THE RELATIONSHIP OF COUNSELOR‘BEHAVIOR T0 CLIENT RESPONSE AND AN ANALYSIS OF A MEDICAL I INTERVIEW TRAINING PROCEDURE INVOLVING SIMULAFED PATIENTS Thesis for the Degree of Ph. D. ~ MICHIGAN STATE UNIVERSITY ARTHUR RESNIKOFF 1958 mun» LIBRARY Michigan State University IIIII III/IIIIIIIIIIIII IISIIII/I ‘1'“ m‘wT-W— —— This is to certify that the thesis entitled THE RELATIONSHIP CE“ COUNSELOR BEHAVIOR TO CLI‘CNT REPONSE AND AN HIALYSIS OF A MEDICAL INTERVIEW TRAINING PRUJEDURE INVOLVING SIMULATED PATIENTS presented by Arthur Resnikoff has been accepted towards fulfillment of the requirements for 1Dh.D degree in Education ’74 // Date AugnaiLfik ABSTRACT THE RELATIONSHIP OF COUNSELOR BEHAVIOR TO CLIENT RESPONSE AND AN ANALYSIS OF A MEDICAL INTERVIEW TRAINING PROCEDURE INVOLVING SIMULATED PATIENTS By Arthur Resnikoff This investigation had a threefold purpose: (1) to study differential client behavior and client perception of his counselor as a result of the counselor's maintain- ing either "high" or "low" levels of therapeutic behavior as rated on the Counselor Verbal Response Scale (Griffin, 1966; Goldberg, 1967; Kagan, e£_al., 1967); (2) to measure increases in the level of therapeutic behavior of second year medical students who experienced special interview training; and (3) to further validate the dimensions of the Counselor VRS by correlating scores obtained on it with a test used in medical school admission, a measure of affective sensitivity, and peer and staff ratings of a studentis effectiveness. While there are a large number of studies dealing with the counseling "process," there is no conclusive research which relates client response to the within inter— view behavior of the counselor. Previous studies have centered either upon pertinent therapist variables or upon Arthur Resnikoff client personality characteristics and their relationship to outcome. Viewing the counseling situation in such a light ignores the two individuals as an interactional sys- tem. The convergence phenomenon has been viewed as a possible explanation for the interaction of the counselor and client by postulating that the client will move closer to the counselor in values and behaviors over the course of treatment. This research was similar to that done in the area of convergence in that it assumed that there would be movement by the client toward the therapist in broad areas of verbal behavior which are delineated by the Counselor VRS. It was assumed that if the counselor "moved in" and commented upon the client's affect, the client would become more affective; if the counselor communicated to the client his understanding of the client's comments, the client would come to emit responses which communicated understanding of himself; if the counselor focused on what appeared to be the client's problem, the client would move with the counselor and would also become more specific and concrete; and finally, if the counselor verbalized the kinds of statements that should encourage client self- exploration, the client would then make self-exploratory statements. This reasoning led to the hypotheses that were tested. Arthur Resnikoff In order to test the hypotheses, a client response scale was developed to correspond to the categories of the Counselor VRS. Reliability estimates were computed and found to be high. Previous research has shown the use of the "coached client" and stimulated videotape feedback to have been effective in the training of counselors. The present research combines both of these techniques. Moreover, to date, there has not been an adequate effort to relate a process measure of effective counseling with such measures as peer and staff ratings, a test of affective sensitivity, and a measure of academic achievement. The investigation was divided into two parts. In the first study, a group of 30 Masters level counselors conducted counseling sessions with high school age clients. The sessions were audiotaped and then rated along the dimensions of the Counselor VHS and the Client VRS by different sets of professional Judges. This procedure was followed in order to examine the interrelationships of the rated categories and to evaluate client movement, one of the dimensions rated by the Judges for each client, as a result of counselor effectiveness. An analysis was performed to specify which categories of the Client VRS would best predict the category of client movement. In addition, each counselor's client filled out the Wisconsin Relationship Orientation Scale to evaluate the strength Arthur Resnikoff of relationship (as perceived by the client) as related to the counselor's level of effectiveness. In the second section of this investigation, an attempt was made to demonstrate the effects of a technique to instruct medical students in interviewing skills and to further validate the Counselor VRS. Treatment consis- ted of each medical student interviewing four different patients over a four week period. The "patients" were simulating different types of individuals with a variety of problems. Following each interview, other students and staff would conduct a stimulated recall of the pre- vious session by use of videotape playback. Analyses were made of the changes interview behavior from pre to post treatment. Peer and staff ratings, a test used for admis- sion to medical school (MCAT Science) and the Affective Sensitivity Scale were administered. The relationships among these measures were then analyzed in support of the validity of the Counselor VRS. The first set of hypotheses postulated in this study dealt with differences between rated responses on the Client VRS of two groups of clients. The clients were divided into two groups on the basis of whether their counselors maintained either high or low levels of the matching variables on the Counselor VRS; the composition of the two groups of clients for the testing of each hypo- thesis was practically identical, due to the high Arthur Resnikoff intercorrelations of the Client VRS. The two groups of clients had also rated their counselors on the WROS. The group having the counselors classified as being more effective were more favorably disposed toward their coun- selors than were the clients meeting with the less effec- tive counselors. No differences were found on the Client VRS categories. An intercorrelation matrix was calculated among all the variables of both rating scales and the WROS. The Counselor VHS and Client VRS each correlated highly within itself, but had low correlations with one another. The dimensions of the Counselor VRS correlated highly with the WROS, indicating a connection between the behavior of the client and the counselor—client relationship as perceived by the client. A descriptive analysis of the Client VRS showed the understanding and exploratory dimensions to be the most important in predicting movement. For the Counselor VRS, only the specific dimension was deleted; the affective, understanding, and exploratory dimensions were all highly related to effectiveness. In a clinical analysis made of these sessions, those clients who were rated as not moving under the conditions of the high-level counselors were rated as such because they became resistant and defensive. These counselors were perceptive in hitting at the client's feeling level; Arthur Resnikoff they reflected the client's nervousness and verbalized hunches directed toward the client's feelings. While no movement appeared to be made in this, the first session, the fact that a relationship was established suggests that movement may take place in further sessions as a result of this relationship and the counselor's function- ing in a mode herein described as effective. Less effec- tive counselors either allowed the client to wander through a range of conversational topics, leading to lit- tle movement, or slowed down the movement that the client attempted to make on his own, in spite of the counselor. While there were variations in client behavior under each condition, there were patterns which were related to the counselor's behavior. In the analysis of the medical school sample, the understanding dimension of the Counselor VRS was found to define the effects of treatment which involved coached patients and the medium of stimulated video playback. In an attempt to provide for further validation of the Coun- selor VRS, the Science score of the Medical College Admis- sions Test, the Affective Sensitivity Scale, and the five dimensions of the Counselor VRS were correlated with peer and staff ratings of the second-year class of the College of Human Medicine, Michigan State University. The corre- lations of the Affective Sensitivity Scale with the peer and staff ratings were found to be either low or negative. Arthur Resnikoff The Counselor VRS also bore either low positive or low to moderately negative correlations with the peer and staff ratings. While replication of these findings are in order, the Counselor Verbal Response Scale, with some modification, appears to be useful in evaluating the interviewing behavior of physicians. THE RELATIONSHIP OF COUNSELOR BEHAVIOR TO CLIENT RESPONSE AND AN ANALYSIS OF A MEDICAL INTERVIEW TRAINING PROCEDURE INVOLVING SIMULATED PATIENTS By Arthur Resnikoff A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY College of Education 1968 . Q; \ ACKNOWLEDGMENTS Every thesis is built upon the efforts of many peo- ple, all of whom played various roles in its development. I would like to share my thoughts and thanks to these people. This study would have been impossible without the efforts of Dr. Norman Kagan, chairman of my committee and director of the IPR project. He pushed, caJoled, urged, aided, and harassed me into completing this work. I thank him for all of that--sometimes it seemed unpleasant, but I think of him for both being a good friend as well as a man with great ideas. My final year at Michigan State University was spent with the project; it was an interest- ing as well as enlightening experience. I have learned many things which will help me in the years to come. Throughout the year, I became more and more convinced that I was in the right profession and more and more anxious to get my degree. And to Bev, for being Norm's wife, a good woman, and a good friend. My thanks go to the other members of my committee. Dr. Hilliard Jason of the College of Human Medicine gave me employment when my financial situation looked a bit grim. He provided for much of the data of which this 11 thesis is composed and for insightful guidance along the way. Drs. Gwen Norrell and Robert Green are people whom I respect and like a great deal. They are two of the "coolest" people I know. A number of people helped greatly in the carrying out of the dissertation. Bob McKie and Al Grzegorek served as raters. While they were sometimes stubborn, i.e., they didn't always want to do things my way, they were indis- pensable. Suzanne, of the Reading and Guidance Clinic, did a good deal of the typing and editing. Alan Goldberg graciously consented to allow me the use of data from his dissertation. And to the CDC 3600, which spewed forth reams of partially intelligible figures which compose my partially intelligible tables. There are others who less directly aided me in the dissertation, but who were indispensable nonetheless. Both Steve Danish and Paul Schauble, my two partners in crime on the IPR project, provided for many interesting discus- sions. So interesting, in fact, they probably held me up a few months in finishing my degree. Seriously, my special thanks go to Paul. If we carry out half of the crazy ideas we have designed, we will end up being very tired, but very well known. Our daily running around the track helped me keep my head clear, but the office somewhat odoriferous with gym clothes. May we "publish or perish" together as 111 we have begun, and may we remain close friends, sharing both our good and our unpleasant thoughts together. My family had a great deal to do with my receiving a Ph.D. Because my brother received his doctorate, I felt that I had to match it. While other boys back in grade school dreamed of becoming policemen and firemen, my parents patted me on the head and remarked that I did not have to, but that I ought to go to college when I grew up. So, I planned to grow up and be a college boy. After nine years at three different universities, I am actually going to take a full-time Job——at another university. And finally to my wife, whom I love dearly. The last two years, since we've been married, have been very full. There are many memories, too numerous and too per- sonal to mention here. Who ever thought that two graduate students could afford to go to Mexico and to Europe in the same year? (Buddy, can you spare a dime?) Our communica- tion and sensitivity to each other has grown, sometimes slowly, but it has grown. Believe me, she has been a brick through it all. We have already enjoyed a lot; it feels that on my finishing the degree, we have opened the door to a somewhat newer, enlarged world. iv TABLE OF CONTENTS ACKNOWLEDGMENTS LIST OF TABLES LIST OF APPENDICES Chapter I. THE PROBLEM. Need for the Study. Theoretical Perspective Definition of Terms Assumptions . Hypotheses Overview II. BACKGROUND OF THEORY AND RESEARCH Types of Therapeutic Process Content Analysis . . Closely Related Research. Further Theoretical Background. Summary and Conclusions . III. DESIGN AND METHODOLOGY OF THE STUDY. Study I Study II Sample Instrumentation. . Judging of the Criterion Tapes. Reliability . . . Analyses of Data Hypotheses Summary IV. ANALYSIS OF THE DATA. Study I Study II Summary Page ii vii ix |._J \OCIJNO‘I-t‘w 10 12 16 21 26 28 28 33 36 37 Al U2 A8 149 51 51 63 73 Chapter Page V. SUMMARY, DISCUSSION, IMPLICATIONS . . . 75 Summary . . . . . . . . . . . 75 Discussion . . . 8O Implications for Counseling and for Future Research in Counseling . . . 89 Implications for Medical School Interviewer Training . . . . . . 91 LIST OF REFERENCES . . . . . . . . . . . 95 APPENDICES. . . . . . . . . . . . . . lOO vi Table 3.1 LIST OF TABLES Sex, level of experience, mean age, mean GPA, and mean number of professional courses of counselor sample. Intraclass correlation reliabilities of average ratings calculated for Judges' ratings of counselors across tapes on each dimension of the CVRS Intraclass correlation reliabilities of average ratings calculated for Judges' ratings of clients across tapes on each dimension of the Client Verbal Response Scale . . . . . . . Intraclass correlation reliabilities of average ratings calculated for Judges' ratings of medical interviewers across tapes on each dimension of the Counselor Verbal Response Scale T tests of differences on the Client Verbal Response Scale and the WROS between clients of counselors who maintained high levels of the matching dimensions on the Counselor Verbal Response Scale . . Intercorrelation of the Counselor Verbal Response Scale, the Client Verbal Response Scale and the Wisconsin Relationship Orientation Scale. Multiple regression equation predicting client movement from the other measures of the Client Verbal Response Scale after deletion of variables with non- significant beta weights. Multiple regression equation predicting counselor effectiveness from the other measures of the Counselor Verbal Response Scale after deletion of vari— ables with non-significant beta weights. vii Page 32 ’45 45 US 55 57 58 59 Table Page A.5 Fisher Exact Test of differences in move- ment between clients of counselors who maintain a high level of effectiveness and clients of counselors who maintain a low level of effectiveness . . . . 61 4.6 T tests between pre and post sessions for medical students on the Counselor Verbal Response Scale-~Total Group . . 66 4.7 T tests between pre and post sessions for medical students on the Counselor Verbal Response Scale-~lst and Ath sessions group . . . . . . . . . 66 4.8 T tests between pre and post sessions of medical students on the Counselor Verbal Response Scale--2nd and Ath sessions . . . . . . . . . . . 66 A.9 T tests between pre and post sessions of medical students on the Counselor Verbal Response Scale--lst and 3rd sessions . . . . . . . . . . . 67 “.10 Intercorrelations of the MCAT, Fall and Spring Peer Ratings, Pre-Clinical and Clinical Faculty Ratings, Affective Sensitivity Scale (Empathy), and the Counselor Verbal Response Scale for the second year medical school sample. . . 7l viii Appendix A. ’IJFIUOUJ LIST OF APPENDICES Counselor Verbal Response Scale. Client Verbal Response Scale Wisconsin Relationship Orientation Scale. Peer Rating Scale Staff Rating Scale Counselor Verbal Response Scale Training Manual Client Verbal Response Scale Training Manual . . . . . . . . Summary of Counselor Levels Case Summaries ix Page 101 103 105 107 110 112 127 138 140 CHAPTER I THE PROBLEM Through previous research within the Interpersonal Process Recall project, a scale had been developed to mea- sure counselor interview behavior. This scale, the Coun- selor Verbal Response Scale (CVRS), has been shown to differentiate between counselors of different experience levels (Griffin, 1966) and between two different methods of counselor education (Goldberg, 1967). While the qual— ities of counselor behavior have been defined by the Scale, client interview responses as a function of coun- selor behavior has not been substantiated. For a counse— lor response to be classified as being effective, one would expect it to have a beneficial effect upon the move- ment a client would make within that interview. At the time the above idea was being considered, the College of Human Medicine at Michigan State University was engaging in an instructional program designed to train medical students how to establish a relationship with a patient, how to attune themselves to a patient's communi- cation. The instruction involved the use of stimulated video playback and the use of actors to simulate patients who had a variety of problems. The Counselor Verbal Response Scale seemed appropriate to measure both the amount of change that would take place as the medical stu- dent gained in interviewing skill and the final level of proficiency that he would attain at the end of treatment. Such a procedure, along with criterion measures of a stu- dent's effectiveness in discussing personal problems, would also provide data with which to further validate the CVRS. In order to provide the maximum amount of informa- tion on behavior within the interview situation, a two- part investigation was conducted: 1. Audio tapes which previously had been analyzed for counseling behavior were re-analyzed along the dimen- sions of an instrument developed for this study, the Client Verbal Response Scale, in order to determine the relationship of counselor to client response. The clients had previously rated their perceptions of the relationship with their counselors. 2. From the knowledge gained in the above investi- gation, medical student--simulated patient video tapes were then rated and analyzed for change over training, and were related to a number of criterion measures of effec- tiveness in interview behavior. This investigation, therefore, has a threefold pur— pose: (l) to study differential client behavior and client perception of his counselor as a result of the counselor's maintaining either "high" or "low" levels of therapeutic behavior as rated on the Counselor Verbal Response Scale (Griffin, 1966; Goldberg, 1967; Kagan, g£_a1., 1967); to measure increases in the level of therapeutic behavior of second year medical students who experienced special inter- view training; and to further validate the dimensions of the Counselor Verbal Response Scale by correlating scores obtained on it with a test used in medical school admis— sions, a measure of affective sensitivity, and peer and staff ratings of a student's effectiveness. Need for the Study In the field of counseling there is a common assump- tion that the expertise of the counselor has an effect upon the client's growth in counseling. Therapists of greater ability should bring about more "growth" in their clients than should counselors of lesser ability. This assumption has been partially examined in the studies which relate different levels of counselor behavior with client outcome (Truax & Carkhuff, 1964), but there have been no conclusive studies which document client process as being related to counselor process. In light of the fact that counseling may have positive as well as negative effects on its clients, it seems important that an inves- tigation be made into the interview interaction itself, in order to more clearly formulate the different levels of counselor behavior and their potential differential effects upon clients. One of the problems in process research is the development of valid and reliable Operational measures with which to describe the therapeutic situation. Toward this end, the Counselor Verbal Response Scale has been created. It was used in the first section of this study as a means of describing the levels of counselor behavior. In the second section (involving the instruction of medi- cal students), the scale was used to determine pre—post change in the effectiveness of interview behavior. Both sections of this thesis, then, contribute not only to a documentation of treatment effects, but also to a further validation of the scale. Theoretical Perspective Counselor and client behavior is to be measured in terms of four major dichotomous dimensions deemed relevant in therapeutic process. These four dimensions-—(1) affec- tive-cognitive, (2) understanding-non-understanding, (3) specificity—non-specificity, and (A) exploratory— non-exploratory, have been developed from previous theory and research. A full discussion of this follows in Chapter II. It has been found that Ph.D. students in counseling receive higher scores on the Counselor Verbal Response Scale (CVRS) than do M.A. students (Griffin, 1966). It has also been found that if M.A. counselor trainees receive instruction through a model of videotape feedback on their counseling behavior, they have higher scores on the CVRS than do counselor trainees receiving traditional supervision. The next question, then, concerns the effects of these differential levels on the interview behavior of clients. The framework being used to account for client behavior is that described by Pepinsky and Karst (196“) as the "convergence phenomenon." This construct postu- lates that the client will shift toward the norms and standards made available to him by the therapist. Speci- fically, for this study, it is postulated that there will be movement by the client toward the therapist in the broad areas of verbal behavior which are delineated by the Counselor Verbal Response Scale. In other words, if the counselor "moves in" and makes reference to the client's affect, the client will emit more affect-laden responses. If the counselor communicates to the client his understand— ing of the client's comments, the client will come to emit responses which indicate self-understanding. If the coun— selor focuses on what appears to be the client's problem, the client will move with the counselor and become more specific and concrete. Finally, if the counselor verba- lizes the kinds of statements that should encourage client self-exploration, the client will then make self-exploratory statements. The above statements serve as the basis for the hypotheses to be tested as part of the first section of this study. A scale has been developed, similar to the Counselor Verbal Response Scale, to measure the behaviors of the client along the stated dimensions. In addition, client attitude as measured by the Wisconsin Relationship Orientation Scale (Steph, 1963) will be related to the dimensions of both counselor and client behavior. The second part of the study, the evaluation of an interviewer training procedure, is based upon past research in the area of counselor education (Goldberg, 1967). The present research has been designed to study the effects of having medical students taught interviewing through a combination of video tape playback and the use of actors as coached patients to insure student exposure to a wide range of patient types with problems relevant to the medical school's curriculum. Definition of Terms Counselor Interview Behavior That part of the counselor's communication during an interview which can be reduced to quantifiable terms. Client Interview Behavior That part of the client's communication during an interview which can be reduced to quantifiable terms. Recall A reviewing of a counseling interview by either counselor or client during which the focus is on the viewer's recreating his thoughts and feelings during the initial interview. In this research, recall is synonymous with the medical student's reviewing of the interview. During the recall session, the viewer is encouraged to stop the replay at any time to comment on covert feelings or thoughts during the session. Coached Patient An actor who is instructed to role-play an individual coming to a doctor to discuss with him a problem needing the latter's care. Assumptions There are a number of basic assumptions which form the foundation of this research. 