the TA's behavior. The significance level for these scales was the same in all cases, i.e., there was a positive correlation between significance on one scale and significance on the other.1 This uniformity meant that separate analyses were not needed. 1In two cases the scale score for View TA was significant at the .01 level while the scale score for View Other was significant at the .05 level. Since both were acceptable significance levels this did not pre- sent a problem in categorizing the episode as Perceived Effective. 90 ggatistical Procedures Due to the small sample size and the exploratory nature of the study, only descriptive statistics were employed. The data were organ- ized into a series of tables according to the demands of the various questions and the dimensions to be explored. These frequency data were used to look for commonalities among effective and ineffective inter- ventions. Commonality was assumed when at least two-thirds of the inter- ventions being considered for a question shared a particular feature. This type of examination was done for all questions with the limitation that there had to be at least five acceptable interventions available for study before any attempt was made to look for commonalities. Differences among the TAs were also examined. These will be discussed where appropriate to illuminate some of the factors that may have contributed to effective and ineffective intervention patterns. Findings in this area are based primarily on Observer ratings and student comments . CHAPTER IV RESUETS Once the intervention episodes (I.E.'s) were identified it became apparent that several research questions would have to be dropped because of inadequate sample sizes. New research questions were added and the methodology was modified to expand the possibilities for analysis of the data that were available. It is therefore necessary to describe the data and the methodological changes before proceeding to the final set of research questions and the corresponding results. Descriptive Statistics Table 10 provides a breakdown of all I.E.'s by critical incident and TA. As shown, altogether there were forty-eight intervention epi- sodes. All of the critical incidents were represented to varying degrees except number 5, "Refusing to be Responsible for Self." This exception is noteworthy since several experienced TAs had identified this critical incident as occurring frequently in groups. It may be that TAs see stu- dents failing to take responsibility for themselves but that they do not address this as a separate issue. Instead it is dealt with in the con- text of other incidents such as "Resistance/Hostility" or "TA put in Authority Position." Several other critical incidents were rare, number 3, "Failure to Own One's Feelings," number 8, "Overtalk/Inability to Facili- tate Self-Closure," number 9, "Failure to Stay with the Here-and-Now," and number 10, "Negative Expectations." In the case of numbers 3 and 9, 91 92 the TA would usually ignore the incident or deal with it so briefly that the interaction could not be categorized as an intervention episode. The small sample size for number 10 is understandable since TAs typically deal with student expectations once, at the beginning of the term, and not after that. As for number 8, it seems this problem is just unusual. TABLE lO.--Frequency of Intervention Episodes by Critical Incident and Group Leader Groups and Leaders 1 2 3 4 Critical Incidents Nat Ken Hank Mark Totals 1) Reluctance to Give Negative Feedback - - 2 2 4 2) Non-Participation/ Withdrawn, Shy, Nontalkers 8 3 - 6 l7 3) Failure to Own One's Feelings l - - l 2 4) Resistance/Hostility 2 l 2 5 10 5) Refusing to be Re- sponsible for Self - - - - - 6) TA Put in Authority Position 1 3 l 2 7 7) an-Facilitative Emo- tional or Nunverbal Behavior 2 - l - 3 8) Overtalk/Inability to Facilitate Self-Closure - 2 - - 2 9) Failure to Stay with the Here-and-NOw - l - - l 10) Negative Expectations 1 l - - 2 Totals 15 ll 6 16 48 93 Locking at overall frequencies, the highest number of I.E.'s involved the critical incident "Nonparticipation/Withdrawn, Shy Non- talkers," followed by "Resistance/Hostility" and "TA Put in Authority Position." Together these accounted for 71% of the total number of I.E.'s. These critical incidents also appeared in several groups indi- cating the generality of the issues. They reflect the basic dynamics of IPL groups: the need to verbalize versus resistance to being evalu- ated and forced to participate, the desire for the TA to take reSponsi- bility for running the class versus the inherent necessity that students take a role in shaping the group if they are to diSplay and practice the skills being taught. It is therefore not surprising that these critical incidents occurred in larger numbers across all groups. The table is broken down by TA to reveal that differences were found in the type and frequency of critical incidents for each group. However, it is beyond the scope of this project to fully explore the significance of these differences. It is the aim of this research to search for commonalities among the TAs in the way they handled the criti- cal incidents and to relate these commonalities to the consequences of TA behavior. Changes in the Methodology Perceived Effectiveness As discussed in chapter 3, determination of effectiveness was based on the analysis of data from the first two scales of the Post-Session Questionnaire (PSQ). As Table 11 shows, there were many sessions where PSQ data were not available, yet intervention episodes occurred. 94 TABLE ll.--Availability of Post-Session Questionnaire Data for Sessions with Intervention Episodes by Group Leader, in Percentages Groups and Leaders Intervention Episodes] 1 2 3 4 PSQ Data Nat Ken Hank ‘Mark Total Sessions with Inter- vention Episodes & PSQ data 6 4 3 4 1? Sessions with Inter- vention Episodes but £2 PSQ data 4 5 2 3 14 Total Nomber of Sessions with Intervention 10 9 5 7 31 Episodes The reasons for this lack of data have already been addressed. In order to extract information from these sessions it was decided that observer ratings of effectiveness would also be utilized. Analysis of intervention episodes was carried out in two ways to account for the possibility that the observer might judge I.E.'s differently from the students. First, only those I.E.'s where PSQ data were available were analyzed for common characteristics. Then I.E.'s categorized using observer ratings were added to increase the size of the sample and another examination for com- monalities was done. Results for each grouping will be discussed sepa- rately unless otherwise specified. Observer Ratings Since observer ratings were used to categorize I.E.'s, it is important to take a closer look at how the ratings were made. Ratings 95 were based on direct observations, listening to tape recordings, and examination of the transcripts for signs of impact on the individuals directly involved and/or the entire class. Impact refers to the immedi- ate reactions of those involved, be they positive, negative or indetermi- nate. The idea was for the observer to try to assess how the students were perceiving the I.E. The impact on the whole class was rated when the class itself was the object of the intervention episode or when the intensity of the episode was such that it had an indirect impact on everyone. Four different ratings were given: (+) clearly positive, (-) clearly negative, (I) indeterminate, meaning the observer could not be sure of the reaction, and (M) for cases where the class was the object of the intervention and there were different reactions. A rating of GM) meant there was some combination of the other three categories. Table 12 summarizes the observer ratings for all intervention episodes. As shown, whenever a rating of CM) is given the underlying ratings are also indi- cated. Table 12 also has two other features that need to be discussed. Each intervention episode is uniquely numbered from 1 to 44. These identification numbers will be used in all tables whenever a par- ticular intervention episode is being referred to. They also appear in the upper righthand corner of each transcript in appendix D.1 This should make it easy for the reader to quickly locate the corresponding transcript for any tabled intervention episode if so desired. The num- bers also allow the reader to see when an intervention episode falls into 1Appendix D can be found in the Reference Library of the Institute for Research on Teaching, College of Education, Michigan State University, E. Lansing, MI 48824. 96 .<.z .<.z .<.z new: .<.z mmoHu ouHuom now mwcHumm uomoEH + H .<.zfi.<.z oo>Ho>cH mucooaum Mom mwcHumm uumoEH mano «H NH mH mH nano mH cH MH no mo mmmHu mH no #0 ommHo mmoHu emcee ma~vmmmao mH Ho ho mnMHo uo\o:m ou>Ho>oH nuaoooum mo uoaaoz mooo OH OH wooasz uaooHooH HmoHuHuu mH MH HH woosoz :OHmmom mH .qH MH NH HH OH w n o m e nooowHom :on:o>uoucH .muoasoz ooouu so moomHom oOHuco>uouoH zoom mo uomoEH may no mwaHumm uo>uomoonl.NH m4m<9 97 .<.z .<.z .<.z .<.z .<. Haw .<.z OHo>cH mucoooum now monumm uuooEH «N NN Nu cm HH oH mm wH «N mN mm «e um mmmHo on on Hmv em HevmmmHo cc mmmHu Hm mano uo\ecm oo>Ho>=H muoooaum mo nooasz ocoo wooeoz uoooHooH HmoHuno wH mH NH oH nH «H MH HH oH nooesz oonmom on mm «H RN 0N mN «N mm mm HN ON mH mH NH oH mooowHom :OHuco>umu=H emscHueoouu.~H mHmHuomoo hHuooHu HIV Ho>Hanoo HHuooHo I H+v unaoHHou no one nwaHumm "H902 o o o o o o o o o o H . + . v o o o o o H . o o o a M . mum my ”PM “O < z < z < z. < z < z H+vz HHvz < z < z < z Hue: < z < z H+V= wwwHuum uwaaam + + H I H H + em>Ho>=H H + I I mucovaum you H H + + H H nwcHumm H + + n .. .<. 2 H H + u I u n H unmaaH mmmHo H~vHs mm HNVms Hm HNVHH Haven mm mmaHo uo\eem HNme He.evme o4 Hm eu>Ho>eH mm Haven om Haves mm co mueoeaum we as He ma Hm co _mmmHu H~.anm an Hm He Haves as me He umaaaz ween N s e H s N ~.H a N a e.e n N N umaaaz ueoeHueH HaUHuHuo oH HH mH NH a m m m m m e e a m umaasz eonmmm so me Ne He as an an Hm on mm an mm Hm Hm mooomHom oOHuco>uoucH emaeHueoouu.~H mHm10 min ** 61% :>10 min TA/Student . . . ... ** 542 25-502 Talk Ratio Affective * 801 0-252 * 85% 0-252 * 77% 0-252 Domain Cognitive * 80% 75-1002 * 85% 75-1002 * 77% 75-1002 Domain Positive * 1002 0-252 * 1002 0-252 * 1002 0-252 Affect Negative * 1002 0-252 * 1002 0-252 * 1003 0-252 Affect ISIA * 2,3,5,6 * ... 2,3,5,6 * ... 2,3,5.6 Categories *' 4,7a ** 7a ** 4,73 Nature of * 73% Overt * 85% Overt * 92% Overt Resolution Number of Episodes 15 20 13 NOTE: A single asterisk (*) means that 67% or more of the intervention episodes shared the characteristic(s). the characteristic(s). NOTE: A double asterisk (**) means that 50-66% of the intervention episodes shared Figures in the column, Z Comm., refer to the actual percentages of intervention episodes sharing the characteristic(s). 112 TA/Student Talk Ratio A double ** rating was found for the value range "25-50%" for question 3. Of greater interest was the Observation that in all but one instance, I.E.'s in this value range also lasted over ten minutes. This relationship was observed in later analyses for both perceived effective and ineffective I.E.'s. Affective Domain A single * rating was found for all questions in the value range "O-25%." Seventy-five percent or more of the I.E.'s for each question fell within this value range. Putting it another way, only 15-23% of the TA codes were rated as having to do with affective issues. Clearly the image of the sensitivity training group leader who always deals with feelings did not apply to these I.E.'s. In part the low percentages are an artifact since not all of the TA codes can be affective. The ISIA categories 4 (Directs and Suggests Solutions) and 5 (Offers Information) can only be cognitive, and Category 5 was frequently coded. Still, the 3 to 1 ratio of cognitive to affective codes indicates that student and observer perceptions of I.E. effectiveness were associated with a high amount of cognitive input by the TA. Cognitive Domain Since the number of affective codes was found to be in the range "0-25%," it necessarily follows that the remaining codes fell into the "75-100%" range for this variable. The underlying percentages of I.E.'s sharing this feature are also identical to the figures for the Affective Domain. 113 Positive Affect A single * rating was found for the value range "0-25%" for all questions. In fact 100% of the I.E.'s fell within this range. Actually, most of the I.E.'s had pp positive affect ratings. Only 5 (22%) of the 23 I.E.'s represented across all questions had any coded remarks that were rated as displaying positive affect. Does this mean that the TAs were not warm, supportive, and caring in their manner? The answer is no, because for the most part these ratings did not take into account the nonverbal behavior that accompanied TA comments. It is well accepted that nonverbals are crucial to interpreting verbal messages from others. Given this it is probable that many remarks that were not rated as ex- pressing positive affect were perceived by students as doing just that. Observations showed that a good deal of positive affect was expressed nonverbally during I.E.'s classified as perceived effective. TAs had what could be called a caring attitude, characterized by smiling, at- tentiveness, body posture and occasionally touching group members. Mbre- over, they showed their caring by trying to elicit student feelings and by checking to make sure student concerns had been resolved. Some good examples of this type of checking are shown by Ken in I.E. number 16. Ken: Daisy, you expressed a while ago some non-closure and not knowing where things are. Where we've gone, have you resolved that? and Jerry just expressed something to you, what's your re- action? Did that make you feel any differently? How does it feel any different now than it did ten, fifteen minutes ago? If so, what's different? Based on this discussion, it may appear that the findings relating to this variable should be dismissed as having little meaning. But in fact they do provide a valuable piece of information because they reveal 114 that TA verbal expressions of caring, liking, and trust are infrequent among interventions perceived as effective. Again this is contradictory to the general stereotypic image of the sensitivity group leader ef- fusively spouting words of caring, trust, and the desire to form intimate, personal relationships with group members. Instead these results and the observer's notations showed the instructor-student role distinction to remain in effect throughout the IPL experience although the boundaries varied somewhat across TAs. Negative Affect Similar to the variable Positive Affect, a single * rating was found for the value range "0-25%" with 100% of all I.E.'s having negative affect ratings in this range. Only 2 (9%) of the 23 sessions represented across the three questions had any coded remarks with attached negative affect. In contrast to the interdependency of percentages for the vari- ables relating to Domain, the ratings on this variable are essentially independent of those for Positive Affect. ISIA Single * ratings were found for the values 2, 3, 5, and 6, corres- ponding to the ISIA categories Active Listening, Eliciting Questions, Offers Information, and Self-disclosure. Questions 1 and 3 had double ** ratings for ISIA Category 4, Directs or Suggests Solutions, while for question 2 a double ** rating was found for Category 7a, Responsible Negative Feedback. Though this gives some indication of the types of communication skills used by TAs, it seemed equally if not more important to know something about how frequently categories were used. Consequently, a rank ordering based on frequency of usage was performed within and across 115 the I.E.'s for each question. It turned out that cagegories 5 and 2 ranked first and second for all three questions. The third rank varied, being 7a for question 1, a tie between 7a, 6, and 3 for question 2, and no obvious third for question 3. The ISIA categories 5 and 2 merit further comment. Category 5 includes many behaviors typically associated with being a teacher, i.e., establishing rules of conduct, offering cognitive infor- mation, providing cognitive frameworks for understanding, summarizing group process, and so on. Issacson (1976) defined it this way in his manual: This type of talk presents responses to questions or self-initiated facts or information concerning the content, subject, or procedures being considered. It consists of facts outside of one's own ex- perience. . . . Very often during a group the facilitator or a group member will comment on the ongoing interaction. These types of statements are explanations or observations of the group beha- vior and are called processing or critiquing in the field of group dynamics. (pp. 219-20) Issacson supplements these criteria with various examples and guidelines to assist the coder in their application. For example, opinion offered as if it were fact is coded 5; so are side comments and social amenities. The inclusion of the latter tends to inflate to some degree the number of 5 codes given. However, the bulk of 5 codes are applied to the types of teaching behaviors just discussed. The author expanded Issacson's definition to include all instances where the TA is providing information in the role of the leader or teacher, as opposed to being just another group member. A few examples should clarify this distinction. In these examples the leader is supplying in- formation about how he wants to lead the group, his style of operation. They tell you something more about him.as the leader, rather than reveal- ing more of his personal self. 116 Well, I'd like to get those things out in the open so we know what we're dealing with, in terms of fears, expectations that you have. I figure that's real. 2:. I would really like to take the time at the end of a Wédnesday or Friday to say, hey, what's going on for you, how did you feel about the experience today. or Something that I hear a lot that really kind of frustrates me is people saying that gee, all you have to do is go in and talk and that is not where I'm coming from. For me it's not the quantity of what we say at all, it's the way we say what we say. Contrast these statements to those that would be considered self-disclosure of a more personal nature. And I believe that people, my bias is that people who believe in themselves can cope in most any situation. People who can be brilliant academically and do not believe in their abilities to cope beyond the academics lack confidence, socially for instance. I also would have some hassles with saying, I wouldn't call them an internally secure person. 0.1: You said something about people don't behave like this or you just don't act like this and I guess I'd have to say I do. I feel that the way I am.in here is the way you'll see me wherever it is. . I live the way I am. Though this distinction seems fairly obvious from the examples, in reality it was often difficult to make. Tone of voice and contextual cues had to be used to decide if a statement should be coded as a 5 or 6. Appendix B includes a summary of the guidelines used by the researcher (coder) to assist in making decisions between ISIA categories. As already discussed, Category 5 was first in the rank ordering of ISIA categories based on frequency of usage in those I.E.'s classified as perceived effective. What can be concluded from this? First, that leaders in these groups do not relinquish their roles as instructors despite the nature of the content and being in a group experience. Rather they spend much of their time providing the cognitive framework students 117 need to understand the experience. Second, a predominance of cognitive input seems to be highly associated with those I.E.'s perceived to be effective. But successfifl.leader behavior in these I.E.'s was not simply a matter of giving information. Ranked second in frequency of usage was Category 2, Active Listening. Here the leader acts to elicit information and feelings from the students. Issacson (1976) defined this category as follows: For a statement to be coded B:2 it must meet two criteria: First the statement should keep the focus of the interaction on the person who has previously been talking and on the subject that person was talking about. Second, the statement must encourage the previous speaker to go further in the interaction, to elaborate on his/her ideas or feelings. Active listening serves as a reinforcement to the speaker in that it is communicating to the speaker that he has been heard and that someone is interested in what he said and would like to hear more. There are generally three types of statements which are coded as active listening: paraphrasing, clarifying statements, and exploratory statements. (pp. 212-13) When actively listening the TA is trying to understand the speaker, to clarify feelings and ideas the person is expressing. As with Cate- gory 5, the high rank position for Category 2 reveals more than infor- mation about frequency of usage; it shows that sincere attempts by the TA to understand group members are an important contributor to student perceptions of effective TA behavior. Nature of a Resolution All three questions had a single * rating for the value "overt." The underlying percentages were high with approximately 73-92% of the I.E.'s in each question qualifying as having an overt resolution. An overt resolution means that some closure has been brought to the I.E. by the TA and/or those involved. In order for an episode to be rated as having an overt resolution there had to be some indication that it was completed. This did not always mean that the issues had been resolved, 118 just that there was evidence to suggest that the episode was ended. Some- times this involved a statement such as "That's all for today," "Let's take a break," or "I want to move on to something else now." Usually an attempt had to be made to check with those involved in the episode to en- sure that some understanding had been reached before moving on to another topic. For example, in I.E. number 9 Nat checks out with Sharrie to see if she understands what he has been explaining to her before allowing the conversation to change direction: Nat: (after some discussion about Sharon's behavior) Does that ‘make sense to you? Sharrie: Yeah, it does . . . (changes topic) Or from I.E. number 21 with Ken: Ken: I guess I want to know where you guys are, do you still want more feedback,do you want to work out some more things? Joshua: No, I want . . . (changes topic) A resolution was rated as implicit when the issues involved seemed to be dropped without any indication of closure. Rather the topic just seemed to drift to someone or something else. Often there was the im- pression that the people in the episode had been left "hanging," their concerns unaddressed. Contextual and nonverbal cues often had to be used to decide if a resolution was implicit or overt. Nonverbal cues were especially im- portant in making these decisions since an episode might be closed by giving nonverbal signals that further discussion on the matter was dis- couraged, for example, shifting eye contact or nodding for someone else to give their input. Returning to the findings for this variable, it can be seen that the highest percentage of I.E.'s with overt resolutions was found for 119 question 3, PSQ data and observer ratings for I.E.'s over five minutes. Two explanations can be given for this. First, longer interventions work to increase the probability of closure since it would be awkward to let the conversation drift off to another topic when so much time had been committed to discussing the initial concerns. Second, the inclusion of observer ratings works to increase the percentage of I.E.'s with overt resolutions as demonstrated in question 2. This seems to indicate that the observer found adequate closure to be one determinant of effectiveness. Questions on Perceiged Effectiveness by Critical Incident To review, a critical incident is the precipitating event that triggers an intervention episode. It is critical because it reflects some underlying issue that needs to be dealt with in a corrective manner so that those involved can have a productive group experience. The issue becomes the focus of the intervention episode. It was felt that there might be some relationship between the issue or topic of an I.E. and the characteristics of TA behavior perceived as effective. Sufficient data were available to examine this hypothesis for two critical incidents, "Nonparticipation" and "Resistance/Hostility." Table 16 summarizes the results. Perceived effectiveness was based on a combination of PSQ data and observer ratings. The findings for each variable will be presented briefly. Discussion will focus on those areas where the results were different from those obtained when all I.E.'s classified as perceived effective were examined together. 120 TABLE l6.--Commonalities Across Intervention Episodes for Questions 4 and 5, by Major Variables. Questions Perceived Effectiveness x Critical Incident 2 1 Perceived Effectiveness x Critical Incident 4 Variables (Question 4) (Question 5) Level % Comm. Level % Comm. Comm. Comm. Value(s)r Comm. Comm. Value(§) Total Time ** 50% > 10 min ** 50% > 10 min TA/Student ** 63% 25-50% . . Talk Ratio Affective * 75% 0-25% * 83% 0-25% Domain Cognitive * 75% 75-100% * 83% 75-100% Domain Positive * 100% 0-25% * 100% 0-25% Affect Negative * 100% 0-25% * 100% 0-25% Affect ISIA * 2,3,5,6 * 2,3,5,6 Categories 7a Nature of * 88% Overt * 67% Overt Resolution Number of E Episodes § 8 6 i NOTE: A single asterisk (*) means that 67% or more of the interven- tion episodes shared the characteristic(s). A double asterisk (**) means that 50-66% of the intervention episodes shared the characteristic(s). NOTE: Figures in the column, Z Comm., refer to the actual percentages of intervention episodes sharing the characteristic(s). 121 Total Time Double ** ratings were found for both questions in the value range ">10 minutes." The underlying percentage of 50% in each case is equivalent to the figure calculated when all I.E.'s classified as per- ceived effective were analyzed. TA/Student Talk Ratio A double ** rating was found for the value range "25-50%" for question 4 on nonparticipation by group members. In contrast, no common- alities were found for question 5 when the critical incident involved resistance and hostility. All of the I.E.'s lasting over ten minutes had TA/Student Talk Ratios in this range for question 4, though this was not the case for question 5. Though commonality was not found for question 5, a noteworthy observation was made. Whereas two-thirds of the I.E.'s for question 4 showed the TA to talk 50% or less of the time, the trend was exactly reversed when the incident involved resistance and hostility. It may be that the leader feels a greater compulsion to take charge and personally resolve the problem when a student challenges him and/or the IPL experience. This response is not evoked when the critical incident involves nonparticipation because the issues involved are less threaten- ing to the leader. Affective Domain A single * rating was found for the value range "0-25%" for both questions. This is in line with the finding already discussed that the ‘majority of leader comments are cognitive in nature. 122 Cognitive Domain Both questions showed single * ratings for the value range "75- 100%." The significance of high cognitive input from the leader has al- ready been addressed. Positive and Negative Affect As shown earlier, 100% of the I.E.'s classified as perceived ef- fective fell into the value range "0-25%" for both these variables. Since the I.E.'s pertinent to question 4 and 5 are a subset of this larger group, it necessarily follows that they also had to fall into the same value range. ISIA Again the findings for this variable paralleled those for the overall group. Both questions showed single * ratings for the categories 2, 3, 5, and 6. Double ** ratings were found for Category 7a. The rank orderings showed a tie between categories 5 and 2 for question 4, fol- lowed by 7a. For question 5 the rank order was 5, 2, 6. Nature of a Resolution Both questions had a single * rating for the value "overt." Though the underlying percentage for question 5 was low (67%), the small sample size makes it doubtful that this difference is meaningful. In conclusion, it can be said that at least in these two cases, the type of critical incident involved did not seem to have much of an impact on the nature of those I.E.'s classified as perceived effective. Questions on Perceived Effectiveness by Group Stage It was originally hypothesized that the group stage would have some impact on member perceptions of effective TA behavior during an 123 intervention episode. Many writers in the field contend that the focus of group dynamics changes over time corresponding to what can be called stages of group life. For example, in the early part of the group life the dynamics involve the development of trust needed for risk taking and self-disclosure. It was felt that TA behavior during an intervention episode would be interpreted in light of the dynamics operating during the stage in which the I.E. occurred. This might mean that in an earlier stage TA behavior that was supportive and helped establish trust would be particularly crucial whereas it might not be as important in later stages. Though the groups were divided into three stages, there was sufficient data to examine this hypothesis for only the first two stages. Table 17 summarizes the results of this analysis for commonalities. As the table reveals, most of the commonalities and underlying percentages were simi- lar to those found when all I.E.'s perceived as effective were examined together. Consequently, instead of reviewing the findings for each vari- able, only those with unusual results will be discussed. Total Time Stage 1 showed double ** ratings in two value ranges, "Oé 5 min- utes" and " > 10 minutes." The underlying percentages showed an exact 50-50 split of I.E.'s in each range. Finding commonality in I.E.'s under five minutes was unusual since the other analyses on I.E.'s classified as perceived effective had not produced this result. No commonalities were found for Stage 2 I.E.'s. A reasonable interpretation of the re- sults for this variable is still unavailable. 124 TABLE l7.--Commonalities Across Intervention Episodes for Questions 6 and 7, by Major Variables. Questions Perceived Effectiveness Perceived Effectiveness x Group Stage 1 x Group Stage 2 variables ,(Question 6) (Question 1) Level % Comm. Level % Comm. Comm. Comm. Value(s) Comm. Comm. Value(s) Total Time ** (50%) 065 min . . . . ** (50%) >10 min TA/Student Talk Ratio . . . * 71% 50-75% Affective Domain * 100% 0-25% * 86% 0-25% Cognitive Domain * 100% 75-100% * 86% 75-100% Positive Affect * 100% 0-25% * 100% 0-25% Negative Affect * 100% 0-25% * 100% 0-25% ISIA * 2,3,5,6 * 2,3,5,6 Categories ** 7a ** 7a Nature of Resolution ** 63% Overt * 100% Overt Number of Episodes 8 7 NOTE: A single asterisk (*) means that 67% or more of the interven- tion episodes shared the characteristic(s). A double asterisk (**) means that 50-66% of the intervention episodes shared the characteristic(s). NOTE: Figures in the column, 2 Comm., refer to the actual percentages of intervention episodes sharing the characteristic(s). 125 TA/Student Talk Ratio Though no commonality was found for Stage 1, Stage 2 showed a single * rating in the value range "SO-75%." Further investigation re- vealed that three of the five I.E.'s in this value range also had total times in the range "0&5 minutes." The significance of this finding will be addressed at a later point. The lack of commonality for Stage 1 de- parts from previous analyses on I.E.'s perceived as effective. Nature of a Resolution Stage 1 showed a double ** rating for the value "overt" while a single * was found for Stage 2. The underlying percentage of 63% for Stage 1 is somewhat lower than that for all I.E.'s classified as perceived effective, and considerably below the 100% for Stage 2. Investigating further, it was noted that two of the three I.E.'s with implicit resolu- tions for Stage 1 also had total times in the range"0£-5 minutes." A survey of other questions produced the same sort of relationship; I.E.'s with implicit resolutions were usually short. Before summarizing the findings related to perceived effectiveness, it seems appropriate to first examine the characteristics of intervention episodes classified as ineffective. In this way a comparison can be made to determine the features unique to both classifications. As will be seen, most of the variables show the same commonalities for both classi- fications, thereby disqualifying them as discriminating features. The next section will address the questions related to perceived ineffective- 11888. 126 Questions on Perceiyed Ineffectiveness - Overall The next two questions examine all I.E.'s classified as perceived ineffective. As before, they are differentiated by the data source used for classification of the I.E.'s as ineffective, with question 8 using PSQ data only and question 9 PSQ data plus observer ratings. Table 18 summarizes the findings. Total Time No commonalities were found for either of the two questions in this grouping. The length of an I.E. did not seem to bear any relation- ship to its classification as perceived ineffective. TA/Student Talk Ratio A single * rating was found for the value range "SO-75%" for question 8, PSQ data only. It was Observed that of the six I.E.'s fall- ing into this value range, four of them had total times under five minutes. An examination of the transcripts for these brief I.E.'s showed that TAs tend to be more controlling during shorter interventions; they do not allow the topic to develop. Implicitly or explicitly they prevent other group members from joining in and expanding the discussion. These behaviors cannot be associated with ineffectiveness.though,for a review of shorter I.E.'s classified as effective produced the same pattern. No commonalities were found for question 9, which utilized PSQ data and observer ratings. It should be noted that there were many dif- ferences in the I.E.'s grouped under each question. Only four of the nine I.E.'s for question 8 were included under question 9. This means that the observer rated six I.E.'s as ineffective that were not classified as such based on PSQ data. The reason for this discrepancy became obvious when 127 TABLE 18o--Commonalities Across Intervention Episodes for Questions 8 and 9, by Major Variables. Questions Perceived Ineffectiveness - PSQ Data only Perceived Ineffectiveness - PSQ & Observer Ratings Variables (Question 8), (Question 9) Level % Comm. Level % Comm. Comm. Comm. Value(s) Comm. Comm. Value(s) Total Time ... TA/Student * 67% 50-75% Talk Ratio Affective * 78% 0-25% ** 60% 0—25% Domain Cognitive * 78% 75-100% ** 60% 75-100% Domain Positive * 100% 0-25% * 100% 0-25% Affect Negative * 100% 0-25% * 80% 0-25% Affect ISIA * 2,3,5,6 * ... 2,3,4,5,6 Categories ** 7a Nature of ** 56% Overt ** 60% Overt Resolution Number of Episodes 9 10 NOTE: A single asterisk (*) means that 67% or more of the interven- tion episodes shared the characteristic(s). A double asterisk (**) means that 50-66% of the intervention episodes shared the characteristic(s). NOTE: Figures in the column, % Comm., refer to the actual percentages of intervention episodes sharing the characteristic(s). 128 the observer ratings were examined in detail. In order to classify an I.E. as ineffective the observer had to make some determination of its impact on the participants. As Table 12 shows, the observer gave a rat- ing of indeterminate to four of the five I.E.'s for question 8. This made it impossible for them to qualify for question 9. In the remaining case, I.E. number 26, the observer gave a rating that was different from the PSQ data. Affective Domain A single * rating in the value range "O-25%" was found for ques- tion 8, PSQ data only. The underlying percentage of 78% is comparable to that for I.E.'s perceived as effective (80%). However, a difference was noted when the data source was PSQ and observer ratings. Though there was double ** commonality, the underlying percentage dropped to 60%, considerably below any of the figures for I.E.'s perceived as effective. No meaningful significance has been attached to this finding as of yet. Cognitive Domain As already discussed, the value ranges with commonalities for this variable are a mirror reflection of those for the Affective Domain. Consequently, a single * rating was found for the value range "75-100%" for question 8, and a double ** in this range for question 9. The under- lying percentages are identical to those for the Affective Domain. Positive Affect Both questions showed a single * rating in the value range "0-25%." Interestingly enough, the percentage of I.E.'s with positive affect codes was higher than it was for those classified as perceived effective. For 129 question 8 it was found that 50% of the sessions had some positive affect codes, while the figure was 30% for question 9. An interpretation of this finding will follow after discussion of the results for the next variable. Negative Affect Again both questions showed a single * rating in the value range "0-25%." The percentage of sessions with negative affect ratings was also higher than it had been for the questions related to perceived ef- fectiveness. For question 8 it was found that 33% of the I.E.'s had negative ratings, while for question 9 the figure was 60%. Looking at all the I.E.'s together, 40% had some negative affect ratings. It was not difficult to understand the higher percentage of nega- tive affect ratings since this was expected given the classification of these I.E.'s as perceived ineffective. However, the figures for positive affect did present a problem. A.review of the transcripts involved quickly brought an answer. The I.E.'s with positive affect occurred im- mediately after some interaction where the TA behaved or was seen as behaving in a harsh or threatening manner. The I.E.'s involved some at- tempt by the leader to deny or compensate for this behavior. Conse- cpently'more positive affect was expressed as a way of apologizing, deny- ing, or in a sense "making-up" for this prior hostility. A good example of this is seen in I.E. number 2 involving Nat. After making several strong, almost harsh and challenging remarks to the group, he asks for student reactions. Byron says he feels pressured and that he disagrees with Nat's implication that extroversion is superior to introversion. Nat, realizing that he may have pressed his point too far, changes both 130 his tone and words. He supports Byron and modifies his position as follows: Part of the message that I think came across and I wasn't, I wasn't intending to send it across so much, an extrovert has some ad- vantages in that people know where they are coming from and that is an advantage most of the time. . . . YOu are introverts, but I want you to learn some extroverted behaviors. There are some very definite advantages to being an introvert. The PSQ ratings which classified this I.E. as perceived ineffective probably relate more to the preceding events than to the I.E. itself. This brings up an important point--once the leader is seen as aggressive or threatening it is very difficult to reverse this impression. Negative comments by the leader are keenly felt and remembered. This brings us back to the original finding that a high percentage of I.E.'s classified as perceived ineffective had negative affect ratings. Students are extremely sensitive to hostility from the leader. This is because he has a great deal of real and psychological power in relation to them due to his role as evaluator and authority figure. In addition,the TA is often the most admired person in the group so that rejection or attacks from him are doubly upsetting to students. Hostil- ity does not always have to be of the aggressive type. It can also be displayed in situations where the leader seems unaccepting of individuals or unwilling to give them a chance to explain themselves. Still, the more blatant the hostility, the higher the threat value. ISIA Both questions showed a single * commonality level for the values 2, 3, 5, and 6, corresponding to ISIA categories Active Listening, Elicit- ing Questions, Offers Information, and Self-disclosure. Question 9 also had a single * rating for ISIA Category 4, Directs or Suggests Solutions, and a double ** for 7a, Responsible Negative Feedback. The results to 131 this point are aimost identical to those obtained for the I.E.'s perceived as effective. However, differences were found between the category rank- ing based on frequency of usage. For question 8, there was a tie rank between categories 5 and 2. On question 9, Category 2 was ranked first followed by number 5. In contrast, Category 5 was undeniably in first place for those I.E.'s perceived as effective. The rank ordering did not reveal a clear pattern of skill usage that could be associated with per- ceived ineffectiveness whereas a pattern was found for those I.E.'s classified as perceived effective. Nature of Resolution Both questions showed a double ** rating for the value "overt." The underlying percentages were 56 and 60%, noticeably lower than the ones found for those I.E.'s classified as perceived effective. Though I.E.'s rated as perceived ineffective are more likely to have implicit resolutions, the association is a weak one since a double ** rating was found for the overt value. Questions on Perceived:;neffectiyeness by Critical Incident The concept of a critical incident and the rationale for exploring its relationship to student perceptions of effectiveness or, in this case, ineffectiveness, has already been presented. When all I.E.'s classified as ineffective were reviewed, sufficient data were available to study this relationship for one critical incident, "Nonparticipation." Per- ceived ineffectiveness was based on a combination of PSQ data and observer ratings. Table 19 summarizes the results. Discussion will concentrate on the differences between the results and those obtained when all I.E.'s classified as perceived ineffective were used. Since this critical 132 TABLE l9.--Commonalities Across Intervention Episodes for Question 10, by Major Variables. Perceived Ineffectiveness x Critical Incident 2 Variables (Question 10) Level % Comm. Comm. Comm. Value(s) Total Time ** 60% £10 min TA/Student ** 60% 50-75% Talk Ratio Affective ** 60% 25-50% Domain Cognitive ** 60% 50-75% Domain Positive * 100% 0-25% Affect Negative * 100% 0-25% Affect ISIA * ... 2,3,5,6 Categories ** 4 Nature of ** 60% Implicit Resolution Number of Episodes 5 NOTE: A single asterisk (*) means that 67% or more of the interven- tion episodes shared the characteristic(s). A double asterisk (**) means that 50-66% of the intervention episodes shared the characteristic(s). NOTE: Figures in the column, % Comm., refer to the actual percentages of intervention episodes sharing the characteristic(s). 133 incident was also examined for perceived effective I.E.'s, comparisons to those findings will be made as well. Total Time A double ** rating was found in the value range "5£:1O minutes" with the underlying percentage being 60%. This is unusual since no com- monality was found when all I.E.'s classified as perceived ineffective were looked at. The value range for I.E.'s perceived as effective was also different, being ">10 minutes." TA/Student Talk Ratio A double ** rating was found in the value range "SO-75%" with an underlying percentage of 60%. This is consistent with the findings for all perceived ineffective I.E.'s. In contrast, a double ** rating was found in the value range "25-50%" when perceived effective I.E.'s were examined. Affective Domain A double ** rating was found in the value range "25-50%." Again this is different from that found for all I.E.'s classified as perceived ineffective and for perceived effective I.E.'s involving the critical incident "Nonparticipation." In the latter two cases a single * comon- ality level was detected in the value range "0-25%." The transcripts reveal that the increased focus on affective issues can be attributed to two factors. First, the nature of the critical incident of "Nonpartici- pation" is such that it naturally fosters discussion about feelings. For the most part lack of participation is related to affective concerns such as fear of self-disclosure, not wanting to be judged, and anxiety 134 about saying the right things. Second, several students became highly emotional during the I.E.'s so that the TA was forced to deal with their affect. Cognitive Domain Corresponding to the higher percentage of TA remarks in the affective domain, the number in the cognitive domain was lower, being in the range "SO-75%." Positive Affect Similar to the finding for all I.E.'s classified as perceived ineffective, a single * rating was found in the value range "0-25," with all I.E.'s falling in this range. However, the high percentage of sessions with positive affect ratings was not replicated. In fact only one of the five I.E.'s had any positive affect ratings. Overall the re- sults were more consistent with those found for perceived effective I.E.'s involving this critical incident. Negative Affect Again the findings were identical to those for all perceived in- effective I.E.'s when it came to the value range, but different in re- spect to the number of sessions with negative affect. A review of the relevant transcripts showed that though little negative affect was associated with the TA's comments, the content message was often threat- ening. For example, consider these two excerpts from I.E.'s number 3 and 10. When we get all through with that, no matter how important that is to you, it doesn’t mean one damn thing as far as graduating, getting out of here, passing Ed 200. . . . I see a whole bunch of introverts sitting here and I'm telling you you're going to 135 learn some foreign behaviors or you're going to be here again. . . yeah, it's hard, but you can do it, if you want to. If you don't, that's cool, but own it. Why invest more money, why invest more time, why set yourself up for pain? If you're not into doing that then bail out cause that's real. There's a lot of things to do in the world besides teach. (I.E. number 3) Such comments could be interpreted as threatening at any point in the group; given they occurred in the first session, it is highly likely that many students were upset by them. Now, from I.E. number 10: I feel a need for a break. I want to pose one thing before we take off. It feels like to me we've gone through some stages in the group and spent some time getting to know some folks. . . . spend a lot of time waiting for me to do some things. I don't feel like too many of you are much further now than you were when you first came in the group as far as passing. . . . I think each one of you is going to take some personal investment, some dis- comfort and accept that that's the way it is going to be for a while. I hope that you can begin to tell each other that is what you're doing and that you appreciate others doing that too, to give them support so you can receive support. But if we don't turn that corner, then you considering that you're going to pass 200 is rather pointless. Naturally several students were seriously concerned about this message even though it was delivered in a nonaggressive, almost matter-of-fact way. The manner of delivery did not mask the fact that the TA was giv- ing the group a warning and a challenge to change. Both of these excerpts come from I.E.'s with Nat, as did four of the five I.E.'s for question 10. Nat's confrontive style did not seem to be very effective as evidenced by the fact that he was still dealing with the issue of nonparticipation on a group level by session 8. Because only one TA is involved it is not possible to conclude from the data that early, aggressive confron- tation of this issue would usually be ineffective. However, logic sup- ports this even if the data are inconclusive. Most group members do not participate because they are afraid of something, trying to threaten them into participation only adds to their fears instead of diminishing them. 136 ISIA Single * ratings were found for the categories 2, 3, 5, and 6 and a double ** for Category 4. Ranking based on frequency of usage showed the typical 5, 2 pattern. The results for this variable did not produce anything new. Nature of Resolution A double ** rating was found for the value "implicit." In con- trast, commonality was found for the value "overt" for both I.E.'s per- ceived as ineffective and those that were effective. Though several noteworthy differences were found in the nature of those I.E.'s pertinent to this question, the small sample size of five makes the results questionable. Questions on PerceivediIugffectiveness by Group_§t§ge Unfortunately data to examine this relationship were only avail- able for Stage 1. The results of the search for commonalities are sum- marized in Table 20. As with question 10, discussion will concentrate on the differences between the results and those obtained when all I.E.'s classified as perceived ineffective were used. Comparisons to the Stage 1 commonalities found for perceived effective I.E.'s will also be made. Total Time A double ** rating was found in the value range ":>10 minutes" with 50% of the I.E.'s falling in this range. This is different from the results for perceived ineffective I.E.'s overall or from that found for Stage 1 when perceived effective I.E.'s were looked at. 137 TABLE 20.--Commonalities Across Intervention Episodes for Question 11, by Major Variables. Perceived Ineffectiveness x Grdup Stage 1 Variables (Question 11) Level % Comm. Comm. Comm. Value(s) Total Time ** 50% >10 min TA/Student ** 507. 25—507. Talk Ratio Affective * 67% 25-50% Domain Cognitive * 67% 50-75% Domain Positive * 100% 0-25% Affect Negative * 83% 0-25% Affect ISIA * . 2,3,4,5,6 Categories 7a Nature of * 67% Overt Resolution Number of Episodes 5 NOTE: A single asterisk (*) means that 67% or more of the interven- tion episodes shared the characteristic(s). A double asterisk (**) means that 50-66% of the intervention episodes shared the characteristic(s). NOTE: of intervention episodes sharing the characteristic(s). Figures in the column, % Comm., refer to the actual percentages 138 TA/Student Talk Ratio A double ** rating was found in the value range "25-50%." Though different from that found for perceived ineffective I.E.'s overall, it is consistent with the observations that I.E.'s over ten minutes are associated with ratios in this range. No commonality was found when perceived effective I.E.'s in Stage 1 were analyzed. Affective Domain A single * rating was found in the value range "25-50%." This differed from the results for perceived ineffective I.E.'s overall or for Stage 1 perceived effective I.E.'s. In the latter instances com- monality was found in the value range "0-25%." A review of the tran- scripts pertinent to this question revealed an explanation for the differences. There was strong emotional involvement on the part of students and the leader in those I.E.'s with a high percentage of TA comments in the affective domain. Students expressed anger toward the TA or other group members in all of these I.E.'s which probably accounts for the increased attention to affect by the leader. Negative Affect A single * rating was found in the value range "0-25%." In this respect the findings are the same as those for Stage 1 perceived effec- tive I.E.'s. However, consistent with the results for all perceived in- effective I.E.'s, a high percentage of the I.E.'s for this question showed some negative affect expressed by the TA. As already mentioned, there were several angry, emotional interchanges during these I.E.'s. When the TA was the target of the anger he tended to retaliate in kind. The negative affect was much weaker when others were the target. 139 ISIA A single * rating was found for the values 2, 3, 4, 5, 6, and 7a. Basically the same results were found for perceived ineffective I.E.'s overall and for Stage 1 perceived effective I.E.'s. The rank ordering based on frequency of usage showed Category 2 to rank first, followed by Category 5. This ordering is identical to that found for all perceived ineffective I.E.'s. It may be that the higher number of instances of negative affect force the TA to deal with the students on a more personal and exploratory level. This means the TA has to question and paraphrase more, behaviors characteristic of ISIA Category 2. Nature of Resolution A single * rating was found for the value "overt." The overt resolution was also found to be common to all perceived ineffective I.E.'s but to a lesser extent. It was also common to Stage 1 perceived effec- tive I.E.'s. §ummary,of Findings on Perceptions of Effectiveness or Ineffectiveness Table 21 summarizes the results for all of the research questions dealing with perceptions of effectiveness or ineffectiveness. As it shows, there were few differences in the commonalities for the overall groupings of perceived effective I.E.'s versus those classified as per- ceived ineffective. There was an increased number of differences when the overall categories were broken down by critical incident and group stage, especially within the perceived ineffective category. It is doubt- ful, though, that these differences are meaningful given the small sample sizes and the inability to generate reasonable interpretations of some 1110 .AmquuuHuuuooumso may ounces nooomHoo GOHuco>uuucH can we NQoIOm unzu memos Harv meuuumo uHoooo < .HmquumHHOuooumco sou nausea mooooHoo :oHuco>HOucH can «0 shoe no "no uuzu memos Hay JOHHOHOO OchHn < "@902 uuo>oe uHuHHeEHae uuo>oa¢ uuu>o«« uco>ca uuu>oa uuo>o« uuo>oe uuo>o« uue>o« coHuoHomox Ho unsunz cm «as o.m omen Ones OH ones o~.ea« ouwuowuuco .m.e.m.~« o.n.n.~. .e.m.~« o.n.m.~e o.m.n.~¢ o.n.n.~e o.m.m.~« o.n.m.~* e.m.n.~« o.m.m.~« «HmH Nm~10« nmwnoe watoa nnmtoa Nmmnos «mNIoe Nmuuoe nnwuoe Nm~aoe wano« uuuuw< O>Hunwoz unmuoa Hmmnoa nmwloa unwaoa Nmmuoa Nmuloa «mule; unmnoa Nmmnoa unmloa nonuu< o>HuHmom Nnnnome NmHlonaa NooHnnna NooHumHa NooHana NooHana NooHnnH« nooHtha nooHlmHu NocHunna cameos O>Huchoo Nemlmmea nonlmmaa Nmmloa unwloa ummuca Nmuloa NmNIoa ummloa anlca waloa cHoson O>Huoon< Nemlmmea NmHIOmaa ... unnuona NnHIoma ... ... Rowanmae ... ... OHuum 3Hc9 ucoooumx<9 :Ha oHAae cHE onmae ... ... ... cHe muoaa owe OHAaa :Ha oHaaa ... cHe oHaaa OEHH Hmuo9 H ammum N uo>uuuoo ch0 One N manum H swoon a N uo>uomoo cho dmm mmqm emote neoeHueH + oma nacho nacho ueoeHueH HmUHuHuo + owe HOOHuHuu HHnuo>o HHmuo>o nuocu>Hu60uuocH vo>Hoouom unoco>Huoouwm oo>Huouom .nooomHom cOHu=o>uoHcH O>HuOOHHOcH ecu O>Hu00wum oo>Houuum co uaoHuoooc you uwcanHh mo suaesomII.H~ mon<9 141 findings. It is therefore tentatively concluded that the modifying vari- ables of group stage and critical incident have little impact on the characteristics that lead to perceived effective or ineffective leader behavior during an intervention episode. This conclusion runs contrary to expectations since it was hypothesized that they would have an impact. Given the disappointing lack of differences between the common- alities for the overall categories, the question now becomes, "Does this study reveal anything of importance about the characteristics of effec- tive leader intervention behavior?" The answer is yes, if one looks be- yond the immediate analyses to the transcripts which provided the raw data for study. Information gathered from this source has been combined with the findings already presented to produce a comprehensive model for understanding the behaviors that lead to perceived, and real effectiveness. This model will be discussed in the next chapter. Looking back to the findings that have been presented, they do provide some interesting information about TA behavior in general. To begin with, the majority or a very high percentage of TA remarks during I.E.'s had to do with cognitive issues. This showed that TAs frequently function in a manner similar to teachers of more traditional subject matter, providing students with a body of knowledge they can use to bet- ter understand themselves and their surroundings. Another general finding was that TAs exercise considerable control over group process, at least during I.E.'s. As the TA/Student Talk Ratios reveal, TA input far exceeded that which would be expected if the TA was acting as an equal member of the group. Students acknowledged the TA's position as controller by deferring to him whenever he attempted to stop or otherwise modify the flow of the interaction. Even though TAs said 142 they would like group members to assume a greater share of the responsi- bility for processing, this usually did not occur. Nor did the amount of TA control change very much over time. Rather TAs seemed to adopt a certain level of control which they kept throughout the term. Ken's group was an exception to this. He did take a lesser role in process- ing as the term went on. Nat and Hank were initially very controlling and remained so throughout the term. In contrast, Mark said from the start that he was not going to take much responsibility for running the group and he stuck to this posture despite student pressure to change. Lastly, as already discussed in the text, several interrelation- ships among the variables were found. During short I.E.'s, under five minutes, the leader dominated the conversation with his input being aimost exclusively in the cognitive domain. With long I.E.'s this was much less likely to occur and leader input decreased substantially, usually to within the 25-50% range. It is interesting to note that the IPLs were, and they remained, classes in teacher education. The touchy-feelly, high risk, encounter group atmosphere was basically absent. One of the IPLs bore a close resemblance to a T-group but none was of the encounter type. Despite this, the findings that caring and cognition were associated with effec- tive leader behavior is just what Lieberman, Yalom and Miles (1973) found when they studied encounter group leaders. They described the most effective encounter group leaders this way: Leaders of this type specialize in Caring and'Meaning-Attribution. . . . These were individually focused leaders who gave love, as well as information and ideas about how to change. They exuded a quality of enlightened paternalism. They subscribed to a sys- tematic theory about how individuals learn which they used in the group but did not press. (p. 243) 143 The parallels between effective leader behavior in the encounter group and the IPL is noteworthy because of the differences in the two settings, as just discussed. It may be that the qualities for effective leadership in human relations training groups are essentially independent of the group's theoretical orientation. The emphasis and goals may change but caring and cognition still need to be given by the leader. The final part of this chapter will focus on the immediate and long-range consequences of I.E.'s on participants. Results for Questions on the Immediate Consequences of ‘Iutervention Episodes and Long-Range Behavioral/Attitudinal Changes One of the unique features of this project was that it sought to examine the characteristics of I.E.'s with real, as well as perceived, impact on students. As discussed, impact would be looked at in two respects: the immediate consequences of the I.E. for those involved and the long-range changes in behavior/attitudes that occurred as a result of the episode. Twenty-five I.E.'s across the four groups were selected for study. The observer rated the immediate and long-range consequences of these I.E.'s according to the criteria outlined on pages and . Table 22 summarizes the observer ratings. Once the ratings were made it was pos- sible to examine the I.E.'s that fell into certain categories. Specifi- cally, a search was made for commonalities among I.E.'s with positive immediate consequences and perceived effective I.E.'s that did and did not result in long-range change in the desired direction. The next sec- tion of this chapter will report the findings from these analyses. 144 .cOHummwo>oou mo muHoou .umz muuoooom wnHuHMuona huHOOHHmHo nH con .oOHuHmoo .uGOHHm uuu£9 .umouu .uuomOHomHn m.umz nOHmHHmHo chawu muunuo .ucuaaoo IHHOO Ho Hmom .mmoonmoco moon .ou economy on monsooum mo moon» seem so nooom chHmmoomHn uHuuHH mH uuoou mHOOH .moooOHHn woOH mum ouoo9 .mmuHu oumcHaoo ou wooHu .o>Hmoouoo umosmaom .wcHuoua vacuum mHou onwou Huzv Iuoo nemesis ooouw mamm sz mH pom .waHwaHw oaom HOHoo awo> some on: mnoHo mmmHo N N m .OHouuuuoHOucho mum .wcmeoom umoa .onocoumuonoo .wonmoom coca umz moumsou xOOH oumOHocH mHmouo>=oc coca umz muuusou MOOH Aozv ou oocHucou muooaua umoz HHV oaom .muooaaoo Ono oz hHHmomo newness mmmHo mmeo H o a .mu:oEuumHm .uoo noon mHo :aso: on menu a: moaHu Hoouo .monHoow umz he coHuoouuoo .ocHHmmmo :mHHomumo ou HHV .mmonH mHo :mcso: moaHuuaom H+v man on ammo mason non OHonmoo no: I mmmHo umuHm con H m m .uooama O>Humucouwcoo m.umz .oo unwoowo hHuouoo um noouuowHum Boom .wcHmeu sou Imam nH uOHumoHuHuumo 30H .uouHHm chaou unmouo mo umoa moon umz .awuu Ho osmmH 6:9 .aoH mchamu .NHO>Hanoo mucus umuH Scum mouOHoaoooH o>mo Hozv mmmHo Ho HO>OH QOHuomuouaH HIV Iaoo scheme moo tho mucououu Hd< I mmmHo umuHm mmmHo H N m owcH pomHEH owaomImaOH mon moonooUOmnoo osHHummm .maauz .mmom a * Iowa Iowa ovauoaaH oooww Ho NH .mooomHom uOHuoo>HoHoH oOuOOHom Ho HomoaH omddMIwnOH new muoooooomcoo oumHoanH moo Ho mwcHumm new moOHumuoz Ho>uumAOII.NN mam<9 145 .ououonou whoa :.mmOchHcmoE: mH uH on ou nous HHHum wnHamm w cOHmmom mo one an nwcHnu when GOHm .aH: Ou mmmHu no use mooun AHHuwo< Imam OH young .suou ucoomow OHoooo HH sumo .monnu mcHuhHmno wcHNHHOHOH>mooo an no: oHo on one» omH< mo nomouoom uoooom has oumuocoo owoa_:H .m OOHmmmm oH oommouoxo ou Boom uoo moon HHHum mommououo announces: Human ooxHH on uHmm awooou mouwoo mumuouoa ou 3o: ouoH uanmcH .uaouxo oOuHEHH m on HIV Ou uOmHouoH Ou mooaHucoo A+V chw on mason uHoM uuumoHOHuumo no: unoM uHoM a w .HOMH muoouo .umz phoneme mucououm meow .OOHu Iomou human m.umz .hHoouw mwmom on menu omH< .oo>Ho>:H .o>Huouoow mmeu menu .mmmHu mummHE .umz ouwa cOHumuaoum hHmem some on: .Ho>OH coco .mmmHo axon uOHoo mH Iaoo wcHuoo human uuouoooa m on waHumo H+v .Howco mmouoxu uoo moon xomh HIV .o>Hmoomoo muow xomh IHOHuumo some on: Home such a H .muooaaoo OH: on Homumou .o>Hmaomon muow .mooouw owHMH OH meu on one: .unuHHm .muooaoa mH pH umz mHHou hum: oHoamH unmouo .ucoaaoo mooww unmcHEoo ecu «0 upon .mooouw HHmam OuaH no mucououm Ho macaw 06mm mum hon9 .HO>OH ouuuoooa xmowo ou muons comma .mouauHHm w:OH mum uuon9 um oomuoucH HHm New: .oOuHoHocH one nowosm .onuooa ocouom nHou wcHuoo Aozv .aouhm .oom .n mH mm mamm HIV mHoow on when oom uuHoo huo> some on: mmMHo mmoHu N o owcH PoOHBH ownumenOH mwcH miscondOmnoo oaHHomom .moaoz .nmom * e name name mpaHeoaaH nacho Ho mH emseHueoouu.- HHHoooo OH: someone ammo no: on oOHmmom on EH: mucouwooo mHeu muHuon .:OHmmom cox cmoz .o>Hmcow ocN mommHz .ouuoa .ooouw sou ou ammo once Ion sou so muoouo muom m.%mcoa mHo muow o: ouom mason Nahum .mHou How mono Wuooooum EOHH HOH>mooo same on mucmB on wcHhmm EH: mooum cox .mmmouooo cu mHo no Homonoom ouHa cox noucoumcoo house H+V mcHwoo HOH>momo moHNHHooonoz H+V moouwmmHo mHHcHuHcH :OHmmom umHHH onHoa human m m 0H .oOHmmmm mo HooCHmaou wcHuon mmOH some use coHuooHoHoH moo onHoo meow munwHw .uuon omHo .oOHm oo he» HHHB .mOOHHOOOHm uHo mH cOHuumuouoH mo Ho>OH one whom .mOwaHw Imuuowsoueo oH mOwaHm A+v no: uo>msoo .wcHwaHw mung H+v one has mONHHmom HHucoooouw OHuwmom OHuHmom HH N NH .oomcmoo o>mo mmOHHoo wcHhHHmooo NH HHOU uoocmo use moHnchHu IOHOH 3m: ou nOOOHo Aozv .OOHumoHOHuumo mo Ho>OH on Nos u: ammo mom o>Huum mcHMuone I coumm AHV one we when I cowhm .HouooHHu .uOHoo H+v .oammIommOm Icoo ewHo m .o>HuOu mH Huzv HHHHmH monaou I ammOm .oumo aouhm .HouonHHuooo 3OH m .cOHucouuu IHcoaeoo cu OHOHHmm HHHum mH use muuoaaoo 3mm Ho Heucoo moaoomo coon no: mo mOOHuoamHoxu u some mm: ammom .sOHnmom couhm .n:OHmmom oau axon wOHHoo m.umz moomumuoooo uan :H mHHmHooomu coumo ammom Auzv OHuuHH mauooHHuooo I OHHumnm A+V one menu I uHuumnm kHHHmm mouooHHuoou OHuumnm mHuwmnm o N m nwcH pooHEH ownumImooq mwaH moonoooomnoo mcHHommm .moauz .muom e e Iowa Iowa opaHemaaH aaouo Ho mH emaeHoeoona.- mHmo OOHmuOo ou mmuum Ou OOmu muOouOum Ho>oaom .ooouw owHOH OH Ooea soOIoOOIouon .BOOIvOmIOHos ou mOHOu Ion OOHmnOomHa .uOua ouuwm soon mHoouo>OOO oaom stOOH uOOHHm mum Hoouo.3OOIOOmIowoo sou ou OuOuou ou OOOO one .oaHu woo Ho umoa soOIuOm Hozv OH OHmaou mOOHanOmHn mmmHo A+v uuoooom OHoooo 3mm < Imam: moo nuHa hmum muooamz mmmHo «H NN .mooOoooomOou O>HuquO OHonmoo may Ho macaw OH .xumoomum msu How ome OH on whom ..nHmOOOOOu .mOOHmoon IOoom ouoa mwOHHoom .mumHouuxo Bum uxoO any um>o hHooxan OH: wOHumnm mo Houseman Ho uxouOoo sou momOOHOOH wOHuoem moo OOHO> moon o: ooOmu OH Homoooom o>mu .OOHumo A+v IOOOuOoom mo HO>OH m.mO£OOH A+v IuHmou HmHuHOH usuw< IHOHuHmo Ho Ho>OH summons: eunuch mH HN .muoo mom monsoon soou .muuoaua umouu HOO moon one oooww Honuo monsoon :uauma -xOHau HHH3 muoouo was» .mwOHuuoa 3mm umun OH .wOmeu mommouoxo HHOuoo COOHHHoa man whom .ouoa oo>uonuu mma one uo>maon .mooouw HHmEm OOO omeH OH wow oxmm .xomoooow “monouw HHmam OH hHHmHooomo H+v OOHuOOHOHuuoo mo Ho>uH stm H+v sou on come msoom .m>Huom mHHHmm Omen mom huuHM HH 0N .mOOHuommH HHunu owns“ uOOOmu .uOOHHm OHOEOH .mwOmoo muoouo .huumh moon on «o uOoammommo quHOOOO ku>HmOoomou once on o» .OHuuHH no>Hooou OOO nuaHu mama HOOOOO on wOHmOHe OH NH» HHHa one when Sam mouou no can once» onu Oo AHV uOOHmmom can uxuO :0 some AHV HOOoomou muooaua 039 xuooooom How nexus no: OuM mmmHU m NH owOH HOOHEH omOOmeOOH mwOH mooOoOUOOOoo OOHHomom .moEuz .unom H * L3H L3H mpaHemaaH 995 Ho 3 omsaHuOoolI.NN mHm<9 148 xomoomow uOooO uOuHuxO hHHOOOO Hozv O>HquOO o>Hw uOO Ooon HHVIHOO OH OoO ONOO I OOOQ .OHOOuo .HOHHOA OH: wOHuHOHH mo HOHHOOH owe—3 OH HFHHH OOOOum .oowOOe—u IOEOO HHHuO OH on Ohmm on ou OnOOO OOHOHO .xomonoom Ixo Ohnm .OOOHOLOOO he Hozv O>HquOO o>Hw uOO Ooon HI: ouOOuOOHou OHOOHIHHOU .OOOOQOS Hons uo>o Hoquoo whoa OOOHOO o>Hw ou OOHmHnoa On OHooaO NmOuOHuO OOO .xomoooow xOOqu OHHE OH xomnooow Aozv O>HquOO o>Hw uOO Ouon HIV O>HquOO OHOOH I goo: .mo memo OOJOu OHHH OH OHOoOoo Ho: HH hOJOHz HHOu uOOOOU .uoouuow OOOn OOOHOuOH OOO Oooo use .OOHOHOOO no mo upon OO memo .xomnooom .uH wOH>Hw uooom OHOO Ho AHOOOOOOuOooO HOOOHO .xomn Hum: Hozv O>HquOO o>Hw uOO moon HHV IOoo OOOOOHoxO I OOOM Inuom o>HquOO OO>Hm O>On OOoz OuOM m mN .uowooou uH o>Hw HHH3 onO OMOO .xomo IOOOH Oou How oooum sou OxOmou OoO OOO .uowoonu OoO Omen sou u< .HOH>O:Oo no: OOOOHO uuoa Omen O>On wow OOOOOOOOOOOO o>Hu has mam OONHHOOH SOO OoO IowOO OOO ou OOHmOo .OHou OnOano OOHuHoxO Ohmm .xomoomow HHoou wow mom :.mOoso: on emu coma OHOOHOHOHOO Oooo mam oooum Oou Oxoonu mom on wOHuOma 30o .uOHHH .uOououOHOHn 36am OOuHo OHOoHO> H+¢ AOOO uOOHv OOOHO uxOO OH A+v um uOOuOHOOu nwoon9 IOoz .HHuOOOooquH OOuOnHquoo upon mH MN owOH HOOHOH OwOOmeOOH mwOH OOoOOoOOOOoo OOHHOOOm .Onamz .OOOm a H name name moaHemeaH aaouu Ho OH emsaHueoonI.