1. A counselor or medical interviewer's within- interview behavior can be measured. 2. A client's within—interview behavior can be measured. 3. Rated statements on either the Counselor Verbal Response Scale or the Client Verbal Response Scale are additive within categories, and a greater number of such statements indicates a greater amount of response behavior on that dimensions. A. Counselor and client behavior can be adequately judged from viewing either audio- or videotaped samples of a total interview. 5. The rating of 20 consecutive counselor respon— ses taken from the middle to the end portion of an ini- tial interview, will be a representative sample of coun- selor performance throughout the entire interview. 6. Initial interview interaction between counselor and client is related to consequent interview interaction. 7. A coached patient portraying a particular prob- lem is an adequate simulation of a "real life" patient displaying that same problem. Hypotheses The hypotheses will be stated specifically in the later chapter on research design. In general form, they are as follows: H1: Clients whose counselors have a high number of responses in each of the categories of the Counselor Verbal Response Scale will have higher scores on similar scales of the Client Verbal Response Scale when compared to clients whose counselors have a low number of respon- ses in each of the categories. H : Clients of effective counselors will rate them- selves as having a better relationship with their counselors than with clients of non- effective counselors. H : Case summaries of interviews of the more effec— tive counselors will indicate more movement on the part of the client than will those case summaries of the less effective counselors. HA: Medical students will make a greater number of affective, specific, exploratory, and effective responses following a period of interviewer training using live simulated patients and stimulated recall by means of video playback than they will before such training. Overview In this first chapter, the need for and purpose of the study, a background of theory, and broad research hypo- theses, have been stated. Chapter II will contain previous research leading to the problems under study in this dis- sertation. Chapter III will contain the sample descrip- tions, research design, methodology, and scale reliabil- ities. In Chapter IV will be the tables of results and their description. In Chapter V will be a discussion and interpretation of the results and possible specific impli- cations for the further use of the operational measures employed in the study-—and, in more general terms, for the future instruction of trainees in the counseling area. CHAPTER II BACKGROUND OF THEORY AND RESEARCH The study is an attempt to examine the common assump- tion that the expertise of the counselor will have an effect on the client's growth in counseling. The basic research statement is if the counselor exhibits certain general patterns of rated verbal behavior, the client's verbal behavior will tend to move in the same direction. For example, if the counselor responds to a client's affect, it would be expected that the client would make a greater number of affective responses. A study of this type is relevant to the present dis— cussion concerning the standards of Master's level counse- lors (Hill, 1967; Hansen and Stevic, 1967; Patterson, 1967). Data seems to indicate that counseling may bring about either positive or negative client outcomes (Bergin, 1963; Truax, 1966). Therefore, it is impossible to look closely at the interview interaction itself in order to more clearly formulate different levels of counselor behavior and their potential differential effects upon clients. Previous studies in this area have centered either upon pertinent therapist variables or upon client person- ality characteristics and their relationship to outcome. 10 11 Few, however, have focused onrthe verbal interaction between counselor and client. In a well-known series of studies, Truax (1964) noted a positive relationship between client self-exploration, measure of successful client outcome, and level of therapeutic conditions, i.e., accurate empathy, communicated warmth, and therapist genuineness. While significant, the above therapist pro- cess variables had a rather low positive relationship to client self—exploratione—UX>low to account for much of the variance. Several less rigorous studies have investigated other aspects of client interview behavior. Kirtner and Cart- wright (1958) found that successfully treated clients, those whose Q—sort real selves approaches their Q-sort ideal selves, tended to deal with feelings to a greater degree. Dittman (1952) demonstrated similar findings; therapeutic movement was associated with therapist response to either feelings or to the interpersonal behavior between therapist and client. In another study by Truax and Cark- huff (196“), clients' tendency toward concreteness, the specificity of interview content-—mentioning of specific events, situations, or feelings--had a high correlation with other process variables. However, one of the prob- lems with this last piece of research is that the authors failed to report any of their statistics, including means and standard deviations. This omission, coupled with only 12 a partial explanation of their operational measures, does not allow for replication of their findings--findings which would seem to have strong implications for counseling. Types of Therapeutic Process Content Analysis The dilemma of the adequacy of operational process measures has constantly plagued this area of research into psychotherapy. There are two basic strategies that have been adapted for investigating the interview process. The first is the classic model, which places a premium on objectivity and is designed so that workers with minimal special training can perform the analysis (Marsden, 1965). This method is juxtaposed against the pragmatic model, which attempts to delve into psychological meaningfulness by working directly with complex clinical constructs. The latter is an effort to use the skills of the clinician while structuring the measures to maintain, as far as pos- sible, some degree of scientific rigor. The Classical Approach Most of the studies in this area deal with client characteristics. Braaten (1961) used a content analysis procedure to test the hypothesis that success in client- centered therapy is correlated with increasing expression of feelings referent to self as opposed to non-self mater— ial. Therapists' ratings of success, the criterion mea- sure, were positively related to references to self. 13 Rogers has developed a seven—point scale of person- ality, a continuum with one end representing stasis, fix- ity, and rigidity, and the other end, looseness and flexi- bility (Rogers, 1959). Findings indicated that successful cases began and terminated at a significantly higher level on the scale than did less successful ones, and that there was a greater movement along the scale during therapy in successful cases (Rogers, 1961). The problems of this research are that (1) success is cast into categories not readily accepted by the field in general, and (2) there is no evidence of the interaction between therapist and client which brings about this change. There have been a number of studies which, while analyzing the types of therapist and client response reli- ably, pay a price for precision through the superficiality of the dimensions measured. Matarazzo and his associates have done a good deal of work in the areas of differen— tiating a number of nosological groups and normal indivi- duals (Matarazzo, 1962), and have generated normative data on the action and slience behavior of subjects (Matarazzo, Hess, and Saslow, 1962). They have found that interpretive interviewer statements can be differentiated from explor- atory or information-seeking statements in terms of inter- viewer response deviation (Kaufer, Phillip, Matarazzo, and Saslow, 1960). Another study along similar lines (Dinoff, Kew, Pickard,and Timmons, 1962) has found that in group 1“ interaction, group members tend to establish a persistent hierarchy of verbal response analogous to the pecking order phenomenon in chickens. While this type of descriptive study may engage one's curiosity, its payoff appears to be in "chicken feed" rather than practical usage. The cate— gories used, e.g., action versus silence behavior of cli- ents, are seemingly considered in a vacuum without any consideration of their relationship to client outcome, either within the interview or upon termination of treat— ment. A number of classical content analysis systems have been developed to facilitate investigation of patient- therapist interaction in an attempt to investigate their characteristics as a communication system. A general para- digm has been developed by Bales (1950) for examining small group interaction. Attempts have been made to apply this model to the dyadic therapeutic situation. Jaffe (1961) has analyzed the speech from a counseling session as if it came from one person. The problem with this type of break- down is that it loses the difference between the partici- pants. A more sophisticated system has been developed by Adler and Enelow (196A) based upon Bale's small group interaction categories. 0n the basis of face validity, it seems to provide a comprehensive view of the therapist- client interaction couched in the terminology of 15 present—day psychotherapy. The instrument is yet to be tested; its reliability has not been established. The Pragmatic Approach Followers of the pragmatic approach to content ana- lysis attempt to uncover the psychodynamic trends in the— matic material; the areas of anxiety, hostility, and severity of schizophrenic disorganization are an example (Gottschalk, e£_al., 1963). The fundamental assumptions underlying these scales were derived from the psycho— analytic theory of defense mechanisms, affects, and primary and secondary thought process (Marsden, 196A). The scor- ing unit is twmagrammatical.clause; relevance of the unit to the different categories is a highly inferential matter, requiring a clinical judgment concerning the suppressed or repressed meanings of the unit to the patient. Coding reliability has been surprisingly high, averaging from the low to mid eighties. Dollard and Auld (1959) have developed a pragmatic model whose main focus in on the patient's dynamic motive states, whether of conscious or unconscious origin. The system is geared to differentiate between neurotic and psychotic patients; little work has been done relative to the validity of the system. A more interactional pragmatic model has been developed by Sklansky, Isaacs, and Haggard (1960), designed to determine the therapeutic effect of psychotherapy by l6 studying the manifest and latent meanings of the patient's communication in relation to certain aspects of the thera- pist's responses. The investigators looked at the depth of consciousness of a client's statement and noted to which level of consciousness the therapist responded. Reliabil- ity was a problem. The authors concluded that When the therapist's response was direct at the manifest level and irrelevant at deeper levels the topic tended to be dropped by the patient at all levels. When the therapist's responses were direct at the manifest level and indirect at the latent levels the topic continued to be discussed even if the therapist made occasional responses irrelevant at all levels. When the therapist responded dir- ectly at the manifest and at some of the latent levels, therapeutic activity seemed to be facili- tated, but when the therapist's responses were directly pitched to latent-meaning levels and only indirectly dealt with the manifest level, there was tendency for the patient to change the topic after a short time (Sklansky, Isaacs, and Haggard, 1960)- The above study seems to suggest that for a thera— pist to keep the client discussing a particular theme, he must respond both to the manifest level of the client's statement and also to some degree below the surface. Neither the manifest nor the latent level may be ignored at the expense of the other. Closely Related Research The Interpersonal Process Recall (IPR) Method uses videotape playback of counseling sessions to stimulate recall of the dynamics involved in this interpersonal interaction (Kagan, et al., 1967). This procedure is 1? accomplished by having a client View his completed session while a trained clinician actively probes for the under— lying affective components of the counselor-client commun- ication. Either the trained clinician (interrogator) or the client may stop the playback to discuss client recal- led feeings and elaborate on meanings. In using this method for purposes of counselor education, the IPR pro- ject was in need of a criterion measure of counselor effectiveness that could be developed and used continually in counselor education research. Scale Development Two procedures were adapted in the development of the Counselor Verbal Response Scale. The first was an intensive examination of video and audio tapes of counselors of known levels of competence. Several audio tapes were available to us of counseling sessions conducted by counselors of national repute employed at the MSU Counseling Center and within the College of Education's Depart- ment of Counseling. Audio tapes were also avail- able of practicing counselors whose repute was less than glowing. Another pool of potential data was available to us on video tapes for by now we had conducted many IPR sessions and had not erased the video tapes of the counseling sessions nor the audio tapes of the recall sessions. This provided us with video tapes which could be differentiated according to client evaluation of the competence of the counselor. Another source of recorded counse— lor behavior (and one which appeared to contain the greatest disparity between groups and hence the best opportunity to identify specific but important behaviors differentiating the two groups) were video— tapes of master's degree candidates and doctor's degree candidates in counseling. 18 These tapes were reviewed and, in almost micro- scopic detail, the verbal and nonverbal behaviors of the more and less qualified counselors were compiled into rather extensive lists. The second procedure was to again review the literature on important elements in counselor behavior. As our lists of observations were collapsed into more and more general headings, four major groupings emerged which seemed to differentiate between the more and less qualified counselors. Three elements of effective counseling had been defined by Truax and Carkhuff (196“) as (l) the counselor communi— cates his understanding, (2) the counselor is spe- cific, and (3) the counselor is exploratory. And the described dimensions--understanding, specifi- city, and exploration--seemed to account quite well for three of the four major groupings we had com- piled. The four grouping was best summarized as an Affective—Cognitive group (Kagan, et al., 1967). These categories are also similar to those developed by Struff (1960). His system tapped the gross behavior of the therapist, the degree of inference and warmth implicit in his communication, and whether he worked from the patient's frame of reference or his own. The above procedures of analyzing tapes and reviewing the literature lead into the construction of the Counselor Verbal Response Scale (CVRS) (Griffin, 1966; Goldberg, 1967; Kagan, et al., 1967). An overview of the Scale follows: Counselor Verbal Response Scalel Overview. The Counselor Verbal Response Scale (CVRS) consists of five forced choice dichotomous dimensions measuring the extent to which counse- lors are characterized by affective, understanding. specific, exploratory, and effective response. The dimensions are defined as follows: An affec- tive response is one which makes reference to or encourages some affective or feeling aspect of the client's communication while a cognitive response refers primarily to the cognitive com— ponent of a client's statement; understanding 19 refers to the counselor's ability to convey to the client his awareness of and sensitivity to the client's feelings and concerns by attempting to deal with the core of his concerns rather than mak- ing vague responses or referring to peripheral con- cerns; exploratory responses encourage the client to explore his feelings and provide him with an opportunity to do so while nonexploratory responses typically restrict the client's freedom to explore. The final dimension, effective-noneffective, is a global rating of the overall effectiveness of the counselor's response in promoting client movement. 1Because only counselors' verbal behaviors are actually rated in the final form of the instrument, we ultimately "settled" for this title although we do not mean it to imply that visual elements are not looked at by judges in determining ratings (Kagan, et al., 1967). The CVRS differs from other rating scales in that it focuses on a series of individual client-counselor verbal units (client statement——counselor response) during the course of an interview, rather than on global ratings of entire interviews or longer interview segments. This for— mat was based on one developed by DeRoo and Rank (1965). Thus, the judge is required to describe every counselor response to a client verbalization on each of the five dimensions of the scale. After 20 counselor responses have been dichotomized on each dimension, totals are obtained. Thus, a maximum score of 20 and a minimum of zero are possible for each dimension. Reliability of the scale using Hoyt's analysis of variance was found to be high in its initial application and in three replications (Griffin, 1966; Goldberg, 1967). There is data supporting the scale's construct validity. 2O Griffin (1966) found significant differences between M.A. and Ph.D. candidates in counseling, the latter having higher scores on all five scales. In research immediately preceding this study, Goldberg (1967) found the scale to differentiate matched groups of counselors-in-training, each group receiving a different training method, IPR versus a more traditional approach. In addition, the scale was found to be sensitive to pre-post change over the course of treatment. Although not a process scale, the Wisconsin Rela- tionship Orientation Scale has been used to tap client attitude toward the counselor-client relationship. It was used in Goldberg's (1967) study as another criterion of counselor effectiveness. Wisconsin Relationship Orientation Scale There exists ample evidence in the counseling lit- erature to suggest that a critical element of effective counseling is the ability of the counse— lor to establish a meaningful personal relationship with the client. If this is so, it seemed likely that the client's perception of the relationship might well serve as a measure of the degree to which the counselor was able to communicate such qualities as empathic understanding and positive regard. The Wisconsin Relationship Orientation Scale (WROS, Steph, 1963) was used to determine the nature of the relationship existing between counselor and client, as perceived by the client. In using the scale, the client is asked to indi- cate how he feels about the counselor in terms of five steps which range from total avoidance of the counselor to the feeling of probably being able to talk with the counselor about almost anything. Clients of counselors in the IPR supervision group rated their counselors as being earier to talk with than did clients in the traditional supervision group (Kagan, et al., 1967). 21 Further Theoretical Background Besides the work accomplished through IPR, a dif- ferent line of evidence comes from previous research in the area of the "convergence" phenomenon in counseling and psychotherapy. Borrowed originally from the study of small group interaction (Sherif, 1936; Asch, 1952), this construct postulates that the client will shift toward the norms and standards made available to him by the ther- apist (Pepinsky and Karst, 196“). Lennard (1962) has summarized his findings relative to therapy as an "inter- actional system," that there is an increasing correlation over time between specific kinds of patient and therapist communications. Lennard and his associates found a great and increasing similarity between patient-therapist pairs in the longitudinal development of the interaction despite major differences in behavior and expectations. They also found that the therapist—patient interactional system con- tained built-in mechanisms for maintaining system equili- brium. For example, sessions exhibiting more than usual amounts of silence tended to be followed by those in which the therapist emitted more evaluative behavior and behavior of greater informational specificity. There has been some research which has been inter- preted to mean that therapeutic interaction must bring about a change in the client as the therapist wants him 22 to change; but there is little agreement as to what areas of client functioning change, what is classified as "good" change, and how the change is brought about. Heller (1963) has clearly demonstrated that the therapist's behavior can shape that of his client. There are diverse hypotheses concerning the specific attitudes, values, and behavior which may be changed via the therapy process, some of which have been described in the above research. Because there is such a wide variety of behaviors being examined, there is very little consistency to be found among these studies. Many content-analysis studies have been done over the past twenty years. One is struck by the lack of replication; a system is used in making an initial thrust at a problem and is then forgotten. Content—analysis research can get tedious; perhaps this tediousness of the research accounts for its relative scarcity. System after system has been developed and presented in one or two demonstration studies, only to lie buried in the litera- ture, unused even by its author. Furthermore, most studies examine either counselor characteristics or behavior, or client characteristics or behavior; but very few have looked at the results of client movement as dependent upon the behavior of the counselor. Validation of counselor behavior regarding its impact on the client is rarely to 23 be found in the literature. The present study has been undertaken as a step in this direction. Research Pertaining to Interviewer Training The beneficial effect of videotape playback and recall in the instruction of counselors in practicum has been shown by Goldberg using IPR methodology. In addi— tion to the Goldberg study, Schiff and Reivich (196A) and Suess (1966) have described the application of video- tape as an adjunct to the supervisory process. The use of role playing in counselor training and the use of videotape feedback has also been documented by Kagan (1965; 1967). Role playing has been used for so long as a way of introducing beginning interviewers to the con- cepts and practices of spychotherapy, that it might be thought of as a "traditional" procedure. More recently, however, the use of a simulated client has been intro- duced. Simulation has the advantages of providing exper- ience under what appears to be a "real" situation without the risks that such an actual encounter might have. It has come to be recognized that skills of great complexity cannot be acquired at once. Separate parts of the total process must be selected out for exclusive attention and mastery before all these components can be effectively fitted together for application in the real sit- uation (Jason, 1968). For example, in the area of medicine, it is felt that a central feature of the physician's role is his 2H responsibility for establishing a productive and thera- peutic relationship with his patients. This relationship should involve for the physician the skills of facilita- ting the emergence of relevant information and feelings, careful listening, observing, systematic questioning, sensitivity to emotional subtleties, and empathy. Through the use of "simulated patients," trained actors and actres- ses to simulate the personality characteristics and medi- cal histories of medically relevant conditions, the follow— ing purposes may be achieved: (1) "patients" with diseases and related problems that are maximally relevant to the material the students are then studying in their other classes and laboratories can be provided. (2) the patients problems can be kept appropriately simple for the level of preparation of students. (3) the patients can be used repeatedly so that after completion of his own interview with a particular patient, each student can observe several other students interacting with the same patient and see the important differences that emerge with different approaches and dif- ferent personalities, and (A) a very real and meaningful clinical experience can be provided for each student (Jason, 1968). Students are then provided with meaningful feedback on their own performance through stimulated recall by means of videotape playback. One of the functions of this pro- cedure is to keep interviewer's anxieties at a helpful and not debilitating level by lowering the amount of evalua— tion and risk involved. 