- NHH<9 149 Hozv Hozv Hozv .OOHuOOuHO Ono Oo>Hw has uOo owcmoo ou uuooxu HOO Oooo OOO uH uooom unwoonu Oo OHOO mo mH OOHOOOm OH .HOH>OnOn OH: owOOOO uOO OOon .xomnomow o>HquOO O>Hw uoO Ooon .xomnomom o>HquOO o>Hw uOO moon HHV HHV .Ohoo noOOO uOO oHooa OOO hOHo OHOH Ono wOHuoo HHHw Oou nooou ou o>Oo uOO OHn o; uoou OOHuHOoo OH: OOHOu IOHOE .OHOOu mo uOHoo cu OOOOO .O>HOOOH Ion huo> Ouow HOOOH .OEOHOOnOOa OOOOHOO wwOOHHOoO ou ooow On HOE uH OxOHnu use Ho>OH uneasy Oo uHoo eBOO OuOO3 .OOOHuoEO noxHE I OHHH .oOOOOHoom OH chooOoo OH: NH HHOu uOOOOo .HOOOH on uOO HHH3 use xomuum OO OxHH OH aoO some IOOOH OHOOH I HOHOHz .O>Huooquoo uow OOOHOOOO OOH uOO HHHa MOO: Oo>OHHom .uH OOH>HO Ho HOHHOOH use xomoomom o>HquoO wOH>HOOOH .OHOEOH Ho OHOB .OOOAOO nooou uOO Oooo .nO>HOOOH hum> OH Heuom House NH WOOOH H HOOD mN owOH Iowa HOOHEH OOOOMIwOOH wOH Iumm OOOOOOOOOOOO OHOHeOsEH OOHHOOOm .0380: .mmwm «u xx ooouw 90 NH OOOOHHOOUII.NN MHH<9 150 .xumoooom O>HquOO OOO o>HuHOoo OO>Hw .Ouonuo OOHHHHOHU .OOHuOOHO .oo>HOOOH EOOO uOO on wOH IxHOu uooom OHOOH no: .Onouo ou Ouoaouum O.xwmz um O>HOOOHOO .OOOHHOOOO o>HuHOwoo ou OOOOOOOH OH hHHHOaHHo .OOHuoO A+V IOano Ho Ho>OH oumuoooz HIV .huwam Oqu I OHOOH IuOuOH Ho Ho>OH 30H I OHOOH .owwOmoo w>Oo OmOHHOOH .o OOHOOOm OOOOHE HOS HH HHOu uOOOOo mam .aooouoo OOOOOHOOO .OHuuHH OOuOQHHu OoO Owes wowOO Ho IOoo OoO .uHOHEOOOHo .nHHsu mOHHOo Oooo .OOOHOOOO uoohoo Ono Ooaoooo OoO .HOOH Ho OHOOOHOOHO oau uOHHm wOHHOn emuonHquoo A+v OOOHOOOO 03H uxOO wOHHOQ AHV HOHHHOH HOOHO I HHHH NHOOOOOOOOOOO uOO On; I HHHH . .Ho OHOO m OBOQ .O>HuouuHo once on OH OOxOu OHO OOHOOOOO m zoom Mum: OuOO3 use wOHOHOOH OH: HH HHOu uOOOOo .OOHOOOO OHou Ho Hume OOOH O30 How OHAHOOOOOOH .HHOHEOOOHO .OOHOOH uOHHm uOHOo OH .HHOO uooom OHOOH on Ou Om: OOOHu Ommm IOoo .OOHuOHuOOHH Ho SHOOHH noxHOu a: uxOO wOHHOQ HHHH Aozv .OHOH O>Huom OOOOHuOou HHV OwOHHOOH OmumoO I H< .OOHOOOO OOo OOOOHS I H< H< m N Hm .OwOOeu Ou EH: OuOms OOOHHO>O uOO .uOO Ho OHOnu HHHHO moans umsu OOOHHOOH HO owOOoo OH OHOOOOHO mo on Ohmm .OOOOHHO wOHHOOH OH: HH OOHHoOuOO Oou HOHHO OOxOHOH ou OHoHOOoo uoz A+v HHHHOH OEOOO Huuom .oOHHoOuOO OH OHoHOOoo uoz HOHOH mH « mN HHV . . owOH uoOHEH OwnumImOOH OwOH OOOOOOOOOOOU OOHHOOOm .Ooaoz .OOOm H H Iowa. Iowa mpaHeoeaH guano Ho OH emseHueoouu.- mHmaH 151 .muONmn mm mwsaaucoo 02‘ cu Momnvmmm m>wuamoa mm>wu .wcwxmmam ma mam umzu Boa uwuuwa mamwm .m:o>um: huw> mumw Hawum tam ou :wxoaw Auzv cumuumn cowumnfiUfiuumm A+v mam m%mm .Ham3 muummm amn3 mxmmmm haco .umfisv hum> cusm N om .wawow ma macaw man hm3 mnu .uuoam5m saga mmafiu um kumuumsuw you macaw w;u .uamm«UHuuma mammm H< .numounmm cu manna .wmaoammu m>wuum I wcaaumma n.3ua: mcoaumuau n.3um: nuwa cmwm wow wuaawnamaoammu wcwxmu AIV Gamma H< :oammmm uxmz aIv Imaumm 80mm uoa mmoa macaw mnu vmuuommsm mm: H< H< c mm .abouw mnu mvum30u mvsuwuum m>Hu IHmom a mm: .mmsouw mwuma :« a: waficmao .mmmHo 50mm :0 xuos ou mucmB A+v chfiusnwuuaou 3mm m mme A+v w:m umnu mmwwmunxm .muamasou o: mvma mm: I name .m=o>uma .vmwamnu macaw m>ms >uw> muww HHHum vam ou mxm< .mamm uponm cu :mxoam saga mxmmam xmmam ou wwxmm sons Aozv haao .umfisv huw> mafimamm AIV mao>uma mum> I nuam .mucmaaou on mva mm: I :u:& .QEHu .mmmau 50mm muamaaou :uwa a: ammo Haas km“ N mwxme nwzonu manoma mawmm .wcfiumnm .mmmau uxms wafiuav coaummwofiuuma mo ma mnm umnu Boa kuumn coauomumuaw mo Hm>mH mumumvoa Aozv Hm>ma 30H um mGHmEmm mamom mnm mmmm I «saw mam .mmmHo uwufim mmmmfia I mama wmaaauaoo an L r _ awed powmsH mmsdmlmcoq mwcfi mmoaoswmmnoo mcaammmm .mnamz .mmmm § § Lam Lam 33355 95.5 Q m: vmsafiuGOUII.N~ mgmHumwmc AIV mmwcmzo m>aufimoa A+v “mumcwaumumvcu AHV muommEH m>fiumwmc maummau AIV “momma“ m>fiufimom %Hummam A+v "maoaaom mm mums mwcfiummn 6me G6 "msoHHow mm mum3 mwcaummm .mumnamz macaw you mwamum .mnamz naouu “scammmm How mvamum .mmmm mumnasz uamvfiuaH Hmuauauo you mvcmum * Ho mumnasz mvomamm aofiucm>umucH you mvcmum § mH "myoz .uamuxm maom ou nanny .mumawoauuma uo: ow 0:3 uamnu uo: m“ u« «H mmosu mvumsou wwcwammm mcowummsv Hagan mam .wcauznfiuucou uoa muma mumSuo human mo c0fimmmuaxm .uoa>m£mn pfimnu uo: umnu moam%o::m vmmmmumxm mace on .cowmmmm mama .maaomq mo vm>ouammmaw asouw ca umwaumm can: waaunu mmquNUNuumm mam udSu xmmnvmmm .mmmua .mwcwammm mmumam A+V .wcaumma uxm: mmmmwz AHV nunmoum >aamwuumm .coaumaauwuumn mo Hm>mH swam mumm 0H N «q .umuwm um uumaxam .QOHummHmwuuma mp Haas mHHaxm magma .mmmcw .mwcfiammm mmumnm A+v mo Hm>ma swan mawwucwmz A+v umsu unmuum ou mammm .cOHquHmwuumn mo Hm>mH swam NH; ma q me .coammnumav mnu vam Ou mxm¢ .hmm .xumnvmmm mumzuo umsz unmuum .xmmnvmmm m>wumwma cam m>aumwma mama mm>fio cu Emma uoa mmon m>HuHmoa mm>fio .coaumawmwuuma AIV .maouv Hm>mH coaummumucH AIV .m>wm:mmmv .muwcm mumo mo Hm>mH m>Humm cu mumumvoz masmm NH H HQ nwcw pommsH mwzmmlwnon mwaa mmocmsummcoo mawammmm .mgamz .mmmm * * Iumm Iumm mpmfimeEH asouu Ho mH cmsamuaooun.- mqm<fi 153 Positive Immediate Consequences Question 12 dealt with all I.E.'s with positive immediate conse- quences. Question 13 only included those that pertained to the critical incident "Nonparticipation." A decision was made to put the results from questions 12 and 13 into one table, number 23, since they both dealt with the area of positive immediate consequences. It was also of interest to compare the findings for the two questions since the I.E.'s for question 13 are actually a subset of those for question 12. As the table shows, the commonalities for both questions are very similar to those found for all perceived effective I.E.'s. This is because 79% of the I.E.'s with positive immediate consequences were also classified as perceived effec- tive. Consequently the results will be only briefly outlined since their significance has already been addressed as part of the discussion for all perceived effective I.E.'s. Total Time Question 12 showed a double ** rating in the value range ">I10 minutes" with exactly 50% of the I.E.'s falling into this range. There were also a sizeable number of I.E.'s under five minutes (43%). It is difficult to interpret the meaning of this split and it may simply be a matter of chance. The split in value ranges was also seen in question 13. However, this cannot be viewed as additional evidence for some trend since the I.E.'s for question 13 are a subset of those for question 12. TA/Student Talk Ratio no commonalities were found when all the I.E.'s were examined together for question 12, but a double ** rating was found in the value range "ZS-50%" for question 13. As observed in the past, there was a 154 TABLE 23.--Commonalities Across Intervention Episodes for Questions 12 and 13, by Major Variables. Questions I.E.'s with Positive Immediate Consequences I.E.'s with Positive Immediate Consequences x Critical Incident 2 Variables (Question 12) 1 4' _fiQuestion 1314 Level 2 Comm. Level 2 Comm. Comm. Comm. Value(sl Comm. Comm. Value(s) Total Time ** 50% >10 min ** 50% 065 min ** 50% >10 min TA/Student Talk Ratio ** 50% 25-50% Affective Domain * 93% 0-25% * 100% 0-25% Cognitive Domain * 93% 75-100% * 100% 75-100% Positive Affect * 100% 0-25% * 100% 0—25% Negative Affect * 100% 0-25% * 100% 0-25% ISIA * . 2,3,5,6 * . 2,5,6,7a Categories ** 7a ** 3 Nature of Resolution * 71% Overt * 67% Overt Number of Episodes 14 6 NOTE: A single asterisk (*) means that 67% or more of the interven- tion episodes shared the characteristic(s). A double asterisk (**) means that 50—66Z of the intervention episodes shared the characteristic(s). NOTE: of intervention episodes sharing the characteristic(s). Figures in the column, 2 Comm., refer to the actual percentages 155 strong relationship between I.E.'s lasting over ten minutes and TA/Student Talk Ratios in this range. In the case of question 13 there was a per- fect correspondence. Affective Domain Both questions had a single * rating in the value range "O-ZSZ." The underlying percentages were extremely high, being 93 and 100%. Of the 14 I.E.'s represented across the two questions, 4 (29%) had no rat- ings in the affective domain. As suspected, it was found that three of these I.E.'s were under five minutes, and their TA/Student Talk Ratios were in the value range "SO-75%." As already noted, this pattern of short, heavily cognitive I.E.'s dominated by the leader appears in many questions. This combination of features seems to be independent of per- ceived effectiveness or ineffectiveness as is the relationship of long I.E.'s to TA/Student Talk Ratios in the value range "ZS-50%." Cognitive Domain Corresponding to the ratings for the Affective Domain, both ques- tions showed single * ratings in the value range "75-1001." Pasitive Affect Both questions had single * ratings in the value range "0-251." Only four (29%) of the I.E.'s for question 12 had positive ratings. For question 13 the number was even less, with only one (17%) of the I.E.'s showing any ratings. Negative Affect Again both questions had single * ratings in the value range "O-ZSZ." Negative affect ratings were found in one I.E. for question 12 and none for question 13. 156 ISIA Single * ratings were found for the values 2, 5, and 6 for both questions and double ** ratings for 7a in question 12 and the value of 3 in question 13. As for the rank orderings, ISIA Category 5 was unequivo- cally in first place for both questions. There was a tie for second place between Category 2 and 7a for question 12, whereas it clearly went to 7a for question 13. TAs seemed to give more responsible negative feedback (7a) during these I.E.'s than was found for perceived effective I.E.'s. Nature of Resolution Single * ratings were found for the value "overt" for both ques- tions with the underlying percentages being 71% and 67%. These per- centages are slightly lower than those found for perceived effective I.E.'s overall but the differences are probably not meaningful. Looking at the data, it was noticed that three of the four I.E.'s with implicit resolutions also had total times under five minutes. Further exami- nation of all I.E.'s in the study showed that 69% of the I.E.'s with implicit resolutions were under five minutes. The relationship was not as strong in the opposite direction, that is, only 53% of all I.E.'s under five minutes had implicit endings. These observations are only of tangential interest to this investigation since no relationship was found between length of an I.E. and perceived effectiveness or positive immediate consequences. Long-Range Changes The ultimate test of the effectiveness of an I.E. with a correc- tive intent is to see if changes in the desired direction occur. Ques- tions 14 and 15 were aimed at detenmining the characteristics of I.E.'s 157 of this type. This could be done only with I.E.'s that were also per- ceived effective since only one perceived ineffective I.E. had positive long-range consequences. Of the twenty-five I.E.'s selected for study, fifteen were classi- fied as perceived effective. Of these seven were found to lead to long- range change in the desired direction, while five produced no change or a negative one. It was not possible to determine if change had occurred for the remaining I.E.'s. It was desired to compare the commonalities found for these two sets of I.E.'s to see if any distinguishing features would be obtained. Table 24 summarizes the results of this comparison. Only those commonalities which are different for the two groupings will be reviewed. TA/Student Talk Ratio A double ** rating was found in the value range "25-50% for those I.E.'s that lead to long-range change whereas no commonality was found for those that did not. It is doubtful that this difference has sig- nificance because of the small sample size for question 15 (5 I.E.'s). ISIA Single * ratings were found for the categories 2, 3, and 5 for both questions. However, a single * rating was also found for cate- gories 6 and 7a in I.E.'s leading to change compared to those that did not. It may be that TA self-disclosure and responsible negative feed- back increase the chances of influencing those involved. The rank order- ing of 5, 2 applied to both questions. 158 TABLE 24.-~Commonalities Across Intervention Episodes for Questions 14 and 15, by Major Variables. Questions Perceived Effective I:E.'s that Lead to Changes Perceived Effective I.E.'s but No Changes Variables (Question 14) (Question 15) Level 2 Comm. Level 2 Comm. Comm. Comm. Value(s), Comm. Comm. Value(s) Total Time ** 57% >10 min ** 60% >10 min TA/Student Talk Ratio ** 57% 25-50% ... .. . Affective Domain * 86% 0—25% * 100% 0-25% Cognitive Domain * 86% 75-100% * 100% 75-100% Positive Affect * 100% 0-25% * 100% 0-25% Negative Affect * 100% 0-25% * 100% 0-25% ISIA Categories * ... 2,3,5,6,7a * ... 2,3,5 Nature of Resolution * 100% Overt * 80% Implicit Number of Episodes 7 5 NOTE: A single asterisk (*) means that 67% or more of the interven- tion episodes shared the characteristic(s). A double asterisk (**) means that 50-66Z of the intervention episodes shared the characteristic(s). NOTE: Figures in the column, Z Comm., refer to the actual percentages of intervention episodes sharing the characteristic(s). 159 Nature of Resolution The most interesting difference between the two groups was found in the nature of their resolution. All the I.E.'s leading to long-range change had overt resolutions. In contrast, four of the five I.E.'s that did not lead to desirable change had implicit resolutions. It may be that lack of adequate closure leaves students with unresolved concerns so that they are less likely to change. Positive Immediate Consequences and Long-Range Consequencesnghanges) A few words are needed on the relationship of positive immediate consequences to long-range consequences. While six of the seven I.E.'s with long-range positive consequences also had positive immediate conse- quences, so did three of the five I.E.'s which did not result in changes. This shows that positive immediate consequences do not always lead to long-range changes in the desired direction. However, only one case, I.E. number 7, was noted where long-range change occurred in the absence of positive immediate consequences. This I.E. involved resistance] hostility and it is clear from reading the transcript that the student changed because he feared the TA. So, it appears that positive immed- ate consequences are a necessary but not sufficient condition for desirable long-range changes in behavior and attitudes. There also seemed to be a strong relationship between positive immediate consequences and perceived effectiveness. Of the fourteen I.E.'s with positive immediate consequences, twelve were rated as per- ceived effective I.E.'s. CHAPTER V DISCUSSION A discussion chapter serves as a bridge between the past and the future. The author looks back over the project and tries to explain why things did or did not occur as expected, as well as summarizing the re- sults and exploring them in greater depth. The focus then turns to impli- cations of the work for future research. This chapter will try to fulfill both these roles. First, discussion will center on the findings for the eight major groupings of research questions outlined in the Results chap- ter. Then additional information and findings based on Observational data and examination of the transcripts will be presented. This quali- tative data will help to clarify and expand upon the quantitative data analysis conducted. Finally, issues of methodology will be addressed along with the implications for research on leadership behavior in human relations training groups. Review of Findings for Researchgguestions Questions on Perceived Effectiveness - Overall The coulnonalities found among the intervention episodes classified as perceived effective showed that leader behavior during these episodes was characterized by high amounts of cognitive input, followed by behavior aimed at eliciting and exploring studentS' feelings and ideas. Little positive or negative affect was expressed in the leader's (TA's) comments. Most of the episodes in this category had overt resolutions, meaning that 160 161 closure had been brought to the episode by the TA and/or those involved. There was a sense of completion although the issues precipitating the intervention episode may not have been resolved. The length of the epi- sode and the proportion of time the TA spoke were unrelated to perceived effectiveness. High cognitive input was seen in the rank ordering of the ISIA categories based on frequency of usage. Category 5, Offering Information, was ranked first. This category includes behavior such as establishing rules of conduct, offering cognitive information, providing frameworks for change, and summarizing group process. These behaviors attach mean- ing to group dynamics and individual behavior. Category 2, Active Listen- ing, was ranked second in frequency of usage. This code was given when the TA demonstrated active listening behavior by asking clarifying and exploratory questions or by paraphrasing student comments. Active listen- ing helped the TA to understand student feelings and thoughts and it let the participants know that the leader was listening to them. The high rank position for Category 2 shows that sincere attempts by the leader to understand group members are an important component of perceived effective leadership behavior. Contrary to expectation, TA verbal expressions of caring, warmth and support were very infrequent. However, the TAs did exhibit nonverbal expressions of positive affect by smiling, being attentive, and occa- sionally touching group members. Questions on Perceived Effectiveness by Critical Incident The aim of these questions was to determine if the nature of intervention episodes and TA behavior perceived as effective would change 162 with the type of critical incident involved. Data were available to exam- ine this interaction for two critical incidents, "Nonparticipation by Group Mambers" and "Resistance/Hostility." Contrary to expectation, no differences were found in the commonalities for both critical incidents. Considering the dissimilarity between the issues at the core of the two critical incidents examined, it is likely that the same result would be found for other critical incidents. The characteristics that are associ- ated with perceived effectiveness seem to be independent of the issues discussed during the episode. Questions on Perceived Effectiveness by Groupggtagg The literature on.human relations training groups suggests that groups go through several stages of development where particular issues are of central concern. The leader is advised to modify his behavior in accordance with the special needs of participants in each stage. It was therefore hypothesized that student perceptions of effective TA be- havior would differ according to the stage in which the intervention episode occurred. Analysis of intervention episodes in Stages 1 and 2 failed to confirm this hypothesis. Like the critical incident, group stage did not seem to affect student perceptions of effectiveness. Some care must be taken in making this conclusion,though, because of small sample sizes and the inability to examine intervention episodes across all three stages for this reason. There is also another factor that may account for the lack of differences and merits discussion. Almost all of the literature on stages in.human relations train- ing groups comes out of the laboratory training movement, i.e., T-groups. Though each T-group is somewhat different from the next, they share a basic approach that identifies them as belonging to this theoretical 163 orientation. The essence of the approach is that the leader fails to accept the traditional leadership role thereby setting certain fairly predictable processes in motion. Given this structure the group will move through various stages, first trying to define its purpose and goals, then attacking the leader, and finally resolving its problems and engag- ing in constructive goal behaviors. Though this simplifies things, the important point is that the structure leads to a predictable sequence of stages. The problem with the Interpersonal Process Laboratory (IPL) is that there really is no IPL structure. The groups studied shared the same goals but the way they were carried out and the overall leadership styles of the TAs were quite different. Nat provided little structure, yet he was very controlling and directive. Ken was also directive but he structured the experience through many large and small group exercises, including didactic presentations. Activities in the early stages were heavily cognitive and didactic. In contrast, Hank concentrated on the development of trust so that the objectives were not even formally pre- sented for several sessions. Finally, Mark ran his group as if it were a T-group, offering little structure or guidance on how students were to proceed. Whereas Mark's group went through a stage where definitional questions were prominent, and then where the leader was attacked, this was not the case for the other three groups. There may be certain issues that all groups confront regardless of their structure, for example, trust. The attempts to resolve such issues could be seen as stages of development. However, it seems probable that the stages through which a group moves are also determined to a large degree by the structure of the group. Accepting this, it may be that there were no differences in student perceptions of TA behavior during 164 each stage because there is no such thing as Stage 1 or 2 in an IPL group. It cannot be defined because there is so much variability in structure across groups. Questions on Perceiyedglggffectiveness - Overall Leader behavior during intervention episodes classified as per- ceived ineffective was characterized by a high level of affective expres- sion, either positive or negative. Analysis of commonalities showed that 40% of the intervention episodes in this category had TA comments with attached positive affect, and the same percentage had attached negative affect. In marked contrast, the figures for perceived effective inter- vention episodes were 22% with positive affect and 9% with negative affect. Analysis also showed that while TAs still provided cognitive input, they spent more time actively listening to students than was true for perceived effective intervention episodes. Though the majority of episodes had overt resolutions, the percentage with implicit resolutions was noticeably higher than it had been for perceived effective episodes. An implicit resolution meant there was incomplete closure because the focus had been inappropriately shifted to another person or topic. The length of the intervention episode, the proportion of time the TA spoke, and the empha- sis placed on cognition or affect were all found to be unrelated to per- ceived ineffectiveness. In order for a TA comment to be coded as having attached positive affect there had to be clear verbal or nonverbal evidence that it expressed liking, caring, trust, support or apologies for behavior. An explanation for the surprising level of positive affect in several of the episodes was found by reviewing the transcripts. They showed that positive affect usually followed an unpleasant interaction between the TA and group members 165 in which the TA had been particularly hostile or disapproving. The positive affect was therefore a way of apologizing or making reparations for the negative behavior. Ratings on the post-session questionnaire (PSQ) dis- tributed at the end of the session seemed to reflect student reactions to the negative interaction, rather than to the intervention episode that followed it where the TA was especially agreeable. Since the PSQ data were used as the basis for making classifications, these intervention episodes were categorized as perceived ineffective. Expressions of negative affect involved hostility, rejection, indifference or nonsupport of some kind. Often there were angry exchanges between the TA and one or more group members and power struggles. Besides being hostile, TAs were often unaccepting, pressuring and insensitive to student feelings. Students reacted with angry words and tears. In the end the TA almost always "won" because of his objective and psychological power over the students as evaluator and leader. The increased attention to active listening probably results from the high level of emotionality that characterized many of the episodes. The leader was forced to deal with strong student feelings which necessi- tated active listening behavior to explore and understand these feelings. Cognitive information was also given but its utility was lessened given low student receptivity to such information in a negatively charged atmosphere. Qgestions on Perceived Ineffectiveness by Critical Incident Questions in this grouping focused on the impact of different critical incidents on the nature of perceived ineffective leader behavior and intervention episodes. Data were available to examine this relation- ship for one critical incident, "Nonparticipation of Group Members." 166 A comparison of the commonalities found in this analysis to those for all perceived ineffective intervention episodes revealed two notable differ- ences. Leaders in these episodes addressed more of their comments to affective issues but expressed less positive and negative affect. The increased emphasis on affective issues is attributed to two factors, the nature of the issues involved and the emotional level of student participation in some of the episodes. Lack of participation almost always stems from affective concerns, not cognitive ones. The individual or group is inhibited by fears of some sort, such as fear of rejection, failure or ridicule. Therefore discussions about the reasons why people are not participating naturally focus on affective issues. The feelings behind the lack of participation are often intense, as wit- nessed by the emotional reactions of some students when the TA confronted their behavior and asked them to change. As for the finding that few TA comments had attached positive or negative affect, this departure from the findings for all perceived in- effective episodes was explained by reviewing the transcripts. There was little positive affect because none of the intervention episodes in this category was of the kind where the TA is trying to compensate for past behavior. The lack of negative affect was more puzzling and disconcert- ing given the categorization of these episodes as perceived ineffective. The transcripts showed that though the TAs did not express negative af- fect, the content message they gave was often threatening. In more than one instance the leader calmly, but pointedly, told students they would have to change their behavior or else,the or else being receipt of an Incomplete and possibly never getting the chance to become teachers. The blunt honesty with which these messages were delivered was almost certainly 167 viewed as pressure and insensitivity to student feelings. One other difference was noted between intervention episodes in this category and the entire group of perceived ineffective episodes. Whereas the majority of all ineffective episodes had overt resolutions, the trend was reversed for episodes in this category with the majority having implicit resolutions. The inadequate closure associated with implicit resolutions and the threatening nature of the content messages seem to be important reasons for the classification of these episodes as ineffective. It is difficult to generalize about the impact of critical inci- dents on the nature of perceived ineffectiveness since only one incident was studied and the sample size was small. What this analysis did high- light is that perceived ineffectiveness can be due to the content part of a message, as well as the manner in which the message is delivered. estions on Perceived Ineffectiyenesg by Group Stage A sufficient number of ineffective episodes were found in Stage 1 to examine the relationship between group stage and the characteristics associated with perceived ineffectiveness. A comparison of the common- alities for episodes in this stage to the entire group of perceived in- effective episodes showed the former to be distinguished by the high number of TA comments that focused on the affective domain. This pattern was noted in two-thirds of the episodes in Stage 1. Review of the transcripts showed these episodes to be marked by emotional exchanges, with student expressions of anger toward the leader or group members in almost every case. The group leaders were equally emotional, especially when they were the target of student attacks. Their reactions are ex- emplified in the high percentage of TA comments with attached negative 168 affect. Given these situations it is not surprising that TAs addressed many of their comments to affective issues. The attention to affect was matched by increases in active listen- ing behavior. The rank ordering of ISIA categories based on frequency of usage showed that Category 2, Active Listening, was in first place, followed by Category 5, Offers Information. This is consistent with the findings for all perceived ineffective intervention episodes. Like- wise, the majority of episodes in Stage 1 had overt resolutions. Again it is difficult to generalize about the effects of group stage on the characteristics of perceived ineffectiveness since only one stage was examined. However, it can be expected that the aggressive, hostile TA behavior exhibited in these episodes would be seen as ineffec- tive regardless of the stage. Of course one can speculate that the con- sequences of this behavior are more severe in the early stages of group life but the data do not present evidence to test this hypothesis. As with critical incidents, the data on the effects of group stage on per- ceived ineffectiveness are inconclusive. Despite this, if the nature of TA behavior perceived as ineffective is considered, i.e., hostility and insensitivity, it is highly likely that these characteristics are fairly independent of both group stage and the critical incident involved. Qgestions on the Characteristics of Intervention Episodes with Positive Immediate Consequences These questions looked at intervention episodes that were classi- fied as perceived effective and also rated as having positive immediate consequences. The observer gave this rating when there was some indi- cation that those involved at least tentatively accepted the information or message the TA was trying to convey. .Acceptance could be shown through 169 verbal acknowledgement or behavioral change in the desired direction. The commonalities for intervention episodes with positive immedi- ate consequences were found to be almost identical to those for the entire group of perceived effective intervention episodes. Only one meaningful difference was noted: TAs gave more responsible negative feedback in these episodes as revealed by the rank ordering of ISIA categories based on frequency of usage. Category 7a, Responsible Negative Feedback, ranked second in usage for one of the questions in this grouping and it tied for second place in the other question. In comparison, Category 2, Active Listening, was clearly ranked second in usage for all perceived effective episodes. This finding suggests that a sufficient amount of responsible negative feedback is a key ingredient in getting students to change their behavior and attitudes. Questions on the Characteristics of Interventigg ‘Episodes that_pgiand_po Not Lead to Long3r§nge Changes in AttitudesAand Behavigg The purpose of each of the intervention episodes studied was to bring about some long-range change in behavior or attitudes. Therefore, episodes that led to such changes would be the most successful. It was of interest to determine if these episodes had any special character- istics that might account for their success. Toward this end a compari- son was made of the commonalities for perceived effective intervention episodes that did and did not result in long-range changes in behavior or attitudes. Long-range meant that change in the desired direction occurred within three sessions following the episode. Two differences in the commonalities merit discussion. The rank ordering of ISIA categories based on frequency of usage showed that a higher percentage of episodes leading to changes had TA comments in the 170 categories of Self-Disclosure and Responsible Negative Feedback. It seems reasonable to assume that these TA behaviors increase the chances that the TA will be able to influence the participants to change. Of greater significance was the finding that while 100% of the episodes leading to changes had 322;; resolutions, 80% of those that did not lead to changes had implicit resolutions. This dramatic difference provides a clue to the importance of reaching adequate closure with a student or group if the TA expects the intervention to have lasting impact. By working with students to achieve closure the leader ensures that all their concerns are addressed and that they clearly understand the reasons why change is needed. Qualitativeégata The quantitative analysis led to several important findings on the nature of effective and ineffective intervention behavior. Perceived effectiveness was found to be associated with high cognitive input and active listening, while ineffectiveness was related to expressions of negative affect. Behavioral and attitudinal changes were more likely to occur in episodes with overt resolutions where adequate closure had been reached. Review of pertinent transcripts shed additional light on the meaning and significance of the findings. Despite this, the findings seemed limited because they failed to reflect many of the insights gained from observing the groups and examining transcripts. There was still much to be said on effectiveness and leadership behavior. The second portion of this chapter will focus on the insights gained from observational data and transcripts. These insights form the basis of a new, more comprehensive descriptive model of effective and 171 ineffective intervention behavior. It is grounded in clinical impressions formed over countless hours of attending group sessions and examining transcripts. The model also incorporates the findings from the quanti- tative analysis. It is proposed that perceived effectiveness is a function of TA behavior in four categories: Caring, Cognition, Clarity, and Communi- cation. Furthermore, in order for an intervention to lead to behavioral and attitudinal change, the TA.must address two other areas, Consequences and Closure. These will be known as the Six C's of Effectiveness. Per- ceived ineffectiveness occurs when the leader omits one or more needed categories of behavior or acts in ways that are antagonistic to the desired behavior. None of the six categories of behavior is mutually exclusive. The leaders gave cognition in caring ways, just as they blended conse- quences and cognition, or communication with clarity and closure. The categories have been separated for discussion and teaching pruposes, but in reality most behaviors served the goals of at least two categories. Still, the primary goal of the behavior could be isolated allowing classi- fication into one or another category. Each of the six C's will be dis- cussed in turn by describing the category, and giving positive and nega- tive examples of representative behaviors from the transcripts. Caring This category deals with the TA's verbal and nonverbal behavior which reflected a genuine desire to understand and assist students. Non- verbal expressions included attending to the speaker, nodding in support, smiling, and maintaining a gentle, accepting tone of voice. Verbal ex- pressions centered on active listening behavior aimed at eliciting student feelings and ideas. 172 This involved paraphrasing for understand- ing, asking clarifying and exploratory questions, and generally showing an interest in what the students had to say. Positive feedback, words of encouragement and concern are also indicative of this category. How- ever, direct expressions of caring, trust and liking were infrequent al- though they would qualify. this category is found in transcript number 13 (lines 1-39). A good example of the use of behaviors in The TA, Nat, is discussing the feelings of a female student who has been extremely quiet in class for most of the team. (1)1 (3) (7) (8) (9) (11) (12) (23) (25) (26) (27) (28) (29) (30) (32) (33) (34) (35) (36) (37) (39) Nat: Joan: Nat: Joan: Nat: Joan: Nat: Joan: Nat: Joan: Nat: Joan: Nat: Joan: Nat: Joan: Nat: Joan: Nat: Joan: Nat: What happens for you here that it is satisfying for you to be quiet? (in a gentle tone of voice) I don't know. It's just always been, just to sit back and listen . . . other than that. I just do that, I do it all the time. Are you comfortable with it? Most of the time, yeah. Does that mean you're comfortable with it other than in here? . . Yeah. Okay, so then I would say that you have a pretty good reward system built in for you to remain rela- tively quiet. Ybu're not punished for being quiet. You probably are given some rewards. I'm not sure what they would be for you but some rewards for being quiet. Family expectations are that you will be relatively quiet and rewards you for being quiet. My family is pretty quiet most of the time too so. Easy to become noisy all of a sudden, huh! (supportive) Yeah. ‘ You're married too, aren't you? Yeah. Is your husband quiet? Not really, he's a lot more extroverted than I am. That makes it helpful. Yeah. You look as though you're nervous. I am. Am I helping you be there, nervous? No, it's just because the focus is on me. red, isn't it? No, you're doing a good job. My face is 1Line numbers here reflect those in the original transcript where there were fewer words on a line. As a result, the numbers do not always correspond to the actual number of lines presented in this text. 173 The leader explored and clarified the student's feelings and behavior at points 1, 7, 9 and 34. He also asked if he was making her nervous, showing concern for her feelings. Lastly, he demonstrated understanding and gave support at points 25 and 39. This conversation » continued with Nat giving Joan positive feedback that she should give herself credit because she had tried and made progress although she was not receiving a passing grade. She thanked him, saying he had helped to raise her spirits. The leader showed keen sensitivity to the student's low self-esteem and her struggle to speak in class. Expressions of caring were not always as obvious. Sometimes they constituted sentences as simple as, "Did that help?" or "That's a good suggestion, glad you said it." Concern for student feelings and needs was shown in other ways, as in Ken's response to a student who felt judged during intervention episode number 18 (lines 29-43). Ken: I've heard that before in some IPLs, a hesitancy to explore some feelings because somebody in the group or Ken might label that as a change of focus or nonownership or something like that. At least I want to share my position with you that the intent is not to be judgmental. I'm sure not going to be locked into that, I don't feel. But I hope you will share any inconsistencies you see happening with me so that I can become aware of that. That's the kind of attitude I have rather than to lock you into a bad place. I just wanted to share that in terms of an awareness thing. Of course verbal statements about being open and supportive can- not stand alone; the TA had to act in accordance with them.or else they became empty words. Students were quick to pick up on such inconsisten- cies as evidenced by the fact that they remained wary of TAs who followed interactions where they had been hostile or unaccepting with especially positive messages. The negativism was not easily forgotten or probably forgiven, which brings up the area of Perceived Ineffectiveness. 174 The most prominent feature of the episodes rated as perceived ineffective by students and the observer was the TA's expression of nega- tive affect. Not only was a caring atmosphere absent from these episodes, it was replaced with hostility, cold indifference and frequently a "win- lose" type of game. episode number 6 (lines 127-60). One of the best examples of this game occurs in Nat tries to tell Mary that she is giving up control of her life to others when she does not talk because she fears they will be bored. He begins to lose patience when she does not understand and he has to repeat himself, as shown by his tone of voice and facial expressions. tears 0 Nat: Mary: Nat: Mary: Nat: Mary: Nat: Mary: Nat: Mary: Nat: Mary: Nat: Mary: Nat: Do you see where you are giving up control of your life to me, Don, Bill? Yes, but I think that's my choice if I want to. (defen- sively) Yeah, I agree with you but I'm wondering if you know what you're doing? Yeah, I know what I'm doing. (defensively) Okay, then I'm wondering then if you're comfortable with the consequences of doing it? I suppose I am or else I wouldn't do it. (nervously) Okay, then you're making an intelligent decision about how you're behaving. I can't say it is intelligent. It may be bad in the end but if I feel comfortable with it. (no longer defensive) Well intelligent means you know what the hell you are doing (laughs softly). Now whether it brings you had or good is a whole different matter. Okay, what I hear you saying is you're content to give up control of your life to other people. (somewhat irritated) (interrupts) I'm content to a certain point but if I feel I have something really important to tell everybody, I will. (defensively) Okay, and if you don't? I'll shut up. Then if you don't and therefore you shut up the majority of the time in here you understand the consequences of that. (threateningly) Yeah, that's what happened last term. That's called a deferred, that's right. (threateningly) This interaction continued with Mary becoming angry and close to As the excerpt reveals, Nat pursued her until he had made his 175 point and "won the game." He wins both logically and emotionally since she was reduced to tears and a child-like defensive stance. The leader failed to show the student that he understood her position and continued to challenge her despite clear signs that doing so was unproductive. This intervention episode had a pronounced impact on the other group mem- bers because it occurred in the second session of a group where many mem- bers had similar problems in speaking. In the third session (I.E. num- ber 7, lines 1,23), Nat actually got into a shouting match with a student. Jack got up in the middle of class, slammed the door, and then sat back down and asked for feedback. Nat: Yeah, it pisses me off. Jack: Okay, my point is, Nat: (interrupts) It better be good cause I'm going to be very angry. (threateningly) Jack: It is, okay, if I can get this down. Last term in my class that happened. A guy came back and asked for feed- back as if he was asking for feedback and showing that he can ask for feedback. Nat: (interrupts) I don't give a shit what happened last term! (angrily) Jack: Can I get to what I said, you asked me to explain it. (angrily) Nat: I want to know what is going on 393! (angrily) Jack: That's what I'm.telling you, right now, that's what's going on in my mind right now. Nat: Jack, you're getting close. You're talking about last term and I want to know why the fuck you went over there and opened and shut that door and what is right now. You got it?! (angrily) Jack: Yup. Nat: Lay it out! This interaction continued in a negative vein for some time. Later students reported feeling frightened, defensive, tense and angry at either Nat, Jack or both of them. Following so closely the interaction with Mary, this episode created an atmosphere of distrust and fear that persisted for several sessions. Fortunately Nat began to modify his challenging type of behavior and he showed greater sensitivity to student 176 feelings during later intervention episodes. Correspondingly, his ratings as perceived effective rose steadily over the term. However, group unity remained at a low level and most of the members received a grade of In- complete because they failed to speak up in class and exhibit the skills. The conclusion must be reached that Nat contributed to this by failing to exhibit appropriate behaviors in this category during the early stages of the group. Cognition Leader behavior in this category helped students to conceptualize and attach meaning to individual and group experiences. The leader identi- fied and labeled events, pointed out the principles governing interpersonal processes, and provided students with the cognitive framework needed to bring about changes in themselves and others. When the leader was exhib- iting these behaviors he was most recognizable as the teacher. This cate- gory is equivalent to what Lieberman, Yalom and Miles (1973) called the Meaning-Attribution function of the leader. They summarized the category by writing that it deals with "the translation of feelings and behaviors into ideas" (p. 238). More specifically, they described the category this way: Meaning-Attribution involves cognitizing behavior--providing con- cepts for how to understand, explaining, clarifying, interpreting, and providing frameworks for change. . . . attaching meaning to a person or a group's behavior. . . . [When leaders display this function] they offer explanations for consideration. These lead- ers name experiences individual members or the group are having, they may suggest that they look into the experience or they may tell a person directly what he's feeling. (p. 238) The authors found that the most successful leaders were high on this variable. Likewise, perceived effective leadership behavior in this study was associated with high levels of cognition as reflected in the 177 rank ordering of ISIA categories based on frequency of usage. The ISIA Category 5, Offers Information, was clearly ranked first in the analysis of all perceived effective intervention episodes. Cognition was also provided in the ineffective episodes, but to a lesser extent. The significance and impact of cognitive input seemed to vary in direct proportion to its specificity and immediacy to the group situation. For example, in episode number 17, Pam.says that she has not been talking because she feels uncomfortable because no one else is talking. Ken gets Pam to see her behavior in a new light by explaining its negative consequences (lines 50-75). Ken: Uh huh, the silence is uncomfortable. Pam: Yeah, I don't want to be the one to break it. Ken: So that reinforces the silence, that the silence remains because that temporarily takes you out of the limelight but the discomfort is still there. I know some of you have said it in your journals, gee, I'd like to become more active and share more responsibility. Ifm.feeling, I feel the best time to do that is when you feel discomfort. Pam: That's fine with me. (seems happy and relieved) Ken: I'm looking at some of the best growth experiences are the result of some discomfort and some anxiety that has been worked through. . . . I'm saying that as soon as you find yourself in an uncomfortable situation, the more you can do to take control, to deal with it, so that discom- fort is gone and you can respond to your needs as well as where the other person is, I think the more you can do to get there is going to be helpful. In this excerpt Ken has done more than explain what is happening, he has given the student suggestions on how to change in the group and outside of it. Cognitive input which included such features had the greatest impact on participants. Another good example of making the information relevant to the here-and-now of group life is found in episode number 9 (lines 106-15, 134-40). Nat intervenes with two students who are having trouble express- ing the reasons why they find it hard to talk about their feelings in a 178 group of strangers. He makes them aware of some of the dynamics which are probably underlying their fears. Nat: I think . . . what you are both saying and maybe I can be helpful there. I think your terms cognitive and affec- tive are getting away from your feelings of if I self- disclose somebody's not going to understand. I think it might be easier for both of you to look at what kind of personal investment you have. There is less personal in- vestment in cognition than there is in feelings. . . . If you say something you have a personal investment in and people kind of look at it and say what the hell is that, so what, there is pain there, cause of your personal investment so where we are is that most of us have learned to keep the personal investment back inside. Both students found Nat's conment enlightening. However, their behavior changed little. Nat had made the information relevant and immediate, but he failed to teach them.how to overcome their fears, or at least to point out the negative consequences of allowing the fears to rule them. Also included in this category are comments that tied group ex- periences to the students' future situation as classroom teachers. These maintained the focus on the long-term goals of the IPL, to train partici- pants to deal with the social-emotional needs of their pupils when they became professional teachers. For example, in episode number 7 (lines 927-32, 984-92) Nat tries to explain that it is necessary for them to deal with anger because it will occur in their classrooms. Nat: Anger is painful, tears are painful. And so a multitude of things teach us those concepts, those values if you like, and we find different ways of being angry, differ- ent ways to confront without being up front about it. . . . YOu're going to get angry. You get angry all the time. There are healthy ways to express that. That is to own that you're angry. You're going to have kids get angry at you when you're a teacher. If you get freaked out by anger, well how are you going to handle that? How are you going to handle your anger and anger coming back at you? Unfortunately this information came after a particularly destructive interaction between the TA and a student so that its impact was diminished 179 in the emotionally charged environment. Still, it represents a good ex- ample of making the bridge between the IPL and the students' future roles. When the leader failed to provide the necessary cognition the group often floundered and negative consequences ensued. Mark was par- ticularly deficient in this category. On several occasions he explored student feelings but did not give information which would help students understand what was happening and how to change things. Sometimes he failed to do either. For example, in one of the early sessions a few students grew impatient with other members for not participating. Two students said they did not see any reason why people should be appre- hensive about talking and appeared very unaccepting. They expressed this in the form of a question to Mark, who answered, "For their own reasons, they have their own reasons." The students' opinions were clearly left unchanged and the feelings of other group members were never probed. A more appropriate response would have been to give some cognitive infor- 'mation about why people do not talk while exploring the feelings of the quiet group members who were being singled out for criticism. Problems arose also when students were given insufficient infor- mation on the goals and procedures of the IPL. Again Mark was the primary offender. He ran his group on the T-group model which meant he was sup- posed to give little guidance initially on how to proceed. This model requires the leader to skillfully make theory and process interventions that gradually reveal the nature of the experience. Unfortunately Mark was not adept at doing this and there were recurrent discussions and problems around the issues of goals and procedures. The confusion gener- ated by the lack of cognition led to diminished student learning. It is interesting to note that at the end of Mark's group, the main thing that 180 people said they learned from the experience was to risk.more, meaning to self-disclose their inner feelings and ideas and to take chances. Few students seemed to have developed the cognitive schema needed to go into other interpersonal situations and understand and affect what is happening. Clarity This category is slightly different in that it does not deal with specific behaviors as much as it describes the leader's verbal ability to clearly communicate his messages, whatever their nature. Clarity therefore cuts across the other categories since one can dis- cuss it in relation to caring behaviors, offering cognition, and so on. In general, leaders who had abilities in this area spoke clearly, logi- cally, specifically and did not get off on tangents. This was true whether or not they were dealing with their own personal feelings or presenting cognitive information. These abilities seemed to give stu- dents a sense of confidence in the leader, that he was knowledgeable and in control of things, which probably contributed to their perception of him as effective. The impact of leader abilities in this area was particularly noticeable in relation to the category Cognition. As discussed, the cen- tral message in the intervention episodes studied was to change some be- havior or attitude. Although the TA did not always have sole responsi- bility for giving negative feedback and handling episodes, he usually had the primary role in facilitating the interaction and students looked to him to furnish needed cognition and guidance on how to proceed. There- fore it was important that he had the ability to clearly state what the problem was, how he felt about it, what needed to be changed and why. 181 Leaders who were able to do this seemed to have more of an impact on students. The detrimental consequences of lack of clarity were also avail- able. ?Mark had the most trouble expressing himself clearly and specifi- cally. He often spoke awkwardly, and in general terms that made it difficult for students to understand. For example, when someone asked him why he felt he had to label things and did he think it would help the group, Mark responded this way: Mark: In a way, some things I think it's important to label, others not so. It's hard to give a clear answer on that. I hears what you were saying about that (pause), . . . have to stop and make sure (pause). I'm not try- ing to label for the reasons that you say. (I.E. number 34, lines 331-37) Mark.never really addresses the student's question. In a sinilar instance a student asked for guidance on what she should talk about in class be- cause she was unclear on this. Mark: Okay, if you can, I hear you sharing your feelings about a question you have, and sharing your thoughts about that, okay. That can help me to clarify some things that can help you make decisions. That's an appropriate thing to say, I would say. Um, it's hard early on too. One thing I'm keeping a lot of responsibility out onto members of the class early on, trying not to let it get too frus- trating. Each person has their own places they need and their own way they are going to develop and places they need to go and steps along the way. I don't know what all of those are. (I.E. number 34, lines 559-73) Contrast this to Ken's response when he was asked by students what he expected them to do. Ken: For me, I would like to see some labeling, that was a change of focus, that was paraphrasing, being able to identify those kinds of skills and being able to use those skills in appropriate times. . . . I would really like to take the time at the end of a Wed. or Fri. to say, hey, what's going on for you, how did you feel about the experience today. Okay. I see that as a pretty ap- propriate time to say, 'Annie, when you were talking to 182 Pam I thought you were really constructive in the way you helped her deal with the question. I felt you really maintained the focus.' (I.E. number 15, lines 303-7, 312-20) Ken provided students with specific examples of what they were expected to do whereas Mark responded with vague, somewhat confusing guidelines. Students frequently asked Mark this question and received the same sort of "nonanswer." The result was that in session 8 students were still confused about what they were supposed to be doing in the IPL. It is highly likely that some group members never did fully understand the purpose of the class. Communication Behavior in this category served to increase the personal nature and the amount of communication among group members. The personal nature of communication increased as students shared.more of their personal feel- ings about each other and themselves. Leader behavior which fostered this included asking people to talk directly to each other instead of to him or the group, making sure they shared their feelings about each other not just their observations, ensuring that others responded to people who make risky self-disclosures, and generally building an atmosphere where it was appropriate and safe to share sensitive impressions and feelings. When the TA acted as a catalyst to increase the amount of com- munication he was also exhibiting behaviors in this category. This was done by questioning students on their opinions, feelings and nonverbal behavior, making sure one or two people did not monopolize conversations and importantly, seeking feedback about the group process and his own behavior as the leader. When a TA responded appreciatively and accept- ingly to this feedback, the students were encouraged to continue giving it. 183 Additional behaviors in this category include suggesting that students share their feelings inside the group, not outside, and providing struc- tured as well as unstructured opportunities for doing this. Intervention episode number 16 provides several good examples of leader behavior in this category. The leader's comments occurred at various points through- out the lengthy episode. They show how he continually and unobtrusively worked to increase the personal nature and the amount of communication. Ken: What happened as a result of the discourse between Jerry and 1? (lines 1-2) Kitty, can you help us identify why that frustration is there for you? (lines 242-43) Why don't you check that out with other folks? (lines 360-61) How can we break the circle? (line 295) Jerry's just expressed something to you, what's your reaction? Did that make you feel any differently? How does it feel any different now than it did? (lines 633-38) Daisy, you were going to say something? (lines 652—53) What I heard Jerry say is that it would be helpful if you feel some distancing is happening due to the way he's speaking to you or discussing things, that it's helpful to him to say, this is happening for me. I want to check it out with other people. I feel very comfortable doing that. I'm feeling some of you might not be as comfortable at least at this point, saying Jerry, that's really starting to put me off. (lines 577-87) These examples also show how the leader enlisted the help of the group in dealing with a particular student. Instead of taking total responsibility, Ken tried to make it the group's problem.by asking for their reactions and using the term "we." Another approach taken by some leaders was to work with troublesome students by themselves for a while, and then ask for input from other group members. This seemed to lessen student input when they were finally given the chance to respond. 184 The group was also reinforced that the leader would take care of their problems; they did not have to deal with them. Overt appeals to the value of communication were useful, but of greater impact were occasions where the leader pointed out how open com- munication might have prevented a problem. For example, in episode number 21 (lines 242-48), Ken pointed out the value of sharing immedi- ately with two students who were trying to work out some old concerns but could not recall the details of the interaction. Ken: I'm feeling you both said you both felt something but you never checked it out with each other. So it's the seventh week and you finally get at it and by the seventh week things have changed a whole lot for both of you so it's hard to be specific. Without the flow of communication undercurrents of feelings re- mained and prohibited student growth. It was therefore a central func- tion of the leader to keep the flow of communication as open as possible. This flow is a fundamental part of a human relations training group since without it there are limited opportunities for students to receive feed- back on their behavior and to learn about group dynamics. Conseguences Whereas the first four C's led to perceived effectiveness, two additional categories were needed to explain why change did or did not occur following intervention episodes. The first category, Consequences, deals with TA behaviors which ensured that students were presented with the negative consequences of their behavior. This could be done solely by the TA or in combination with feedback from other students. Partici- pants had to be shown why change was needed and just how their current behavior or attitude prevented them.from obtaining some desired goals, such as developing relationships with others or being a responsible 185 professional. Making students fully aware of the negative consequences was found to be strongly related to changes in the desired direction. In episode number 24 (lines 286-307), Ken points out to Dora how her inconsistent participation has created problems for others and herself. Ken: The trouble I have with highs and lows and activity and nonactivity and ups and downs is that god, I never really know how to react cause I don't have that ongoing infor- mation like Jerry was talking about. You say that's really important, that you get off on sharing spontane- ously with people in a very honest, right on way, but then sometimes I don't have that information either and that doesn't seem to be an honest, open, ongoing way so that leaves me with a question mark. . . . I want to open a lot of doors so that we can become more aware of our thoughts and things and it seems like you don't choose to walk through a lot. Ken goes on to explain that though he respects Dora's choice, it has consequences in the group and may also later as a teacher if she remains inconsistent. This excerpt reveals several features that added to the effectiveness of giving negative consequences. The TA gave his personal feelings, showed how the behavior or attitude was actually hurting the person, but still left the recipient with some degree of choice on whether to change or not. As with the category Cognition, the more inmediate and specific the feedback, the greater its impact was. This meant the recipient was given specific instances where their behavior or attitude had caused difficulties in the immediate group situation. Though some leaders took total responsibility for doing this, the most effective technique in- cluded soliciting examples from the group members as well. The recipient now had concrete, real examples to deal with rather than postulated ones of future negative consequences. While it was easy for students to re- fute conjectures about what was likely to happen, it was much harder to deny present realities. For example, a student in Ken's group firmly 186 denied there were negative consequences to his sporadic participation and passive nonverbal behavior because he had never been criticized for either in the past. However, once group members told him they had felt hurt and confused by this behavior, the student said he would reconsider his position. His behavior underwent a noticeable change in the next few sessions. Students were also more likely to consider changing when given specific information on what they were doing incorrectly, when they did it, and how it impacted on others. General comments on the order of "You haven't participated enough" or "You don't seem interested in what others have to say" were ineffective with most students. Often group members were unable to recall the details of past interactions and it was up to the leader to supply the missing data or help students to re- member. Ken gives an excellent example of specific feedback in this excerpt from episode number 21 (lines 1-21): Ken: I'm feeling at times I see you physically withdraw, like you'll be a foot and a half back from a lot of folks in the circle. The consequences for me is that I sense in- tuitively less involvement. . . . it's a distraction in the sense that I'm concerned about what's going on for you and up to the last couple of times you haven't been very vocal about what's going on for you. So it seems like for someone who does not know your intent or does not know where you are coming from, that could really be misinterpreted and it might be a distancing factor. . I see those consequences as being something you might really want to look at. It was very important that students were presented with the nega- tive consequences but not made to feel guilty about past behavior. Lead- ers who showed some understanding for why the person may have behaved in a certain way made it easier for the recipient to accept the negative feedback. 