25 Enelow (1968) has used trained actors and actresses to simulate psychiatric and psychosomatic disorders and has prepared a series of training videotapes and films using principles of programmed instruction. The use of coached clients in the training of counselors has been a method also used in previous IPR research (Kagan, et al., 1965). Other Relevant Descriptive Measures The Affective Sensitivity Scale (A.S.S.) (Campbell, 1967) has been shown to be a predictor of counseling suc- cess--success as defined by ranked judgments of a class of individuals of the competence of their fellow class— mates' ratings or ratings by individuals in a supervisory capacity (Kagan, §t_al,, 1967). This type of rating has been used as a validation technique in the initial stages of scale deveIOpment and as a subjective measure of effec- tiveness (Stefflre, 196A). Affective sensitivity has been shown to be a predictor of peer and staff ratings, and has also been found to have a low, positive relationship of the dimensions of the dimensions of the Counselor Verbal Response Scale. However, the interrelationships of the three types of operational measures has never been estab— lished. 26 Summary and Conclusions While there are a large number of studies dealing with counseling "process," there is no conclusive research which relates client response to the within-interview behavior of the counselor. Previous studies have centered either upon pertinent therapist variables or upon client personality characteristics and their relationship to outcome. Viewing the counseling situation in such a light ignores the two individuals as an interactional system. The convergence phenomenon has been viewed as a possible explanation for the interaction of counselor and client. This research is similar to that done in the area of convergence because it assumes that there will be a move- ment by the client toward the therapist in broad areas of verbal behavior which are delineated by the Counselor Ver- bal Response Scale. It is assumed that if the counselor "moves in" and comments upon the client's affect, the client will become more affective. If the counselor com— municates to the client his understanding of the client's comments, the client will come to emit responses which communicate understanding of himself. If the counselor focuses in on what appears to be the client's problem, the client will move with the counselor and will also become more specific and concrete. And finally, if the counselor verbalizes the kinds of statements that should encourage the client's self-exploration, the client will 27 then make self-exploratory statements. The above state- ments are the basis for the hypothesee which will be tested. In order to test these hypotheses, a client response scale will be developed to correspond to the categories of the Counselor Verbal Response Scale. Manuals for rat- ing both the counselor and the client scales may be found in Appendices A and B. Reliability estimates will be com- puted and included in Chapter III. The use of the "coached client" and of stimulated videotape feedback has been effective in the training of counselors. However, to date there has not been an effort to relate a process measure of effective counseling with such measures as peer and staff ratings, a test of affec- tive sensitivity, or a measure of academic achievement. CHAPTER III DESIGN AND METHODOLOGY OF THE STUDY An experimental research design was formulated which would permit the testing of the research questions posited in Chapter I, namely that: (a) the client's movement and perception of the counselor is related to the level of therapeutic behavior maintained by the counselor, (b) the stimulated recall by means of video tape feedback produces change in interviewing skill of second year medical stu— dents, and (c) various measures correlate with the Counse- lor Verbal Response Scale, adding to the construct valid- ity of the scale. The dissertation will be discussed in two parts, Study I related to question (a) above, and Study II related to questions (b) and (0). Study I The data for this study was derived from thirty audio taped interviews gathered by Goldberg (1967) in his study of new techniques in counselor supervision. Following a training period of five sessions, each counselor saw a new client; this sixth session was the criterion session used for analysis. It lasted approximately one-half hour and was audio-recorded. Following the interview, the 28 29 clients completed the Wisconsin Relationship Orientation Scale. All counselors received supervision; the super- visors focused on those dimensions of counseling which were characterized by specific counselor behaviors. These behaviors had previously been deemed important for novices to learn and were those categories which make up the Coun- selor Verbal Response Scale. The discussion with the supervisor focused on whether the counselor: (l) dealt with the affective or cognitive concerns of the client, (2) followed the client in an attempt to (a) under- stand his mood, feeling, and emotion; (b) recog- nize the underlying content of his statement; and (0) respond in the appropriate manner and appropriate level, (3) allowed and encouraged the client to explore his feelings, (A) concentrated on the central or core issues, help- ing the client to specify the focus of her con- cerns rather than on peripheral or incidental matters, (5) maintained the focus on the client and the inter— action between client and counselor, and (6) conveyed by his verbal and non-verbal behavior that he understood or was attempting to under- stand the client (Goldberg, 1967). The above six points relate directly to the Counse- lor Verbal Response Scale, i.e., whether a response (a) referred to affect or cognition, (b) reflected the counselor's understanding or non-understanding, (c) focu- sed on specific or peripheral concerns, and (d) encouraged or limited client exploration. These counselor behaviors described above are not tied to any one theoretical posi- tion but instead focus on the essentials of interpersonal 30 communication rather than on technique (Truax and Carkhuff, 1963; Griffin, 1966). If client-counselor behavior is viewed as an inter- actional system in which the counselor is somewhat able to determine the direction of the interview and is attempting to respond in a manner which reflects the dimensions of the CVRS, it is felt that part of the client's response system will be in reaction to that of the counselor. Spe- cifically, if the counselor reflects client affect, the client will talk about and emit more affect-laden respon- ses. If the counselor communicates his understanding of the client's comments, the client will begin to emit responses which indicate or strive for self-understanding. If the counselor focuses on what appears to be the client's problem, the client will move with the counselor and become more specific and concrete in his responses. If the counselor verbalizes the kinds of statements that should encourage client self-exploration, the client will then make self-exploratory statements. And finally, if the counselor makes the kinds of statements that are clin- ically judged as being effective, the client will emit responses which are clinically judged as being indicative of movement. Rating Procedure The recorded sessions were then rated for the counse- lor's behavior on the Counselor Verbal Response Scale. Four raters were used, three rating at any one time; this procedure was adopted to ease the time pressures on the judges. All judges were Ph.D. students in counseling and had either completed or were in the process of completing doctoral counseling practicum at Michigan State Univer- sity. Ratings were made of twenty counselor statements from the middle of the interview. Using a different set of three raters, the sessions were rated along the dimen- sions of the Client Verbal Response Scale. In this way, the relationship between the counselor response and its effect upon the manner of client response could be studied. Summary scores were used for this analysis to tap the general tenor of the interview. Both the scales and the scale reliabilities will he discusses later on in this chapter in further detail. In addition to these ratings, the second group of raters each wrote a case summary on the recorded material which they had rated; the composite of these summaries serves as a global measure of client movement. Sample Counselors.--The counselors participating in this study were Master's degree candidates in the Department of Counseling, Personnel Services, and Educational Psy- chology at Michigan State University. All degree candi- dates as of Spring, 1966, were informed of an opportunity 32 to receive additional supervised counseling experience by participating in a research project. The variability of the subjects in their counseling skills provided a good range of behavior necessary for an analysis of good ver- sus poor counselors. The original sample from Goldberg's dissertation (1967) consisted of thirty-six graduate stu- dents over three quarters of instruction. The group is described according to (1) level of experience, (2) sex, (3) age of subjects, (A) grade point average in guidance courses, and (5) number of professional courses in Coun- seling and Guidance. Summaries of this data are found in Table 3.1. TABLE 3.1.——Sex, level of experience, mean age, mean GPA, and mean number of professional courses of counselor sample. Mean Mean Mean Number of Pro— Level Of Experience Age GPA fessional Courses Sex 25 Pre—practicum 3A.06 3.Al 10.1 Males 19 ll Practicum Females l7 The sample appeared to be typical of graduate stu- dents in counseling at Michigan State University. Thirty from the original sample of thirty—six were involved in the analysis. Six were lost because recording of their 33 sessions were either inaudible or missing. There is no reason to believe that these six were any different than the other members of the sample. Clients.—-Clients for the recorded interview were tenth grade students from local high schools. A parti- cular age-grade level was selected since it was assumed that these clients would be facing relatively similar developmental tasks and would therefore be presenting somewhat similar problems to the counselor. It was hOped that this approach would at least minimally equate clients across counselors. Study II This section of the investigation was designed to validate the Counselor Verbal Response Scale on a group of second year medical students who received a treatment consisting of stimulated recall of their interviews with coached patients. The structure of the study is not a true experimental design as such but a pre-experimental design, a pre—post no—control group study (Stanley and Campbell, 1963). All subjects received a treatment of four interviews with different actors portraying patients who had physi- cal problems accompanied by strong emotional overtones. Following each interview the subject received instructions via stimulated videotape recall by instructors in the College of Human Medicine, physicians in the community, 3A and a member of the Office of Medical Education who had experience with the recall technique. The recall consis- ted of a review of the interview by the student, the above- mentioned instructors, and some members of the student's class. The student was encouraged to stop the replay at any time to comment on his reactions, feelings, and thoughts during the session. The focus of the instruction was on the student's learning how to establish a relation- ship, to make the patient feel comfortable, and to be sen- sitive to the patient's particular problems. The students were not required to obtain case histories or to come up with diagnoses during these interviews. Because of their nature, these sessions were assumed to be similar in sev- eral ways to initial counseling interviews and, to that extent of similarity, amenable to evaluations via the Coun- selor Verbal Response Scale. All subjects conducted their four interviews within a one-month period in the Fall of 1967. All sessions were video—recorded. The first and fourth interviews were to be rated as the criteria sessions for the pre and post measures, respectively.* In most cases, the actor used for each interview was the same for all students, i.e., the actor for the first interview was seen by everyone for his first interview; a different actor was used for *Explanation for the loss of part of the sample may be found under Sample (Study II) in this chapter. 35 the second interview. A total of one actor and seven actresses were used. Problem topics ranged from an about- to-be-married male who feared he had venereal disease, to a Negro woman suspicious of doctors and complaining of a chronic stomach pain, to a pregnant woman with a history of miscarriages. All problems had emotional overtones. Three raters, each a doctoral candidate in counsel- ing, performed the ratings. All ratings were made without the rater's knowledge of which was a pre and which a post interview. In addition to the ratings described above, the students were administered the Affective Sensitivity Scale (a measure of empathy) and subjective ratings by peers and staff of the students' competence in dealing with others' personal problems. Pre—post differences on the tape—ratings above pro- vide further construct validity of the scale, and demon- strate how effective the treatment just described is for instructing medical students in the fine art of interview- ing. Due to the lack of a control group, interpretation of whatever treatment differences have occurred may be limited; a group, by virtue of its exposure to interview- ing for the first time, may gain over time. Comparisons of scores on the CVRS to other studies using the rating scale cannot be made; the instrument may be reliably used only within a study to look at differences between groups or within a group over time. It is important that the 36 same set of raters by used for a set of comparisons. No matter why differences might exist, those which do (e.g., ones uncovered by the CVRS) would add to the construct validity of the rating scale. Intercorrelations will be computed among the peer and staff ratings, scores on the CVRS, and scores on the Affective Sensitivity Scale. Using either peer or staff ratings as a dependent variable, a multiple regression equation will be constructed with the empathy and rating scale measures to "predict" the subjective class standing measure. The Medical College Admission Test Science Score will also be included as an achievement-aptitude measure of possible significance. Sample All subjects were members of the Year Five (second year) Class of the College of Human Medicine, Michigan State University. There are twenty-three males and three females in the class. All had either very little or no experience at interviewing. Two members of the class were also soon to be receiving their Ph.D. degrees in the phy- sical sciences. Because of a limited tape library and technical difficulties, treatment comparisons could only be made on seventeen students. There is no reason to believe that these seventeen are any different on the dimensions being used in this study than is the class as 37 a whole. For two students without pre-tapes, the post- tapes were rated and included in the correlational analysis. The one actor and seven actresses used in the treat- ment condition were either students in drama at Michigan State University or were otherwise contacted and trained to participate as part of the treatment conditions. Instrumentation In Study I, four measures were used: (1) the Coun— selor Verbal Response Scale* which measured trainee effec- tiveness, (2) the Client Verbal Response Scale** which measured client "movement," (3) the Wisconsin Relationship Orientation Sca1e*** (WROS) which measured the nature of the relationship existing between counselor and client as perceived by the client, and (A) case summaries of client movement as seen by clinical judges. The Counselor Verbal Response Scale and the WROS were described fully in Chapter II. Client Verbal Response Scale The Client Verbal Response Scale was developed for use in the present study. With it, the client's responses are judged along similar categories as those of the coun- selor. This scale consists of five forced choice *See Appendix A. **See Appendix B. ***See Appendix C. 38 dichotamous dimensions measuring the extent to which clients are characterized by affective, understanding, specific, exploratory, and movement responses. The dimen- sions are defined as follows: (a) the affective-cognitive dimension indicates whether a client is referring to his own feelings or to content-oriented concerns; (b) under- standing responses communicate to the counselor that the client understands the nature of his own communication or is clearly seeking enough information of either a factual or affective nature to gain such understanding; (0) speci— fic responses indicate whether the client is delineating his problems by focusing on his core concerns rather than on peripheral issues; (d) exploratory responses tell whether a client is exploring his cognitive or affective concerns; they demonstrate his latitude and involvement in what he is discussing. The final dimension, movement- non—movement, is a global rating of the judge's profes- sional impression of the quality of the client's responses; that is, how important is the communication in arriving at or working with the client's own concerns. This rating is not based upon how the response has been judged along the above-mentioned dimensions, but on how well the client may be differentiating stimuli, lowering his defenses, or owning responsibility for his actions and discomfort. This scale is similar to the Counselor Verbal Response Scale and different from other rating scales in 39 that it required judges to rate on a series of individual client—counselor verbal units during the course of an interview rather than on global ratings of entire inter- views or longer interview segments. The judge is there— fore required to describe every response which the client makes to a counselor verbalization on each of the five dimensions of the scale. The fourth measure, the case summary, was written by each of the three judges, and it described the client's movement throughout the rated segment in an overall global fashion. In Study II, the Counselor Verbal Response Scale was used pre and post treatment to detect change in coun— selor effectiveness. For purpose of scale validation, other measures were included which were expected to corre- late with the CVRS scores. The Affective Sensitivity Scale (A.S.S.) Form B Developed through IPR research (Campbell, 1967), the Affective Sensitivity Scale is an instrument designed to measure one's ability to detect and describe the immed- iate affective state of another. Kuder-Richardson Formula #20 reliability is .76 for a group of “00 individuals, mainly practicing school counselors and groups of people entering counselor education programs. The Scale measures changes in an individual's affective sensitivity as it is 40 associated with participation in a counselor training program or intensive group sensitivity training. A mod- erate relationship exists between scale scores and sub- jective measures of counseling effectiveness; a high affec- tive sensitivity score may be a necessary but not a sufficient condition for counselor effectiveness. Peer and Staff Ratings At the end of their second year, each member of the class of medical students ranked the others in hierarchi- cal order on whom he would go to in order to discuss a personal problem. For this type of ranking to be effec- tive, the individuals must be a fairly collective unit; and the class met this criterion. Measures of a similar nature were made in the beginning of the second year, and these were somewhat different, simply asking each medical student for one person with whom he would want to talk over a personal problem. Faculty members who had contact with the class were asked to rank the individuals as to whom they would refer someone to if they wished the person to be seen by a sen- sitive, compassionate physician. Both rating scales may be found in Appendixes D and E. Medical College Admissions Test The science score of this test was used as it is heavily relied upon as a screening device for acceptance to medical school. Al Judging of the Criterion Tapes For the judging of counselors in Study I, nine doc— toral students in both counseling and clinical psychology, all with prior counseling experience, served as judges for the rating of the criterion tapes. To insure maximum commonality in rating for all judges, each judge was trained in the use of the CVRS with a series of pre- selected counseling tapes. During the training sessions, the tapes were stopped at various points and the ratings of the judges were compared and discussed. This procedure was continued until there was ample agreement among the judges. The video-tapes were randomly presented to the judges during the rating sessions; and the procedures for rating tapes by Griffin (1966) were used.* Tapes were started at the fifth minute of the interview. Prior to rating, judges viewed two or three minutes of the tape in order to gain an understanding of the pace and content of the interview and to become accustomed to the sound and visual quality of the videotape. Judges then rated twenty consecutive counselor responses. For the ratings of client movement, three doctoral students in counseling participated. One of these raters, the investigator, had participated in the previous rat- ings. All ratings were made on a blind-random basis, and these raters had no knowledge of the counselor—ratings *See Appendix F. A2 made previously. A similar training procedure to the above was pursued.