187 Leaders who were inadequate in this category failed to make sure that students were presented with the negative consequences of their behavior or cast the consequences in the guise of, "If you do not do this you will not pass the course." Such threats had marginal success, especially when basic personality changes were being discussed. Failure to make examples specific and immediate to the group situation was also a problem, as was admonishing students for their past behavior without showing some understanding of why they felt or acted that way. Closure This category includes all TA behavior which ensured that the central cognitive and affective needs of students during an intervention episode were adequately addressed if not resolved. The importance of TA behavior in this category increased with the length of the episode and its affective quality. Completeness of closure was found to be associated with long-range changes in behavior and attitudes. Leader behaviors that addressed cognitive needs included check- ing to see that students knew what they were or were not supposed to do in the future, asking for content questions, summarizing what had hap- pened, explaining the relationship of an episode to the IPL goals, and giving students a reasonable chance to continue discussion if they re- quired additional information and feedback. In dealing with affective needs, TAs asked students how they were feeling, if they had other feelings they would like to share, and whether they felt ready to move on to another topic. They also asked for reactions from silent members and questioned participants on the mean- ing of their nonverbal behavior. Leader behavior aimed at achieving closure on affective needs was especially crucial during episodes where 188 the discussion had involved highly emotional or sensitive issues. Review of the transcripts showed that the process of bringing closure to an episode actually began some time before its end. By con- tinually addressing the students' affective and cognitive needs, in addi- tion to questioning for understanding, the leader would slowly bring the intervention to an end and closure. At that point any questions or statements falling into this category of behavior were almost formali- ties because the leader was already fairly certain that closure had been reached and it was appropriate to move on. However, if the leader had not done these things all along it was imperative that he do so in the final part of the episode if adequate closure was to be reached. Inadequate closure was characterized by several features: pre- mature changes in focus, dealing only with cognitive or affective needs, and failure to make sure understanding had been reached. The impact of inadecpate closure was shown in the quantitative analysis where 100% of the episodes leading to changes had overt resolutions, while 80% of those not leading to changes had implicit resolutions, i.e., poor closure. An example of inadequate closure occurs in intervention episode number 8. A student, Kurt, says that he does not see why they have to analyze people all the time and that he wants the group to be more struc- tured so he can get something out of it. Nat tries to help him.see that he can.make the group process concrete and orderly by looking at in in terms of reinforcement theory. {After Nat presents an example of this, Kart gives some evidence that he gets the point by saying, "I see what you mean." Nat then continues for several minutes on the general prin- ciples of reinforcement theory and how it applies to their group and classroom teachers, but he never returns to Kurt to see if his initial 189 concerns are still there, nor does he ever deal with the feelings behind the concerns. The type of situation just described was characteristic of the intervention episodes not leading to change. It was difficult to tell if the recipients of negative feedback in these episodes did accept or fully comprehend the feedback. Their comments were usually brief, on the order of "Yeah, I see what you mean" or "I get it, yeah, you're probably right." Unfortunately the leader did not follow up to make sure the person did understand the feedback in its entirety. Since the individuals continued to behave in the same way it is likely that there was some misunderstanding or nonacceptance. This might have been avoided if the leader had acted differently. The Six 0'3 and Modifying Variables In the quantitative analysis, the factors contributing to per- ceived effectiveness and ineffectiveness seemed to be fairly independent of the group stage or the type of critical incident involved. Clinical impressions of the data supported this conclusion. Behavior in each of the categories outlined was required to achieve perceived effectiveness regardless of the group stage, critical incident, or even the student type. However, it did appear that it was more detrimental to omit cer- tain categories or exhibit undesirable behaviors at some points than at others. For example, the exhibition of hostile and noncaring behavior by Nat in the early sessions of his group seemed to create an atmosphere of distrust that was never fully overcome. In contrast, when Hank acted similarly during an episode at a later point in group life, the negative effects on his group were short in duration. Students had already de- veloped a trusting atmosphere and positive regard for Hank so that they 190 could weigh his behavior against their prior experiences. Caring behavior was also particularly needed with students per- ceived by their classmates as insecure and passive. There were angry reactions to any signs that the leader was pressuring these students or was insensitive to them. In truth, these students did respond best to interventions with a preponderance of caring behaviors, whereas students who were secure and verbal responded better to interventions which empha- sized cognition and consequences. Integration of the Six 0'3 and Strategies Perhaps due to the overlap of categories, it was hard to pick out any particular leader strategies of combining the categories of behavior that proved more or less successful. No special sequences of usage could be found either, apart from frequently addressing the category Closure toward the end of an episode. However, as already discussed, there were some strategies that seemed to further or hinder the goals of behavior in a category. For example, working with students one-on-one and then asking for student input was a less effective way of increasing communi- cation than encouraging and integrating student input throughout the epi- sode. Similarly, getting students to give examples of negative conse- quences was better than having the TA alone provide them, Using examples from the group situation was also more effective than hypothetical ones. Though an in-depth search for strategies was not undertaken, this would appear to be a promising avenue for research efforts. Though no overall strategies of combining strategies could be discerned, it was possible to identify cases where the leader exhibited behavior in all the categories to achieve a successful intervention. 191 Episode number 16 provides the best example of this and has been included in appendix C for reference and illustrative purposes. Generalizability of the Model This model of effective leadership behavior is based on studying intervention episodes where the aim.was to change behavior or attitudes. Just how generalizable it is to other episodes with dissimilar themes is still unknown. Asking someone to change involves different dynamics than discussing a philosophical or procedural issue, for example, and the latter may therefore require other types of behavior. It might also be hard to set criteria for objective success in such episodes, though measures of perceived effectiveness could still be obtained. In the episodes studied objective success was defined in terms of immediate and long-range changes in behavior or attitudes. Though the evidence is not yet available, it seems reasonable to suppose that five of the C's--Caring, Cognition, Clarity, Communication, and Closure-~are fairly central to effective leadership in a wide variety of episodes. The sixth category, Consequences, is very specific to cases where in fact there are negative consequences to some behavior or attitude. The issue of generalizability not only applies to other types of intervention, it also relates to different types of group experiences. Some evidence of generalizability comes from the work of Lieberman, Yalom and.Miles (1973) who also found that caring and cognition were important components of overall effectiveness. Despite this, the extent to which the model is generalizable to other group experiences is undetermined. One other facet of the model needs discussion. It is based pri- marily on the analysis of leader behavior during intervention episodes 192 where the focus was on one or two students. It is possible that a slightly different model would be developed if one examined effective leader be- havior during episodes where the group was the focus. Group problems are harder to deal with since the leader is often alone in his perceptions of the causes and is usually a major contributing factor himself. Un- fortunately this study produced few examples of effective leader behavior in dealing with group problems so answers on this issue await further research. Discussion of the Mbthodology Many of the methods used in this project to study leader inter- ventions are uncommon. Therefore it is appropriate that these methods be discussed to assess their utility for further research. Five aspects of the methodology will be dealt with: critical incidents, intervention episodes, the variables, ISIA, and the stimulated recall technique. Critical Incidents The assumption behind the interest in critical incidents is that how the leader handles an incident will have a significant impact on the individuals involved and the group. If this is true, then it is crucial to identify critical incidents and evaluate alternative leader responses. Doing so should provide valuable information on the nature of effective and ineffective intervention behavior. The findings from this study support the validity of this assumption and the utility of following this course of action. Looking at the critical incident from the standpoint of methodol- ogy, one realizes that not only is it important to identify critical incidents so that alternative leader responses can be examined to see 193 what factors account for effectiveness, but it would be almost impossible to make this comparison without a concept like the critical incident. Making comparisons between intervention responses demands that they be directed toward a common source or type of stimulus. The concept of a critical incident provides these sources, for it implies that there are certain kinds of events which occur with regularity across groups that can be equated on the basis of the issues involved. Once this assump- tion is accepted, it becomes possible to compare intervention responses to these varying events as if they were all directed toward some common event. The methodological benefits of doing this are obvious. Since identifying critical incidents promises to be a central component in further research on leader intervention behavior, it was comforting to find that the critical incidents in this study were readily identifiable. Also leader responses in relation to them did seem to have a profound impact on the group, especially when the incidents reflected the basic issues of the IPL, such as fear of participation or resistance to being judged. Even when the issues were less central to group process, the manner in which the incident was handled often had a profound effect on the group. The best example of this is episode number 16. Here the critical incident was "Overtalk/Inability to Facilitate Self-Closure." The TA.masterfully guided the group through a potentially disastrous confrontation with a group member. By carefully working through the feelings of group members, giving constructive positive and negative feedback and ensuring that adequate closure was reached, he modeled all of the IPL skills and demonstrated their value. As Ken put it at the end of this episode (lines 1134-39, 1149-57): We might not have resolved where we began, but at least there is more understanding. And the only way to facilitate that 194 understanding is having the patience and the skill to methodi- cally work that through, and sometimes that's a lengthy process but that's the only way that I'm aware of where we can construc- tively come to an understanding. . . . to skillfully work through what we did and if it ever arises again, to more skillfully deal with it before it ever reaches that point is what we're all about. And to learn concepts to be able to label what went on, what ex- actly happened, what people did that was skillful and not skill- ful, that's what the class is about. Other examples could be given, many negative as well as positive. Perhaps the best evidence of the impact of leader behavior is that inci- dents that were not dealt with effectively kept recurring. This was true in both Nat's and Mark's groups, where the incidents involving par- ticipation and resistance/hostility kept coming up, session after session. One is forced to speculate that this would not have happened if the leader had taken a different approach to dealing with these issues. In summary, identifying critical incidents and evaluating al- ternative responses is a fruitful method of studying intervention beha- vior and should be pursued. As will be discussed in chapter 6, the study of alternative responses to critical incidents can also serve as the core of a training program for future group leaders. Intervention Episodes To review, the intervention episode is a sequence unit of analy- sis which conceptualizes an intervention as a series of verbal moves re- lated to a single topic that together have cumulative value or impact. These verbal moves form an episode, the boundaries of which are speci- fied according to established criteria. The idea of an intervention episode grew out of the limitations of previous conceptualizations. Being a new concept, a research question was included which called for an evaluation of the intervention episode as a sequence unit of analysis and a method of viewing leadership behavior. Such an evaluation can be 195 made on logical and practical grounds. The two flow together as one examines the advantages of utilizing this concept. Observations of groups show that much of the interaction can be seen as a series of discussions, each with its own central topic. It therefore makes sense to study these discussions to determine the pat- terns that emerge as participants attempt to deal with an issue. Exam- ining the sequence of verbal moves within each discussion seems logi- cally superior to treating comments as if they were unrelated. The intervention episode is really a more formalized version of the dis- cussion. By looking at an intervention as an episode composed of verbal moves, it should also be possible to uncover the strategies different leaders use when addressing similar topics. The consequences of these strategies can then be compared to see which approaches are beneficial or harmful. As discussed, certain strategies were found to be more effective in facilitating the goals of the six C's. An intervention episode should not be equated with a strategx,though,since an episode is a unit of analysis, not an approach to dealing with some issue. Actually, a leader can employ several strategies within an intervention episode. Another advantage of the intervention episode is that it has reliably distinct boundaries. Previously the question of boundaries was often ignored or decisions were made on the basis of ease of coding. In contrast, the intervention episode has a definite set of criteria for determining its boundaries. It is necessary to remember that the intervention episode is a 423;; of analysis, not a method of analysis. This means that the researcher can use various methods to study the interaction of verbal moves within 196 the episode. Thus s/he has the advantages of a sequence unit with reliable boundaries and the flexibility to apply different methods of analysis to the content. Overall, the intervention episode is a valuable concept and should be used by other researchers in the study of leadership behavior. The Variables There are numerous dimensions along which interventions vary. It was therefore a difficult task to select which variables would be best to include in this study of the characteristics of effective leader interventions. While not wanting to take a shotgun approach, it was felt that several variables should be included to maximize the chances that some would prove to be good discriminators of effectiveness. The final choices were based on suggestions from the literature on leader behavior, intuition, and the desire to concentrate on the objective characteris- tics of interventions. This would reduce the need for high-inference judgments when classifying events. The fewer inferences the coder had to make, the more reliable and less biased the categorizations were likely to be. The requirement of high-inference judgments was a flaw of past research where raters were asked to categorize interventions on variables such as "Ego State of Trainer," "Quality of Enactment," or "Distribution of Power and Control" (O'Day, 1968). Based on these criteria, eight variables were chosen: Total Time of the Intervention Episode; the TA/Student Talk Ratio; Domain - Affective or Cognitive; Affective Qual- ity - Positive or Negative; Communication Skills - ISIA categories; and Nature of Resolution - Overt or Implicit. The results showed most of the variables to be poor discrimi- nators of effective versus ineffective interventions. Observed differences 197 on some variables did hint at discriminating characteristics. These hints were_followed up by studying the transcripts to further identify, define and elaborate these characteristics. Three variables of this type were ISIA, Affective Quality - Positive or Negative, and Nature of a Resolution. The ISIA will be discussed separately. In the newvmodel proposed, Affective Quality has been incorporated into the category Caring, while Nature of a Resolution went into Closure. It is suggested that any refinements and development work be done on these new categories of behavior, instead of the original variables. The detailed quantitative analysis pointed out some of the limi- tations of using variables that call for low-inference judgments. In the quest to make the coder's judgments more reliable, there is the risk that the phenomenon under study is broken down into categories that have little meaning. This problem of interpretation was true for all of the variables chosen. It was necessary to constantly return to the original raw data and.make inferences about the meaning of the results. Using high-inference variables may decrease reliability, but it aids interpre- tation since they usually correlate better with the criterion. lfilé The ISIA deserves special mention because of its complexity and the fact that it is a system in itself. It was picked for several reasons. First, consistent with the criteria for selecting variables, it required low-inference judgments of the communication skills being used by the TA. Second, the instrument is reliable and valid in the sense that it was developed specifically for the IPL situation. Its validity as an instru- ment to study leader interventions would be tested in the study. Previous research had never intensively examined the leader's communication skills, 198 probably because reliable instruments for this purpose were unavailable until recently. Finally, it seemed reasonable to ask the question, what does the leaderIQQ during effective and ineffective intervention episodes, does he self-disclose, provide cognition, give responsible or irresponsible feedback, etc.? It was decided to concentrate on the behavior since the overall topic of the intervention episode would be known. As it turned out, simply identifying the communication skills used by the TA provided little information on the characteristics of ef- fective leader intervention behavior. Leaders exhibited the same skills in perceived effective and ineffective intervention episodes. A.rank ordering of the skills based on frequency of usage proved more fruitful in revealing patterns associated with effectiveness, i.e., the heavy emphasis on cognition followed by active listening (caring behaviors). The ISIA should not be used in future studies to detenmine the charac- teristics of effective leader behavior. Instead a content analytic sys- tem is needed that would reflect the Six C's of Effectiveness. One can envision a system where each of the C's would be a major variable with its own set of values. Criteria for making judgments about the classi- fication of events would be established. Researchers could then look at the level or type of caring behavior exhibited, for example, and begin to identify which types were most facilitative. An instrument developed along these lines would call for higher-inference judgments since the focus would be on the goal of the leader's behavior, e.g., to increase communication, show caring, ensure closure. 199 Stimulated Recall Technigue The stimulated recall technique (SRT) was found to have several advantages and disadvantages as a research tool. 0n the plus side, it provided the TA with specific and direct feedback on his behavior during the intervention episode being reviewed. Both TAs who used the SRT said they valued the experience and saw things in their behavior they were previously unaware of. They also modified their behavior toward the individuals involved in the intervention episode by demonstrating in- creased understanding and sensitivity to their needs during future inter- actions. The SRT gave students a chance to share negative feelings about TA behavior in a more socially acceptable and therefore less threatening environment. Whereas few had given negative feedback to the TAs at the time of the intervention episodes, several students did so after the tape recordings were played. The questions on the form students filled out following the recall also served as the basis for giving specific, re- sponsible feedback, as well as helping students pinpoint the reasons for their negative or positive reactions. Overall the students had little trouble reinstating their feelings and seemed to enjoy the recall experi- ence, with almost all calling it worthwhile and interesting. Several disadvantages were also noted. The audiotape recordings provided only partial information about the intervention episodes since nonverbal behavior was missing. Several students remarked that this omission probably influenced their responses to some degree though they could not specify in which direction, more or less favorable. The im- portance of nonverbal behavior cannot be overlooked since it plays a crucial role in our interpretation of events as helpful, threatening, 200 and so on. If such techniques are used in the future, videotaping should be done. Another disadvantage is the length of time it takes to rerun the tape, fill out the forms and discuss reactions. It is questionable if the time investment is worth the gains. Such recalls might prove benefi- cial during the early stages of the group to encourage open expression of feelings, particularly negative ones. However, later in the group life this encouragement is needed less and the time could be better used in other ways. The problem is that though it could be helpful at times to replay and process episodes, having this option necessitates recording all sessions. This is difficult and costly when many groups are involved because of equipment and additional personnel requirements. Though it does seem worthwhile to continue using the SRT for research purposes, at present the evidence does not suggest that IPLs should be taped as standard procedure. Implications for Research on Leader Intervention Behavior Reviewing, the following aspects of the methodology show promise for research on the characteristics of effective intervention behavior: 1. identifying critical incidents and evaluating alternative responses and their consequences 2. the intervention episode as a unit of analysis and a way of con- ceptualizing interventions 3. a content analytic system.based on the Six C's of Effectiveness which would replace the ISIA and the remaining variables used in the study 4. the stimulated recall technique 201 Other suggestions for future research investigations emerge out of the limitations of this project; consequently both will be discussed together. 1. Replication of the Results with Larger Samples. Sample sizes were very small for many of the questions which lessens the reliability of the findings. Several of the questions originally posed could not be answered completely because of insufficient data, i.e., the interaction of all critical incidents, group stages, and student types with group member perceptions of TA behavior. It is therefore especially imperative that attempts be made to replicate the outcomes of the quantitative analysis and assess the validity of the clinically based descriptive model in other situations. Use of High-Inference Variables and Elaboration of the six C's. The findings from this investigation suggest the abandonment of low- inference variables in research on leader behavior. Almost all of the low-inference variables were poor discriminators and it was hard to make meaningful interpretations of the differences that were found. It seems unlikely that future research using low-inference variables would prove more fruitful. Instead research should concentrate on elaborating and defining the six categories (C's) in the new descrip- tive model proposed. Doing so would substantially increase their utility in teaching and research efforts. As it now stands, they are only rough guidelines for intervening effectively. Further Use of the Stimulated Recall Technique. Unfortunately only two stimulated recalls were carried out. A more thorough test of this technique's research potential is needed. This should include various ways of using the SRT, for example, with 202 videotapes, immediate versus delayed recalls, open discussions versus questionnaires. Developing Post-Session Questionnaires. Such questionnaires can be a valuable data source for the study of leader behavior if care is taken in selecting the questions. Of course there are limits on the number of questions that can reason- ably be asked and the areas covered on one form. In this investigation the forms were well designed to single out problems of TA overmanipu- lation, aggressiveness, and excessive pressure. On the positive side, they looked at trust building, acceptance and helping students gain insight into their behavior. However, the questions were not sensi- tive to TA behavior that was detrimental in a passive sense. One of the leaders was extremely nondirective and this often led to hostil- ity, confusion, and other negative consequences. This was not de- tected by the post-session questionnaires because they focused on assessing active behaviors. Given that it is unreasonable to include questions on several areas because of length restrictions, the follow- ing suggestion is made. Since students tire of answering the same questions session after session, it is suggested that some of the questions on the forms be switched on an alternating basis to increase the amount of information gathered. This would allow the researcher to collect data on a wider range of TA behavior. Study of Leader Impact on the Total Learning Environment. Research on the characteristics of effective interventions is a worth- while pursuit. But it is important not to take a narrow view of the leader's influence as occurring only during these intervention epi- sodes. The leader's influence extends far beyond these interactions. 203 As Lieberman (1976) suggested, research should concentrate on the leader's impact on the total learning environment. This is especially true in the IPL because the lack of volunteerism and the evaluative component vest the leader with more power than in other types of human relations training groups. They were the ones who set the norms via their position as teacher and authority figure. Like any other class, students came in, wanted to know the requirements and then how they should behave to meet those requirements. The TA could create what- ever type of learning environment he wanted to. Though it cannot be explored in detail here, the data showed that not all of the environ- ments were equally successful. The leaders were tremendously influ- ential in determining the course of each group and TA effects were far greater than those due to individual student differences. More needs to be done to establish which learning environments are most conducive to the attainment of the IPL goals and how to help TAs create them. CHAPTER VI SUMMARY AND CONCLUSIONS The chief purpose of this dissertation was to answer the question, "What are the characteristics of effective leader interventions at cer- tain critical points in a human relations training group for future teach- ers, the Interpersonal Process Laboratory?" A related purpose was to detennine the characteristics of ineffective interventions. In order to accomplish this a complex quantitative analysis of leader interventions at these points was performed using several variables including a content analytic system which examined the leader's communication skills. A new unit of analysis was developed, the intervention episode, to isolate the time in which the leader's remarks would be examined. Effectiveness was determined by analyzing student self-report data and observer ratings of the impact of the intervention. As reported, the research had limited success in answering these questions. The quantitative analysis did suggest that certain leader behaviors were associated with effective and ineffective interventions, but the results were not at the level of specificity expected. Further examination of the data based on clinical impressions from reading transcripts and observing sessions led to the formulation of a new de- scriptive model of effective leader behavior which incorporated and ex- panded upon the findings from the quantitative analysis. The original analysis indicated that student and observer per- ceptions of leader effectiveness depended upon adequate amounts of 204 205 cognition and caring. Cognitive information helped students increase their awareness, understanding and control over individual and group dy- namics. Caring behaviors were needed to develop an atmosphere of trust and concern so students would be willing to accept the feedback and infor- mation being given. This was especially important since the intervention episodes being studied all had a corrective aim, that is, there was some attitude or behavior the leader (TA) wanted to modify because it had detrimental consequences. It was therefore necessary to give negative feedback to those involved. Transcripts and observer notations showed that two more categories of behavior were needed to account for percep- tions of effective intervention behavior, clarity of expression, and communication. Together these categories form.part of a six-component model of effective leader behavior, called the Six C's of Effectiveness: Caring, Cognition, Clarity, Communication, Consequences, and Closure. Interventions that included behavior from the last two categories had real as well as perceived effectiveness, i.e., students were more likely to change their behavior or attitudes after the intervention episode. A brief description of each category reveals its distinguishing features. The Six C's of Effectiveness Caring This category deals with the leader's verbal and nonverbal beha- vior which reflected a genuine desire to understand and assist students. NOnverbal expressions included attending to the speaker, nodding in sup- port, smiling, and maintaining a gentle, accepting tone of voice. Ex- amples of verbal expressions included paraphrasing for understanding and asking clarifying questions. As if following the old bromide, "Listen to 206 what a man does, not what he says," so students appeared to be influenced more by the leader's honest effort to elicit and appreciate their feel- ings and ideas than they were by verbal expressions of caring. This was confirmed by the observation that students continued to be wary of lead- ers who expressed positive intentions and support following angry or otherwise hostile actions. Failure to exhibit caring behavior was strongly associated with perceived ineffectiveness. Leaders not only mmitted this category of behavior, they frequently replaced it with hostility, cold indifference, and verbal games aimed at "winning" rather than helping. Cogpition Leader behavior in this category helped students to conceptual- ize and attach meaning to individual and group experiences. Examples of behavior in this category included labeling events, providing frameworks for change, pointing out the relationship of group process to the stu- dents' future responsibilities as teachers, and explaining group pro- cedures and goals. Offering information or cognition was the most fre- quently used communication skill for all interventions rated as having real and perceived effectiveness. Cognition was also provided in the ineffective episodes, but to a lesser extent. Lack of cognition led to confusion about group process and why the Interpersonal Process Laboratory (IPL) was being included as part of the preparation to become classroom teachers. The end result of inade- quate cognition was that fewer students grasped the basic, underlying principles of interpersonal communication, principles that could be applied to almost any situation. 