* Following the rating of each tape, each judge then wrote a short case summary, encapsulating the movement or lack of movement which the client made. Tapes of the medical student interviews were judged on the CVRS by the same three raters used in the analysis of client movement. The usual training procedure was fol- lowed, and a similar segment of the interview was rated for each tape. Rating began after approximately five minutes of the interview had elapsed. The twenty counselor- statement segment (forty-statements in all) covered approx- imately a six to fifteen minute period, depending upon the pace of the interview. For most sessions, this twenty- statement segment covered the middle and terminal phases of the interview. Reliability In determining interjudge reliability, Griffin applied Ebel's analysis of variance technique, the intra- class correlation (1950), to test the agreement (reliabil- ity) among two sets of three judges who had rated the videotaped interviews of fifty inexperienced M.A. candi- dates in Counseling and Guidance and thirteen experienced Ph.D. candidates in Counseling and Guidance at Michigan State University. Corresponding four—minute segments were rated for fifty—threeof the counselors who had interviewed *See Appendix G. ”3 the same coached client. Because there were unequal num- bers of responses in this last set of scores, all ratings were pro-rated to twenty-statement segments. Griffin reported coefficients of average tape interjudge relia— bility of .8“, .80, .79, .68, and .79 for the affective- cognitive, understanding-non-understanding, specific-non- specific, exploratory-non-exploratory, and effective-non- effective dimensions of the scale respectively. A series of studies was then undertaken to deter- mine whether similar results could be obtained using the same videotapes with different raters as well as with different counseling tapes. Three separate sets of coun- seling tapes were rated using different combinations of judges for each rating: (1) raters judged six audio tapes selected from the department of counseling tape library representing counselors at different levels of prepara- tion, (2) raters judged fourteen tapes of initial inter- views of high school counselors, and (3) the fifty-three videotapes developed by Ward (1966), used in the initial IPR study (Kagan, g§_§1., 1965) and rated by Griffin, were re—rated by a different set of judges. Similar reliabil- ity indices to those reported as having been found by Griffin were found in this series of three studies. For this investigation as with the others, Ebel's intraclass correlation technique was used. This technique offers the reliability of each judge's rating as well as AM the average ratings for each set of judges across tapes and across individual responses. Since the total number of responses for a tape used used as the unit of analysis for either counselor effec- tiveness or client-movement, the across tape (interview) reliability rather than the reliability of individual responses was calculated. Also, for every twenty-statement segment which was rated, the average rather than the sum of the three judges' responses was employed; and these average ratings were used as the criteria for testing the hypotheses. The reliability of raters for pre and post tapes for the group of counselors in Study I is presented in Table 3.2; reliability of raters for the thirty clients involved in Study I on the Client Verbal Response Scale is presented in Table 3.3; reliability of raters for the seventeen medical students involved in Study II on the Counselor Verbal Response Scale is presented in Table 3.u A review of Tables 3.2, 3.3, and 3.“ indicates that all ratings are sufficiently reliable for further analy- sis, ranging from the lowest rating of .89 to .97. Analyses of Data The following statistical analyses were performed on the data which emerged from Study I: “5 TABLE 3.2.—-Intraclass correlation reliabilities of aver- age ratings calculated for judges' ratings of counselors across tapes on each dimension of the CVRS Intraclass Correlation Dimension Aff. Und. Spec. Exp. Eff. Tapes N Cog. N. Und. N. Spec. N. Exp. N. Eff. Averaged Rat— ings, Pre-post 72 .93 .95 .92 .9“ .9“ combined Legend: Aff.—Cog.=affective-cognitive Und.-N.Und.=understanding-non-understanding Spec.-N.Spec.=specific-non-specific Exp.—N.Exp.=exploratory-non—exploratory Eff.-N.Eff.=effective-non-effective TABLE 3.3.--Intraclass correlation reliabilities of average ratings calculated for judges' ratings of clients across tapes on each dimension of the Client Verbal Response Scale Intraclass Correlation Dimension Aff. Und. Spec. Exp. Move. Tapes N Cog. N. Und. N. Spec. N. Exp. N. Move. Averaged Rat- ings 3o .95 .93 .93 .93 .93 Legend: Move.—N.Move=movement—non-movement TABLE 3.“.--Intraclass correlation reliabilities of average ratings calculated for judges' ratings of medical interview- ers across tapes on each dimension of the Counselor Verbal Response Scale. Tapes N Aff. Und. Spec. Exp. Eff. Cog. N. Und. N. Spec. N. Exp. N. Eff. Averaged Pre- tape Ratings 17 1'11* '91 -86 .91 ~86 Averaged Post— tape Ratings 17 '97 '89 '97 -9LI -9LI *Reliability was over 1.00 because of the large number of zero ratings. “6 1. For each dimension of the Counselor Verbal Response Scale, the thirty counselors were ranked from low to high. The top and bottom thirds were separated at appropriate breaking points. The clients of the upper and lower counselor groups were compared by using 2 tests on each category of the Client Verbal Response Scale. 2. The brief case summaries of the clients were pre- pared and given to three trained clinicians in a random, blind manner. Judges will rate each summary on the basis of whether the client either did or did not make therapeu— tic movement. These judgments were then statistically com- pared by the Fisher Exact Test (Siegel, 1956) according to the following groupings: Client Client Movement Non-Movement Counselor high effectiveness Counselor low effectiveness 3. A 3 test between clients of effective and clients of non—effective counselors was computed, testing for dif- ferences on the Wisconsin Relationship Orientation Scale. “. For the total sample of clients and counselors, intercorrelations of the ratings of counselor responses with that of the client responses were computed. 5. A multiple regression equation with the depen- dent variable of counselor effectiveness and independent variables of the other four dimensions on the Counselor “7 Verbal Response Scale were computed. Beta weights were analyzed to determine which of the categories, affect, specific, understanding, or exploratory, load more heavily on the effectiveness factor. This will determine the cate— gories on which clinicians, in using this scale, seem to place the greatest amount of weight. A similar analysis was conducted for the client responses with the dependent variables of movement being predicted by the four other categories. The following analyses will be made for Study II. 6.- Ratings on the CVRS were gathered on the pre and post medical student interviews with a coached patient. The first and fourth interviews were rated when available. There are a number of recordings that were lost; provision was made to rate those students on other sessions for which there is information, as the first and third, or second and fourth interviews. Analysis will be computed separately for those students not having the appropriate criteria ses- sions. The statistic used was the t—test for correlated samples. This test is apprOpriate where the same set of sub- jects is tested twice and the investigator wishes to deter- mine whether the mean obtained on the second testing differs significantly on the first testing. Because hypotheses related to analyses 1, 2, 3, “, and 6 are directional, a one-tailed test of significance at the .05 level was used “8 in all cases. Analysis #5 is descriptive and used the two-tailed .05 level. Hypotheses The specific hypotheses generated by this study are presented in research form. H 1: Clients whose counselors more often refer to affective components of the client's communi- cation will more frequently make affective responses than will clients whose counselors less often refer to affective components. Clients whose counselors more often communi- cate understanding of the client's basic communication will more frequently indicate understanding of themselves than will clients whose counselors less often indicate under- standing of the client. Clients whose counselors more often make re- sponses that delineate the client's problem will make more responses which serve to delin- eate the problem than will clients whose coun- selors less often serve to delineate the problem. Clients whose counselors more often encourage the client to explore his cognitive or affective con— cerns will make more exploratory responses than will clients whose counselors less often encour- age the client to explore his concerns. Clients whose counselors more often make ef— fective responses will make more responses indic- ative of movement than will clients whose coun— selors less often make effective responses. Clients whose counselors more often make effec- tive responses will be judged on the basis of case summaries as having made more movement than will clients whose counselors less often make effective responses. Medical students will make a greater amount of affective, understanding, specific, exploratory, and effective responses following a period of interviewer training using live simulated pa— tients and simulated videotape playback. “9 Descriptive Analyses H8: Some categories on the CVRS will bear a greater relationship to the effectiveness dimension than will the other categories. H : Some categories on the Client Verbal Response Scale will bear a greater rela- tionship to the movement dimension than will the other categories. Summary In the first study, a group of thirty Master's level counselors conducted counseling sessions with high school age clients. The sessions were audiotaped and then rated along the dimensions of the Counselor Verbal Response Scale and the Client Verbal Response Scale by different sets of professional judges. This procedure was followed in order to evaluate client movement as a result of coun- selor effectiveness and to examine the inter-relationship of the rated categories. An analysis was performed to specify which categories of the CVRS could best predict counselor effectiveness, and which categories of the Client Verbal Response Scale would best predict client movement. In addition, each counselor's client filled out the Wis- consin Relationship Orientation Scale to evaluate the strength of the relationship (as perceived by the client) as related to the counselor level of effectiveness. In the second section of this investigation, an attempt was made to demonstrate the effects of a technique to instruct medical students in interviewing skills, a 50 method of stimulated video feedback through recall with the use of coached patients, and to further validate the Counselor Verbal Response Scale. Analyses were made of the changes in interview behavior from pre to post treat- ment. Peer and staff ratings, a test used for admission to medical school, and the Affective Sensitivity Scale were administered. The relationships among these measures were then analyzed in support of the validity of the CVRS. CHAPTER IV ANALYSIS OF THE DATA In Chapter IV, an analysis of the data is pre— sented based upon the methodology and statistical treat- ment outlined in Chapter III. Audiotapes were rated for both counselor and client response by professional judges in Study I. Each client also rated the counselor after the interview, using the Wisconsin Relationship Orienta- tion Scale. In the portion of the study dealing with the medical school sample, pre and post videotapes were rated on the Counselor Verbal Response Scale. In addition, scores were obtained on the Medical College Admissions Test, the Affective Sensitivity Scale, ratings as to whom in the class each student would seek to discuss a personal problem and ratings by the staff of a student's sensitivity. Results of the analysis of this data are reported in the following sequence: 1. differences between scores on the Client Ver- bal Response Scale and on the Wisconsin Rela- tionship Orientation Scale for clients whose counselors maintained "high" levels of condi- tions versus clients whose counselors main- tained low levels of conditions on the Coun- selor VRS; 2. interrelationships among the variables of the Client VRS, the Counselor VRS, and the WROS; 51 52 prediction of client movement on the basis of the four other categories of the Client VRS; prediction of counselor effectiveness on the basis of the four other categories of the Counselor VRS; comparison of case summaries of those clients ' whose counselors were rated as being more effective versus those clients whose counse- lors were rated as being less effective; comparison of pre-post medical student scores on the Counselor VRS; intercorrelation of the Counselor VHS and the other collected measures; and prediction of peer and staff ratings on the basis of the Affective Sensitivity Scale, the test of medical school science aptitude, and the scores of the Counselor VRS. Difference in Client Response py Counselor Level The null hypotheses for the comparison of the two groups of clients were: H O 2: There will be no difference in the number of affective statements made during the rated portion of the interview between clients whose counselors more often refer to affec- tive components of the client's communication and clients whose counselors less often refer to affective components of the client's com— munication. Symbolically: HOL; M1 = M2 HA1: Ml > M2 Legend: Ml: mean on the affective dimension of clients whose counselors more often refer to affect. M2: mean of clients whose counse- lors less often refer to affect. 53 There will be no difference in the number of statements indicating self-awareness between clients whose counselors more often communi- cate understanding of the client's basic con- cerns and clients whose counselors less often communicate understanding of the client's basic concerns. Symbolically: H02: M1 = M2 HA2: Ml > M2 Legend: Ml: mean of clients whose counse- lors more often communicate their understanding. M2: mean of clients whose counse- lors less often communicate their understanding. There will be no difference in the number of statements indicating a client's focus on his problems between clients whose counselors more often make responses that delineate the client's problem and clients whose counselors less often make responses that delineate the client's problem. Symbolically: H03: M1 = M2 HA3: Ml > M2 Legend: Ml: mean on the specificity dimen— sions of clients whose counse- lors more often delineate the problem facing the client. M : mean on the specificity dimen- sion of clients whose counse- lors less often delineate the problem facing the client. There will no difference in the number of exploratory statements between clients whose counselors more often encourage the client to explore his cognitive or affective con- cerns and clients whose counselors less often encourage the client to explore his cognitive or affective concerns. H 6: 5“ Symbolically: HO“: M1 = M2 HA“: Ml > M2 Legend: Ml: mean on the exploratory dimen- sion of clients whose counse- lors more often encourage the client to explore his concerns. M : mean on the exploratory dimen- sion of clients whose counse- lors less often encourage the client to explore his concerns. There will be no difference in the number of responses indicating client movement between clients whose counselors more often make effective responses and clients whose counse- lors less often make effective responses. Symbolically: H05: M = M l 2 HA5: Ml > M2 Legend: Ml: mean of clients on the move— ment dimension whose counse- lors more often make effective responses. M2: mean of clients on the move- ment dimension whose counse- lors less often make effective responses. There will be no difference in scores on the Wisconsin Relationship Orientation Scale between clients whose counselors more often make effective statements and clients whose counselors less often make effective state- ments. 0 l 2 Symbolically: H 6: M = M HA6: M M 1>2 Legend: Ml: mean of clients on the WROS whose counselors more often make effective statements. mean of clients on the WROS whose counselors less often make effective statements. 55 Six 3 tests were computed to examine the above hypo- theses, testing differences between the two groups of clients on the five dimensions of the Client VHS and on the WROS. The results of these analyses are summarized in Table “.1. TABLE “.1.--T tests of differences on the Client Verbal Response Scale and the WROS between clients of counselors who maintained high levels of the matching dimensions on the Counselor Verbal Response Scale.+ Counselor Levels pass... N N 32:22:“ Affective 10 8.23 10 5.33 1.81 1.60 Understanding 10 11.00 9 8.30 3.69 0.73 Specific 11 9.15 11 7.85 1.9“ 0.67 Exploratory 11 10.15 9 7.81 2.03 1.15 Movement 11 9.00 11 7.63 1.73 0.79 WROS ll “.72 11 2.28 0.““ 5.55* *Significant at the .001 level +See Appendix H for counselor level means While all the results fell in the expected direction, the only significant difference is that found on the Wis- consin Relationship Orientation Scale. From this analysis, the relationship of client response along the affective- cognitive, understanding-non-understanding, specific-non- specific, exploratory-non-exploratory, and movement-non- 56 movement dimensions, to similar responses along the Coun- selor VRS, ratings of the counselor's behavior, are not confirmed. It is apparent, however, that the client's feelings about the relationship are related to the level of counselor effectiveness. Those clients of more effec- tive counselors feel they could talk to their counselors about almost anything, while those clients of the less effective counselors would rather avoid personal exposure and not seek out the counselor.* Intercorrelations of the Scales Table “.2 contains the intercorrelations of the Client VRS, the Counselor VRS, and WROS, for the full sam— ple of 30 clients and 30 counselors. The correlations among the variables of the Client VRS are very high (.76 to .96), as are the correlations among the variables of the Counselor VHS (.77 to .95). The correlations of the scales with each other are low to moderate (.06 to .3“). The Client VRS, which measures the client's behavior throughout the interview, bears low to moderate relation- ship to the WROS (.16 to .32). However, the Counselor VRS, the counselor's rated behavior, bears a high rela- tionship to the WROS. While it is not confirmed that the counselor's rated behavior does have an effect upon the client's similarly rated behavior, it appears to be *See Appendix C for the Wisconsin Relationship Orientation Scale. 57 mamom coapmpcmfino aacchHpmamm camcoomfiz u momz mmmcm>fipommmm poaomcsoo u mmmIoo pcoEo>oz unmaao u o>ozIHo mnoumcmadxm Noammcsoo u axMIoo ANOBmpoaaxm pcmfifio u meIHo camaomdm noaowcsoo u oQOIoo oaufioodm pCmAHo u oonIHo mcflocmpmnmocb noammcsoo u oCDIoo wcflocwpmpmoc: pcoafio u UQDIHO pommma noaomcsoo u mmoz I Ho oo.H mm. mm. as. dam I Ho oo.H em. aw. 666m I Ho oo.H we. on: I Ho oo.H mm< I Ho mmmIoo oxMIoo oonIoo UEDIoo mm¢Ioo o>ozIHo dmeHo omdmIHo UCDIHQ mm pcmfiao esp .mamom oncodmmm.amnpo> noammczoo on» mo :ofipmHmNLoomopsHII.m.z mqm¢e 58 confirmed that the counselor's behavior does have an effect upon how the client views the counselor, regardless of the specific behavior of the clients during the ses- sion itself. Predicting Client Movement from the Other Scale Categories A multiple regression equation was constructed in which the client movement category was the dependent var- iable and the other four categories of the Client VRS the independent variables. Since the movement dimension is a global judgment of a client's response, this analysis was computed to determine which of the other four categories were most important to such movement. In the initial equ- ation all categories were included; a new equation was then formed deleting the weakest variable. This procedure was followed until the remaining variables had beta weights which were significantly different from zero at the .05 level. TABLE “.3.--Multiple regression equation predicting client movement from the other measures of the Client Verbal Response Scale after deletion of variables with non- significant beta weights.* Under- Explora- Dimension Affective standing Specific tory Multiple B B B B B Movement delete 0.567 delete 0.“13 .97 *All betas significant at the .05 level. 59 The understanding and exploratory categories are related highly to client movement, the multiple correla- tion being .97. The affective and specific dimensions were deleted. It should be kept in mind that all the categories are correlated with movement at very high levels; and that, therefore, the multiple equation does not add much more information. Either of the two varia— bles left after the deletion could be used in the place of the movement dimension. Predicting Effectiveness from the Other Counselor VRS Categories A regression equation similar to the one above was computed for the Counselor VRS. TABLE “.“.-—Multiple regression equation predicting coun- selor effectiveness from the other measures of the Coun- selor Verbal Response Scale after deletion of variables with non-significant beta weights.* . Under- Explora- Dimension Affective standing Specific tory Multiple B B B B R Effective- ness 0.381 0.298 delete 0.3“0 .98 *All betas significant at the .05 level. In this equation, the multiple correlation of the affective, understanding, and exploratory categories with effectiveness was .98; the specific dimension was deleted. 60 Again, the intercorrelations of the Scale are quite high; not much more can be added by compounding the variables. It should be noted, however, that close to perfect predic— tion of judges' ratings of effectiveness can be accounted for by the use of three of the four other scale variables. Comparison of a Global Measure of Movement by Counselor Effectiveness The hypothesis for this analysis is: H07: There will be no difference in the frequency of clients judged as having made therapeutic movement on the basis of case summaries between clients whose counselors more often make effective responses and clients whose counselors less often make effective respon- ses. Symbolically: H07: Fl = F2 HA7: Fl > F2 Legend: Fl: frequency of clients making movement whose counselors more often make effective responses. F2: frequency of clients making movement whose counselors less often make effective responses. Brief session summaries were prepared by each judge following each rated tape. These summaries note the qual— ity of the counselor-client interaction.* Each summary was then ranked by three additional doctoral students as to whether the client did or did not make therapeutic movement, and a single ranking was agreed upon. The com- parison of differences in client movement between clients *See Appendix I. 61 of counselors maintaining high effectiveness and clients of counselors maintaining a low level of effectiveness is found in Table “.5. TABLE “.5.--Fisher Exact Test of differences in movement between clients of counselors who maintain a high level of effectiveness and clients of counselors who maintain a low level of effectiveness. Counselor Levels High Low Total Movement 6 6 12 Non-movement 5 5 10 Total 11 11 22 p = not significant The results of this global analysis support the results found in using the Client VRS. That is, there appears to be no difference in the movement of clients of the high and low level counselor groups. Since 12 of the 22 summaries were indicative of movement (as rated by clinical judges who were doctoral students in counseling), it can be concluded that some of the clients whose coun- selors were all M.A. students in counseling, were indeed making what could be considered therapeutic movement. The clinical summaries after each tape rating bring the variation found in Table “.5 into focus. The behavior 62 of clients under each of the two conditions of counselor functioning was quite different. An interpretation of these summaries is as follows: 1. High counselor leve1—-client movement: These counselors functioned well, appropriately prob- ing and reflecting. They were quite active and seemed to make leading statements which helped the client to explore. The clients reacted well to this condition; they were able to express affect and to explore well. 2. High counselor level--1ack of client movement: These clients reacted to the high activity of the counselor with defensiveness and resistance. The counselor moved in quickly to the client's concerns and then spent most of the session responding to the anxiety of the client gener- ated by the counselor's behavior. 3. Low counselor 1evel--client movement: In this situation the counselor tended to remain rather inactive and responded primarily on the cogni- tive level. Some counselors appeared to show signs of annoyance and hostility. At times, some engaged in lecturing behavior. The client tended to make some degree of movement in spite of the counselor's behavior. Clients sometimes either ignored the counselor's verbalizations or responded to them but then added more perti- nent information. The counselors did not gen- erally deal with what the client way saying. “. Low counselor 1evel--1ack of client movement: These sessions were typified by passive counse- lors who responded very infrequently and/or only dealt with fairly irrelevant details. The tenor of these sessions was a conversation on a superficial level, with little productivity, both parties seemingly content to maintain such conversation. It is interesting to note that those clients who did appear to be defensive in this first interview rated their counselor as someone whom they could talk to, someone in whom they had confidence. In other words, it appears that 63 clients whose counselors were highly rated, regardless of the client's behavior during the session, had the feeling that they could talk with the counselor about almost any- thing, as rated on the WROS. Moreover, clients of the lower-rated counselors would either not seek out the coun- selor or would refrain from discussion of their more per- sonal concerns. Study II This section of the investigation is concerned with testing the effects of a training procedure involving coached clients and stimulated videotape recall and with validating the Counselor VRS on the basis of the inter- viewing behavior of medical students. Scores from the rated tapes will be related to a medical school admissions test, peer and staff ratings of the student's ability to deal with the personal problems of others, and a test of empathy known to predict counselor success. Pre—Post Differences for Medical Students on the Counselor Verbal Response Scale The null hypotheses for this section of the study for the counselor VRS are as follows: H08: There will be no difference on the affective- cognitive dimension between the pre- and post- treatment scores for the medical school sample. Symbolically: H08: M1 = 2 M M HA8: 1 > M2 10: ll: 6“ Legend: Ml: mean of post-treatment scores on the affective-cognitive dimensions. M2: mean of pre—treatment scores on the affective-cognitive dimen- sion. There will be no difference on the understand- ing-non-understanding dimension between the pre- and post-treatment scores for the medical school sample. Symbolically: H09: M1 = M2 HA9: Ml > M2 Legend: M : mean of post-treatment scores 1 on the understanding-non- understanding dimension. M2: mean of pre-treatment scores on the understanding-non- understanding dimension. There will be no difference on the specific- non-specific dimension between the pre- and post-treatment scores for the medical school sample. Symbolically: H010: M1 = M2 HAlO: Ml > M2 Legend: Ml: mean of post-treatment scores on the specific-non-specific dimension. M2: mean of pre-treatment scores on the specific—non-specific dimension. There will be no difference on the exploratory- non-exploratory dimension between the pre- and post-treatment scores for the medical school sample. Symbolically: H011: l M M HAll: Ml > M2 2 65 Legend: M : mean of post—treatment scores on the exploratory-non- exploratory dimension. mean of pre-treatment scores on the exploratory-non- exploratory dimension. H 12: There will be no difference on the effective- non-effective dimension between pre- and post- treatment scores for the medical school sam- ple. Symbolically: H012: M1 = M2 HA12: Ml > M2 Legend: Ml: mean of post-treatment scores on the effective-non—effective dimensions. M2: mean of pre-treatment scores on the effective-non-effective dimensions. Five 3 tests for correlated data were computed to examine the above hypotheses, testing differences between pre- and post— treatment mean scores on the five dimen— sions of the Counselor VRS. It should again be noted that the ratings were conducted on a random blind basis. The results for the total sample of fifteen students on whom there was available data appears in Table “.6. Table “.7 contains the scores and analysis for the group of students having the first and fourth sessions available for rating; Table “.8 contains scores for those students having the second and fourth sessions available; Table “.9 contains the scores of those students having the first and third tapes available for rating. TABLE “.6.--T tests between pre and post sessions for med- ical students on the Counselor Verbal Response Scale-~Total Group Dimension N=15 Pospeiission PreMgzision Spipgird t Affective 2.16 1.28 0.50 1.76 Understanding 8.53 6.67 1.02 1.9“* Specific 5.59 3.85 1.21 1.““ Exploratory 6.“1 “.9“ 0.98 1.50 Effective 6.26 “.73 0.9“ 1.63 1‘Significant at the .05 level TABLE “.7.--T tests between pre and post sessions for medi- cal students on the Counselor Verbal Response Scale-—lst and “th sessions group Post Session Pre Session Standard Dimension N=8 t Mean Mean Error Affective 2.20 1.61 0.80 0.7“ Understanding 8.8“ 7.22 1.90 0.85 Specific “.5“ 3.91 1.78 0.35 Exploratory 6.“2 5.63 1.52 0.53 Effective 6.00 5.29 1.61 0.71 TABLE “.8.--T tests between pre and post sessions of medi- cal students on the Counselor Verbal Response Scale--2nd and “th sessions Dimension N=6 Post Session Pre Session Standard t Mean Mean Error Affective 3.00 0.83 0.u8 14.50+ Understanding 8.61 5.22 1.12 3.01+ Specific “.89 2.50 1.83 1.31 Exploratory 6.78 “.17 1.70 1.53 Effective 6.83 3.77 1.28 2.39* +Significant at the .01 level *Significant at the .05 level TABLE “.9.--T tests between pre and post sessions of medi- cal students on the Counselor Verbal Response Scale--1st and 3rd sessions. Dimension N=3 Pospeiission PreMgzision Spgpgird t Affective 0.33 1.21 1.15 -0.77 Understanding 7.56 7.“5 1.82 0.06 Specific 6.““ 3.00 “.10 0.8“ Exploratory 5.67 “.67 2.“0 0.“2 Effective 5.78 5.11 2.0“ 0.39 From an examination of the tables, it can be seen that all differences are in the predicted direction, except for one comparison of the group of three individuals, rated on sessions one and three only. For the total sample, dif- ferences on the understanding-non-understanding dimension are significant. A breakdown into sub-groups was made in order to examine the extent to which the particular ses- sions used for comparison may have weighted the final results. It appears that the comparisons made of the sec- ond to the fourth session lent the most weight to the scores for the total group, with significant differences on the affective, understanding, and effective dimensions. It is apparent that the post session scores across all three sub-groups were similar; the group being rated on the sec- ond and fourth sessions had lower ratings on the pre-test, perhaps indicating that the actress they were interviewing was particularly difficult to converse with. This 68 contention is affirmed by the overall perceptions of the raters, who mentioned that this actress delivered rather short, curt, and somewhat hostile responses and would not carry any of the interview on her own. For the total group, there was a difference from pre- to post-treatment on the understanding-non-understanding dimension. Over the course of treatment, each of the med- ical students communicated his understanding to the patient more frequently. It might be implied from this that the interviewer was more able now that at the outset of the training, to understand the patient and to commun- icate this understanding to the (coached) patient. No comparisons were computed between these medical student scores and scores taken from the sample of counse- lors in a Master's practicum. The average rating for this sample can be found in Appendix H. Generally, the scores of the counseling sample are higher than those of the medical student sample, as would be expected; the counse- lor group would have had more background and training in interviewing and would thus be more attuned to the dimen- sions used in rating them. The possibility that the focus of the medical student and counseling student might be different will be discussed in Chapter V. 69 Intercorrelation of Variables to Establish Validity A number of variables other than the five dimensions of the Counselor VRS were collected from the medical school sample: the Science subtest of the Medical Col- lege Admissions Test, the Affective Sensitivity Scale, and a before-treatment hierarchical rating by the students themselves of whom in their class they would go to in order to discuss a personal problem; also, another hierar- chical rating by the students of whom they would see to discuss a problem, and a rating by the staff of which stu- dent they might refer a patient to, were gathered after the training period. The ratings are considered a measure of success in interviewing. The staff rating was divided into two parts--ratings by pre-clinical and ratings by clinical faculty. Although subjective, this type of oper- ational rating is a good predictor of success at this level of training (Stefflre, 196“). In was believed that the test of affective sensiti- vity (an empathy scale) would be positively related to the rating measures as well as to the dimensions of the Coun- selor VRS; and assuming the subjective rankings to be a criterion variable, that empathy, plus the dimensions of the Counselor VRS, would permit accurate prediction. The Science subtest of the MCAT was included because it is heavily relied upon for admission to medical school. 70 Assuming the doctor-patient relationship to be an import- ant outgrowth of medical school training, the relation- ship of the MCAT to actual interviewer behavior and rat- ings of effectiveness as a person to whom individuals can turn to discuss their problems, should be explored. The interrelationships of these variables are found in Table “.10. The correlations found here are for the most part contrary to expectations. It was expected that the Affec— tive Sensitivity Scale would be positively related to both the peer and staff ratings. These relationships were not found; the ASS has a correlation of .11 with the peer ratings made in the Fall, of .05 with the Spring peer ratings, of —.0“ with the staff ratings of pre-clinical faculty, and of —.17 with clinical faculty. The ratings themselves appear to be consistent. The correlation of the Fall rankings with the Spring rankings is .6“, with pre-clinical staff ratings, .“6, but only .28 with clin— ical staff. The correlation of the Spring rankings with the pre—clinical staff rankings (also made in the Spring) is .69, and with the clinical staff, .“5. The Spring peer rankings and both sets of staff rankings would, therefore, be more appropriate indicators of outcome than would the Fall rankings. The intercorrelations of the Counselor VRS within itself are high, and the Scale bears either low or negative correlations with all the class ranking measures. 71 .pom mpmp some CH po>Ho>cH washes: uconmmmfip can go mmBMOmn mammanm> anHmEm esp mo 2 one Heads Hafiz cofipwamnnoo mam now 2 039* mm. mm. mm. em. mm.I OH.I mH.I Hm.I eo.I H:.I mfi.o ma msapoecum oo.H mm. mm. am. Hm.I HH.I 0H.I Hm.I oo. Hm.I mm.m ma shoemnoaaxm oo.H am. am. mm.I HN.I mm.n o:.I mo. Hm.I mm.m ma eacfiomam oo.H mm. mm.I OH.I mm.I ::.I SH.I mm.I A:.m 0H mefiecepmseeea oo.H mm.I mo.I mH.I mm.I mH.I ao.I mo.m ma esfipomeea oo.H mH.I =O.I mo. Ha. :m.I mo.m: mm snpmaem oo.H :5. ma. mm. mH.I me.m em emceemm , meeem Heeficaao oo.H me. we. AH.I wm.m mm mwcaeem heeem oo.H so. om.I mm.m em sweepem wcanam pCmUSBm oo.H :N.I mm.H om mmeaemm flame eceespm oo.H oo.mmm mm Baez axm 66am ems ee< mmwm emmww eceem wmmwmm mmmm e hoammczoo on» cam .Asneeaemv maeom spfi>aeamcem msauoeeu< .mwefiemm endgame Heoaeeao e Heoacfifio Imam .mwefieem seem weasam eee flame .eaoz we» go chASmHeseoohechII.OH.z mqmae 72 The Science score from the MCAT bears low negative corre- lations to all the other variables. It was thought that the tape ratings along the dimen- sions of the Counselor VRS and the Affective Sensitivity Scale, would bear positive relationships to peer ratings (a measure of students' feelings of whom they could dis- cuss personal problems with), and staff ratings of whom in the class they would refer individuals to if they wished a sensitive physician. The ASS has been found to relate to peer and staff ratings in samples of counseling students (Kagan, gp_§1., 1967) .“2 and .52 respectively; the Counselor VRS has been found to have a low positive relationship to peer and staff ratings, .23. These find- ings were not upheld, however, on the medical student sample. A discussion of the meaning and possible reasons for this lack of relationship may be found in Chapter V. Multiple regression equations with stepwise dele- tion of variables were computed in an attempt to develop the best possible prediction of both peer and staff rat- ings. There were no instances in which the variables Predicted the criterion to the degree that any of the beta 'Weights would be significant from zero. That is, the tintercorrelations are too low and unstable to provide for any reliable prediction; each stepwise deletion proceeded t0 the point where there were no variables remaining. 73 Summary The first set of hypotheses postulated in this study dealt with differences between rated responses on the Client VRS of two groups of clients. The clients were divided into two groups on the basis of whether their coun- selors maintained either high or low levels of the match- ing variables on the Counselor VRS. The composition of the two groups for the testing of each hypothesis was prac- tically identical, due to the high intercorrelations of the Scale. The two groups of clients had also rated their counselors on the Wisconsin Relationship Orientation Scale. While no differences were found between the two client groups on the dimensions of the Client VRS (on the affec- tive, understanding, specific, exploratory, and movement dimensions), the two groups were clearly differentiated on the WROS. The group having the counselors classified as being more effective were more favorably disposed toward their counselors than were the clients meeting with the less effective counselors. An intercorrelation matrix was calculated among all the variables of both rating scales and the WROS. Each scale correlated highly within itself, but had low corre- lations with the others. The dimensions of the Counselor VRS correlated highly with the WROS, indicating a connec- tion between the counselor's behavior and the relationship as perceived by the client. 7“ A descriptive analysis of the Client VRS showed the understanding and exploratory dimensions to be the most important in predicting movement. For the Counselor VRS, only the specific dimension was deleted; the affective, understanding, and exploratory dimensions were highly related to effectiveness. In the analysis of the medical school sample, the understanding dimension of the Counselor VRS served to show the effects of the treatment, which consisted of coached patients and the medium of stimulated video playback. In an attempt to provide for further validation of the Counse- lor VRS, the Science score of the Medical College Admis- sions Test, the Affective Sensitivity Scale, and the five dimensions of the Counselor VRS, were correlated with peer and staff ratings of the second-year medical students. The correlations of the ASS with peer and staff ratings and with the Counselor VRS dimensions were found to be either low or negative. The Counselor VRS also bore either low positive or low to moderate negative correla- tions with the peer and staff ratings. Because of these findings, there were no multiple regression equations which combined the criteria variables to predict peer and staff ratings with any degree of meaningfulness. CHAPTER V SUMMARY, DISCUSSION, IMPLICATIONS Summary This investigation had a threefold purpose: (1) to study differential client behavior and client perception of his counselor as a result of the counselor's maintain- ing either "high" or "low" levels of therapeutic behavior as rated on the Counselor Verbal Response Scale (Griffin, 1966; Goldberg, 1967; Kagan, 23:31., 1967); (2) to measure increases in the level of therapeutic behavior of second— year medical students who experienced special interview training; and (3) to further validate the dimensions of the Counselor VRS by correlating scores obtained on it with a test used in medical school admission, a measure of affective sensitivity, and peer and staff ratings of a student's effectiveness. While there are a large number of studies dealing with counseling "process," there is no conclusive research which relates client response to the within-interview behavior of the counselor. Previous studies have centered either upon pertinent therapist variables or upon client personality characteristics and their relationship to 75 76 outcome. Viewing the counseling situation in such a light ignores the two individuals as an interactional system. The convergence phenomenon has been viewed as a possible explanation for the interaction of the counselor and client by postulating that the client will move closer to the counselor in values and behaviors over the course of treat- ment. This research was similar to that done in the area of convergence in that it assumed that there would be move- ment by the client toward the therapist in broad areas of verbal behavior which are delineated by the Counselor VRS. It was assumed that if the counselor "moved in" and com- mented upon the client's affect, the client would become more affective; if the counselor communicated to the client his understanding of the client's comments, the client would come to emit responses which communicated understanding of himself; if the counselor focused on what appeared to be the client's problem, the client would move with the counselor and would also become more specific and concrete; and finally, if the counselor verbalized the kinds of statements that should encourage client self- exploration, the client would then make self—exploratory statements. This reasoning led to the hypotheses that were treated. In order to test these hypotheses, a client response scale was developed to correspond to the categories of 77 the Counselor VRS. Reliability estimates were computed and found to be high. Previous research has shown the use of the "coached client" and stimulated videotape feedback each to have been effective in the training of counselors. The present research combines both of these techniques. Moreover, to date, there has not been an adequate effort to relate a process measure of effective counseling with such measures as peer and staff ratings, a test of affective sensitivity, and a measure of academic achievement. The investigation was divided into two parts. In the first study, a group of 30 Master's level counselors conducted counseling sessions with high school age cli- ents. The sessions were audiotaped and then rated along the dimensions of the Counselor VHS and the Client VRS by different sets of professional judges. This procedure was followed in order to examine the interrelationships of the rated categories and to evaluate client movement, one of the dimensions rated by the judges for each client, as a result of counselor effectiveness. An analysis was performed to specify which categories of the Client VRS would best predict the category of client movement. In addition, each counselor's client filled out the Wisconsin Relationship Orientation Scale to evaluate the strength of relationship (as perceived by the client) as related to the counselor level of effectiveness. 78 In the second section of this investigation, an attempt was made to demonstrate the effects of a technique to instruct medical students in interviewing skills and to further validate the Counselor VRS. Treatment consis- ted of each medical student interviewing four different patients over a four week period. The "patients" were simulating different types of individuals with a variety of problems. Following each interview, other students and staff would conduct a stimulated recall of the previous session by use of videotape playback. Analyses were made of the changes in interview behavior from pre- to post- treatment. Peer and staff ratings, a test used for admis- sion to medical school, and the Affective Sensitivity Scale were administered. The relationships among these measures were then analyzed in support of the validity of the Counselor VRS. The first set of hypotheses postulated in this study dealt with differences between rated responses on the Cli— ent VRS of two groups of clients. The clients were divi- ded into two groups on the basis of whether their counse- lors maintained either high or low levels of the matching variables on the Counselor VRS; the composition of the two groups of clients for the testing of each hypothesis was practically identical, due to the high intercorrelations of the Client VRS. The two groups of clients had also rated their counselors on the WROS. While no differences 79 were found betWeen the two client groups on the dimensions of the Client VRS (the affective, understanding, specific, exploratory, and movement dimensions), the two groups were clearly differentiated on the WROS. The group having the counselors classified as being more effective were more favorably disposed toward their counselors than were the clients meeting with the less effective counselors. An intercorrelation matrix was calculated among all the variables of both rating scales and the WROS. The Counselor VRS and the Client VRS each correlated highly within itself, but they had low correlations with one another. The dimensions of the Counselor VRS correlated highly with the WROS, indicating a connection between the behavior of the client and the counselor—client relation- ship as perceived by the client. A descriptive analysis of the Client VRS showed the understanding and exploratory dimensions to be the most important in predicting movement. For the Counselor VRS, only the specific dimensions was deleted; the affective, understanding, and exploratory dimensions were all highly related to effectiveness. In the analysis of the medical school sample, the understanding dimension of the Counselor VRS was found to tap the effects of treatment, which involved coached patients and the medium of stimulated video playback. In an attempt to provide for further validation of the 80 Counselor VRS, the Science score of the Medical College Admissions Test, the Affective Sensitivity Scale, and the five dimensions of the Counselor VRS were correlated with peer and staff ratings of the second-year class of the College of Human Medicine, Michigan State University. The correlations of the Affective Sensitivity Scale with the peer and staff ratings were found to be either low or nega- tive. The Counselor VRS also bore either low positive or low to moderately negative correlations with the peer and staff ratings. Because of these findings, there were no multiple regression equations which combined the criterion variables to predict peer and staff ratings with any degree of meaningfulness. Discussion The assumption that counselors will elicit from their clients responses along the dimensions on which the coun- selors are focusing, seems not to be confirmed, even though the client categories are assumed to indicate positive growth. Moving away from a statement-by-statement analy- sis, the global analysis, via the brief case summary data, also did not confirm the assumption that the better counse- lors would generally induce greater client movement. Could it be, then, that the measures of counselor effectiveness bear no relationship to the behavior of the client? ‘The results of the WROS tend to support the fact that the 81 behavior of the counselor does, in fact, make a differ- ence. If the assumption is to be made that a critical element of effective counseling is the ability of the coun- selor to establish a meaningful relationship, then these results have significance. It would seem likely that the client's perception of the relationship might well serve as a measure of the degree to which the counselor was able to communicate such qualities as empathic understanding and positive regard, a willingness to help, "to get in there and grapple" with the problems and feelings of the client. Those clients who were rated as not moving under the conditions of the high-level counselors, were rated as such because they became resistant and defensive. These counselors were perceptive in hitting at the client's feeling level; they reflected the client's nervousness and verbalized hunches directed toward the client's feelings. While no movement appeared to be made in this, the first session, the fact that a relationship was established sug- gests that movement may take place in further sessions as a result of this relationship and the counselor's function- ing in a mode herein described as effective. At the same time, there appears to be a good deal of variation in client behavior, which must be attributed to individual client differences. 