207 Clarity Clarity refers to the leader's verbal ability to clearly communi- cate messages to the students, be they of an affective or cognitive nature. Behaviors in this category were needed to give direction to group process and to instill confidence in the leader as someone who was knowledgeable and in control of things. Leaders with abilities in this area not only spoke clearly, they were logical, specific, and stayed with the topic instead of getting off on tangents. Lack of clarity diminished student trust in the leader. It led to confusion, frustration, interruptions in the flow, and a generally less productive experience. One of the four leaders studied was notice- ably deficient in this category and his group floundered for the majority of its existence because he was unable to clearly explain what he wanted them to do. Communication Leader behaviors in this category served to increase the personal nature and the amount of communication among group members. The personal nature of discussions was increased as students shared more of their personal feelings about each other and themselves. Some of the leader's behaviors which fostered this included asking people to express their feelings, praising and protecting students who made risky self-disclosures, and bringing up issues which focused on feelings. The amount of communication increased with the frequency of stu- dent participation. Leaders encouraged participation by asking students questions about their ideas and feelings, soliciting and rewarding feed- back on the group and their own behavior, and providing structured oppor- tunities for many people to express themselves. Leader behavior in this 208 category served a vital purpose since open.communication is a fundamental part of having a successful human relations training group experience. The personal nature and amount of communication were decreased when leaders consciously or inadvertently punished students for expressing themselves, or allowed other students to do so. Other leader behaviors which depressed open communication included allowing the group to spend time discussing outside, irrelevant issues, taking care of problem.situ- ations himself instead of encouraging student input, and failing to pro- vide structured opportunities for participants to share feelings and ideas about each other, themselves, and the group experience. Conseguences This category deals with leader behavior which ensured that stu- dents were presented with, and fully understood, the negative conse- quences of their current behavior or attitude. The leader could furnish examples of negative consequences himuelf and/or encourage other students to do so, the combination being the best approach. Specific examples drawn from.the group experience were more effective in getting students to change their behavior and attitudes than were hypothesized examples of future negative consequences. Pointing out the rewards for changing was another powerful inducement. Leaders who were inadequate in this category failed to make sure students were fully aware of the negative consequences of their behavior. They concentrated more on giving examples of possible negative conse- quences instead of looking for the ones that had already occurred. In addition, they often admonished students for their past behavior or sittitude without demonstrating understanding of the reasons why the students felt or acted that way. 209 Analysis of transcripts showed that making students aware of the negative consequences of their behavior, particularly within the group, was strongly related to changes in the desired direction. It was one of two features which distinguished intervention episodes that led to changes from those that did not. The other feature was the degree to which closure was reached. Closure Included in this category are leader behaviors which made sure the central cognitive and affective needs of students during an inter- vention episode were adequately addressed if not resolved. Examples of leader behaviors which addressed cognitive needs included summarizing what had happened, asking for content questions, and explaining the re- lationship of the episode to the goals of the Interpersonal Process Laboratory. Examples of attempts to deal with affective needs included asking students how they felt as a result of the episode, if they still wanted to share feelings, and whether or not they felt ready to move on to another topic. The importance of leader behavior in this category increased with the length of the episode and its affective quality. IAt- tempts to reach adequate closure was one of the features which distin- guished intervention episodes with real effectiveness from those with only perceived effectiveness. Behaviors which led to inadequate closure included allowing or personally making premature changes in focus, failing to deal with both cognitive and affective concerns, and failing to check that students fully understood what was expected of them. The leader might ask the student(s) if they understood, but he settled for a yes-no answer instead of probing to make sure understanding was reached. 210 Integration of the Six 0'8 and Strategies Most behaviors served the goals of more than one category since the categories were not mutually exclusive. For example, a piece of cog- nition could be given in a clear and caring way. However, it was possible to identify the primary goal of the behavior thereby allowing classification. Strategies of combining the categories did not seem to affect per- ceived or real effectiveness to any significant extent, although some strategies within a category appeared superior to others in furthering its goals, e.g., to increase communication, present consequences. Addi- tional work is needed to identify and study the effects of varying strategies. The Effects of Modifying Variables One of the objectives of the study was to assess the effects of different modifying variables on student perceptions of leader behavior. Three modifying variables were looked at, the type of student(s) involved in the intervention episode, the critical incident, and the group stage. Student type was determined by having the group members rate each other on a series of semantic differential questions twice during the term. The data were analyzed to enable classification of students into several types. Unfortunately it was not possible to systematically examine the effects of student type on perceptions of leader behavior because of small sample sizes when the data were broken down by type and episodes perceived as effective or ineffective. In spite of this, clini- cal impressions did indicate that student type was significant when the individual was perceived as insecure and passive. As for the effect of critical incidents and group stage, the quantitative analysis showed the factors contributing to perceived 211 effectiveness and ineffectiveness to be fairly independent of these modifying variables. Clinical impressions supported this conclusion in the main. Still, it is somewhat premature to conclude that the critical incident and group stage have little impact because of small sample sizes. However, the findings indicate that their impact is far less than many writers in the field have suggested. Methods for Studying Leader Intervention Behavior Several different methods or approaches to studying leader inter- vention behavior were used in this project. Those with promise for future research include: the critical incident, the intervention episode, the stimulated recall technique, and post-session questionnaires. Critical incidents are events occurring throughout the group life that involve issues which are important to the successful functioning of the group. Since these incidents can vary with the type of group, efforts were made to identify some of them for the IPL with its unique demands. Ten were finally selected. It is assumed that how the leader handles these incidents will have a significant impact on participants and will therefore further or impede attainment of group goals. The findings support this assumption and argue for continued work to identify critical incidents and evaluate alternative leader responses. The intervention episode is a new unit of analysis created for this dissertation. It looks at the intervention as an episode composed of verbal moves related to a single topic that together have cumulative impact. The topics in this case were the critical incidents. Criteria were established which enabled reliable determination of the boundaries of the episode. This new way of conceptualizing an intervention has 212 logical and practical advantages and should be used in other investigations. The stimulated recall technique is based on the work of Kagan (1969a, 1969b, 1972). Students were presented with a tape-recorded seg- ment of a previously recorded intervention episode and asked to give their reactions to the leader's behavior in written and oral form. This recall session provided the opportunity to gather specific, direct feedback on leader behavior. An evaluation of the advantages and disadvantages of the technique argued for its continuation as a research tool, but not as a routine part of human relations training groups like the IPL. Post-session questionnaires were distributed at the end of each session to assess student reactions to the leader's behavior during that session and the IPL in general. The fonms were designed to focus on leader behaviors which were likely to occur in intervention episodes and influence student perceptions of effectiveness or ineffectiveness. The data from these forms were used to classify intervention episodes into one of three categories: perceived effective, perceived ineffective, or neutral. A comparison was made of answers to these questionnaires to those given during recall sessions. Few significant differences were found indicating the questionnaires were sensitive to the phenomena they were designed to detect. Implications of the Results Training_Group Leaders One of the reasons for this study was the need to upgrade the training given to IPL group leaders. Training was left to modeling ex- perienced leaders CTAs) and general staff discussions of the "what to do when" type. These practices had limited utility because of the lack of 213 knowledge about effective leadership in the IPL. Without this, there was little.guarantee that new TAs would.mode1 the essential elements of effective leader behavior or that training sessions would impart needed information and skills. This study partially remedies this situation by providing information on the factors associated with effective leader interventions. It also suggests methods of training group leaders which are superior to those currently used. The study of critical incidents and alternative ways of handling them can serve as the basis for a systematic training program for IPL group leaders. They could view videotaped critical incidents and then discuss alternative ways of approaching them. All kinds of critical incidents could be shown, those that are problematic and others that simply reveal opportunities to facilitate attainment of group goals. The training program would have several objectives. First, it would help TAs learn to recognize problems and opportunities as they arise. Then it would assist them to identify possible underlying causes or is- sues behind the surface behaviors, and finally, the program would train them to intervene in a way that maximized their chances of being suc- cessful, thereby moving the group/individuals in the desired direction. Besides critical incidents, the videotapes could show effective and in- effective intervention responses, pointing out the behaviors that lead to both. These behaviors would be the ones identified in this study as the six C's. Role-playing exercises could supplement viewing tapes and discussions. Another advantage of pre-recorded videotapes is that they can be designed to slowly increase the complexity of the issues and fac- tors to be considered in making an appropriate intervention. A training program designed along these lines would far surpass current practices. 214 It would help develop a shared conceptual framework that would give new meaning to.staff discussions and co-training experiences. A comprehensive training program might also include reviewing tape recordings of the TA leading his own group. Every so often a TA could tape-record a session and bring this back to a supervisor or sup- port group for discussion and analysis. Though somewhat threatening, if handled correctly this would give the leader valuable feedback. To- gether they might decide that the TA should carry out a stimulated recall with the entire group. Reviewing tape recordings with supervisors is a standard part of many counselor training programs. There is every reason to believe it would be useful for training group leaders as well. Occa- sional observations of the TA, which is the technique currently used in IPLs, does not provide the leader with the in-depth kind of feedback on intervention style that is needed if improvements are to occur. Also, listening to a tape-recorded interaction can be a beneficial experience in and of itself, as the two TAs who were part of the stimulated recalls found out. This type of training program is based on the belief that train- ing someone to intervene can be done in a similar fashion to teaching other complicated skills. The discriminations, rules, and responses re- quired may be exceedingly complex, but the standard methods of teaching can still be applied. Even if one accepts that modeling and learning by doing are necessary to becoming an expert leader, this does not diminish the value of formal, systematic training to help leaders identify the behaviors to model, or the factors that account for effectiveness in one case and failure in another. Certainly this is what occurs when the TAB teach the IPL communication skills to the students in their groups. 215 Students are taught when and how to respond to numerous verbal and non- verbal cuea. They learn concepts, discriminations, principles and rules of generation. The skills required by a leader go beyond those dealt with in the IPL, but the basic approach to imparting these skills should still be the same. Of course a TA.has to knowwmore than how to intervene effectively. As discussed, the learning environment established by the leader is also crucial. Research is needed to determine the type(s) of learning en- vironment(s) that best lead to attainment of the IPL goals. Once known, leaders can be trained to create this atmosphere. Lodking beyond the immediate situation, the IPL is similar in its goals and many of its methods to other human relations training groups in career preparation programs for the health sciences, social service fields, and business. The idea of imparting interpersonal communication skills through participation in a group experience is widespread in most fields where the job requirements entail dealing with people. Developing methods to ensure the competency of the leaders of these groups is there- fore something of interest to responsible professionals in many areas, not the least of which is education. The training program proposed would be a major step toward meeting this end. Teaching Communication Skills It may be time to seriously examine the approach of using human relations training groups as the sole vehicle for teaching interpersonal communication skills and group dynamics. Just as group leaders can bene- fit from more formal, systematic training which guarantees that certain things will happen and be taught, so can teacher trainees benefit from 216 this type of training. This training would be similar to that given to group leaders except in this case the critical incidents would be replaced by examples of regularly occurring classroom situations that require the teacher to effectively utilize interpersonal communication skills. As before, there would be discussions and examples of effective and inef- fective responses. Training of this type would be more likely to transfer to actual classroom behavior when students entered the field. It should be viewed as a supplement to participation in a group experience rather than a substitution since the latter is needed to give students feedback on their behavior and increase self-awareness. Finally, it would ensure that all students were receiving the same core body of knowledge and training, instead of having the leader's abilities and decisions determine what is learned. To conclude, it is hoped that this study will serve as the impetus for the development of training programs along the lines discussed, eSpecially for group leaders. It is incumbent on those who require group experiences as part of career preparation courses of study to pro- vide the most competent personnel possible. Not only do students deserve this on moral grounds, the instructional benefits of the experience will increase proportionately to the skills of the leader. Efforts to in- crease the effectiveness of human relations training group leaders will surely have positive consequences. APPENDIX A STUDENT FORMS Directions: For all questions above the dividing line (nos. 1-19), make sure to base your answers on today's session only! For all questions below the dividing line (nos. 20-23), 218 Post-Session Questionnaire (PSQ) J .‘ MU! NI MIMM a ' _ basegyour answers on your feelings in general. Please use the following scale when making your decisions: Strongly Agree Agree Unsure Disagree Strongly Disagree IS. 16. 17. 1 2 3 h S The TA seemed accepting of viewpoints different from his own........... At times the TA acted overly critical and disapproving... ...... ........ The TA made a real effort to help others gain insight into their own behaVior and/01" feelings.............."...............o......o...o...o At times the TA pushed too hard, beyond a point where it was productive The TA did a great deal to help build trust in the group today......... At times the TA seemed too manipulative in getting others to react the way he wanted...u.u............o.o....”...o.......o..............o.. The TA made a sincere attempt to understand why people reacted the way they did.......OOOOOOOCOOOOOIOOO.........0.0.000.000.0000.............O The TA's actions did 222 help build trust in the group today..........- Some people seemed to feel overly controlled and manipulated by the TA. Host members seem to trust the positive intent of the TA............... Some members seemed to agree with the TA just for the sake of it....... Most members seemed receptive to feedback from the TA.................. Most members seemed to be guarded and hiding their feelings............ Host members seemed to feel that the TA made an honest attempt to understand and appreciate differences of opinion....................... Many members seemed uninvolved with the group.......................... There is a strong sense of unity in the group.......................... Many people seem to identify with one another's needs and/or problems.. 18. 19. 20. 21. I feel a low amount of trust towards this group........................ I feel I can rely on the TA to be understanding and constructive tom ”......OOOOOOOCCCOO......OOOOOOOO0.0.........CCOOOOOOOOIOIOOOO I am hesitant about expressing my feelings in class because I don't think the TA will give them fair consideration.........................1 I feel positive towards the IPL experience so far...................... Mat" VF” ."uo wanna. ”er 1 is :- 219 Stimulated Recall Questionnaire (SRQ) Directions: You have just heard a portion of the previous meeting. Try to recall the feelings you had at the time the incident occured and answer the questions based on your reaction at that time. Ca Please use the following scale when making your decisions: Strongly Agree Agree Unsur Disagree Strongly Disagree 10. 11. 12. 13. 1 2 3 h S The TA seemed to have a good understanding of the issues/problems in\'01ved in title situationQOOOOOOOOOO......I0......OOOOCOCOOOOOCOOOOOO The TA seemed overly controlling and manipulative in getting those involved to react in the way(s) he wanted............................ The TA seemed overly critical and judgmental at times during the interactionOOOCOOI0.00.I......OOOOOOOO..............OOCOOOOOOOOOOOOOO The TA made an honest effort to understand and appreciate the feelings and ideas Of those invOlved.0.0.000.........QOOOOOOOOOOOOOOO The TA's actions seemed very helpful in getting those involved to gain meaningful insight into their behavior and/or feelings.......... The TA put too much pressure on the individual(s) to change.......... I felt the TA handled the situation very skillfully.................. The TA seemed to be sensitive to the feelings of those involved as the interaction proceeded......................................... The TA pushed too hard, beyond a point where it was productive....... I think this interaction ma have some destructive conse uences for Q those invol'ledIO0.0.0.0.0...I0.00............OOOOOOIO......OQOOOOOOOO I felt those involved in the interaction had a positive experience... The issues/problems being discussed were ones I could personally relate to.0.0.0.0...0....O.......ICUOOOOIOOO...I...‘.........OO...OOO This incident produced strong feelings in me when it occured......... 221) Semantic Differential Peer Perception Form Directions: TA # SC I Classmate's Name: For each of your classmates, please decide which of the adjectives best describes that person using the seven point scale below. Bess your answers on the way you have seen the person behave in the group plus your general impressions of the person. Confident 1 2 3 b S 6 7 Lacking in Confidence Introverted 1 2 3 h 5 6 7 Extroverted Secure 1 2 3 h 5 6 7 Insecure Outgoing 1 2 3 b S 6 7 &w Quiet 1 2 3 h 5 6 7 Tslkative Leader 1 2 3 h 5 6 7 Follower Aggressive l 2 3 h S 6 7 Unsggressive Passive 1 2 3 h 5 6 7 Active APPENDIX B DATA ANALYSIS FORMS AND MATERIALS 222 Observer Rating Form TA # Session 5 Critical Incident Issue lo. of Students Directly Involved Unusual Occurences During the Session Non-Verbal Reactions - Those Directly Involved Positive Impact for how many students Behaviors indicating this reaction: leutral Impact for how many students Behaviors indicating this reaction: Negative Impact for how many students Behaviors indicating this reaction: Non-Verbal Reactions - for Entire Class Positive Behaviors indicating this reaction: Neutral Behaviors indicating this reaction: Iegative Behaviors indicating this reaction: mouoo <9 mo .0: .2309 9 mmooo <9 9o .0: 1309 9 mouoo avenue 99.3%: <9 mo .0: mop—9 u moz mlfluoo pooufim m>3amoa <9 mo .0: 2.9 u mom mmeoo E oo .o: 132. a woven <9 .3 .8 H38 9 .mloooo :HeEdlo m>3ficmoo <9 mo 6: o5. n .ol mmooo cameo—o 92.39am <9 mo .oc 299 ul<| .ommmmos 338.38 .330 oz I “:52 ”pooh: 95.3me n moz 30mm: 953mg a mom 3388 oificmoo n o 232.8 3.389.: n < "3.02 I\| $43 5. so ..2 30? ab 223 .92: ......P .F a. .9. L. er 3. u .slz .. mm s s M. m 1 L. :r r r l IUl I ”q n. m .n d. I 23a «as? o: o: I +£34$

10 Min. x x - - x x - x x - - x - - x ** TA/Student O - 25% - - - - - - - - - - x x - - - - Talk Ratio 25 - 50% - x x - x x - x x - - - - - x - 50 - 75% - - - x - - x - - x - - x x - - 75 - 100% x - - - - - - - - - - - - - - - O - 25% x x x x x x x x - x x x - x - * Affective 25 - 50% - - - - - - - - x - - - x - x - Domain SO — 75% - - - - - - - — - - - - - - - - 75 - 100% - - - - - - - - - - - - - - - - 0 - 25% - - - - - - - - - - - - - - -,fi - Cognitive 25 - 50% - - — - - - - - - - - - - - -I> - Domain 50 - 75% - - - - - - - - x - - - x - x - 75 - 100% x x x x x x x x - x x x - x - - 0 - 25% x x x x x x x x x x x x x x x * Positive 25 - 50% - - - - - - - - - - — - - - - - Affect 50 - 75% - - - - - - - - - - - - - - - - 75 - 100% - - - - - - - - - - - - - - - - O — 25% x x x x x x x x x x x x x x x * Negative 25 - 50% - - - - - - - - - - - - - - - Affect 50 - 75% - - - - - - - - - - - - - - - - 75 - 100% - - - - - - - - - - - - - - - - la - - - - - x - x - - - - x - x - 1b - - - - - - - - - — - - - - - - 2 x x x x x x - x x x x x x x x * ISIA 3 x x x x x x - x x - - x x x x * Categories 4 - x x x x x - - x - - x - — x ** 5 x x x x x x x x x x x x x x x * 6 x x x x x x - x x - - - x x x * 7a - x x x - x x x x - - x - - x ** 7b - - - - - - - x - - - - - - x - Nature of Overt - x x x x x x x x - - - x x x * Resolution Implicit x - - - - - - — - x x x - - - - 8A single asterisk (*) means that this feature. A double asterisk (**) share this feature. 67% or more of the episodes share means that 50-66% of the episodes 2311 TABLE C2.--Comparison of Intervention Episodes Perceived as Effective based on Post-Session Questionnaire Data and Observer Ratings - Question 2. Intervention Episodes Lo Variables 8 9 ll 12 13 15 16 17 18 20 21 23 24 27 29 30 37 38 43 44 mmonalitya O ‘ Shin. - - — - - - - x x x - - - x - - x - x - Total Time 5 510M1n. - - x x - - — - - - - - - - x - - x - - >10Min. x - - - - - - x - x ~ - - - - . - I O - 25% - - - - - - - - - - - x - - - - x - - - - TA/Student 25 - 50% - x x - - x x - — - x - - x - - - - - - Talk Ratio 50 - 75% - - - x - - - x - x - — - x - - - x x - 75 -100% x - - x - - - - x - - - - - - x - - - - - O - 25% x x x x - x x x x x x x x x - x x x - x * Affective 25 - 50% - - - - - - - - - - - - - - - — x -‘f - Domain 50 - 75% - — - - - - - - - - - - - - - - - — - - I - 75 -100% - — - - - - - - — - - - - - - - — - - - 1 - i O - 25% - - - - - - - - — - - - - - - - - - - - ‘ - Cognitive 25 - 50% - - - - - - - - - - - - - - - - - - - - . - Domain 50 - 752 - - - - x - - - - - - - - - - - - x - ff - 75 -lOO% x - x x x x x — x x x - 34;, * 1 0 - 25% x x x x x x x x x x x x x x x x?r * Positive 25 - 50% - - - - - - - - - - - - - - - - - - - — — Affect 50 - 75% - - - - - - - - - - - - - - - - - - - - - 75 -100% - - - - - - - - - - - - - - - - - - — - - 0 - 252 x x x x x x x x x x x x x x x x x x x * Negative 25 - 50% - - - - - - - - - - - - - - - - - - - - - Affect 50 - 75% - - - - - - — - - - - - - - - - - - - - - 75 -100% - - - - - - - - - - - - - - - - - - - - - 1a - - - - x - x x — - x x x - - - - - x - - 1b - - - - - - - - - - - - - - - x - - - - - 2 x x x x x x x x - - x x x - x x x x x x * ISIA 3 x x x x x x x x x - x x x x x - - - x x * Categories 4 - x x x x x x - - - - - - x x x - - - - - 5 x x x x x x x x x x x x x - x x x x x x * 6 x x x x x x x - x - x x x x x x - x x x * 7a - x x x x - x x x x x - x x x — — - - - ** 7b - - - - - - - - - - x - - - - - — - - - - Nature of Overt - x x x x x x - x x - x x x x x - x x x * Resolution Implicit - - - - - - - - - - - - - - - - - - 8A single asterisk (*) means that 67% or more of the episodes share this feature. A double asterisk (**) means that 50-66% of the episodes share this feature. 235 TABLE C3.--Comparison of Intervention Episodes Perceived as Effective Last- ing Over Five Minutes, based on Post-Session Questionnaire Data and Observer Ratings - Question 3. f Intervention Episodes Common- Variables 8 9 ll 12 13 15 16 21 23 24 29 3O 43 ality? 06 5 Min. - - - - - - - _ - _ - _ - _ Total Time 5e10 Min. - - x x x - - - - - - x x - >10 Min. x x - - - x x x x x x - — ** O - 25% - - - - - - - - x - - - - - TA/Student 25 - 50% x x - - x x x - x x - - ** Talk Ratio 50 - 75% - - - - - — — - - _ - - x - 75 -100% x - - x x - - - - - - x - - 0 - 25% x x x x - x x x x x - x - * Affective 25 - 50% - - - - x - — - - - x - x - Domain 50 — 75% - - - - - - - — - - - - - - 75 -100% - - - - - - - - - - - - - O - 25% - - - - - - - - - - - - - - Cognitive 25 - 50% - — - - - - - - - - - - - - Domain 50 ‘- 75% - - - - x - - - - - x - x - 75 '-100% x x x x - x x x x x - x - * O - 25% x x x x x x x x x x x x x * Positive 25 '- 50% - - - - - - - - - - - - - - Affect 50 '- 75% - - - - - - - - - - - - - - 75 '-100% - - - - - - - - — - - - - - 0 '- 25% x x x x x x x x x x x x x * Negative 25 - 50% - - - - - - - - - - - - - - Affect 50 - 75% - - - - - - - - - - - - - - 75 '-100% — - - - - - - - - - - - - - la - - - - x - x x x x - - x - 1b — - - - - - - - - - - x - - 2 x x x x x x x x x x x x x * ISIA 3 x x x x x x x x x x x - x * Categories 4 - x x x x x x - - - x x - ** 5 x x x x x x x x x x x x x * 6 x x x x x x x x x x x x x * 7a - x x x x - x x - x x - - ** 7b - - - - - - - x - - - - - - Nature of Overt r- x x x x x x x x x x x x * Resolution Implicit x - - - - - - - — - - - - - 8A single asterisk (*) means that 67% or more of the episodes share this feature. A double asterisk (**) means that 50-66% of the episodes share this feature. 236 TABLE C4.--Comparison of Intervention Episodes Perceived as Effective based on Post-Session Questionnaire Data and Observer Ratings for the Critical Incident "Non-Participation" - Question 4. Intervention Episodes Variables 9 ll 13 20 21 24 37 Commonalitya 0g 5 Min. - - - x - - x - - Total Time 5510 Min. - x x - - - — _ - >10 Min. x - - - x x - x ** O - 25% - - - - - - x - - TA/Student 25 - 50% x x - - x x - x ** Talk Ratio 50 - 75% - - - x — — - - _ 75 -100% - - x - - - - - - 0 - 25% x x - x x x x - * Affective 25 - 50% - - x - — - - x _ Domain 50 - 75% - - - - - — - - - 75 -100% - - - - - - - - - O - 25% - - - - - - - - - Cognitive 25 ' 50% - - - - - - _ - - Domain 50 v 75% - - x - - - - x — 75 '100% x x - x x x x - * 0 - 25% x x x x x x x x * Positive 25 - 50% - - - - - — - - _ Affect 50 - 75% - - - - - - _ - _ 75 -100% - - — - - - - - - 0 - 25% x x x x x x x x * Negative 25 - 50% - - - - — - - - - Affect 50 '- 75% - - - - — - - - _ 75 «—100% - - — - - - - - - la - - x - x x - x - lb - - - - - - - - - 2 x x x - x x x x * ISIA 3 x x x - x x - x * Categories 4 x x x - - - - x - 5 x x x x x x x x * 6 x x x — x x - x * 7a x x x x x x - x * 7b - - - - x - — x - Nature of Overt x x x x x x - x * Resolution Implicit - - — - - - x -L — 9A single asterisk (*) means that 67% or more of the episodes share this feature. A double asterisk (**) means that 50—66% of the episodes share this feature. 237 TABLE C5.--Comparison of Intervention Episodes Perceived as Effective based on Post-Session Questionnaire and Observer Ratings for the Critical Incident "Resistance/Hostility" - Question 5. Intervention Episodes Variables 8 23 29 3O 36 44 Commonalitya 0g 5 Min. — - — - x x - Total Time 5510 Min. - - - x _ - _ >10 Min. x x x - - — ** O - 25% - x - - — - - TA/Student 25 - 50% - - x - — - - Talk Ratio 50 - 75% - - - - x x - 75 -100% x - - x - - - O - 25% x x - x x x * Affective 25 - 50% - - x - - — - Domain 50 - 75% - - - - — - _ 75 —100% - - - - — - - Cognitive 0 - 25% - - - - - — _ Domain 25 - 50% - - - - — - - 50 - 75% - - x — - — - 75 -100% x x - x x x * Positive 0 - 25% x x x x x x * Affect 25 - 50% - - — - - - - 50 - 75% - - - - - - - 75 -100% - - - — — - - O - 25% x x x x x x * Negative 25 - 50% - - - - - - - Affect 50 - 75% - - - - - _ - 75 -100% - - - - - - — 1a - x - - - - — 1b - - - x - - - 2 x x x x x x * ISIA 3 x x x - — x 9: Categories 4 - - x x - - _ 5 x x x x x x * 6 x x x x - x * 7a - - x - - - - 7b - - - - - - - Nature of Overt - x x x - x a Resolution Implicit’ x - - - x - - 8A single asterisk (*) means that 67% or more of the episodes share this feature. A double asterisk (**) means that 50-66% of the episodes share this feature. 238 TABLE 06.--Comparison of Intervention Episodes Perceived as Effective based on Post-Session Questionnaire and Observer Ratings for Group Stage 1 - Question 6. Intervention Episodes ' Variables 8 9 15 16 17 18 27 37 Comonalitya O 5 5 Min. - - - -- x x x x ** Total Time 5 £'10 Min. - — - — - - - - - >10 Min. x x x x - - — - as 0 ' 25% - - - - - - — x _ TA/Student 25 - 50% _- x x x - - - - — Talk Ratio 50 - 75% E - - - - x - x - - 75 - 100% ;x - - - - x - - - l O - 25% {,x x x x x x x x * Affective 25 - 50% E- - - - - - - - - Domain 50 - 75% L: — - - - — - - - 75 - 100% E - - - - - - - - - 0 - 25% '~:7 - - - - - - - — Cognitive 25 - 50% L: - - - - — - - _ Domain 50 - 75% L7 - - - - — _ - - 75 - 100% : x x x x x x x * O — 25% x x x x x x x x * Positive 25 - 50% Z- - - - - — - - .- Affect 50 - 75% 3 - - - - — - - - - 75 - 100% 5- - - - - - - - - I 0 - 25% fjx x x x x x x x * Negative 25 — 50% 1- - — - - — - - - Affect so - 757. l- - - - - — - - - 75 - 100% - - - - - - - - - 1a - - - x x - - - — lb - - - - - - - - — 2 x x x x x - - x * ISIA 3 x x x x x x x * Categories 4 - x x x - - x - — 5 x x x x x x - x * 6 x x x x - x x - * 7a - x - x x x x - ** 7b - - - - - - - - - Nature of Overt - x x x - x x - ** Resolution Implicit x - - - x - - x - 8A single asterisk (*) means that 67% or more of the episodes share this feature. A double asterisk (**) means that 50-66% of the episodes share this feature. 239 TABLE C7.-Comparison of Intervention Episodes Perceived as Effective based on Post-Session Questionnaire Data and Observer Ratings for Group Stage 2 - Question 7. Intervention Episodes Variables ll 12 20 21 38 43 44 Commonality? 0 g 5 Min. - - x - x - x .. Total Time 5 g10 Min. x x - - - x - - >10 Min. - - - x - - _ .. 0 - 25% - - - - - — - - TA/Student 25 - 50% x - - x - - - - Talk Ratio 50 v 75% - x x - x x x * 75 - 100% - - - - - - - - O - 25% x x x x x - x * Affective 25 ' 50% - - - - - x - - Domain SO - 75% - - - - - - - - 75 - 100% - - - - - - - - O - 25% - - - - - - - - Cognitive 25 r 50% - - - - - - - - Domain 50 - 75% - - - - - x - - 75 - 100% x x x x x - x * 0 — 25% x x x x x x x * Positive 25 - 50% - — — — - — - - Affect 50 - 75% - - - — - - - - 75 — 100% - - - - — - - - O - 25% x x x x x x x * Negative 25 - 50% - - - - - - - - Affect 50 - 75% - - - - — - - - 75 - 100% - - - - - - - - la - - - x - x - - lb - - - - - - - - 2 x x - x x x x * ISIA 3 x x - x - x x * Categories 4 x x - - - - - - 5 x x x x x x x * 6 x x - x x x x * 7a x x x x - - - ** 7b - - - x - - - - Nature of Overt x x x x x x x * Resolution Implicit - - - - - - - - 8A single asterisk (*) means A double asterisk share this feature. this feature. that 67% or more of the episodes share (**) means that 50~66% of the episodes 240 TABLE C8.--Comparison of Intervention Episodes Perceived as Ineffective based on Post-Session Questionnaire Data only - Question 8. ’ Intervention EpisSEes Variables l 2 3 4 6 25 2 28 Commonalitya O 9 5 Min. - x x x - - - x - _ Total Time 5 610 Min. - - - - x x - — - - >10 Min . x - - - - - x — x _ 0 - 25% - - - - - - x - — - TA/Student 25 - 50% x - - x - - - - _ _ Talk Ratio 50 — 75% - x x - x x - x x * 75 - 100% - - - - - - — - - _ O - 25% x x x x - - x x x * Affective 25 - 50% - - - - x x - - - - Domain 50 - 75% - - - - - - - - - _ 75 - 100% - - - - - - - - - - O - 25% - - - - - - - — - - Cognitive 25 - 50% - - - - - - — — - - Domain 50‘ - 75% - - - - x x - - - - 75 - 100% x x x x - - x x x * O - 25% x x x x x x x x x * Positive 25 - 50% - - - - - - - - - - Affect 50 - 75% - - - - - - — - - - 75 - 100% - - - - - - - _ - - O - 25% x x x x x x x x x * Negative 25 - 50% - - - - - - - - - — Affect 50 - 75% - - - - - - - - - - 75 - 100% - - - - - - — — - - la x x - - - - - - — - 1b - - - - - - — - _ _ 2 x - x - x x x - x * ISIA 3 x x x x x x x x x * Categories 4 - - x - - x x - x - 5 x x x x x x x x x * 6 x x x - x - x x x * 7a - - - x - x x - x - 7b - - - - - - - - - - Nature of Overt x - - - - x x x x ** Resolution Implicit - x x x x - - - - - 2A single asterisk (*) means that 67% or more of the episodes share this feature. A double asterisk (**) means that 50-66% of the episodes share this feature. 241 TABLE C9.-—Comparison of Intervention Episodes Perceived as Ineffective based on Post-Session Questionnaire Data and Observer Ratings - Question 9. . ’ Intervention Episodes Variables 3 6 7 10 14 25 28 32 33 41 Commonalitya O: 5 Min. X " -' - X - - - — x - Total Time 5: 10 Min. - x - x - - - - x - - 1>1O Min. - - 'x - - x x x — - - O - 25% - - - - - x - - - - - TA/Student 25 - 50% - - x - x - - x x - - Talk Ratio 50 - 75% x x - - - - x — _ - _ 75 ’ 100% ’ - - X - - - — — x - O - 25% x - — x x x x - - x ** Affective 25 - 50% - x x - - - - x x —; - Domain 50 - 75% - - - — - - — - _ -f - 75 - 100% - - - - - - - - - -? - o — 257. - - - - - - - - - F. - Cognitive 25 - 50% - - - - - - - - - -75 - Domain 50 - 75% - x x - - - - x x - - 75 - 100% x - - x x x x - - x, ** T 0 - 25% x x x x x x x x x x * Positive 25 - 50% - - — - - - - - - :4 - Affect 50 - 75% - - - - - - - — — -1 - 75 - 100% - - - - — - - — _ -7 - 0 - 25% x x - x x x x x x x * Negative 25 - 50% - - x - - - - - - x - Affect 50 - 75% - - - - - - - - - - - 75 - 100% - - - - - - - - - - - la - - x x - - - - x - - lb - - - - - - - - - - - 2 x x x x - x x x x - * ISIA 3 x x x x x x x x x x * Categories 4 x x x x x x x - - - * 5 x x x x x x x x x x * 6 x - x x x x x x x x * 7a - x x - x x x - x - ** 7b - - - x - - - - - - - Nature of Overt - x x - x x - x - ** Resolution Implicit x - - - x - - x - xl - 7A single asterisk (*) means that 67% or more of the episodes share this feature. A double asterisk (**) means that 50-66% of the episodes share this feature. 242 TABLE lO.--Comparison of Intervention Episodes Perceived as Ineffective based on Post-Session Questionnaire Data and Observer Ratings for the Critical Incident "Non-Participation" - Question 10. Intervention Episodes a Variables 3 5 6 10 3 Coggpnglitx 0 9‘ 5 Min. x - - - — _ Total Time 5 g 10 Min. - x x x - ** >10 Min. 1r - - - — x .. O - 25% l- - - - - - TA/Student 25 - 50% - - - - x - Talk Ratio 50 - 75% x x x - - ** 75 - 100% - - - x - _ 0 - 257. x - — x - - Affective 25 - 50% - x x - x ** Domain SO - 75% - - - — - - 75 - 100% - - - - - - 0 - 25% - - — — - - Cognitive 25 - 50% 7:7 - — - - - Domain 50 - 75% - x x - x ** 75 - 100% x - - x - _ 0 - 25% '1x x x x x * Positive 25 - 50% - - - — - _ Affect 50 - 75% - - - — - - 75 - 100% f- - - - - - 0' - 25% "E x x x x * Negative 25 - 50% - - - — — - Affect 50 - 75% - - - - _ - 75 - 100% - - - - - _ 1a ' - — - - — — lb ' — - - - - - 2 3“; x x x x * ISIA 3 1‘3 x x x x * Categories 4 Y x - x x - ** 5 x x x x x * 6 x x - x x * 7a - - x - - - 7b - - - x - - Nature of Overt - - x x — - Resolution Implicit x x - - x ** 8A single asterisk (*) means that 67% or more of the episodes share A double asterisk (**) means that 5066% of the episodes share this feature. this feature. 243 TABLE C11.--Comparison of Intervention Episodes Perceived as Ineffective based on Post-Session Questionnaire Data and Observer Ratings for Group Stage 1 - Question 11. Intervention Episodes Variables 3 25 32 33 Commonalitya O 2 5 Min. x - - - _ _ _ Total Time 5 £10 Min. - x - - - x — >10 Min. - - x x x - ** 0 — 25% - - - x - - ea TA/Student 25 - 50% - - x - x x - Talk Ratio 50 - 75% x x - - - - - 75 -100% — - - - - _ _ 0 - 257. x - - x - — - Affective 25 - 50% - x x - x x * Domain 50 - 75% - - - - - - _ 75 -100% - - - — _ - - O - 25% - - - - - - — Cognitive 25 - 50% - - - - - - g - Domain 50 - 75% - x x - x x A: * 75 -100% x - - x - — - 0 - 25% x x x x x x l * Positive 25 - 50% - - - - - _ _ Affect 50 - 75% - - - - - _ ‘ _ 75 -100% - — - - - - - 0 - 25% x x - x x x * Negative 25 - 50% - - x - - - - Affect 50 - 75% 75 —100% la x x - lb - - - - - - - 2 x x x x x x * ISIA 3 x x x x x x * Categories 4 x x x x - - * 5 x x x x x x * 6 x - x x x x * 7a - x x x - x * 7b - - - - - - - Nature of Overt x x x * Resolution Implicit - — - 8A single asterisk (*) means that 67% or more of the episodes share this feature. A double asterisk (**) means that 50—66% of the episodes share this feature. 244 TABLE ClZ.--Comparison of Intervention Episodes with Positive Immediate Consequences based on Observer Ratings - Question 12. Intervention Episodes Common- Variables 3 8 9 12 16 20 21 22 23 24 26 29 37 44 alitya Os 5 Min. x - - - - x - x - - x - x x — Total Time 5910 Min. - - - x - - - — — - - - - _ - >10 Min. - x x - x - x - x x - x - — ** 0 - 25% - - — - - - — — x _ _ _ x - - TA/Student 25 - 50% — - x - x - x - - x - x - - - Talk Ratio 50 - 75% x - - x - x - x - - x - - - - 75 -100% - x — - - — — - - - _ - - - - 0 - 25% x x x x x x x x x x x - x x * Affective 25 - 50% — - — - _ - - - - - _ x - - * Domain 50 - 75% - — - — - - - - - - - _ _ _ - 75 -100% - - - - - - - - - _ - - - -1 - T l . 0 - 25% T; - - - - - - - - - - - - -i - Cognitive 25 - 50% 7; — — _ - - _ _ - _ - _ - -: _ Domain 50 - 75% 75 -100% 0 - 25% Positive 25 - 50% Affect 50 - 75% 75 -100% 0 - 25% Negative 25 - 50% Affect 50 - 75% 75 -100% la - x x lb - - - -’ 2 x x x x ISIA 3 x x x x Categories 4 x x - - 5 x x x x 6 x x x x 73 x x x - 7b - - — — Nature of Overt x Resolution Implicit 8A single asterisk (*) means that 67% or more of the episodes share this feature. A double asterisk (**) means that 50-66% of the episodes share this feature. 245 TABLE Cl3.--Comparison of Intervention Episodes with Positive Immediate Consequences based on Observer Ratings for the Critical Incident "Non- Participation" — Question 13. Intervention Episodes a Variables 9 20 21 22 24 37 Commonality__ _ 02 5 Min. - x - x - x . ** Total Time 56 10 Min. - - - - - - - ‘>10 Min. x - x - x - ** 0 - 25% - - - - - x - TA/Student 25 - 50% x - x - x - ** Talk Ratio 50 — 75% - x - x - - - 75 - 100% - - - - — - - o, - 25% x x x x x x * Affective 25 - 50% - - - - — _ - Domain 50 - 75% - - - - — - - 75 - 100% - - - — - - - O - 25% — - - - - - _ Cognitive 25 - 50% - - - - — - - Domain 50 - 75% - - - - — - _ 75 - 100% x x x x x x * 0 - 25% x x x x x x * Positive 25 - 50% - - - - - — — Affect 50 - 75% - - - - - - — 75 - 100% - - - - - - — O - 25% x x x x x x * Negative 25 - 50% - - - - - - - Affect 50 - 75% - - - - - — _ 75 - 100% - - - - - - - la - - x - x - - lb - - - - - - - 2 x - x - x x * ISIA 3 x - x - x - ** Categories 4 x - - - - - - 5 x x x x x x * 6 x - x x x - * 7a x x x x x - * 7b - - x - - — - Nature of Overt x x x - x - * Resolution Implicit - - - x - x - 2A single asterisk (*) means that 67% or more of the episodes share this feature. A double asterisk (**) means that 50-66% of the episodes share this feature. 246 TABKE Cl4.--Comparison of Intervention Episodes Perceived as Effective based on Post-Session Questionnaire Data and Observer Ratings which also lead to Long-Range Behavior/Atttitudinal Changes - Question 14. Intervention Episodes Variables 12 16 20 21 24 44 45 Commonalitya Oé 5 Min. - - x - - x - - Total Time 52 10 Min. x - - - - — - _ >'10 Min. - x - x x - x at O ' 25% — - - - - - - - TA/Student 25 - 50% - x - x x - x ** Talk Ratio 50 — 75% x - x - - x - - 75 - 100% — - - - — — _ - 0 - 25% x x x x x x - * Affective 25 - 50% - - - - - - x - Domain 50 - 75% - - - - - - - _ 75 - 100% - - - - - - — - 0 - 25% - - - - - — - _ Cognitive 25 - 50% - - - - - — - - Domain 50 - 75% - - - - - — x - 75 - 100% x x x x x x - * 0 - 25% x x x x x x x * Positive 25 - 50% - - - - - - — - Affect 50 - 75% - - - - — — - - 75 - 100% - - - - - - - - 0 - 25% x x - x x x x x * Negative 25 - 50% - - - - — — _ - Affect 50 - 75% - - - - - - - _ 75 - 100% - - - - - - - - 13 " x - x - - x .. lb - — - - - - - — 2 x x - x x x x * ISIA 3 x x - x x x x * Categories 4 x x - - - - x _ 5 x x x x x x x * 6 x x - x - x x * 7a x x x x x - x - 7b - - - x - - x - Nature of Overt x x x x x x x * Resolution Implicit - - - - — - - - 9A single asterisk (*) means that 67% or more of the episodes share this feature. A double asterisk (**) means that 50-66% of the episodes share this feature. 247 TABLE C15.--Comparison of Intervention Episodes Perceived as Effective based on Post-Session Questionnaire and Observer Ratings which did not result in Long-Range Behavioral/Attitudinal Changes - Question 15. Intervention Episodes Variables 8 9 36 37 42 Commonalitya 0 9 5 Min. - - - - - - Total Time 5 9 10 Min. - - x x - - >10 Min. x x - - x ** O - 25% - - - x x _ TA/Student 25 - 50% - x - - - - Talk Ratio 50 - 75% - - x - — - 75 - 100% x - - — - - O - 25% x x x x x * Affective 25 - 50% - - - - - - Domain 50 - 75% - - - - - - 75 - 100% - - - - - - 0 - 25% - - - - - - Cognitive 25 - 50% - - - - - - Domain 50 - 75% - - - - — - 75 — 100% x x x x x * 0- - 25% x x x x x * Positive 25 - 50% - - - — — - Affect 50 - 75% - - - - - _ 75 - 100% - - - - - - 0 — 25% x x x x x * Negative 25 - 50% - - - — - - Affect 50 - 75% - - - - - _ 75 - 100% - - - - - — la - - - - - - lb - - - - - - 2 x x x x x * ISIA 3 x x - x * Categories 4 - x - - x - 5 x x x x x * 6 x x - - - - 7a - x - - x - 7b - - - - - - Nature of Overt - x - - - - Resolution Implicit x - x x x * 8A single asterisk (*) means that 67% or more of the episodes share this feature. A double asterisk (**) means that 50-66% of the episodes share this feature. 248 mo so» :6cp one \eopnmpm 69:» 690969 65969 Hm 6:» HH6 ooOpmaooq: H 6xHH 9H69 H "HmHmn om \.eon9 oopnopm mH H 696g3 Moon pane a.H one m6HonHo mH cH undone wcHom 69H oxHH H669 amen 9H H .pH cocoon H 6xHH H669 p.noo .nHH66.H mH H ens weeps ozone enemas» was Has no me mo pHemon 6 66 . . . 6MHH H669 H .moHonHo 4H oH venous wnHow 69.63 6MHH H669 H xenon MH A6HpHesosHv . . . . . NH . . . . . . . . mmwmmfl \Aw6nv .wanon on 69 on wnHow 699 one .wcoH Om 636p on wnHow 69H .pH 6mH96o op me63 69.63_oeos.6HmmHm 996>6 no .669w6 p.noe H .pHmh .966 can move mo 0» 6>ms 63 99 one .H669 H pmmoH 9< . . . 6866 63» 6xHH 68666 ween pH .moeoh 939 wnHoHo6e .chHpHnH96o 039 wnHoHooo 9303 :6 9:696 63 35H 65666 #659. H “Hg: mo \Aps6av «H was 9996h c66kp6p 669506 lawn 63» 9o pHam69 6 mm omc6mmmn 96:3 HNMJMWb—QQOH .mmmHo 6:9 09 6:959 :63» com .30: non 93p .6EHp 96zpoc6 96 SH: apHs nonmsomHn 6939 656959 HHHk 6: 9689 6966 zHHmnH9 cox .6on6HpmmaH 90 manm H6996>non macaw mmmHo 6:9 man wnHoe no 6966& 6a 6< .G6M nee? m6swe6 one mcoHpm65d 9HH6H66966 6m .o6mmSomHo mnH6p mpm6onoo one mono: 6gp HHm 96>o mcH6:MHm n669 mos hae6h Am oesmoeoueeem oespeeeoae ob apeeeeeee . m bemeeoee escapeeo An museums mm “o m eoemmwm Am use A< wH 969552 6UOmHmw noHpC6>96vnH 249 mo w999069 90 0699 69.63 6996a 699H H669 H \.69dm mo wo no no mo mo m< w< \.999 6H999H 6 69 999H69os o903 699 999 .99 69 09 >99 6963H6 0H5096 909 99999 H . . . . . . . .o999 w9969 .966909 wn969 .69668 9HH669 99 9693 9993 669w6 H 9&5099 m6>6 . . . 90 6999 69 999H69oa U903 699 969 \ .6096906900 930 9509 wn96D .69966596 699 909 H690996969 9608 69 6H503_9699 660 9993 9: 6600 09 wn9999 696 . . . .9099 160566 90 69609 699 696 H00906 699 69Hn 996ew05n :30 9909 .6096906900 9509 mn965 99096 wn9on69696nnn 6906 90m H 99999 H . . . .969969 9669 6>69 99w9a 6903 996969 .996 .6 CHE... H \395 9963866 ens 886 meme 9:89 6.93: 96 9966 8.9 we: 66:8 9.96606 9699 @903 6 6.99 .9HH669 909 69H .50» 09 6509 6900 9HH669 9.96606 99 6999 669.990-6636 .8236? 996. 699999 s 669506 999H69oa @903 699 6609956 H “6669 \.6609956 H 999H69oa 6660690 was 69:09 66999 909 6969 996 :60 63 959 99 09 6908 90H 6 6.69699 3069 H \Am9wd6H 666H0V .8969 9699 95096 306 3099 3 966: H 996 .569 H 8:9 9669 H :55 face 9 £5 99:88 peace 6:26. 6.93 as 66966H0 690969 6966 669 9693 ”69nn< \99H996wv 9:09 90 9669 699 668.6 .669 face 9 $666868 99933 a; so: 959 .6963_9699 9693 666H0 6999 0999 wGHOMEGMH..........uflIOIeM \.o966 695w :09 9693 09 96969H 09 w999n63 9o mm9on69696omd 663 668666 99 66H6 960906 6666069 6H969909 -865 696.9 639. H 6.9 369 pom been one 250 :0 No no \w99h66 696 699069 9693 . . . \9999669996696 9693 600969669: \.9690 .99 9609 996 w999 I99w69 699 96 M999969696699 6906 669 696 9699 .666 9969 696 966 m6: 69669 H . . . . . . . w999996o 6963 699069 996 @6699 I906 6908 90m 696 9699 966 96969 69669 H .90 9963 9699 90996696999 699 99096 9969 909.309 96 9009 09 m99999 99996 9.9 .96 9009 09 909 66966 H 9693 9099 9636 mo .8 To mo mo 9w90999 96m 966 H 06 6966 9009 699 9693 3099 09 9963 9699 H 9963 9699 69 w99b66 696 699069 699 9693 90 90996969996999 9: \96999 09 6909669 H 960 mummmm \.966n996 699 990 90w 90 6999 63 6999 69666 "6909 \.6fi6m 6903.6 999 .99996908 9699 90 99996909 6999 909 663 99096 m999969 6963 99966 996 909 9693 99999 H .6696 w99996906 663.90 90m 63 92.3 9.99 9 Be. 52.96 66398.8 6.929 999 6999 69093.98 9696906 09 9963 H 9996906969 563 699 6.9699 996 .69 99 963 29 3 693 69639899696996 923 6.59 .9699 6369 9 .9090 9699 69696669 90 909 6 6.69699 99999 999669 H 996 9669w 69 9966 -68 96.9 9629 H 65 9996989 3 629966 9009 699 9693 3099 H 99999 H .99filmw 66699906 9699999 9.H 99999 9.906 H ”969 \AQHPHQSUGHV.oooo.o....... \.A6999©9699v .90 m990w 663 9693.6009696999 H 99w9099 H O Amnflpfiugdgfl v C O O O O O O O O O O O O 0 O O O O O O O O O O O O O O O O I "aha 251 mo mo mo m< mo H .pmmp map you 9“ ande on wnfiow a.H «an» mafia» map map as mom 9.:ou H .pmop vmnfim map commwm hcdmhdw H . . . . . Umpmohmpna you a.H tong H \.0c 0» wnfiow mh.m3 pug: on ma cnfia ammo whoa_d and wnfi Icqumuqus pcmhmmmfiv a £9H3.v>cmH hHHpm tome: and mdeo mwnp cw psonm mafinhme mn.m>_9633 psonu vnmaonp hHHdcfiwHuo m3 Hag: opna muduomhoonfi ow .aonp 9:09» xnfinp on was mama on and name use mono? mmmnp Hug? ow mocfis use avmo ma ow Op wnwom mH mmeo many was» wcfimon mh.hmnp pan: wad munch pamuommau mmwnp HHd abogu muoofi n30 AHmnp spat mmeo wasp ova“ memo machhm>m .dmcw n30 nfimnp man muchhm>m .n: 95m .mndma aoapdponnoo who; Hag: ma page .mugoa mo acafix Haw hp_uouugpop a“ machho>m HHvk..mUA03 :Hdphvo hp umnonvon mH machho>m mxHH .pH Mo #50 wnHQde HdQOmnvm d mxwa on wafiow ma machhm>v and machhm>w op wGHQdoa HdGOmnmm hhw> a mcHon pfionm wnfixadp ou.m>.w:HSphhm>m pap xoop an» GH 8039 mwm ado a: mafia .pnonm mnwxdap mn.w3 mwGHSp was moHHOp amuse HH< \.p.cmws H hoax H .HHd pd wasp wafihdm ma: ocohqa Adan» p.noc H "achdo Amoco pm hmsmqw on mmfiup HHd mdeo easy \.ms pa pnmeHQ H mg: common man n.9anp wad ma mumnpop thaou was» mmmsm H .08 you can \nwsonsp pmw Op scan on cmon H pug: nun» whoa wcfinphqw ma nodmp 09 my» p.nov can we mamman p.aoc on gun» npfia.hngus a.H and .ma mama HHfiu mane .pumunoo hwwhuH map aw w20h3 no pswfiu mpfi 9H wade p.noc H can ouwn no 990 90w and mmhsoo on» 252 No \Apzwnv .mdk wasp Hmmm p.:on so» mo mo mo no mo who «3 mac cosaoo one m>an HHw o3 mmmHns was mammonoo may mo waficcdpmnmuc: owmdn u om on was .hwwOp ow on wufiow mu.m3 wfiwm so» was: .mwk mfinp we wanna 08p pawnonp H mmswomp smwmem hppmnm . . . . . mo m: amazoom do» was: xdwnv H .HHm3 Honda \.02 mummmm \wpxmv ms» oven :0» w>dm knoHpmmsw w nhhhohv do» xmd H aw: "mmmm \.psp page unchep poH a m.mhmsa .pH on xofipm canonm 03 now aaHmo he a“ and pxwp map mafia: mu.m3 pap . . . . . . . map .mmsmmH Hth mud vhmnp .mSmmH cu m« «awn» ma and pump 0:9 0p Moapm oHsonm or waxy xnfinp H .gs .pmop a w>w£ ow 03 monfim .mdeo many mo mmmom than 0:» mom asp .thHduoE m.mmHm hflOphud op :mpmHH ov ammo E.H and #009 ms» GH mm: pmn3_:dsp me you mnowvdoHHgafi whoa vOH a was thHdnoa .hpfidauoa no nowpomm on» damn can #009 map nwsohnp yam? H qmsz .vnHE ha a« mammH as you mu .hHHdQOmnmn MHmmnsoh how mammfi nu ma .thHdhoa an no» you wanna nu mp baa pug: Mafia» H mmmm \.nmapflm has Qua» pamm hmnp Haas» v.cow H .mem mqohho>o cubfimohvm H son and \th3 pan» Hmmm v.aoc H .on .3: \o o o o o o o u.“ ”no Ihhm>m has no qOHPdNHHdeqmw mmouw a vman m.pwnp xcfin» H \.oop ma vacuum>m “a song p.nou H .qowdmu pump you ohms Poq_au omH mma and Hma mma mma and mmH mma Hma oma maa mafl Had mad :aH mzH mad Hga oafl amH mma ~MH mma mmH :MH mMH mma Hma omH mmH 253 No \Apzmcv wpwzp ham machqd Hum: 90% cHQ mo 2b .8 mo mo who \. . . . pxmv on» qH m: on am>Hw qupHnHmmo man Hwnpmgs mum on MOOH 0p pxwpcoo dwhspdao A50 Ho vac mnow pom m>w£ as no pmoa was» mH Ode MoHHmp ha cad pampcoo Handeso a scam maoo 03 pan» mH HmHHmp Ha .HHHaHpnam pH 9» HooH H .HmHHmp as mpH .panu .pasp cHam H mummmm \AhHw>Hmcmmch .pso wcHxOOH damn n.0mHm thQHQw mchcH pom mh.:oh magma HmHHwD :30 Huck mpH mmswo span» 03p m.pwsp ham hHmen p.qao sch cad wnHme mud mHmowm stpo map 9633 wGH>Hmonm wh.:oh so: on ma Hmapqaw use usoH m.ausp xaHn» H mummmm \.pw£v mufiqwoomh p.nou hwnp mundomp ha IHonmHomnm asap ham 9.:oo mHmomm mHMMMW \AnOmHnfi GHV oz ”mvnwcSpm \.und 9639 mnwnwoomh p.nou «Hmomm pump we on mammm pH uq< .pwzp waHoc pom mh.m3_GQd moSHdb an3 oo 0p man pwzp pmmonoo u abopm mpqwampdpm mst> vsonpfik hHmpwpowwpo wzwmmSUmfiu «A.ms bum mHmoom mmsmo .xowp va ov me How and: Ho qupmoumeqH Amanda mpH .mvH and mpnmampwpm mst> can: m>ds m5 Ho HHd pH wnHmmSoch qH um>m .ohmn mm: ow and: so» who: Hw>mpw£3 .pzopu mnHMHdp 0H.03 Hm>mpws> no mmst> Ho moanm Ho thHuhoa ans was» mchmmHman mpH mumsw H mummmm \.meonHo mmuHmmp mumsshqa vmw Hobo: HH.m> mmHm Ho hp Mao: 0p pump ma» «>6: . . . .mudon NHm you pH vawhd v.03 .mmno pconHmHu macs 0m m>dn on maHow mad as no HHo HH omsmoop pH apHa ago: p.quo HHHaun mmH :9“ mg” Nod H9” 09“ mg” me Fwd mg“ mg” :QH mmH Nod HmH OQH a: 95.. CH whH m5” inn MPH NEH HNH ONH mmH mmH Fwd mmH mmH :mH mmH NmH HWHW 254 mo own» yonpdy xdmyn w oxdp 0p mp cHsok ow o» mcHsp pmonoo onp vow mxHOH msom you son pany oprpyoyaooq: oprHH a on pnmya pH was» wayHmmm s.H .mpsnye a pyaz wHo go no no mo mo mo \.coow .gdmw "mpoovnpm \éowpmomwafi coco "mm-MM \.oyo£§hqw wcHonw oy.mk AGHSp p.aoc H mmdwoop yw>o pympm can Mdoyp d mxdp pmdfi 03 p.aoc has om oouH hHHdo IHpomym mpH .coHpmmwwSm d o>m£ H mNmmmm >ng IoSOm ham on wand: hyymm annp H "awn \.Aoopwyp Imsymv .thHwyoa oHyo>.oHog3 ms» oHdeNm op . . . . £03238 a uni. PpH ”mwfl pooymo yoy pmsn an odds oy.o3 MaH£¢\p.ooc H wasp mooHPwNHHoyooom meow was yocyo 080m 959 op pmamppo ad mpH .mooc pH pap .hyvcsoo yospoow yo myopHSO tho yonaH mounds yo 0.3qu goyymglnmfindmm aH 9.:moow HHonpymo pH .oyszzo yonvoow oH wofinphqw name no: hwa 9H . . . "mmom \.w:H£phcm ooaHyoo hHHomy o>.o> xcHsy p.206 H van» mH wthwm B.H pug? and waynpyqa mgame HHHamy aoypynyymc mynp «H use ccyy op paw» may my am>yw noyuynyymu 029 pm xOOH op pkopooo deSpHSO yso no 950 now p.om>wn m3 yam .mep on» wavy p.noHu H cme H can: agave H pugs n.90ne .mowwm om on» wavy ¢.ao>wn H .myonpo wad yHom psogw thmwnos vzoao manv many yon» toqx op nmsonm pump may no coHOOH o>.H .uooomm d prz Ampmbyyovafiv mHMMMN \cooy p.no>wn och pom "oan< omm mmm mmm Hum mmm mmm :mm mmm mmm me omm mHm me Sm mHm mHm :Hm mHm , «Hm HHm OHN mom wow wow wow mom How mom mom How oom amH me 3H me 255 N¢ no \.p:prQs 50% oy< \.p£prms HHw manpmm mH hoophym>m \.nwmy \NSOH yom myonp mH GOprypmSyH pump Ha: yyypcooy w: mng soy ado .prHH \Apdmcv .wdHHmmm he m.pdnu .n: was .wGH tom my.o> myo£3 op ma oydmoao wEOm nomoy m< w< m4 m< mo $4 w< mo mo mo \.£de "mvomodpm \.oa yom wqudypmSym thwoy mpH cod mmHoyHo :H codoyw maHom my.m3 cod mEHp ydo maprma my.m3 mammm pmsn pH was away ow p.2do vmsfi 50h mamoom home man fingdmooooococooowHHHyAdeoH .zoh vyos hym>o pdonw oswyd .Uyok hyo>m mofimmo .mmmv hHmey ow .pH oucH pow .GOm Iyom ymspo moo npwk mwcfinp ammo ouaH pom H .oamomm hows mHsp Sufi: mw:H£p mood ovofl pow on pod: ¢.coc H .90: mH mmoHo mynp pap pH ow ucw snow pHm ow paw: pmsn H mmoHdd mwcan mood opaH pom 09 and: p.auyu H mmswomp pH opaH pmm op and; p.uoyo H .H0 mmswomp HHa any and .gmmy \Amnwde wmeov .mnH> man now H \.uca pgmypms HHw moHpnom mH hoophym>m .hdxo .mmswoom \pr HHHpcmuH H cHsonm \.qy Ham 3 $33 33. PH {aux fnou H “wmlaqum \.dopwypm:yh quppwm S.H mHMMMM \eaw H “Sam \.pomm: HHHamy mH yuonyqa Maya» p.qoc H mammmm \.:3oo onoom maHdo pH mmmmmfl \.quoHo>w pmdn .md 08 @oxHyvm was: m.uwne .mdmmH may UHo>w . . . mummmm mmm :wm mmm mmm How owm mmm mmm Hmm mmm mmm 2mm mmm mmm Hmm omm mam mzm Ham m:m mqm :zm m:m mzm Hzm cam mmm mmm Hmm mmm mmm :mm mmm mmm Hmm 256 No no no mo \meoyHo may xwoyn wk :60 Bow \. . . . . . . HH.w3 om hHo>Hposypmooo myoa HHHwyow>wnon pow coo o: to: pd MOOH a wcyxmp and pwsp wanHpHsoym my wayew lymmo ma msoym on» has on» pwnp wcHHooy B.H woflhdm Ho maymp GH mowono a o>os m3 xcwnp.mm H .hwxo .pdona oyd mpaoocoo omosp was: 0p mo wdecwpmyodqd ed on mcHaoo HHo>HpomHHoo and mpmoocoo omonp :ka Homo op wcH>wn ms mom H ma yom .hwxo \.ma yow wcfizp msoyw w m.pw£p .pH nowoymmw op wdHow o3 oyd 30m Hoymnp aoym m>oa and some omonp was: no wchcdpmyoch mo HpHHdoanoo w on oeoo and mammoGOU yOon meow .modmmH yowwa oEom mmoywvw 0p onom m3 my< onwm mHoo< mo yawn wchpHHmm .odoe mayos map was: 09 mo mosaw> Hacomyom ovoH pow cod chow pHm ov wowow o3 oyd .msoyw w m< \.o3mmH mop mm: psmsogp H pan: m.pw:p cad HpHprop cH mooym a mo opwyomo ow moHow my.o> Son yo odmmH ad apHs onmeo m: mom H pony mH ma you no onow m.pwnp wownvoaom .hoxo no no mo “mqwnpmaom :pw: mooywmeo huonoBOm HH can . . . . . . chHomc mHmoom 03p ozp pug: Uwoy HUODmEOm can and nwdoynp onow pmHH may Ho Hep may pa umpyapm pmsn .omcym umu pmsq .cmmaopm o: my guys» H .Hmmmm \.wcHHomm may do: H .mmHoyHo GH undoyw mcHom my.o3 mxHH mammm pmdn pH .panm \chwvodpmymund Ho MomH monmmymxmv .oquHpso HHd mH mmeo yso .man CH moHono w pow m3 .hdxo \wmdoym a mo mpwymmo op wGHow my.m3 tom ”Hp HM oom mmm mmm Ham mmm mmm :mm mmm mmm Ham omm mam mam Hmm wwm mam zmm mmm mmm Ham omm mHm mHm HHm mHm mpm :Hm MHN mHm HHm OHm awm mmm Hmm mom 257 no mo mo m< no mo mo no mo mo \.3oa oyw wt to: thodxm m.pwna .pmoa oy.o3 omwkymnpo .monnu mmdomHo ado o3 cone .co ooyww hHHwyocom 03 yoga mzfinp Imsom on maoo m3 HHaq: poosmpwpm poop eoym hmsw mxwp m3 yo dopono on ddsonm pH yo .quspoEom vow H yo denpmSOm cow oddonm 30h McHsp :oh how swap and mfinp mH mpcoawpwpm deoH>dan xcan H .deo .hwm cod stow mpHm hoonhyo>m mNmmmm \.gsn g< “Hmmmm \.mpomao9wum deOH>w£on mxdp m3 mom m.poH .hwxo .okoo pHm 03 and nmdoynp om o3 HH \.ommdmzoo hHHdpov mm: H .pswdonp hoonhyo>m was: :H yonpww p.qudoo H\.h©onhyo>m cad soquy hHHdpop mm: pH .Eocody mm: mome xoov was: mNmmmm \.wcynp mawm map mm Hosp mom H \woode xoop 96:3 vow .hws ySOH mH wcthm my.305 pas: .553 ySOH cooSPop moooyoHHHc map m.pwn3 «NMMMW \.oHnoHHoynd ow mH has mHne .mxdwzp omHo hoophyo>m Hogs mo wofiuodpmymocd 080m m>m£ HH.@3 9‘33 pm psm .222 H .mflp 935. “Hm-mm \.mmyww Op hoop thyo>o pom op mcHow ym>on my.sow «mmmmm \.ms Ho HHw .waHvowpmyooq: ad on 0800 o: mmmaqs mSOHSoHUHy no oaHx mammm pH and pooch mwcHHmmy pcmyommHo hHHdeammmo m>w£ no mo HHw pomp mayo» yw>o wdeswyd hHHdoy myma m3 vow thHwyoa on now: o3 cos» .hwxo .cOprHdeoda yo thdeoa pdonm maHwap my ompympm pmsn m3 .ccHHp my way Ipywpm my.m3 pom \.»H mmdomwd ado 03 now» mmm :mm mmm Nmm Hmm 0mm mmm mmm Nmm mmm mmm :Nm mmm mmm Hmm omm on wdm Pam mHm mam :Hm mHm NHm Ham OHM mom mom Pom mom mom :om mom mom Hom 258 do \ymxHoo “capo map spy: #50 pmnp Homno 30» p.aou Has No no mo mo mo mo mo mo wo m< mo \wzmsp mH on: mummmm \wm: pom m>ws honp pampooo map canH: mayspyym>m mama 0p m>an so» .HdHyopwa map wan swsoynp pom p.:deo> o3 mowow mo: pH hos map pH 6H6 m3 HH .ym>oo op HdHyopwa Ho pOH m o>w£ w: .omysoo Moms amp 6 m>wn wt . . . . mummmm \Amnm56H mmeov medp pH HHHz w20H 30m uwHGq< \Adopwypmdyt Hoox ..HHm3 goo» "mmm \HHHpqs omHo wdanmaom psopd wcHxawp myok o3 cams H H30: pnwwy pwnp yokmod H oHdozm Homx 0pv mummmm \thm op onhyp so» myd past .thHwyoE no poHom ydOh m.pd£3 p59 .hdxo mmMm \.¢0Hpomm oHont map wavy p.co>wn pmdn H .thdeoS yo noprnHmmo x009 map omoy ohms H Ampmsyympnpv mummmm \.yo aoypynymoc Moon map dwmy p.:m>wn :0» song H wth IHmyoE ysopw chom ySOH mo: posh pan \pH op m20HpmoHHQEH myoa hows oyd oymnp ponp chpmyocod H cod Moon map op mszyooow mH thHwyoE pom: codpmymons H and do lopmms was oaHp ma yom GOHnHmo as :H pan .zwm 0p moo uydn w m.pwnp .mfiap moHpmw3 HHoz \.pwnp Hays» H and mayp wcypmws my.03 mmswo owe wcwppmw myon wanpHm mm: H oxHq .owE hym> pom hHHMQOmymm H wad npyom and xowp mepwn a mo 3mm H pas: no cH pom m3 own: van konx p.coo H hoax H \.thHoyoS psonw was qu09 y50h pug: muonx Hconyuw Hays» p.aou H .pnmym «mmm OHm mmm mmm Hmm mmm mmm :mm mmm mmm Hmm omm mmm mmm Hmm mmm mmm :mm mmm mmm Hmm omm mam mgm Ham wzm mam Ham mam mam Ham 02m mmm mmm Hmm mmm 259 mo mo mo no mo wwo mo \Aompwypmdymw .wcwzwm E.H pan: po: m.pdnp .oz "wH:n< \.m>dm :0myom yonpoqd pop: psopw mywo p.cov 50h mH mnHme my.50h pan: moyok yonpo :H mNMMMW \.mcHH map :30@ om pmdn ado pH and .ooHGHmo yHmnp was we 0p pxmc :Omyom map .GOHGHQO he v>wn H .HHm> mthwm hoonhym>m mxHH mo vqfix mo: pH pun coon o>w£ UHdOo pH .umcpm moo mo oofix mo3 pH .mdoyw oHogs map wooed cowpmmym>ooo Hazy a poo was pH mmwmmfi \.pH monmdome on p.2oddos o: onm 50% .wcfipmmyopzw m.pwnB .pH mchmSoch hvomwa mywb m3 mummmm \.pH mmdomHu op oHQa on p.odeoz o3 om mmyww Had p.qudos 03 HHHpm cod .noHcpmo yHmnp pow acmymm myo>o pwnp om pyomw mammp mepHH zoom wawxowm oEHp nods 0m wofloaomm mp v.03 omswo mcpnp Imam yo moHpowym och swap pan ms mHmn op opr on poo dado: com and syop mHons 23 Sons 33. SE» 3:03 pH 323 Hwy 0m on 6.03 .mpmmocoo mmonp Ho mono do you moo x00p ms HH \.hw© woo oxHH oxmp cod .pmmoooo mHMGHm hyobm .mdmmH mHmch hym>m mxwp pocodo 50» .thpm pnmccom . Imqu mxHH wnpnpoEOm on vHsonm mymnp ha Immmo man myoagxm dado: pagp mmydoo a do Ip:d3 50% HH .poqqdo 50H .ywoh oHogz map mxwp cHsox mmydoo map pso onme hym>o mayo .onSphym>m mxwp 50% HH \.50h 9H0: 33% £33 £3: ymfi pump H3533 Ho pesosw x nwsoynp pow 30h 9H0: op maydov w m: pom mo: pomapywmou mpna mNmmmm mo: :3 mo: m3 Ho; 8: mmm mmm Ham mmm mmm :mm mmm mmm Hmm omm mmm mmm Ham mam o mwm :wm mmm mmm gm 0% opm mpm Hpm mHm mHm ppm mpm mwm me 260 2b o5. 3.0 \.so£p op ooocommoy no» box \Awocv .onoom yonpo oEOm yoy wcpoommon on pan8 pomp so» 0p voocommoy mxHom osom ho: opp mo pHdmoy o mo wcHHoom E.H 3oz .Hmo o8 oydp 0p yo o8 ooqopmHo 0p o8 0p poop mooo hoopoEOm con: hocoooop o mos pone \.omcommoy ySOH up mcoHp Iopoomxo yo mayop up :30 ydoh Ho momoHp osom coooo 505 poSp .hHHosoH>Ho:H yo: 0p voommoy sonp yozpoy ponp pHoH H .mmoooym msoyw onp mo mayop CH wownpoSOm hmpyoHo 0p mathp mo: oHod< waHHoom B.H .mcH Imam mH ooooaom pop: yom :OHpopoymyopoH coo . . . mp pcoaBOo ydoh yo>opo£3 wcpooo HHHdoHpmooyom Ho UGHx mp oa op mmoyoo onBoo m.pm£3 and .oyo hogp oyos> Ho pdo myonoos o monoa oy.505 woo so» an3 mmoH IHooH pmooon oaom wcHyonm oyo onoom ponp my soy opp: 98% op 85 PH moEpoEom mo .32. 00 mo wo .mo>Homyso psopo ooxHop o3 con: mcHochop ozp p< .oyos moan ooo ham H moo "opond \AyopoHsvv .50» Ho HHo onxooppo Ho pyOm mo: H mmosw H «HMMMW \.©oxooppo Hoop o6 oooe wcHxHop mos oHcd< oHHgs oooa so» omoommoy one mummmm \whoSp yo EoSp op woooommoy H hon zoos so» «NMMMN \.ooymo H \Aoopoypmsym .hywqov .oz mummmfl \. . . . coop o>os HHH> o3 .mpooaoyHSon omydoo onp mmom HHH3 o3 .myonooop hand: on HH.o3 dam SooymmoHo man mo poo hHo>Hmmom ow coo up pH oxop HHH3 o3 coo Moon opp Ho owoyOpdp opp yous: .xoon osp yous: oyosp ppm HHH3 o3 coo mmoHo mpsp up thyOszo yso on HHH> xoon opp .moyoz yonpo qH mummmm oz: mm: mm: Hm: mm: mm: :m: mm: mm: Hm; om: mm: mm: Hm: mm: mm: :m: mm: mm: Hm; om: mH: mH: pH: mH: mH: :H: mH: NH: HH: 0H; mo: mo: Ho: mo: 261 oH< N< No 0p poo: p.ooo 50h poop .onHooy poop oxHH p.ooHo och poop oa dHOp yo>oo oow \Awoov .HHHooomyom pH 0p moHpoooy oyos 30% 30: o8 HHop p.ooHo so» coo do» opp: xoonwoom oEOm coyoSm H \wonooopmHo Ho mooooodomooo mo oopx omoop poooo Hoop soy oo pom poHooo .HHombE yop mooooodomooo oop too: so» poop Boo 30o pomHy onHooH do» oyo ponz \wooooowomooo poop Ho oyooo 50% oyoz AQHQflUSdCflVOQ000000000...o m< wo oH< mo owo mo mo \ooa onQQOpm yooh oopoHoomeo H UHom H \.oHoHooop opp xoyop H \O O o O O . . . opom H poo3_3oox p.ooo H .oo ooHp mpH .HHHooy poop .oyo poop mo mooooodom Iooo oop poopo MoHop coo 90pm H oo£3 poh poo so: pomHo HHpooooHHo poop How mo oopx ooo H .moh HHo3 .oHom H coo .ooyp poop mH .onxmo yooh woo oEHp poop po pooaopopm poop opus oHooo H .Hoomooooz poop poo oopopom H moHopoEOm mo: poop o8 0p HHo3 \Aoox OpV .powHy hHopoHOmno oy.30H .50» opp: oowyo 0p moHow poo E.H .powHy oy.oow .oooo pmon o> .ooh poo: m.poop .onhom coo poosos poop po o8 onQQOpm yoom opoHooynmo H mNMMMm \.pomHo ofooH onop H .oooH «Hmmmm \.onoo oyok ooh poo: mmoH oo oooa mo: poop onop H .HHo: mmmmmfl \.oooH «NMMMW \Hokoc xOOH coo ooop o>opo ooopm op coop och .okoo onoom yoopo poo 0p coop so» poop zoo ooh p.odHo .Hoomooooz oo poop hob oMHH m»: pH: m»: NH: H»: or: am: am: pm: we: no: om: mo: mo: Ho: ow: mm: mm: Hm: mm: mm: on: mm: mm: Hm: om: mg: on: pg: mg: ma: 2:: mg: mg: Hz: 262 mo owo \. . . . . . mopoo on koooaom poops proo poop pHoo soy poop opom so» \Amoov .uHoth yo hmpom yo opoo< oooopmpo mo m0 w< m< mo w< .mpop oMpH oooo mot poop moon mpop yoooo wopyoopow m: uo HHo 3mm H AmowooH mmoHov =.o&oa oop yoooD poo: you HopoyooEoU poop opo pHop H ooo .xoon pooym mop poooo opom so» poo? .opo mopHoou sonoEOm mo3 \.onoom yoopo yo>o no uHom Ibo ompoy op mpmop_o mo poop omo p.ooo H pop you proopyomno oop o>oo yo>oo pompo onoom yoopo poop wopHo>oyp uo pop o oooo o>.H .moQon uo pOH o ooon o>.H .ooom p.oo>oo onoom yoopo uo pOH o poop pOH o ooom coo mwopop uo pop o oooo o>.H Hooop>poop oo mo poop Hoou H omooo oop uHomho uo oooym E.H ooos H \.oooymmoHo mpop op oom H poo: ooop oyoa do» koox p.ooo H omooo wooyk hHHopop on oHooo H .oOpoon HE oH .omHo mopopooom on oHooom pp 0m yOpyomom Hoou 0p o>oo ooh oooa thyommoooo poo mooo oooym pom mmmmmfl \.y0pyomom Hoou H .3oo mom ooo H supp: woo hoomooooz oo opom H mo moooa supp: mmoow H .oooym E.H .opom H mo soz .hwa poop Hoou pmon H .zoox p.ooo H wNmmmm \ppoop opp: oo op mopopoo o>oo pp mooo yo osooyo>o op wophyp oy.ooh opo wopHoou oyooomop opmop o pp mH \onmoom yoopo op poop oo 0p poo3.oo> moods poop ooh uo oopmop pp mp pops pop mo>Homooop poopo mpoommo opopyoo oyOHon p.ooo onooH uo pOH o omoooop zoom poo pompa_oow .ooop wopopooom ooh xmo H ooo HHoz "opoo4 \.p.ooo H HHoz OHm mom mom Hon mom mom pow mom mom Hom oom mm: mm: Hm: om: mm: om: mm: mm: Hm: om: mm: mm: Hm: mm: mm: om: mm: mm: Hm: om; mp: mp: pp: 2.; 263 oHo mo mo mo mo wo oHo mo Iyompo oo pmow m.poop woo .oopo> yooh op ooOp pooyouupo o oom H .oooppoo H .ooop moo y0p>ooop ySOH poo: aoyu pooyouupo mo: poop woo yoo ooyommooy .Hoou hppmooos H .505 coo uo oyom p.omo3 oom ooyo oo oohoa m.pH .myoopo 0p mmoyoo wopEOo mo: oom to: poooo oyooooo oEOm commoymxo oom .ooo oopop p.oopo sop poop poo oopHHop pmow .oooe o>.oo% opoooBOo o>pmoyno yoopo oop uo oSOm Boyu pooyouupo wopop IoEOm mo poop zoo H coo oOppoyoopmooo Opop mmopHoou yon MOOp pmow so» ”mmom \.