82 A visual inspection of the standard errors involved in the analyses of difference between the client group whose counselors were functioning at a high level and the client group whose counselors were functioning at a lower level, indicates a large amount of variance with the groups. In other words, while all comparisons were in the predicted direction, there was a wide range of scores within each group. A statistical description of these sum- maries showed six clients of highly-rated counselors as having made therapeutic movement and five as not having made such movement; these results were duplicated exactly for clients of counselors functioning at a lower level. There are a number of questions which are raised by the lack of confirmation of theory. Is the Client VRS an appropriate operational measure for judging the various aspects of growth that may take place during a counseling session? The categories do make logical sense. Because counselors are trained to react to a client's affect, to point out the client's feelings (Goldberg, 1967; Kagan, gp_§1., 1967), a client's bringing out of his feelings would seem to be a desired behavior. The counselor is trained to help the client focus on the crucial aspects affecting him, to help the client become concrete rather than abstract in his problem areas. The counselor should allow the client the room to explore his concerns; and client exploration is considered to be desirable behavior 83 (Truax and Carkhuff, 196“). The Client VRS, therefore, at face value, appears to be describing some aspects of client interview behavior deemed important for client growth. There are other factors which are considered important, as, for example, client insight, lowering of defenses, owning discomfort, relating to the therapist, committing to change, differentiating stimuli, and, finally, behaving differently (Kagan, gp_§l., 1967). While these dimensions are also rated from observation of client-counselor interaction, they are the results of the interaction rather than mea- sures of the actual interaction process, which is the intent of both the Counselor VRS and the Client VRS. In this light, the categories of the Client VRS might seem appropriate as intervening variables between the interview and those qualities of client growth listed above. What are, then, possible alternative explanations for the lack of relationship among the rated counselor and client dimensions? The hypotheses assume a form of model- ing behavior to be taking place on the part of the client toward the role of the counselor. That is, good clients may not necessarily act in similar ways to good counselors. The counselor is taught to verbalize his feelings, to discuss the counselor-client relationship openly, as he perceives it. He is taught to try to probe different areas that might be of concern to the client. While the counselor may make advantageous use of silence, he must be 8“ active, quick—thinking, listen closely to the client, while at the same time, attempt to make clinical sense out of the data with which he is confronted by the client. To a counselor who is functioning well in an inter- view, according to the dimensions of the Counselor VRS, a client may respond in a number of ways. He may be some- what passive, mulling over insights he might have had but not yet verbalized, or might simply be thinking over some- thing the counselor had said. The counselor, on the other hand, does not traditionally have this time during the interview for slowly considering the course of events. It may be appropriate for the client to do so, but not for the counselor. Looking at the counselor—client interaction from another perspective, it appears that if the counselor sets Up a particular theme which he feels is important for consideration, the client may not necessarily follow the lead of this theme. From past research (Griffin, 1966; Goldberg, 1967), and from the scores obtained by clients on the Wisconsin Relationship Orientation Scale in this study, the dimensions of the Counselor VRS, by which the counselors' interview behaviors are rated, seem to be appropriate. Other dimensions for rating client behavior may have to be adopted. For the medical school sample, the understanding dimension of the Counselor VRS was the only measure which 85 was found to discriminate between the pre-post behavior related to the simulation training procedure. Although all differences were small, they were in the predicted direction. There was a great deal of variation within the change scores of the group. Perhaps an increased sample size would help to overcome this variation by pro- viding for greater statistical sensitivity. There are other possible alternative explanations for the above findings. It is possible that the simple experience of interviewing may produce gain. Informal impressions by both the students and faculty members, however, mitigate against this as a sufficient interpre— tation. Does the Counselor VRS function as an applicable instrument for measuring the interview behavior of the medical school sample? Positive change on the understand— ing dimension was noted. The Scale would seem somewhat applicable for the treatment goals--he1ping the student to establish a productive and therapeutic relationship, to develop the skills of facilitating emergence of relevant information and feelings, careful listening, observing, systematic questioning, sensitivity to emotional subtleties and empathy (Jason, 1968). These are qualities that counselor-educators would want to develop in their stu- dents. There are differences between medical and counsel- ing interviews which were found in these tapes and which 86 must be taken into consideration in interpreting the find- ings. One of the major differences is that of the neces- sity for diagnosis. Diagnosis is a factor which fits into the medical model but not necessarily that of counseling. Towards this end, the physician may need, during the inter- view, more cognitive behavior on the part of the patient, more answers to direct questions aimed at information gathering, and delimiting the possible alternatives so that a diagnosis might be made. While each of the problems presented to the medical students had some degree of emo- tional overtone, for example, a hostile housewife, a former nurse with a proctological problem, a somewhat guilty young man about to be married, having a probably case of venereal disease, and a pregnant woman-with a history of miscar- riages, the raters had some difficulty in deciding at which points information-gathering was an appropriate activity, and when it was more important to deal with the emotionally- 1aden aspects of the interview. While the physician may recognize a patient's anxie— ties, he might, at certain points in time, choose not to deal with them directly. Rather than encourage explora- tion by the patient in areas quite unrelated to the physi- cal problem at hand, he might choose to direct specific questions aimed at making an accurate diagnosis. It is interesting to note that the understanding dimension of the counselor VRS reflected pre-post treatment differences for 87 the total group of medical students. If the student was cognizant of an emotional problem but did, in fact, not choose to deal with it, he would be rated as communicating understanding, but not as discussing affect, being speci- fic, or allowing for exploration. This idea was supported by the results found in this study. The results from the intercorrelation of variables designed to establish validity are must less clear. Posi- tive correlations for both the peer and staff ratings with the MCAT, the ASS, and the dimensions of the Counselor VRS, were expected but not realized. One of the problems of a correlational analysis is that the measures being used to effect high correlations must have a fairly wide range. Both the scores of the ASS and the dimensions of the Coun— selor VRS had fairly narrow ranges. While the treatment appeared to be effective, the post-treatment means were still fairly low. Another problem may lie within the peer and staff ratings themselves. Each student was to rate each of the other class members in ascending order of their conduciveness to the discussion of personal prob— lems. The staff was to rate each class member in order of whom they would refer a patient to, if they wished a sen- sitive, compassionate physician. Could it perhaps be that other criteria were used in determining the rankings other than those listed for judgment? Feedback from the staff indicated that some may have used their feelings of the 88 student's overall competence as a physician or his aca- demic accomplishments as a basis for rating. What "com- petence" might be composed of, is a moot point, a stu- dent's activity in the lab or excellence in academics may be just two of many possible examples. If this were the case, the lack of relationship between the ratings and the other measures would come as no surprise. The staff ratings were divided between staff having strictly didactic encounters with the students and those staff who had more clinical contact. There was, however, very little differ— ence in ratings of the two staff groups. The second year medical students were initially resistant to rating the other members of their class as to whom they would be most and least likely to seek out and to discuss a personal problem. Some students did not respond to the questionnaire. However, the peer and staff ratings correlated with one another to the same extent as from previous research with students in a counseling insti- tute (Kagan, e£_§g,, 1967). One result difficult to justify is the lack of a positive relationship and, indeed, a negative relationship of the ASS and the Counselor VRS dimensions. Restriction of the range of scores has been offered as one possibility; in addition, the sample was quite small. Affective sensi- tivity has been shown to have a low positive correlation with the dimension factors for a sample of students in 89 counseling (Kagan, gp_§l., 1967). Affective sensitivity, or empathy, as it is conceived of by the ASS, is a person- ality trait. For it to be an effective agent in a counsel- ing or therapeutic situation, its effect would have to be mediated through some form of behavior; and whatever the form of behavior, there is the possibility that these stu- dents have not learned how to get it across, how to act on their hunches. This possibility is evidenced by the expected lower overall group mean, lower than Masters' students with more experience. Implications for Counseling and for Future Research in Counseling Clients of effective counselors rate the counselor as one in whom they have confidence. If a counselor is not very effective, those clients who seem to make gains may do so in spite of the counselor, but the amount of movement is not as great as if the counselor were effec- tive, and, in addition, the client does not rate the coun- selor as one with whom he cares to discuss his concerns. While the relationship of client response to counse- lor behavior was not confirmed, clients of counselors rated as being effective, rated themselves as having more confidence in their counselors than did clients of less effective counselors. What meaning do these observations have for counseling? If one assumes that the relationship is the foundation on which client movement is based, then 90 there is support for a counselor to move in on his percep- tions and not to wait until "the client is ready for them." When is a client ready for perceptive reflections of inter- pretations? It appears that a client has confidence in a counselor who moves quickly, even though the client ini- tially reacts anxiously. The counselor may have to "prove his worth," therefore, before the client will feel the confidence and trust to allow the counselor to see more of him. Considering the above results, further study should move in two parallel directions. There is a large degree of client variability; perhaps over a long-term study this factor would not affect outcome. However, the results of this study imply that further consideration of client per- sonality characteristics must be taken. Are there certain personality types who react better to what is considered as a "good" counselor. Do clients having low to moderate anxiety function better with an active counselor than do high anxiety clients? Secondly, this study should be carried out over the full course of counseling, on a session-by-session basis. In this way, client movement could be studied both as a function of rated counselor effectiveness and of the developing relationship. This study used the first ses- sion only. Does a client of an ineffective counselor make therapeutic movement in spite of his counselor? Does a 91 client who had confidence in his counselor after the first session but who reacts defensively in that very session, become less resistant over time gpd gain in therapeutic movement? These questions may be answered in a further study viewing therapeutic movement over the course of counseling--from the first interview until termination. Because this study suggests that client and counse- lor behavior may be quite different from one another, per- haps different scales might be used for measuring counse- lor behavior and related client behavior. The next step that another study might take would be to use the Client Growth Scale and Client Observation Sheet (Kagan, gp_gl., 1967), which include more global aspects of client move- ment in conjunction with the Client and the Counselor VRS. A possible obstacle to this procedure may be the way each scale correlates highly within itself; the correlations among the various client dimensions, therefore, would all be expected to be quite high. Implications for Medical School Interviewer Training The simulation procedure for instruction into medi— cal interviewing provides a unique method for adjusting the training to meet the needs of the individual student. It seems to provide a "real" situation in which to learn, apparently more adequate than simple role-playing but with less risk and less initial anxiety than an actual situation 92 would provide. After a four-session, one-month sequence, there were signs of students becoming significantly better at important behaviors involved in the doctor-patient relationship. At the present time, there are no operational mea- sures in use which quantify the quality of a doctor's skill in interviewing. Since the relationship between patient and doctor seems so crucial, an evaluation of a medical student's skill at interviewing should be a pri- mary concern of the medical school. A measure of this skill is therefore sorely needed; the Counselor VRS offers promise as such an instrument. For its future use in a medical setting, however, provision must be made to judge the appropriateness during the interview of series of interviews, in the timing of data-gathering by the medical student or physician. While some patients may be quite emotionally involved with a real or psychosomatic physi- cal problem and this involvement should be approached by the physician, the patient might only allow such discus- sion after divesting himself of those physical symptoms tnhich he expects the physician to inquire about. In addition to a judgment of the timing of data- gathering, the expertise with which it is done while main— ‘taining the relationship between doctor and patient Ishould also be considered. While the efficacy of the Coun- selor VRS may be seen with the type of simulated patients 93 used in this investigation (that is, patients with prob- lems having strong emotional overtones), the Scale may not be as effective with problems having a clear-cut physical basis. Even in the latter situation, however, the doctor-patient interaction still remains important as a necessary condition for effective diagnosis and treat- ment. Is there an optimal ratio of time spent by the phy- sician in gathering data for a diagnosis and time spent in dealing directly with the patient's feelings? A model used in the development of the Counselor VRS might be appropriate to the construction of a scale more conducive to measuring physician interview behavior. Good and poor interviews might be rated by experts as to their differ- entiating qualities. These qualities could then be incor- porated into a scale such as the Counselor VRS and then validated upon other samples. Because of the variability that the type of patient may cause in such ratings, dif- ferent norms for a variety of patient types could be established. Appropriateness of the timing of, for exam- ple, statements directed at information gathering and statements directed towards the exploration of the patient's feeling could be formulated. The fact that none of the variables used in this investigation had positive correlations with the criteria of peer and staff ratings was indeed a surprising finding. lIlllIIIIlII‘! 9“ Could it be that medical students have different social expectations than do students in counseling that lead to different ratings? The sample size itself may be another factor in the instability of the results. Are the criteria on which the peer and staff ratings made completely tangential to the criterion measures? The results are especially confusing in light of previous work completed by Campbell and by Resnikoff (Kagan, g£_§1., 1967), showing the Affective Sensitivity Scale to be positively related to peer and staff ratings and to the dimensions of the Counselor VRS. Also interesting is the lack of positive relation- ship between the Science sub-test of the MCAT and the rating measures; the measure which is heavily relied upon for admission to medical school does not appear to be an apprOpriate criterion measure for this one sample. In summary, significant findings and also confusing ones were found in this investigation. A replication of the study conducted here is certainly in order. LIST OF REFERENCES 95 LIST OF REFERENCES Adler, Leta M., and Enelow, A. J. A scale to measure psychotherapy interactions. Unpublished paper, University of Southern California, 196“. Bales, R. F. Interaction process analysis. Cambridge: Addison-Wesley Press, 1950. Bergin, A. E. Some implications of psychotherapy research for therapeutic practice. Int. J. Psychiat., 8, 1967, l36-l“9. Braaten, L. J. The movement from non-self to self in client-centered psychotherapy. J. counsel. Psychol., 8, 1961, 20-2“. , Campbell, D. T., and Stanley, J. C. Experimental and quasi-experimental designs for research. Handbook of Research on Teaching. Edited by N. L. Gage. Chicago: Rand McNally, 1963. Deroo, W. M., and Rank, R. C. Counselor response system of the behavior interaction description system. Paper presented at the American Personnel & Guidance Assn. Convention, Minneapolis, Minn., 1965. Dinoff, M.; Kew, J. K.; and Timmons, E. O. The stability of group verbal behavior. Psychological Record, 12, 1962, 323-325. Dittman, A. T. The interpersonal process in psychotherapy: development of a research method. J. abn. soc. Psychol., “7, 1952, 236-2““. Dollard, J., and Auld, F., Jr. Scoring human motives: a manual. New Haven: Yale University Press, 1959. Ebel, R. L. Estimation of the reliability of ratings. Psychometrika, 16, 1951, “07-“2“. Enelow, A. J. and Mckie, R. Audio-visual aids in psycho— therapy and therapist training. Unpublished paper, Michigan State University, 1968. 96 97 Goldberg, A. D. A sequential program for supervising counselors using the Inter-Personal Process Recall technique. Unpublished doctoral dissertation, Michigan State University, 1967. Gottschalk, L. A., and Hambidge, G., Jr. Verbal behavior analysis: A systematic approach to the problem of quantifying psychologic processes. J. projective Techniques, 19, 1955, 387-“09. Griffin, G. G. An evaluation of the objectivity of an audio-visual counselor scale which compares the behaviors of M.A. candidate counselors with the behaviors of Ph.D. candidate counselors. Published doggoral dissertation, Michigan State University, 19 . Hansen, J., and Stevic, R. Practicum in supervision: a proposal. Couns educ. Superv., 6, 1967, 206-207. Heller, K. Interpersonal style in an interview analogue. In R. A. Myers (Chmn), Social influence, counseling, and psychotherapy, Symposium, American Psychologi- cal Assn. Convention, Philadelphia, 1963. Hill, G. E. The profession and standards for counselor education. Couns educ. Superv., 6, 1967, 130-136. Jaffe, J. Dyadic analysis of two psychotherapeutic inter— views. Comparative psycholinguistic analysis of two psychotherapeutic interviews. Edited by L. A. Gottschalk. New York: International Universities Press, 1961. Jason, H. Programming and simulation: The discovery of the student in medical instruction. Universiteit en Hogeschool, Januari 1968, Nr. 5, Jaargang 1“. Paper prepared for the Boerhave Conference on Medi- cal Education, Leiden, Holland, December 1“-16, 1967. Kagan, N.; Krathwohl, D. R.; and Farquhar, W. W., et a1. IPR--Inte§personal Process Recall: Stimulated by Videotape. Educational Research Series, No. 2“, Bureau of Educational Research Services, Michigan State University: E. Lansing, March, 1965. Kagan, N., and Krathwohl, D. R. (Directors). Studies in Human Interaction: Interpersonal Process Recall stimulated by videotape. Educational Publication Services, College of Education, Michigan State Uni— versity, E. Lansing, December, 1967. 98 Kanfer, R. H.; Phillips, Jeanne S.; Matarazzo, J. D.; and Saslow, G. Experimental modification in interviewer content in standardized interviews. J. consult. Psychol., 2“, 1960, 528-536. Kirtner, W. L., and Cartwright, D. S. Success and failure in client-centered therapy as a function of initial in—therapy behavior. J. consult. Psychol., 22, 1958, 329-333. Lennard, H. L. Some aspects of the psychotherapeutic system. Research in Psychotherapy II. Edited by H.6H. Strupp and L. Luborsky. New York: Harper, 19 2. Marsden, G. Content-analysis studies of therapeutic inter- views: 195“ to 196“. Psychological Bull., 63, 1965, 298-321. Matarazzo, J. D. Prescribed behavior therapy: Suggestions from interview research. Experimental foundations of clinical psychology. Edited by A. J. Backrach. New York: Basic Books, 1962. Matarazzo, J. D.; Hess, H. F.; and Saslow, G. Frequency and duration characteristics of speech and silence behavior during interviews. J. clinical Psychol., 18, 1962, “16-“26. Patterson, C. H. The practicum requirement in "Standards—- 1967 Edition." Couns. Educ. Superv., 6, 1967, 352-353. Pepinsky, H. B., and Karst, T. 0. Convergence: A pheno- menon in counseling and in psychotherapy. Amer. Psychol., 19, 196“, 333—338. Rogers, C. R. A tentative scale for the measurement of process in psychotherapy. Research in psychotherapy. Edited by E. A. Rubinstein and M. B. Parloff. Washington, D.C.: American Psychological Assn., 1959, 96-107. Rogers, C. R. A process conception of psychotherapy. Op becoming a person. Edited by C. R. Rogers. Boston: Houghton—Mifflin, 1961. Schiff, S. B., and Reivich, R. Use of television as an aid to psychotherapy supervision. Archives gen. Pschiat., 10, 196“, 8“-88. 