>ooM p.ooo H mHMMMw \pooh 0p appooyouupo mmoyoo oooo pp opooom pox mummmfl \.omHo boophyo>o op mmoyoo oooo pp ho: oop m.poop o>oppop p.ooc H pop .os op mmoyoo oooo opo pp hoz oop m.poop wophom E.H \.ooxomm moo oOmyom oooo ho: oop woo .ooxoom o>.:op ho: oop poop .oyoo moppooyopop ooop o>oo onoom poop Ho: oop soyu pooh o>opHop H .omHo hoopkyo>o 0p mmoyoo mopEOo oyok oom ho: oop mo: poop o>oppop p.ooo H .oooH mwmmmm \.505 0p pmofi mp mpop yo omHo hoop Jhyo>o 0p mmoyoo mopEOo ooon o>oo H ho: oop mp poop up zoox p.ooo H pom ”opood \.mmoyoo mop Iaoo mo: pp Ho: oop m.poop .ofi 0p mmoyoo mopEoo mo: pp no: oop m.pooB .powpy m.pooB .hywoop ooom m.poop .ooow «Hmmmm \.ooo :0» 0p mmoyoo mopooo mo: H to: m.poop pom .Hom 0p poooa H poo: poo m.poop omooo .moppom S.H poo: poop mp .opom H coo .mopopooom yo ooyooo o oxpp mom ppm mom mom ppm oom mmm omm pmm omm mmm :mm mmm mmm Hmm omm mom omm pom mmm mmm :mm mmm mmm, Hmm omm mpm mHm ppm on mHm :Hm mHm «Hm ppm 264 mo mo m< opo no mo o.pp poop .mwopop wopomoompo yo :0» 0p mopxoomm m.oo »o: oop 0p ooo wopoommoo mp wopooopmpo ooom Hoou so» up pounpoo on opook pp poop mp »om »yyow oyooo H pops \ppoo poop poooo H ooo \.oo» moppooyu Iooo oppopyouaoo mo Hoou poops . . . . . . . mopsoox .pHHopppop poooH po .oooo op oxHou ooom on poopo oyoop poop oyom 0m poo E.H 302 \.»oxo .y0p>ooon uo om»p poop moppooyuooo oppopyouooo »ppoym poou H \.mooop yoo» uo ooom poopo coo uHom Iyoo» mmoymxo :0» »ok oop poono pooSooo> »yo> on 0p ooom do» coo Hooo> »yo> oy.oo» \.oo wopow mopopoeom . . . . . coo oa 0p osoo mopxpop oy.oo» moppoou opp mopppom E.H .opoope o ppoz boo .»om coo :0» 90pm op o8 oHoo @Hook pp o>ppooypmooo pmoa on 0p poop wopHoou E.H .uHomyoo» moppoomoym oy.oo» »o3 oop uo pHomoy o mo so» Eoyu »oso wopkoym S.H yo ooooopmpo wopop E.H poop ooppoop a.H pp .poop mopHooo a.H \.wopooommoy oyop oo» »o: oop »p opoo< yo uHom»E yo »mpoQ wopooop Impo oyok so» moooyom poop oyozo p.ooyo> oo» .oo wopow mo: pp oops y0p>ooop yoo» uo mooooovomooo oop uo oyoko p.ooyo3 oo» poop »om so» oyooo pmon H .wopoyoop op oopmoyopop oy.oo» .onoom op oopmoyopop oy.oo» poop mp »om so» oyooo o>.H pop: pHoumHoo op opook poo? wopyooooz E.H mo m¢ mo \.pooo \.ooos om o8 mopQQOpm yoo» oopopo onmmo H oomooy oop m.poop .ooo» .pooyw Epoopo m . poop "Home \.o>pppmoo opp oopop H .ooppo> own mpm mpm ppm wpm mpm 2pm mpm mpm ppm opm mwm mom pom mmm mmm 4mm mmm mmm Hmm oom mmm wmm pmm wmm mmm 4mm mmm mmm Hmm 0mm mam mam pan mam 265 mo mo no mo m< mo \.oydm \Apyooo opv ppooopoo op poop deoz \.pp mopmmdompw oy.o3 zoo uo pHdmoy o mo o5 uo owpmop wopooo loos m.poo3 mp mpop pdp uuo do» pom yo do» oupopppyo 0p oOppoopop »E poo m.pH \.mopoomooo m.poo> Boox »yyow pop Hprm woo poop »om ooo do» »o3 o m.oyoop wopHoou E.H \.uuo o8 pdm 0p mop Ipyopm »HHooy m.poop .»yyoh mop»om .popom mpop po pmooH po .oppopyousoo mo on poo powpe do» uo oSOm wopHoou E.H .poop mop low oppopyouooo »yo> Hoou H .onoom oEOm opp: pdo pp xoooo 0p poos H \.oa you wopoommoo mp mpop .»om 0p 8p: 0p Hdumpon mo mo md oowwdw o uo HHo omdoo pp opp: Hoow opop ooo o3 ooop .do» soyu woooopmpw wopHoou E.H .»om 0p oa you omoo mpop op oo .»om 0p do» you pdp.mmN_woooopmpw H .oowwdm o uo HHo poop .oo wopooopmpw oy.do» .opoo< .wow»om ooop yoopoy woooopmpw moppoou Bo H poop wop>pooyom so H .wopHoou_ao H opoo< .»om woo do» QOpm 0p oyoz H uH .oo oooopmpw 0p oyok do» omommdm .do» 90pm 0p oyok H uH .»oxo .oook mp Hdumpoo mp .oo pdono »Hpoopupoomm wopMHop oyok do» pdn .»won»oo wopQQOpm uo »ok ooo poop Mopop H \pppp o poop oowooyp H ooo aooo» mHMMMm ooo“; \.o0ppoopop »E poo m.poop omdooop moopHoop mp: pyoo op poo: p.ooo H .apo mopppdmop yo .Epo wopMooppo »HHooomyom E.H Hoou poops oo wpoyuo E.H omdooop pp mopoo oHooppoosoo Hoop p.ooo H mmpmmp mHo 2pm mpm mHo HHo oHo mom wow pom mow mom oom moo moo How, ooo mmm mmm pmm mmm mmm :mm mmm mmm Ham 0mm mom oom pwm oom mam own mmm mmm me 266 m< \ppoo Iyouupw m.poo3 .om uH powo mopdopo ooop nppp .oop opo pp ooop ooo poopopppo poo Hoop pp oooo ooo ppppoopopppo poo poop do» oxos poop wpm poOppoooy ydo» o.poo3 .do» 0p wopoposom wommoymxo pmdn m.»yyoh no mo .wopoopn woo mopoko .wyoon oop oo moOppp Iopuow 03p opp mp poOppodppo uo wopx »oo opp: mopHoow uo »o: o>ppodypmooo pmoa opp pHoppoppoo op poop poop .aoopooopo mpop uo owpmpdo oo>o woo .odoyw oop uo pmoy oop woo »yyow opp? mopoommoo mo: pop: popop H \.ooo ooo poop op popop oop pop: wopoko .»mopoou woo wopoko mp .»owpym depo mopxpop oyok o3.po£3 mpp omd< mmmm \.wop30Hu woo yoopooom oprpH o mooom pp woo oyouop on»: uo wopx wo looom pH .»wyooo uo »ox yokoH o oo mmwmmfl \.poop pp poo poo ppo po Hoop p.oopooo H oo os mop Ixooppo p.omp pp om .pdo pp xyos.ooo o3 om .md oooopmpw op poo: H ow yo do» 0p omopo on 0p poo? H ow . »op HHo3 .xopop H oooB .do» op omOHo on op youoyo de03 H .mpop oxpp p.oow H mop»om oy.do» poop mop»om »m .»o: Hpod .303 no .ow H woo o8 mQOpm pp HHo uo pmypp .woooopmpw yo »ywoo yo uuo woxopp moppoou »HHooy E.H .»yyoh .»om 0p do» you .do» op poouuo wooypm o wop>oo »HHooy oyo mop»om B.H mmopSp opp .do» mopooopmpw E.H oyooz .omoo mpop op .»om 0p do» you 0m .mwopop opopyoo wpom do» mmooyom omdooop ooHpoym ydo» OmHo mpp poop .»HpopOn uo wopx .yo>oo Impw OmHo o3_woo soHpoyH »E wopyoom E.H 0mm ooo wow poo wow moo oom mow Now How oom mmm mmm pmm wmw mmm omw mmw mmo Hmo omo mmm mmm poo wmw moo :mm mmm moo poo ooo mpw wpm ppm mHm 267 No No No m< 0p do» poo: H .ooo» .mop»oo op wo>po>op »ppoyo m.oo omdooop pooopmo>op Hooomyoo o owoo oo Mopop H .»yyoh 0p woommoy 0p do» wopxmo op boo powpy ow 0p wopooa H poo: .pp uo . . . . . . . . . . . oop m.poop oopop H .mwopop ooom moppppoo uo .do» you omooopoo uo pyOm o ooooy wompoo poop uH .oo wopow mo: poo? »uppoowp 0p woo pp po Moop o oxop woo pp o0po op wooo o pHou H .poop Hoou H .oOppoypodyu 0p owwo poop opo Hoou H . . .yoopo oooo oo mdoou moppupoo oyos o3 pHou H woo »wyooo op popom ompo o owdoyop poo3.o3 poop oupHooy 0p oxpp omHo H \.00p 0p opo w.H \psoo mwoopmyowod oo Hoou do» on \.poop ooopopoooo p.ooo H \poo op ozow wopxpop oy.do» Hoou H .»oo .wop»oo woo wopQQOpo 0p o>ppooooy oyoo op wpdob »yyoh Hoou do» on \ppooyouupw »oo wopHoou do» oyo .mopopoEOm »om 0p wopow oyok do» .»mpoa m< mo oo mo oo oo o< mo \.oooyp o mopxop popoo mpop po opnopyouooo Hoou deok H\.ooo» \.woopmyowod p.owpw »yyoh oppoo poop oydm poo 8.H pdm \.»o3 poop wopHoou o.»wop»yo>o »oo woo onQOpm w.do» »oo opopmxo deoo do» .woopopopo do» woo »yyob onQOpm do» oops \o o o o o . . . HHos poop pp ooopopoxo H pp .ooop \.Hoop H ooo owpoow 0p oppos o oo ooop 0p wopom m.pH .ooppopop ooo oopoopopp opoo ooppppo poop o.H\.soop p.ooo ppHoo.H H ooo powpo "Hmpoo \.pooppooop ppo> o.poop popop H poo .oo oo ooo poo ooo opo opo ppo opo ppo opo mpo Npm ppm 0pm poo ooo poo ooo poo poo moo moo poo ooo opo ooo poo ooo .oo ooo mo Npo oo 268 oo mo oH¢ opo mo \.pdo poop pow op woo H .oopp poop po oo you pop o mo: poop omdoo .do» moppooooym oyok do» too uo pHdmoy o mo as you oo wopow no: poo: do» opp: wopyoom mo: H Hoou H \ o o o o o o o o o o o o o o pgmavdmhd oo op wo>Ho>op op w.o> do» opp? ooywo p.oopo H pH .pppopoppp pp opop op poz \Amoov .oo you oHoopyouooo »Hpooy m.pH . . . . . . . . . yowpmooo w.do» oowoa Iwoop ydo» Opop powdop H oooHoD .oyou loo myoow ooom mopooop oyok do» pHou H .yoopydu ow ooo o3 Hoou H . . . . . . 0p o>ppooooy »HHooy o8 ooooo poop .do» oopm op opoooo pooh »Hpooy do» woo ooopwoou uo wopo poop oEooHo3 w.do» mop»om oy.do» pyoomdm »HHooy ooo H pdp \OOp oE mop Iooopmpw mob pH .pmypu po mmoyoo wopooo oyop do» boo pyooodm p.ooo H A»yyoh OpV \.do» op od uo wopo »ppprpoooomoy oop uo pmoy oop oom H zoo woo mmoyoo wopooo m.oo boo 0p mo ooopwoou opo o>pw op do» you yoow o wopoooo opo oom H .»ooo .ooyu wopooU o.H oyook owopo m.poo9 .ow 0p do» oopH w.oo wopopooom m.poop wpom oo wopoom .3oo de03 do» Hoou do» poo yo yoopooz Epo opp: »prmmom oyoom 0p do» you pooppoosp op poops pp poppp H .po: poop mooyoo wopooo oo oom do» ooo: oo HHop no No pdooo wopopooom woooppooo do» "odomOH \popoooo . . . . . . . . oyooz oOppodppu oop 0p ooom o3 wpw ooo om oo opp: ooywo p.oopo oop .ooooop poop oo p.oopo sop \poowokwoon ydo» Opop »dp .oooo do» ow poo? ”an... omp mpp opp pHp opp mpp opp mpp me HHp oHp mop mop pop oop mop pep mop mop Hop oop mmo ooo pmo omo mmo ooo mmo mmo Hmo omo moo ooo poo ooo 269 No No No \.poop 0p oOppoooy o>ppomoo o woo do» om \ooop \ppp op oOppoooy ydo» mos poo: w< mo oo w< mu mo oo mo mo po opop poooH po Ppoop 2: pop pop oop uo pyoo o p.omp poop do» yoo \.poop odHo> p.oow do» poop wpom do» pmooH pd \.oHoooo yoopo pmoo »o wodHo> poo mp oo uo uHoo oopH mpp omdoooo wooooppo mopoo mp odHo> Hooomyoo »z .Aopoo< OpV poop 0p omOHo mpp .»ppydoomop wyok oop womd do» .wpom do» woo .poop oopH pow H »oo uo oOppmodo oop myozmoo mpop oopoz .H omdoooo pooSooo> »yo> oooooo ow H »oo oomooy oop uo pyoo m.poop mmodw H \.pydo poop 0m .pdo soyop deoom H oo uo uHoo poop oo wHOp pmdn do» oopo .oo Opop ooow »yo> mpdo poop mxmmmm \. . . . . mpoooodwyo HoopooomoHpoo ydo» Opop pom poo yoopoy w.H pdo Amowdopv »ooo m.pooB .oo 0p osow wopopop do» wopo p.oow H .oooyp uuo pow do» mpooo Iodmyo Hoopooomoppom opop mopppow pyopm do» ooo: pdo oopppopuow ompoooo oopo o powwghvpddhofi.......o.... . . . . . . . . . . . . . . . . opooa nodwyo Hoopooomoppoo uo 0p womooxo oooo o>.H poo: .uHom»o p.ooo H omdoo .ooo» \Jpoop .po ooooo. o poop o.pH \.pp opop poo poo poopop o.H pmpoooodwyo HoopoQOoOHpom .od .oo \. . . . . . . o o .090 .OQFHHAO aphubfiOOm .Hoopooomoppoo »Hoydo oyoh poop mpooo Iodwyo Opop mopppom pmdn oyoo oh.oooyp mpp oop oyouom .mpooaodmyo Hoopomoo tOHpoo »Hoydo mo omoop oom H .pdooo mop IoHop wopyopm do» .oowoswoop oop oo madn mmp amp mmp Nmp Hmp omp mpp oop pop oop map oop mop mop Hap oop mmp mmp pmp mmp mmp amp mmp mmp HMp omp amp mmp pmp mmp mmp Amp. 4 mmp mmp Hmp 270 mo \.o0pmmdompw HoopoHOm IOHpoo uo pOH o opop pow 0p oyoo Hduomd pp oopp p.omooo on poop .mppapp oop ooppo oOppodppm opupooom mpop op .uHomopo you mo mo mo mo m0 m0 m0 mo N< o yo omHounodyp o uo oopooo ydo» woo do» woo mmoHo oobpm »oo op pmop o ooop 0p woo do» uH .oOppmodu o do» omo H ooo "mmom \.oyoo3»oo pow yo>oo deoa o3 yo . . . oooHo oop poo mp mpop poop pmdn mpp .modHo> ydo» mopoooppo mp odom0h oopH poo mpH poooo \.modmmp HoopooomOHpom . . . . . mmoHo mpop uo pxopooo oop op pmdw pdo moOppmodd omoop po oOOH H do» oydmmo H .ow H .moOppmodu omoop ad sooo woo oooHooEOm mwook oop op ppm p.ooo H poop pom p.ooo H pmmmmmm \pmmoHo mpop op pmop pp ppopH oop opo \.mp oo »oo oop poop .uHom nopo you moppom mp oo ooop .»ooo "Hmyoh ooo» "mpoowdpm \.uHomopo you pmdn wpom oo .poop »om p.owpw oo pdm pooop \.mmoHopyo> »HHopop oy.»oop .oooyp oop uuo md pom »oop .opmos o oy.»oop .mmoHopyo3 oy.»oop »om op oOmyoo o you 0m wo< .oo uo pyoo m.poop woo oo 0p poopyooop »yo> oyo omooe .puoH oop 0p woo pompy oop op .oyoop ozow .oyoo no mwopop .modHo> .»ooo »oQOmOHpon oop .ooyo poop .ooom oop mpp .mpoppop oop p.omp pH "Hmpop \pow do» mo muoppoo ooom oop o>oo p.omoow »popoOm op Hodwp>pwop yoopooo poop oop poopop pp opooom poo poo ”opooo \.mpydo poop .Om woo oo oop mwp wmp pmp oop mwp oop mwp mop Hmp oop mpp mpp ppp mpp mpp opp mpp mpp pr Opp mop mop pop oop mwp oop mop mop Hep omp mmp mmp pmp mmp 271 mo m< No w< mo mo pogo pmop m.poop .soop p.ooo H .poooa o3 poop mo ooop 0p do» you wowoopop »woo |»oo poo:_m.poop oopop p.oow H .mooyb mo: wpom do» poo3.poop mo .wpom do» poop oyoo p.oopo o: poop mo .pp poop oop oop mp poop oopop H woo »HHooomyoo do» oooppo op wop»yp mo: »woo»oo oopop p.oow H .oou yo .»pppm mpp yoopook yo do» mpp yoopooz mmopo mpop op »om 0p moo »woo»oo poo: op wopmoyopop E.H .»om do» poo: op wo upmoyopop o.H .oto ydo oooo o>.o3 omdoooo mydo» omoooo woo »yp op wopow poo o.H woo oOpopoo HooOmyoo »a omoooo op woppyp do» poo: p.oow H .ow ooo do» HHo m.poop oOpopoo ydo» wopopm o>.do» oooo woo .oOpo Ipoo ydo» wopopm do» pdo .mob H .oOpopHo ydo» op wopmoyopop p.omo3 H poop p.omo3 pH .woow »oo »woo»oo wopow p.omo3 pp woo poonodm oop uuo wopppow »HHooy oyo3.o> poop ppop H .pmop mpp .ppo po poop p.omoo pH \.»HHooomyom do» wyo30p »ppprmoo woo odoyw oop pHou do» ppou H pdo »HHooomyoo do» wyokop »ppprmoo poo mpH .oprmoo ppo> ppop H pppopppop .opopop .oz ”ooom \pzoo pmop. .ooo3 ”a \.oprmoo »yo> »yo> ppou H .wopopop mo>.oo ooo: pHou H zoo opp ppop op ooo: oop poop pmop H "ooom \.wopop»oo »om yo>oo 50hP5Q............vaxppO~n .»om 0p poo: do» ow pooh .oooh mHmwmm \.»ommo oop ooop w.H moooo pmoE op oopp ooppoppoo ooo H pp .ooop powwow x.»ommo oop ooop »Hmd0p>oo 0p oopH wpdob do» .»ommo oo yo o0pooo onppHda mmm omm mmm mmm pmm omm mpm mpm ppm opm mpm opm mpm mpm ppm opm mom mom pom oom mom oom mom mom pom oom mmp mmp pmp omp mmp amp mmp mop pmp 272 opo \Awoov .wopmooo yoo puoH woo, do» 0p oooo mdoou oop oOOp do» oopp Hoou, H woo oyooooo HooOmyom o wommoyoxo pmdfi, oom .oopp moOH o you poop oo moppppm ooooU m.oom do» wHop pmdm oom .ospp woOH o youm op wopopoEOm wopwpoo oooo moo ooom oopHM Hoou H .mopopoEOm »om H ooo Ampodyyopopvm o M “ 00 mo mo mo oo deoom pp oopop H woo mpop wpom poo o>oo »oop oopop H pdo .moooo omHo »woo nppo>o popop H poo: mp mpop .mp ooos H poo: .oooyp oop uuo wopppom .»ooo ”ooom \.oydm mNMMMW \pdo» 0p mopop oyoo ooo pmdfi »om H ooo "oooh \. O O . . wopomppoooooo p.omo3 H poop upom»o you ppou H .ompHQSOooo op wopoop o3 poo: wopomppoooooo poo .poonodm oop uuo wopp Ipom .oooyp oop uuo wopppow mayop wopoH on oop womd o>oo opooom uo poH o poop moppmopopop mp poop popop H .ooppmooo mpop omo 0p pow »ppooy o>.H mummmm \.oopop p.oow H HooOmyoo wopopoo mpH .»o: oEOm op do» mmpoo poop omoo H .wop>os oo deoom o3 powdoop o3 oOppooypw oop op wop>oo p.ooyo3 o3 ppou o3 pmdn m.pH .»HHooOmyom do» oooppo op p.omo3 pH .»owOp wopopooom omppmooo loo 0p wopoos o3 .oooyp oop uuo oyo? o3 poop mo: pH .oOpopoo ydo» 0p oopmpH 0p poo: p.oopo o: poop p.omo> pp .opom poop I»yo>o .»om wpw o3 pdo mo: omooydm ydo» poo: Booo p.oow H .Op wopooz do» yoopooz yo wopoopmpp p.ooyo3 do» up booo p.oow H .opowo oopp woo oopp wpom »woo»yo>o popop H \.oopp mpop ppo moppoop ooop o>.H oom mmm mmm pmm omm mmm :mm mmm mmm pmm omm mom mom pom oom mom oom mom mom pom oom mmm mmm pmm omm mmm omm mmm mmm pmm omm mmm mmm pmm omm 273 mo mo mo mu mo mo opo do» .wpom H poo: 0p woommoy p.owpw do» oo 0p pdo .wpom »HHooOmyoo H poo: do» HHop 0p wop»yp mo: H .owo opdopo o wpom oom mo pmdw \.po» oopooyo oop wo>pomoy o>.os popop p.ooo H pop pp poooo mopppop »o oyoosoSOm mopppow oy.o3 oopop H ”ooom \.ooop mpmmmm \pwpom do» poo: poop mp .oyoop pom 0p woppyopm mpp oopop do» on ”ooom \poyoop pom 0p . . . . . . . mpp popop oop on ppmmmm \.wo>Homoy mpp oopop p.oow H pdo oyoop wopppow mpp oopop H "ooom \Awopoypmdyuv pmop uppop mpop ppo omoopop ooppomoy mopppom mp ooppooppm oop popop oop on ppmpmm \.oHoooo wopmduooo pmdw mo: pp .wopmpoo poo mo: poop woo yoopo oooo mpoo 0p o>oo 0p wopow HHo oy.o3 .»om 0p mopppp ooo: o: poo: mp poop popop H oo< .mpoom omoop ooooy 0p »woo»yo>o you moOp ImmoooOo opus 0p o>oo 0p wopow HHo oy.o3 up woo mmopo mpop you pom mpoow o>oo o3 omdoo . . . . . . ydoo oo uHoo yo ydoo oo wooom 0p oo you ompuHom oo de03 pH .odoyw mpop op ooo»yo>o you omp IuHom mp poop omdoo oOmyom ooo pmdn you ydoo oo wooom p.ooo o3 .oOppooypw ooom oop op o>oo 0p o>oo HHo o3 woo odoym o mp mpop omdoo mmopo mpop uo oOppooypw oop owpoow pooooo ooopo do» .oooyp oop uuo wopppow »o poooo o3 poo: mp poop oooo H .ooopo oOpopoo ydo» you poo pmdn mp mmoHo mpop .wopop»oo yo »prOpowoyow mpop oooa p.ooo H ooo .mmopo mpop .poop mp opom mmm pmm mmm mmm pmm 0mm mmm mmm pmm mmm mmm 4mm mmm mmm pmm 0mm mpm opm ppm opm mpm_ opm mpm Npm pr 0pm mom mom pom mom oom mom mom pmm 274 mo mo mo m< mo mo mo oo opm m< oo moon .oooowoou do» oo>pm HHo o>.o> oooo H .wpom »woo»yo>o poop wopop»yo>o .oz "ooom \pmopopooom wpom do» poop .pooz mNMMMW \poooo pp moow poo: .oooo H pooa m.poop .mopop»oo oooo pp moow yo ooopm poo mopoopp pmop 2» o2 pooppoopopop mpop HHo opp: ow op wopom do» oyo pooB \.3ooo 0p wopoos H poo: m.poop .»ooo ”ooom \.yooooo yoppoo o op moppooyopop pyopm HH.H woo oyoo od yoopom lop pp pdo HH.H »HHduomom .woppoooy oy .»oop 30o woo upom»o pdooo oyoo oom 0p oo mopoo pp woo odyp mo3_wpom »oop poo: omdoo .wpom myoopo oop poo? wopopooyomo H oopH .wpom do» poo: wopopooymmo H 0m mopop ooom oop wpom woo onoom yoopo HHok .pHou H omdoooo poop mo: woommoy p.owpw H »oo oOmooy ooo mmodm H \.ozow oo wopppdm mo pp opop fooo H ooo soap H .222 H "a \.mHoo 0p mop»yp oy.o3 .oaow do» mopppdo poo oy.o3_pdm “ooom \.oo oo mpoooooo oop mopmdoou uo wopo mp oyoo »woo»yo>o woo onooo uo pOH o oyo oyoop omdoooo op wooomoy op ooppooppm ppooppppo o mp mpop ”a \.pdo pp pdooo oopop ooo do» .pomm oop oo do» pdm op poo: p.oow H "ooom \podoyw OpV H .»ooo .do» oooop mmodm H .oox oo mopomopm you moooop .»ooo mNMMMW \.oooo H poo.3 m.pooe ppp ooop do» ow oom pHoou »HHooOmyoo do» ow poo: oooo H \.wpom moo »woo»oo wopop»oo pdooo oopop do» poo: woopmyowod »Hpooy p.oow H .oowp oo o>oo »Hpooy H woo .wpom H poo: pdooo pHou do» poo3 oo HHop p.owpw omm mmm mmm pmm omm mmm pmm mmm mmm pmm omm mpm mpm ppm opm mpm opm mpm mpm ppm opm mom mom pom oom mom oom mom mom pom oom , mmm mmm pmm mmm 275 mo mo mo mo mo ‘Ixo op oOppopyommpw o oo om do» Hoou H poop mp do» wooo0pm H »oo oomooy »Hoo ooB \ppmppp poop oo oop ooo \popo: o ppppooop do» ooo pmopHoou do» oyo 30m Ampmdyyopopv \.ompp poop moppop do» 0p pooopmo>op HooOmyom »8 mp mpop .poop 0p oOppoooy »8 mp mpop .oo wopo Io0pm op ompy o oOOp do» up .»oo .»om op mpyopm pooe .»yyow pp momppooomyoo poop \pooow pdooo wopopop oy.do» poo: poop mH \.opoo ooym 0p oyoo onoom obpm deo> mwyoo wopHoou uo wopo omoop oopop H pdo» opp: oooowoou woyoom »HHoopupooom ooo opoooo oop 0p yomOHo Hoou do» ow .mwopop oEOm oo oyo: . . . . . . . do» ow .mpop uo pHdmoy o mo odoyw oop ooyu poopmpw Hoou do» ow .mdoym oop 0p yomOHo Hoou do» ow .o>pmoouow Hoou do» on ppdo mwopop oEOm soyop H ooo oo oo oo mo mo wo mo mo m0 m< \.pompp ppo .oo \.poop op oooopo pom pp.H popop H \.mp pp opp; oo ppp: H mpop pmppp oop pppopopo .ppoz \.oydm "Hmyoh \.pp opop Iooymoo do» poop mp kooo o3 HH< "odomow \.oooH umpoowdpw \p»o3 poop mHoou omHo »woo I»oo up oooo p.oow H \.pp pdooo Hoou do» poo: oE mopHHop »Hdooy poo oy.do» ppp pdooo Hoou do» poo: kooo 0p poo: H oooo H pyoopo oop pdo woo yoo ooo op wopow pmdn pp mp yo do» 0p wopopooom oooo pp mom oom mom mom pom oom mmm mmm pmm omm mmm omm mmm mmm pmm omm mom mom pom oom mom oom mom mom ppm oom mmm mmm pmm omm mmm omm mmm mmm pmm 276 mo mo :0 mo opo \.poop pp oomppo Iydm oy.do» poop moppoou oop pow pmoopo H \ppmpoo opop poop opm \A»ppoomv .woooo ow .ooo» \A»yyom OpV pdo» wpdoo \.pmypu mwyoo oop »uppoowp 0p do» you wooo o o>oo H »o: m.poop 0m yoppyoo do» oppz mopyoom moo ooo poo: m.poop poop H ooo popoo pooxo oop op wopppow ooop yoopoy wopopopm op mm oo oo mo mo m< N< mo mo mo mo oo op mo mo Nd \.owpmpdo oodo 0m poo .yom0po .woow poou oo moooo poop «Hmmmm \ppoou do» oooo poop moow sou «Nmmmm \.os pdooo oyoo ow do» poop omppooy H poop mp wopop pmowwpo oop »pooooym pmdn mmodm H «HMMMW \.ooop mpmpmm \Aupom 0pv.yom0po poou do» moooo poo: .»ooo mummmm \pyom0po poou do» moooo pooz mmwmmfi \poo yo ooo» ooop oyoo wopopoaom oopp do» wpdoz \.ooo» mNMMMN \po>pmoouow poou p.oow do» ooopm yom0po poou do» om "Hmmmm \.oz mNMMMN \pzoo powpy o>pmoouow poou do» ow .»ooo \Awompymydmv pH wpdoo \pospp o po ooo do» woomo oooooom up moOppmodm omoop yoSmoo do» wpdoo mNmmmM \.mwy03 moppoou oyo omoop poop poou paCOUH\ooooooooooooo 303 .poou H .poo »popdpomoo .oz \pmd ooyu woooopmpw poou do» on \Aupom OpV ppoou H ow tom \ppoou do» ow tom 3517.21 ooop mmm mmm pmm omm mmm omm mmm mmm pmm omm mmm mmm pmm omm mmm omm mmm mmm pmm omm mpm mpm ppm opm mpm opm mpm mpm ppm opm mom mom pom oom 277 opo mo opo mo \.do» opp3 poop oyoom op wopoos H woo . . . . . . . oyo: popooov do» ppou H \.ow 0p poo: do» you boo woo ooow ooo .oyooa woo 30o opoppppoop op oop poo: H pom .popopoo op powpo oo you mopoooooo m.poo3 .oo wopom m.poo> woopmyowod do» opoo ooo H »oo oop woo A»yyowv mdoou yOwoo oop oy.do» poop woppoou E.H \.pduopoo ooop mopmduooo oyoo oo pompo poop »popmmoo woOpyo>o oOppoa Iyouop oo oo 0p mopow m.oyoop woo o>oo do» poop mopyooop ooom woo mopooomOHpoo oEOm wopwwo »o ooda om moOppmodd omoop odo ooo do» .moOppmodm omo ooo do» poop poou H .oOppo>yomoo oo oooo H ooo Ampodyyopopv \.oyooooo oEOm soom 0p owdooo oyoo powpo »woooEOm poop wompym Iydm oy.do» poop woppoou oop pow pmooo H mo 00 oo mo mpo pdooo yo opo pdooo wopop»oo wopaooo poo pp oo opop o pom opp opop opooo H up wopoom mo: H woo \m»o> ypou o ooop ow pmdo aopooyo oop om pp pdooo powdoop woo oo owdoop oo>o pp pdooo wopop»oo ooow poo woo »ppooyommo oo oopooym mpop woo oo poop wompomoooy oo uH \.>ooo o» .oopooyo o uo wopo oEOm 0p opopoy wpdob poop Amowdop mmopov wooy yo>o o>.H ooop »w0pooo»mo »oo ooyu opoooxo uo wopo »oo o>pw pmdw 0p mop»yp mo: H kooo p.owpw oo wpom oo woo opo woomo H ooo: .ppo3 ”opoo< Amopdopo ooyop pdooo r mpop pdooo momo opood .soopmoiupom uo mwoppoou .pmoo mpo .wdoyo mp »yyow »o3 pdooo opop »oop .Epo wopoouuo pp 30o .mmooo>ppoHSOo .»ppaou m.»yyoh uo oOpmmdompw o Opop mpom odoyov . . . . . ooop p.ooo H .mooo pH opop: pm mmop omop mm0p mmop pmop omop mmop moop pmop omopf moop pmop moop moop pmop omop mpop mpop ppop opop mpop opop mpop mpop ppop opop moop moop poop ooop moop ooop moop moop poop 278 mo no mo Iwwo ooow woo .Ho>oH poopomomOHpom o oo Toow 0p wooo pdo» poop wok onQOpm pmopm H omo3 »HHopppop mopop ooop pow do» oooowoop o>ppomoo oop po ooom .ompo ooooooo mo oooomxo oop po poo pdp pdoop pooomooo ado» oodo »ho> mopodHo> ooo wpooo H .meHw oop op omHo ooo»oo no MHooooo mo omoomxo oop po mos pp pp pooop poop oopo> p.oopo ooo poop pdn pdmop ydo» wodpo> oom poop mop»om ooo Boo H .poo oop on wpdos pp Hdm Impoo wooo oop .»HHopOp Adepw o mo mooon o>pppmom wooo woo mopwoopmoowod ooow 0p wopaoo »po>ppooHHoo woo mwooo podwp> Ipwop mop»Mmppom opoppHpoop woo .mdoow o mo »po>ppooHHoo omppmooooo 0p ooop o pow o>.o> .»HHodwp>pwop ompHmooooo 0p ooop o o>oo o: poop oo3_poop woo Opop woodp pppoop p opom wooo poop wopopoaom \.wopop mo ppom poop woo mo wdoom on 0p woo do» poo: woo oppozlppom woo mpompdo» wo I3op> do» zoo pdooo wopopop do» mo: »ooow mo pom Hdpwopoomo mo: poop wopopooom \Awoz 0pv.woooo 00 \.oodm oHo opo oHo .womHoo »HHoop pH mmmm \.oodo pom mmmmmfl \.do» pdooo wooo ppp prppp o woopmomwod mo wompoo pp mo opp: oopop H .pooo »Hpooo mos poop powdoop H . . . . . mpoopom pdo» pdooo »Hpopoommo .do» wooomdppop poop boo woo oppp pdo» op mwopop Hoooooom oooom 0p wooo poop mpH . . . . . . md opp: wopoom do» poop ooow pppoop mo: pp pooooop p puma \.oo op oopop ppoopoopo o o.poop ooo pooo ooop moop ooop poop ooop moop ooop moop moop poop ooop mmop mmop pmop omop mmop omop mmop moop pmop omop moop moop poop opop moop poop m30p moop poop ooop “mmop omop pmop omop 279 mo \.oo 0p wopooz »oop ooopmoEOm p.omo3 poop oapp opop po pop poop oopoo oopo> p.oopo poop poop poz .ooopo opop op .ospp mpop op mopom oom opopAQOMQQo poo mo: .mpop woomoo oo po oo oo no mo mo op< mo \.oopoo p mmmmmm \oCOQ>OE@-D.0~H\ooooooooo.- . . o o oOpppoppow oop pdooo mopwoommoo Opop oo om ooop woo pmopm poop do» too oopopo oop pp pp oppp opoos p ”ooomoo \.wodopo oopp poopp poop wopoposoo ooomps p popop p.mpmmmm \.pp oppos op opop .ooop " \Amowdop mmoHov poppos 00.00pe odo» pom do» oopop do» on ”ooom \.»om0moppom oop mo oop mo »ppwoo 0p poomppo oppoppow o wooo ppp: H pdo »Hpoopppoomm wooo odoow m.mdopw oop ooo poo: zooo 0p oopH deoz Hprm H .ooo mHoom o.odoow oop po>Mpoo3 ow pp.m3 .pmdw pH.o3 »omomOHpom Opop pom ppoos o3 ouppoop H oOpmmdompw opop no omdoooo MGHgfihmflpogooooooooooooo . . . . 08 mo poom o »ppooo woo woNpHoo loopop omoop pom ooo H pH \.mwoppmop »E opp: Hoow poop moOppmodd mo pmpp o .pdo ppopopooesp pp poopop p .pmpp poop oa wop>pm odo» wopopooommo H \.oo0ppomdd mopHomp mo mom»p omoop wopookmoo .moooo omoop op modo wooo mop>pm .woomooo ppp: H pdo oOppmodw poopooo opop ppo ow p.o03 H mo mo oOppmodv o momo »wooMSOm Gmnkfooooooooooooooama op wowoooo mo oop mo .Op »op HH.H .»ooo .»Hwopwpoooo o0p>oooo »E »ppwoo ppp? H oo poop woopmoowod ooo H .»ooo mwmmmm mopp popp mopp mopp popp oopp mmop ooop poop ooop moop ooop mQOp moop poop ooop ooop mwop poop noop moop :wop moop moop poOp ooop mpop opop ppop opop mpOp opop mpOH mpop ppop 280 oH< no odd mo Iodopmooo oodo mpop o>oo 0p wopppoxo m.pH .woom.HmmmMM moppoom S.H \.mopwoopmoowod oo Op @800 »po>ppodopoooo ooo m3 ooooh p0 ooozo o.H »o3 »Hoo oop m.poop pdo moooooo »opmoop o o.poop mooppooom woo owdooop poop ooo: »Hpoopwoopoo 0p HHpoo oop woo ooooppoo oop wopboo mp mopwoopmoowod poop opoppppoop op poo ppoo oop oo< .oopooopo Ioowod wooo op whoop pmoop po pdo .oowoo o3 wooo: wo>Homoo o>oo poo powpo m3 mo mapop op popppooop mp poppp p opom pmpoo wopopooom \.mwopop owdooop wopopos p0 mapop op wooo op oooommoopo o>pppmom »H IHooo o Hoop pmdn H .pooppo 0p wopwoommoo oooo o>.do» oopp Hoop H .womooomopom do» oopp Hoop H .mopoOppmodw »o0poo0Hon opop wopppom ooo opmoom oopH poop H .mp0 poopo ooom pdo o>oo opoooo oopp Hoop p.oow H .oOppoooo>ooo odo p0 mompoodwloooop pmop oop oom woopopopoo ooo odoom oop popop p .oopoopoop oop po opopom oooop odOpooo Hoop ooom p0 owpmpdo .omOHompw Impoo oHooom oEOo ooom o>.H .mpop owdooop wooooa do» »oo oop .oooo oo poo: poo: pdooo woppoxo woo woow moppoom »Hpooo o.H wepoom omoHo wopop o oom .oo opp: oo mopom m.poop wopHoop o opooo 0p pow o>.H \ppoop wo>H0m loo do» o>oo .ooom o>oo o3 ooooz .ooo mwopop wooo: wopkooo poo woo oodmOHoIooo oSOm owo oppoa o wommoooxm do» .»mpon mo \.wopwoopmoowod wooo pdo .wo>Homoo poo mo»oo .wookop »o: woOH o ooom o>.o3 .30o po E.H popog oop 0p ooop po mo: H popom oop ooom "Hmmmm oopp mmHH ompp pmpp ompp mmpp pmpp MMHH mmpp pmpp ompp ompp ompp pmpp ompp mmpp :mpp mNHH mmpp pmpp ompp mppp oppp pHHH oppp mppp pppp mppp mppp pppp oppp mopp oopp popp mopp 281 mo \.pdooo mp moopo oop poo: o.poop .popppppm poo ooo popppppo mo: poop opo opoooo poo: .ooooo looo »Hpooxo poo: .oo poo: poo: HoooH 0p opoo mo 0p opmooooo ooooH 0p wo< .pdooo ppo op.o3 poo: mp popom poop ooooooo popo pp opopop pp opp: pooo pppspppppm wooo 0p .opowo momppo po>o pp pp woo wpw o: poo: oosopop ppo: pppopppppm op . . . \.w0ppom wopop o oom mmoom>pp mu oo oo o< \.oooop o oom »woop op.o3 oopop H "Hmpon \Amowdop mmopov .pdooo ppo op mmopo mpop poo: mo mopwoopmoowod ooppoo o o>oo H «Nmmmm \oooooooooooMn—Hmmgwqwrhmh pdooo wooo woopooowod H oopop H mama \.00p »ooohlmw oppppmom wooo pOH o Hoop H oopop H "oom meH pmpp ompp mmppo :mHH mmpp mmpp pmpp ompp oppp mopp pppp wppp mppp :npp mzpp «ppp pppp REFERENCES 282 REFERENCES Argyris, C. Intervention theory and method. Reading, Mass.: Addison- Wesley, T971. Back, K. W. Beyond words: The story of sensitivity training and the encounter movement. New York: Russell Sage, 1972. Banet, A. 6., Jr. Therapeutic interventions and the perception of process. In J. W. Pfeiffer & J. E. Jones (Eds.), The 1974 annual hagdbodk for groupfacilitators. LaJolla, Ca.: University Associates Publishers, Inc., 1974. Berzon, B., & Solomon, L. The self-directed therapeutic group: Three studies. Journal of CounselingPsycholggy, 1966, 12(4), 491-497. Bolman, L. Some effects of trainers on their T-group. Journal of Applied Behavioral Science, 1971, Z, 309-325. Bradford, L. J., Gibb, J. R., & Benne, K. D. (Bds.). T-group theory and laboratory method: Innovation in re-education. New YOrk: John Wiley, 1964. Campbell, J. D., & Dunnette, M. D. Effectiveness of T-group experiences in managerial training and development. Psychological Bulletin, 1968, ZQfiZ), 73-104. Cohen, A,‘M., & Smith, R. D. The critical incident approach to leader- ship intervention in groups. In W. G. Dyer (Bd.), Modern theory and methods in group training. new Ybrk: van Nostrand Reinhold, 1972. Cohen, A. M., & Smith, R. D. The critical incident in growth group : Theory and technique. LaJolla, Ca.: University Associates Inc., 1976. Cooper, C. L. The group leader and training effectiveness. In M} L. Berger & P. J. Berger (Eds.), Group trainingtechniques. NEW’YOrk: John Wiley & Sons, 1972. Culbert, S. A. Trainer self-disclosure and member growth in two T-groups. Journal of Applied Behavioral Science, 1968, 2, 47-74. Culbert, S. A. Accelerating laboratory learning through a phase progres- sion model for trainer interventions. Jburnal of Applied Behavioral Science, 1971, 6(1), 27—37. 283 Dies, R. R. Group leader self-disclosure scale. In J. W. Pfeiffer & J. E. Jones CBds.), The 1977 annual handbook for group facilitators. LaJolla, Ca.: University Associates Publishers, Inc., 1977. Dies, R. R., & Cohen, L. Content considerations in group therapist self- disclosure. International Journal of Group Psychotherapy, 1976, 26, 71-88. Duncan, M} J., & Biddle, B. J. The study of teachiqg. New Ybrk: Holt, Rinehart and Winston, Inc., 1974. Fiebert, M. 8. Sensitivity training: An analysis of trainer inter- ventions and group process. Psychological Report, 1968, ZR: 829-838. Flanagan, J. C. The critical incident technique. Psychological Bulletin, 1954, 21f4), 327-358. Gordon, W. P., Jr. On becoming a trainer. In W. G. Dyer (Bd.), Modern theory and methods in gpoup training. New York: Van Nostrand Reinhold, 1972. Golembiewski, R. T., & Blumberg, A“ (Eds.). Sensitivity trainingand the laboratory approach: Readings about concepts and applications. Itasca, 111.: P. E. Peacock, 1970. Golembiewski, R. T., & Blumberg, A. Learning and change in gppups. Baltimore, Md.: Penguin Bodks Ltd., 1976. Harrison, R. Problems in the design and interpretation of research on human relations training. In R. T. Golembiewski and A. Blumberg, CEds.), Sensitivity training and the laboratoryppproach: Readingp about concepts and applications. Itasca, Ill.: P. E. Peacock, 1970. Hartley, D., Roback, H. B., & Abramowitz, S. Deterioration effects in encounter groups. American Psychologist, 1976, Mar., 247-255. Issacson, R. M} Development of the interpersonal skills interaction analysis: An interaction analysis technique to measure interpersonal communication skills in small groups (Doctoral dissertation, Michigan State University, 1976). Dissertation Abstracts International, 1976, 37, 5702A. Kagan, N., & Danish, S. Emotional stimulation in counseling and psycho- therapy. Psychotherapy: Theory, research and practice, 1969, 9(4), 261-263. Kagan, N., Krathwohl, D. R., & ward, G. R. An attempt to measure and facilitate counselor effectiveness. Counseling Education and Effec- tiveness, 1972, Mar., 179-186. < Kagan, N3, & Schauble, P. G. Affect stimulation in interpersonal process recall. Journal of Counseling Psycholpgy, 1969, lpfl4), 309-313. 284 Kaplan, H. I., & Sadock, B. J. (Eds.). Sensitivipythrough encounter and marathon: Modern group books, vo1. 4. New York: J. Aronson, 1975. Lakin, M} Interpersonal encounter: Theory and practice in sensitivipy training. New York: McGraw Hill, 1972. Lieberman, H; A. Change induction in small groups. Annual Review of Psychology, 1976, 31, 217-250. Lieberman, M3 A., Yalom, I. D., & Miles, M3 B. Encounterpgppups: First facts. New York: Basic Books, 1973. Lippitt, G. Selected guidelines for laboratory group training. In W. G. Dyer (Ed.), Modern theory and methods in group training. New York: Van Nostrand Reinhold, 1972. Lundgren, D. C. Trainer style and patterns of group development. Journal of Applied Behavioral Science, 1971, 1, 689-709. Mann, R. D. The development of the member-trainer relationship in self-analytic groups. Human Relations, 1966, 12, 85-115. McLaughlin, P. E., White, B., & Byfield, B. MOdes of interpersonal feedback and leadership structure in six small groups. Nursing Research, 1974, 32, 307-318. O'Day, R. Training style: A content-analytic assessment. Human Relations, 1971’.Z§(5)’ 599-637. Psathas, G., & Hardert, R. Trainer interventions and normative patterns in the T-group. Journal of Applied Behavioral Science, 1966, 3, 149-170. Rogers, C. R. Carl Rogers on encounter groups. New Ybrk: Harper & Row, 1970. Shulman, L. 8., Loupe, M: J., & Piper, R. M. Studies of the inqpiry process. E. Lansing, Mich.: Educational Publication Services, College of Education,‘Michigan State University, 1968. Smith, B. 0., & Meux, ML 0. A study of the logicof teaching. ‘Urbana, 111.: University of Illinois Press, 1962. Stock, D., Bradford, L. P., & Horowitz, M3 How to diagnose group problems. In R. T. Golembiewski 6: A. Blumberg (Eds.), Sensitivity trainipgand thpplabogppory approach: Readin a about concppts and ppplications. Itasca, 111.: F. E. Peacock, 1 70. Tannenbaum, R., weschler, I. R., &‘Massarik, F. The role of the trainer. In R. T. Golembiewski 6: A. Blumberg (Eds.), Sensitivity training and the laboratory approach: Readings about concepts and appli- cations. Itasca, Ill.: F. E. Peacock, 1970. 285 Vincino, P., Crusall, J., Bass, B., Deci, E., & Landi, D. The impact of process: Self-administered exercises for personal and inter- personal development. Journal of Applied Behavioral Science, 1973, 2, 737-756.