99 Siegel, S. Nonparametric statisgips for the behavioral sciences. New York: McGraw-Hill, 1956. Sklansky, M. A.; Isaacs, K. S.; and Haggard, E. A. A method for the study of verbal interaction and levels of meaning in psychotherapy. Scientific papers and discussionsyydivisional meeting, Mid- West area district branches. Edited by J. S. Gottlieb and G. Tourney. Detroit: Amer. Psychia- tric Assn., 1960, 133-l“8. Stefflre, B., and Leafgrenn, F. A. Mirror, mirror, on the wall. . .a study of preferences for counselors. Pers. & Guid. J., “3, 196“, “59-“62. Steph. J. A. Responses to hypothetical counseling situa- tions as a predictor of relationship orientation in school counselors. Unpublished doctoral disserta- tion, Univ. of Wisconsin, 1963. Strupp. H. H. Psychotherapists in action. New York: Grune & Stratton, 1960. Truax, C. B.; Carkhuff, R. R.; and Douds, J. Toward an integration of the didactic and experiential appro- aches to training in counseling and psychotherapy. J. counsel. Psychol., 11, 196“, 2“0-2“7. Truax, C. B., and Carkhuff, R. R. Concreteness: a neg- lected variable in research in psychotherapy. J. clin. Psychol., 20, 196“, 26“-267. Truax. C. B. Counseling and Psychotherapy: Process and outcome. V.R.A. Research and Demonstration Grant #906 — P: Univ. of Arkansas, 1966. . Ward, G. R. Educating counselors by videotape using the interpersonal Process Recall Technique. Unpublished doctoral dissertation, Michigan State University, 1965. APPENDICES 100 APPENDIX A COUNSELOR VERBAL RESPONSE SCALE 101 1(12 AQBOH AdhOfi mmwcomm Iwm no a mm «N .HNM‘PIDWFW I H N n v 0>auuoumecoz II 0>wuuwwum >N0umuonxw Icoz >u0umuonxm oeufluwmm :02 oauauwnm wcfivcuum Iuwucsusoz mcflvcuum Iuopnn m>fiuwcmou uuomw< ,mwmcommmm mcoumcoan "mama . "cane HH‘Um 02Hh KOAMMZDOO KAN APPENDIX B CLIENT VERBAL RESPONSE SCALE 103 10“ 41309 A4309 mennomn Iflu we a ‘0th IHNM'm .J a N, U E 05>?G02 n v II H§§>Ox uncuuuoHaxm Inez aneuuuo~mxu 338m .82 uwuwoomm mcwnzuuu Inevasuzoz mcwunauu o>uuwnu00 uoowu< inc-noa-um Iuovca Inca-cuiaa 98‘9 . aim-B .00u55 HA¢UD uzuktl flmIOhIfl“ d‘dflflb FIHHAU Ian I l l. [I‘ll l llllllllll III... 1 I ll." APPENDIX C WISCONSIN RELATIONSHIP ORIENTATION SCALE 105 I. I I ll| ll ll. . ‘II II- WISCONSIN RELATIONSHIP ORIENTATION SCALE DIRECTIONS: Check the item which best describes your 1. feelings toward the counselor. I would attempt to avoid any kind of interaction or relationship with this person. If no one else were available, I might consult this person for specific information of a fac- tual, e.g., education or vocational nature, but I would avoid any personal exposure. I would be willing to talk with this person about factual, e.g., educational or vocational, concerns and some of the personal meanings connected with these. I would be willing to talk with this person about many of my personal concerns. I have the feeling that I could probably talk with this person about almost anything. 106 APPENDIX D PEER RATING SCALE 107 Office of Medical Education PEER RATINGS Year 5 Class This form is designed to reveal your reactions about the other members of your class as persons to whom you would go if you had a problem. This questionnaire is aimed at being a measure of your feelings of your classmates' com- petence in the area of discussing problems. These results will be compared to empathy scores and ratings of your interviews conducted with coached patients, in order to set up better methods of evaluation. All ratings will be kept strictly confidential. On the next page, you will find a list of all the members of your class with a number in front of each name. First find your name and draw a line through it. Next notice that underneath the list of names is an approximate nor- mal curve turned on its side, consisting of a number of rows. You are to fill in the rows with the numbers in the list to indicate the extent to which you would be apt to go to the various members of your class for any prob- lem you might have. In front of each row are listed the minimum and maximum total of numbers that may be included in each row. The numbers of the people you would be most apt to consult would be placed at the top. The numbers of the people you would be next most apt to consult would be placed in the next row. The order with which you place the names in any one row does not matter since everyone placed in that row are all in the same position. Continue placing numbers, remembering that the further down the page you place the number the less apt you would be to consult him (her) for a problem. You may find it easier to do this task if you draw a line through each name when you place its number in a row. Make sure you have used all the rows (but not necessarily every box within a row) and have used all the numbers associated with the students. Try to find individuals whom you could easily use as anchor points on the scale and build from there. All ratings will be kept confidential and used only in the development of procedures for evaluation. Hillard Jason, M.D. Norman Kagan, Ph.D. Arthur Resnikoff 108 \OGDNOUTSWNH HPJFJ [UFJO }_.I LA) List List List List List List 109 PEER RATINGS Adam 1“. Maledon Alguire 15. Molstad Baker 16. Oberstein Campana 17. Olson Feldman 18. Parker Denslow l9. Potter Dorey 20. Rathbun Eidt 21. Ruggles Gendleman 22. Schoenkerman Gillespie 23. Victor Gottschalk 2“. Weiss Hanson 25. Williams, D. Lawrence 26. Williams, M. STUDENTS TO WHOM YOU WOULD LIKE TO GO from 1-3 from 2-5 from 3-7 from 3—7 from 2-5 from 1-3 STUDENTS TO WHOM YOU WOULD BE LESS LIKELY TO GO APPENDIX E STAFF RATING SCALE llO 111 STAFF RATINGS 1. Adam 1“. Maledon 2. Alguire 15. Molstad 3. Baker 16. Oberstein “. Campana 17. Olson 5. Feldman 18. Parker 6. Denslow 19. Potter 7. Dorey 20. Rathbun 8. Eidt 21. Ruggles 9. Gendleman 22. Schoenkerman 10. Gillespie 23. Victor 11. Gottschalk 2“. Weiss 12. Hanson 25. Williams, D. 13. Lawrence 26. Williams, M. STUDENTS TO WHOM YOU WOULD REFER SOMEONE List from 1-3 List from 2-5 List from 3—7 List from 3—7 List from 2-5 List from 1—3 STUDENTS TO WHOM YOU WOULD BE LESS LIKELY TO REFER SOMEONE Name of respondent Please return to: Arthur Resnikoff OMERAD 371 Olin Health Center Campus APPENDIX F COUNSELOR VERBAL RESPONSE SCALE TRAINING MANUAL 112 MANUAL FOR TRAINING JUDGES IN USE OF COUNSELOR VERBAL RESPONSE SCALE This manual is intended to standarize procedures for use of the IPR Counselor Verbal Response Scale in the judging of either video or audio tapes and to provide some guidelines in the training of judges in the use of the scale. I. What is a scorable response? Scorable responses include words, phrases, or senten- ces used by the counselor which interrupt the flow and momentum of client communication or which are in direct response to a client's statement. Although such responses normally occur at the conclusion of a client's communica- tion, there are times when a counselor will deliberately interrupt a client's statement. In either case, such counselor statements are treated as scorable responses. Responses which are part of a counselor's unconscious mannerisms, and do not interrupt the flow and momentus of client communication, e.g., "Umm," "Okay," "Hmm," "Yes," and "I see" are not scored. Responses such as "continue" and "go on" are scorable responses. In essence, they urge and encourage the client to continue with his flow of 113 11“ communication and reinforce the client's pattern of com— munication. There are times when counselors make two distinct responses to a single client communication, these respon- ses normally being separated by a pause: Example: CL: "It's good to get rounded especially if I decided to teach which you really can't decide until you get into college, I guess." CO: "Kind of hard to figure things out, isn't it?" "Do you think student teaching is helping you get the feel of teaching?" In this case, there are clearly two responses differing in nature which must be scored separately to accurately evaluate the counselor. There are other instances in which there is a clear shift in the content of the coun- selor's response without a distinct pause. In such cases, as in the preceeding example, counselor statements are treated as two separate responses and are scored individ- ually. N.B.: Each scorable counselor response must be rated on each of the five dimensions of the scale. 11. How many responses are scored? For purposes of accurate evaluation, twenty conse- cutive counselor responses are scored from each counseling session rated. These responses should be drawn from the middle portions of a counseling session, avoiding both the beginning and the terminating segments of the interview. Judges should, however, be given an opportunity to listen 115 to a few responses prior to the start of judging so that they may become acclimated to the voices and pace of both client and counselor, and may gain some familiarity with the general tone of the interview. 111. When is rating done? Each response is rated at the conclusion of the coun- selor's statement (word or phrase). The tape (audio or video) should be stopped after each scorable counselor response and scoring should be completed by all judges prior to the playing of the next unit. As judges become more comfortable with the use of the scale, they should require no more than thirty seconds for the scoring of each response. As in all judging procedures, it is desirable that communication among judges be minimized during the rating session. Thus, judges should be so placed that, while they all have adequate View of the video monitor or can clearly hear the audio tape, there is a minimum of contact among judges. IV. How are judges trained? It is important that adequate time be given to the training of judges. The success of any evaluation of this sort depends upon the agreement reached by the judges in defining the dimensions of the scale. Training must involve actual rating of practice tapes. The number of 116 tapes used will, of course, depend on the needs of the judges. However, the tapes used should represent a variety of counseling interviews, i.e., experienced, inexperienced or beginning, etc. During training, discussion should follow the rating of each counselor response (obviously in the later stages of training, this is not as crucial and a group of respon- ses may be rated before discussion occurs) until agree- ment about interpretation of the dimensions of the scale is reached. Prior to the actual rating, judges should again go over the definitions of the five dimensions to insure complete understanding. V. How much time does rating require? It has been found that at least one hour is required for the rating of two audio tapes. It is clear that more time would be involved in the rating of video tapes due to the time required for changing tape. These time allow- ances should be noted prior to establishing a rating ses- sion.* It has also been our experience that approximately three hours is necessary for adequate training of judges and discussion of the definitions of the five dimensions. *Care should be taken in scheduling rating sessions so that judges are not required to rate too many tapes at any one session. If this is allowed to occur, judges may acquire a "set" which will affect later ratings. 117 General Questions: The most frequent question which occurs centers about whether responses can be judged independently of prior content in the interview. The intent of this scale is to focus primarily on a single client communication and counselor response. It is obvious, however, that many counselor responses take into account material which has been elicited in prior portions of the interview. Most interviews also present a general theme within which indi— vidual interactions occur. The judge must clearly be aware of this larger framework in making his rating. How— ever, the emphasis still remains on the individual response to a client communication. In this context, it is important to note that ratings of responses take into account the appropriateness of the response at a given moment in time. For instance, while a response may be specific in the early portions of the interview, the same response coming later in the interview may not only be nonspecific, but also inappropriate in moving the client to a further understanding of his own concerns. To this extent, the unfolding theme and the appropriate timing of responses must be considered by the judge in making his rating. IPR COUNSELOR VERBAL RESPONSE SCALE The Counselor Verbal Response Scale is an attempt to describe a counselor's response to client communica- tion in terms of four dichotomized dimensions: (1) affect— cognitive; (b) understanding-nonunderstanding; (c) specific—nonspecific; (d) exploratory-nonexploratory. These dimensions have been selected because they seem to represent aspects of counselor behavior which seem to make theoretical sense and contribute to client progress. A fifth dimension—-effective-noneffective-—provides a global rating of the adequacy of each response which is made independently of the four descriptive ratings. The unit for analysis is the verbal interaction between counselor and client represented by a client statement and counselor response. A counselor response is rated on each of the five dimensions of the rating scale, with every client-counselor interaction being jud- ged independently of preceding units. In judging an individual response, the primary focus is on describing how the counselor responded to the verbal and nonverbal elements of the client's communication. 118 119 Description of Rating Dimensions I. Affect-cognitive dimension The affective-cognitive dimension indicates whether a counselor's response refers to any affective component of a client's communication or concerns itself primarily, with the cognitive component of that communication. A. Affective response.--Affective responses gener- ally make reference to emotions, feelings, fear, etc. The judge's rating is solely by the content and/or intent of the counselor's response, regardless of whether it be reflection, clarification or interpretation. These respon- ses attempt to maintain the focus on the affective compon— ent of a client's communication. Thus they may: (a) Refer directly to an explicit or implicit refer- ence to affect (either verbal or nonverbal) on the part of the client. Example: "It sound like you were really angry at him." (b) Encourage an expression of affect on the part of the client. Example: "How does it make you feel when your parents argue?" (0) Approve of an expression of affect on the part of the client. Example: "It doesn't hurt to let your feelings out once in a while, does it?" (d) Presents a model for the use of affect by the client. Example: "If somebody treated me like that, I d really be made." Special care must be taken in rating responses which use the word "feel." For example, in the statement "Do you 120 £231 that your student teaching experience is helping you get the idea of teaching?", the phrase "Do you £321 that" really means "do you think that." Similarly, the expres- sion "How are you feeling?" is often used in a matter-of- fact, conversational manner. Thus, although the verb "to feel" is used in both these examples, these statements do ppp_represent responses which would be judged "affective." B. Cognitive Responses.--Cognitive responses deal primarily with the cognitive element of a client's commun-. ication. Frequently, such responses seek information of a factual nature. They generally maintain the interaction on the cognitive level. Such responses may: (a) Refer directly to the cognitive component of the client's statement. Example: "So then you're thinking about switch- ing your major to chemistry?" (b) Seeks further information of a factual nature from the client. Example: "What were your grades last term?" (c) Encourage the client to continue to respond at the cognitive level. Example: "How did you get interested in art?" II. Understanding—nonunderstanding dimension The understanding-nonunderstanding dimension indi- cates whether a counselor's response communicates to the client that the counselor understands or is seeking to understand the client's basic communication, thereby encouraging the client to continue to gain insight into the nature of his concerns. 121 A. Understanding responses.--Understanding respon- ses communicate to the client that the counselor under- stands the client's communication——the counselor makes appropriate reference to what the client is expressing or trying to express both verbally and nonverbally--or the counselor is clearly seeking enough information of either a cognitive or affective nature to gain such under- standing. Such responses: (a) Directly communicate an understanding of the client's communication. Example: "In other words, you really want to be treated like a man." (b) Seek further information from the client in such a way as to facilitate both the counselor's and the client's understanding of the basic problems. Example: "What does being a man mean to you?" (c) Reinforce or give approval of client communica- tions which exhibit understanding. Example: CL: "I guess then when people criti- cize me, I'm afraid they'll leave me." CO: "I see you're beginning to make come connection between your behavior and your feelings." B. Nonunderstanding responses.--Nonunderstanding responses are those in which the counselor fails to under- stand the client's basic communication or makes no attempt to obtain appropriate information from the client. In essence, nonunderstanding implies misunderstanding. Such responses: 122 Example: CL: "When he said that, I just turned red and clenched my fists." CO: "Some people don't say nice things." (b) Seek information which may be irrelevant to the client's communication. Example: CL: "1 seem to have a hard time get- ting along with my brothers." C0: "Do all your brothers live at home with you?" (c) Squelch client understanding or move the focus to another irrelevant area. Example: CL: "I guess I'm really afraid that other people will laugh at me." CO: "We're the butt of other people's jokes sometimes." Example: CL: "Sometimes I really hate my aunt." CO: "Will things be better when you go to college?" III. Specific-nonspecific dimension The specific-nonspecific dimension indicates whether the counselor's response delineates the client's problems and is central to the client's communication or whether the response does not specify the client's concern. In essence, it describes whether the counselor deals with the client's communication in a general, vague, or peripheral manner, or "zeros in" on the core of the client's communi- cation. NB: A response judged to be nonunderstanding must also be nonspecific since it would, be definition, misunderstand the client's communication and not help the client to delineate his concerns. Responses judged under- standing might be either specific (core) or nonspecific 123 (peripheral) i.e., they would be peripheral if the coun- selor conveys only a vague idea that a problem exists or "flirts" with the idea rather than helping the client delineate some of the dimensions of his concerns. A. Specific responses.--Specific responses focus on the core concerns being presented either explicitly or implicitly, verbally or nonverbally, by the client. Such responses: (a) Delineate more closely the client's basic con- cerns. Example: "This vague feeling you have when you get in tense situations--is it anger or fear?" (b) Encourage the client to discriminate among stimuli affecting him. Example: "Do you feel in all your classes or only in some classrooms?" (c) Reward the client for being specific. Example: CL: "I guess I feel this way most often with someone who reminds me of my father." 00: "So as you put what others say in perspective, the whole world doesn't seem so bad, it's only when someone you value, like Father, doesn't pay any atten- tion that you feel hurt." B. Nonspecific responses.--Nonspecific responses indicate that the counselor is not focusing on the basic concerns of the client or is not yet able to help the client differentiate among various stimuli. Such respon- ses either miss the problem area completely (such respon- ses are also nonunderstanding) or occur when the counselor is seeking to understand the client's communication and 12“ has been presented with only vague bits of information about the client's concern. Thus, such responses: (a) Fail to delineate the client's concern and can— not bring them into sharper focus. Example: "It seems your problem isn't very clear--can you tell me more about it?" (b) Completely miss the basic concerns being pre- sented by the client even though the couselor may ask for specific details. Example: CL: "I've gotten all A's this year and I still feel lousy." 00: "What were your grades before then?" (c) Discourage the client from bringing his con- cerns into sharper focus. Example: "You and your sister argue all the time. What do other people think of your sis- ter?" IV. Eyploratory-nonexploratory The exploratory-nonexploratory dimension indicates whether a counselor's response permits or encourages the client to explore his cognitive or affective concerns, or whether the response limits a client's exploration of these concerns. A. Exploratory responses.--Exploratory responses encourage and permit the client latitude and involvement in his response. They may focus on relevant aspects of the client's affective or cognitive concerns but clearly attempt to encourage further exploration by the client. Such responses are often open-ended and/or are delivered in a manner permitting the client freedom and flexibil- ity in response. These responses: (a) (b) (C) B. 125 Encourage the client to explore his own con— cerns. Example: Cognitive--"You're not sure what you want to major in, is that it?" Affective--"Maybe some of these times you're getting mad at yourself, what do you think?" Assist the client to explore by providing him with possible alternatives designed to increase his range of responses. Example: Cognitive--"What are some of the other alternatives that you have to history as a major?" Affective--"In these situations, do you feel angry, mad, helpless, or what?" Reward the client for exploratory behavior. Example: Cognitive--"It seems that you've con- sidered a number of alternatives for a major, that's good." Affective--"So you're beginning to wonder if you always want to be treated like a man." Nonexploratory responses.--Nonexploratory respon- ses either indicate no understanding of the client's basic communication, or so structure and limit the client's responses that they inhibit the exploratory process. responses give the client little opportunity to explore, expand, or express himself freely. Such responses: Discourage further explorations on the part of the client. Example: Cognitive—-"You want to change your major to history." Affective-~"You really resent your parents treating you like a child." These 126 V. Effective-noneffective dimension Ratings on the effective-noneffective dimension may be made independently of ratings on the other four dimen— sions of the scale. This rating is based solely upon the judge's professional impression of the appropriateness of the counselor's responses, that is, how adequately does the counselor's response deal with the client's verbal and nonverbal communication. This rating is ppp dependent on whether the response has been judged affective-cognitive, etc. A rating of “ indicates that the judge considers this response among the most appropriate possible in the given situation, while a 3 indicates that the response is appropriate but not among the best. A rating of 2 indi- cates a neutral response which neither measurably affects client progress nor inhibits it, while a rating of l indi— cates a response which not only lacks basic understanding of the client's concerns but which in effect may be detri— mental to the specified goals of client growth. APPENDIX G CLIENT VERBAL RESPONSE SCALE TRAINING MANUAL 127 MANUAL FOR TRAINING JUDGES IPR CLIENT VERBAL RESPONSE SCALE The Client Verbal Response Scale is an attempt to describe a client's response to counselor communi- cation in terms of four dichotomized dimensions: (a) affect-cognitive; (b) understanding-non- understanding; (6) specific-non-specific; (d) exploratory-non-exploratory. These factors have been used effectively in the description of coun- selor behavior; high levels of these factors are related to counselor effectiveness. It would seem that client response along these dimensions would be a necessary condition for growth in counseling. A fifth dimension, movement—no-movement, provides a global rating of the adequacy of each response rated independently of the first four factors. The unit for analysis is the verbal interaction between counselor and client represented by a coun- selor response and consecutive client statement. A client response is rated on each of the five dimen- sions of the rating scale, with every client- counselor interaction being judged independently of preceding units. In judging an individual response, 128 o-v l w“-.. . ‘ _ .' 3,. l "*1. 129 the primary focus is on describing how the client responded to the verbal and non-verbal elements in the counseling situation. This implies that the interaction between the client and the counselor should be viewed as such, and that counselor behavior should have a strong effect on client response. A counselor response may not determine a client's response, but the client's response will bear strong dependency to that of the counselor. Description of Rating Dimensions I. Affective--cognitive dimension The affective-cognitive dimension indicates whether a client's response refers to any expressed affec— tive component by reference to his own feelings or remains primarily on a cognitive level. Special care must be taken in rating responses which use the word "feel." For example, in the statement, "I feel that the school should change its policy," the phrase "I feel that," really means, "I think that." This statement does not represent a response which would be judged "affective." A. Affective responses.--Affective responses gener- ally make reference to emotions, feelings, fears, etc. They are determined by the content, the intent of the client's response, and/or by the II. 130 manner of delivery. These responses attempt to maintain a focus on the affective components of the client's own communication. Thus they will be an explicit or implicit reference to affect (either verbal or non-verbal) on the part of the client. Example: "I was really angry at him." B. Cognitive responses.—-Cognitive responses deal primarily with the cognitive element of a client's communication. Frequently such respon- ses are information of a factual nature. They generally maintain the interaction on the cogni- tive level. Examples of such responses might be: Example a) "My school courses are difficult this term." Example b) "I have two brothers and a sister older than I." Understanding--non-understanding dimension The understanding--non—understanding dimension indi- cates whether a client's response communicates that he understands or is seeking to understand his basic concerns leading to increased insight into the nature of these concerns. A. Understanding responses.-—Understanding responses communicate to the counselor that the client. I" 1*- .V' Laden. -;x~ 1!"; name} 1'". I ‘K. _\z.-‘ II 131 understands the nature of his communication or is clearly seeking enough information of either a cognitive (factual) or affective nature to gain such understanding. Such responses might be: Example a) "You really think that I am angry with my mother," or "I think that I may be angry with my mother." m...” A-.. ,7.— 1 Example b) Following a counselor probe: "You want to know what I feel about that. :rmr—a-u. It. ‘94....— Well, it's like this In this type of response, notice will have to be kept on the quality of the counselor communica- tion. The client may stray from what the counse- lor is aiming towards, or may not get to his concerns because the counselor is straying. Non-understanding responses.--Non-understanding responses are those in which the client fails to understand his own feelings, thoughts, or concerns or makes no attempt to obtain help (in terms of appropriate information or reactions) from the counselor. In essence, non-understanding implies misunderstanding. Such responses: (a) Communicate understanding of himself. Example: CO: "You seemed very angry there." III. 132 (b) Seek information which is irrelevant to the ongoing interview process. Example: CO: "How do you feel towards your brother?" CL: "Do you think that I should tell my parents about what I did?" The client may occasionally move away from the topic. This can reflect non-understanding. Specific--non-specific dimension The specific--non-specific dimension indicates whether the client's response begins to delineate his problems and is central to the counselor's communication or whether the client's response does not serve to help him specify his concerns. In essence, does the client deal with his concerns in a general, vague, or peripheral manner, or does he "zero in" on the core of the problem or communica- tion? A response judged to be non-understanding must also be non—specific since it would, by definition, miss the ongoing communication and not help the cli- ent to delineate his concerns. Responses judged understanding might be either specific (core) or non—specific (peripheral), i.e. they might be only a vague idea that a problem exists or such responses u.‘ .. [.17‘; .v;‘11m.lfi 133 might begin to specify some of the dimensions of his concerns. Those responses categorized as "trying to understand" would be clearly non-specific (peripheral) since they imply that the client has not been able to delineate his concerns and is clearly seeking further information. A. Specific responses.-—Specific responses focus L fl on the core concerns being presented either explicitly or implicitly, verbally or non- verbally, by the client. Such responses: 'I (a) Delineate the client's basic concerns. Example: "This vague feeling that I have, I think it is fear." This statement also shows the client to be discriminating among the stimuli affecting him. Non-specific responses.--Non-specific responses indicate that the client is not focusing on his basic concerns or is not yet able to differen- tiate among various stimuli. Such responses either miss the problem area completely (such responses are also non-understanding) or occur when the counselor is seeking to understand the client's communication and has been presented with only vague bits of information about the client's concerns. Thus, such responses: IV. 13“ (a) Fail to delineate the client's concern and do not bring them into sharper focus. Example: "The problem isn't very clear. I can't seem to understand what I was feeling." (b) Move away from a client's talking about him- self or his relations with others. I «a Example: "My sister has many friends." I Exploratory--non-exp1oratory dimension I 8“.“ “.9 I . l‘l‘4_“ The exploratory--non-exploratory dimension indicates whether a client's response demonstrates exploration of his cognitive or affective concerns, or whether the response is not exploring these concerns. A. Exploratory responses.--Exploratory responses demonstrate the client's latitude and involvement in what he is discussing. They may focus on relevant aspects of the counselor's statements but are clearly attempting to further explore his problems. Such responses: (a) Demonstrate exploratory activity. Example: Cognitive - "I think there are other alternatives in choosing a a major than the one that I chose." 135 Affective - "I wonder why I felt angry in that situation." (b) May follow a counselor reward Example: Cognitive: CO: "That was a good choice to make." CL: "Yes, because by doing that I was able to take more courses that interest me." Affective: CO: "So you're begin- ning to see that you don't always want to be treated like a child." CL: "1 really think so. It really bugs me that I am that way." Non-exploratory responses.--Non-exploratory responses demonstrate little or no exploration by the client of his basic concerns. The client does not explore, expound, or express himself freely. These may often be a function of the counselor's squelching or inhibiting the explor- ation process. Such responses: (a) Indicate a lack of exploration on the part of the client. 136 (0) Example: CL: "1 am not sure what field to change into; I have been thinking of history. 00: "You want to change your major to history?" CL: "Well, I have been thinking of history." Movement—non-movement dimension Ratings on the movement dimension are made indepen- dently of ratings on the other four dimensions of the scale. This rating is based solely upon the judge's professional impression of the quality of the client's responses, that is how important the communication is in arriving or working with the client's own con— cerns. This rating is not based upon whether the response has been judged affective-cognitive, etc., but how well the client may be differentiating stim- uli, lowering defenses, or owning responsibility for his actions. While this is certainly not easy to evaluate over a few statements, the judge is urged to try and determine client responses which indicate his progress in the context of the counseling session. A rating of “ indicates that the judge considers this response as being highly appropriate in the given situation while a 3 indicates that the response 137 is appropriate but could be better. A rating of 2 indicates a neutral response which neither measur- ably demonstrates client progress nor the lack of it. A rating of 1 indicates a response which not only shows little progress, but which demonstrates a loss of movement or decided lack of progress, e.g., indicating defensiveness, resistance, etc. APPENDIX H SUMMARY OF COUNSELOR LEVELS 138 w.‘ If;.‘.. - nan-am- nun-AT— _ _ a 139 Means and range of the high and low counselor groups and total sample on each dimension of the Counselor Verbal Response Scale Mean Counselor Levels Dimension High Range Low Range Total .50 (0.67—5.00 6.86 [U Affective 11.00 ( 9.67—13.33) Understanding 16.80 (1“.67-18.33) .“1 (1.33-6.33) 11.01 LU Specific 13.67 (11.00—17.33) 3.11 (0.00-6.00) 7.96 Exploratory 15.88 (13.33—18.33) “.81 (0.00—8.33) 10.96 Effective l“.61 (12.67—17.33 3.30 (0.33—“.33) 9.66 APPENDIX I CASE SUMMARIES 1“O 1“1 The following brief case summaries are in their original form, as written by each of three judges follow- each session. These summaries have been discussed in Chapter V. A code number precedes each session summary. The agreement among the judges is quite high. The ses- sions are grouped according to counselor effectiveness as judged by the Counselor Verbal Response Scale, and as to whether the client made or did not make therapeutic movement. I. High Counselor Effectiveness--Client Movement (136.8) Judge #1: The counselor is trying to probe and explore. The client appears a bit frigh- tened and hesitant, but nevertheless tries. It is somewhat vague as to whe- ther the client is expressing affect or cognition, but I think that he is trying to stay with it. Judge #2: The client is both willing and able to express his feelings. He is exploring himself well and seems to be picking up insights along the way. The counselor is right with the client; the client follows the counselor's leads and adds to them. Judge #3: The counselor is primarily clarifying for the client. The client is exploring himself, and the counselor is doing a good job of keeping him on the right track. The client seems to engage in a lot of explanation of his own responses, which tends to take the feeling tone out of them. (135.10) Judge #1: Judge #2: Judge #3: (127.9) Judge #1: Judge #2: Judge #3: (127.10) Judge #1: Judge #2: Judge #3: 1“2 The client is emitting a lot of affect here but doesn't really specify it or verbalize it. The counselor is good; he tries to focus on the client's feel- ings and deal with them. He helps the client begin to explore them. The client appears very nervous; conse- quently, the counselor is trying to put his finger on it. He is staying with her and is helping her to express her- self. He is doing a good job, and she is working with him. Overall, this is a good session. The counselor is excellent here, sensing the client's anxiety and responding to it. The client in turn responds to the coun- selor and they both interact well with each other, laying the groundwork for a meaningful relationship. Good interview. Client and counselor are moving together; just beautiful. Couldn't be better; both are really moving. No comment. Counselor is doing a really good job; he is focusing very well for the client. The client is very defensive but is taking some risks. Counselor is focusing very well. The client is anxious and sometimes seems to agree with the counselor's interpreta— tions but does not follow them through to generate further material. (135.15) Judge #1: Judge #2: Judge #3: (121.2) Judge #1: Judge #2: Judge #3: l“3 The counselor is not with the client here. The client makes a little movement toward the end of the session but does this on her own. The counselor is reflect- ing content; the client is not moving deeply into anything. The client is not moving anyplace in this session. She drops some hints that she has some things to discuss, but just talks about superficial areas. The client is nervous, but the counselor isn't respond- ing to it. Near the end, the client moved the session into an area of importance to her. The client at first wanted to have a nice social discussion. Part of the way through the session, however, she began to lead it into an area of importance upon which the counselor helped her focus. The client begins with a vocational area problem but seems to want to shift the focus to her feelings about her vocational choice and some personal concerns that may go with it. The counselor ignores her feelings and sticks to the areas of voca- tions and jobs. The counselor is letting the client do all the leading; as a result, the client is wandering all over the place. The counselor is not helping the client to focus. The client does some exploration, but it doesn't seem to be new material to her. Whatever little the client get to, she does it by herself. Nothing of importance seems to be taking place in this session. The counselor doesn't seem as if he's listening at all. The client mentions something about her view of the world which may have some rele- vance to her, but the counselor seems unable or unwilling to deal with it. II. luu Low Counselor Effectiveness-—Client Movement (136.2) Judge #1: Judge #2. Judge #3. (121.8) Judge #1: Judge #2: Judge #3: (127.u) Judge #1: This is a very difficult session to rate. The counselor is responding poorly and gives the client few good leads. The client tries to express some of her feelings but the counselor does not take notice. The client is striving and moving on her own. The counselor is holding the client back; he is not accepting at all and con- sequently tends to turn the client off. The client is being put on the spot; she is feeding out information that the coun- selor isn't picking up. The counselor is slowing down any pro- gress that is being made. For every step the client makes, the counselor pulls her one step back; but the client keeps try- ing to move. Terrible interview; the counselor might as well not be there. The client isn't discussing anything of relevance. The counselor is terrible, not seeming to pick up anything the client says. I think the client has Just given up on him. Poor session; the client isn't going any- place. Even if she were to discuss some- thing of importance to her, the counselor would not be able to respond adequately to her. The counselor sometimes appears to be trying to move things here, but the cli- ent isn't very receptive. The client's orientation seems to be toward thinking and content rather that to his feelings. Judge #2: Judge #3: (127.6) Judge #1: Judge #2: Judge #3: 1&5 This is a fair--not a poor—session, with both partners contributing to the wander- ing. The whole interview is vocationally orien— ted, although late in the session the tone changes to include some personal concerns. The counselor seems to pick up the client's concerns, but the client seems to stand still or move away from him. There is a bit of movement, but not very much. The pace of this interview is quite slow. After a while, the client changes the subject to involve more of himself; the counselor picks this up, but any attempts to move with these areas are then spurned by the client. There was a slight bit of movement; but this was very small. The client appears to be making some move- ment. The counselor seems to be pretty good, although he doesn't help to focus as much as he could. The client seems to be exploring his concerns quite a bit. The client is moving but is somewhat held back by the counselor. The client is about to get into some important con- cerns even though the counselor is not quite with it. The client began with a vocational problem but has switched the tenor of the discussion to her own personal—social concerns. The client seems to be producing a lot. The counselor could reflect much more than he does, but the client seems to move forward nevertheless. The client does get slowed down and has to deal with the counselor's statements, which for the most part are not relevant. (136.1) Judge #1: Judge #2: Judge #3: (121.6) Judge #1: Judge #2: Judge #3: 1&6 The counselor is not understanding the client at all during this session. The client is moving, however, in spite of the counselor. The counselor misunder- stands the client, who nevertheless pushes ahead by herself. The counselor appears nervous and not at all with this client. The client is mak- ing some movement, but all on her own. The client is discussing material import— ant to her, concerning her relationship to her divorced parents. The counselor does not seem to be listening to her, for he isn't responding. The counselor sometimes seems to respond to the client's concerns, but at other times moves away from them. The client is a bit afraid, but is willing to dis- cuss. The counselor probes and focuses. This is a good session although the client is very hesitant. The client seems to be quite naive. The counselor is trying to focus and pull the client along but is not clear as to what he wants the client to do. There is some movement. The client is nervous and resistant and tries to fend off the coun- selor by constantly asking for clarifica- tion. There is some movement which appears to be due to the counselor's behavior. There is some movement here. The client did some exploring but was generally resistant to the counselor's leads. III. 1H7 High Counselor Effectiveness-- Lack of Client Movement (121.7) Judge #1: Judge #2: Judge #3: (121.5) Judge #1: Judge #2: The counselor is somewhat aggressive here and seems to be turning the client off. The client is exhibiting some affect, but it is in the form of hostility toward the counselor, with no understanding of what is going on (i.e., the hostility is fairly subtle). The client attempts to bounce back from some negative feelings toward the counselor. However, the coun- selor is simply too verbal, and the client raises her defenses too often. The counselor seems to be getting a little annoyed as well as the client. The cli- ent is closing off any leads and getting hostile while not recognizing it. She tries to get off the hook from the coun- selor's direct questions but comes back to them at times. She then closes off and intellectualizes. She is exploring, but very reluctantly. The session is like pulling teeth. The counselor seems to be doing a pretty good job, although he sometimes tends to keep the client boxed in, something which she fights against. The client wanders through this session and makes responding difficult for the counselor. The client is not very aware of any of her own definite feelings. The counselor tends to repeat what the client has said, asks questions that are much too broad, and generally leads the client away from the central issues. The client is standing still. The client is extremely verbal and tends to wander all over the place. The counse— lor isn't very active and allows her to keep talking, which is very unproductive. They both seem bored with what is taking place. Judge #3: (128.3) Judge #1: Judge #2: Judge #3: (136.7) Judge #1: Judge #2: Judge #3: 1&8 This client appears to be confused and gets more so as the interview proceeds. The counselor constantly retraces and makes extraneous comments which would be better off not said. The counselor seems to be doing a good job, but the client is resisting all the way. The counselor seems to stay right with the client and keeps him on his cen- tral concerns, even though the client keeps trying to change the subject. The client doesn't seem to be using what the counselor is giving him. The client is extremely defensive in this interview; he is clamming up and not giv- ing much of himself. The counselor moves in well but apparently gave too much too soon to the client. The counselor does try to help the client with his nervous- ness, but the client is really closed to it; he simply doesn't want to talk. The client is extremely defensive: He denies much of what the counselor reflects and runs away from any self-involvement. The client is playing a game here, but the counselor does not see that he is being played with. The client is not wil- ling to look at herself and contradicts herself constantly, even though the con- tent of what she is giving might sound important. I doubt whether it is true. The counselor isn't bad although the client is playing games with him. She is hostile but very subtle, a first—class bitch. I feel that the counselor does an excel- lent job with a very defensive, intellec- tualizing client. Any movement she may be making, is quite shallow. It is difficult to believe her. IV. (128.5) Judge #1: Judge #2: Judge #3: 1149 The counselor doesn't appear to be very interested in the client. He doesn't interact with her very much; she therefore wanders and accomplishes little. Nothing much is going on here. The coun- selor is supportive but very passive and the client is just making pleasant con— versation. The client initiated a descriptive of her school's social world. The counselor bought this wholesale and didn't pursue anything, simply allowing the client to go on and talk. Low Counselor Effectiveness--Lack of Client Movement (127.12) Judge #1: Judge #2: Judge #3: This is a generally poor interview. The majority of the interview is a conversa- tion based upon content matter. The counselor doesn't pick up any of the indi- cated affect at various points. The client is just trying to talk without giving her own feelings and does not seem to recognize much of her nervousness. The counselor keeps the conversation on a very general level. The client expres- ses some nervousness, but the counselor foes not seem to know in which direction to go. It's hard to tell whether the client would follow his lead, anyway. The interview is generally poor, both individuals contributing to the lack of movement. The client exhibits nervous- ness in a number of instances, mostly having to do with some self-conscious con— cerns, i.e., "Am I saying the right thing," "Will anyone view this tape?" (128.1) Judge #1: Judge #2; Judge #3: (121.4) Judge #1: Judge #2: Judge #3: (135.16) Judge #1: 150 The counselor is rather poor and doesn't seem with it at all. He fouls the client up. Nothing is going on here. The interview is generally vocationally oriented. The counselor is slowing up the processs, not helping the client to move ahead. It is a very boring session. The counselor seems to follow up blind areas; he has no idea of where the client is going. The client doesn't seem to make any progress. The counselor is very poor here. He tends to take the client away from her feelings and her central concerns. The counselor doesn't relate the client's verbalizations about others back to the meaning they might have for her. The client is very verbal in this session but isn't focusing at all. The counselor is not giving her very much help, either, simply letting her wander. Nothing much is going on; they both seem to be trying hard just to maintain a pleasant conver- sation. The client doesn't discuss either herself or her problems. The coun- selor stops any movement at might take place. The client stays away from discussing her— self. They discuss a lot of extraneous material. The move around and around in the same areas but do not deal with any- thing. The counselor is quite poor in this ses- sion; he seems to have virtually no understanding of the client, nor does he attempt to move in the client's feeling level or help the client explore. The Judge #2: Judge #3: (128.6) Judge #1: Judge #2: Judge #3: 151 client simply engages in conversation and answers the counselor's questions, which just maintain the conversation. The counselor here is not pushing the client at all. The client is not expres- sing anything that might be of concern to her. There is very poor interaction between them. It is a slightly nervous general conversation; a quick question- and-answer period. The client acquiesces to the counselor's lead, which is very poor. The counselor seems to be dealing with his own con- cerns; the client spends the session try- ing to deal with the counselor but not the things that might be troubling her. The counselor is functioning well here. He focuses on feelings and lets the client work through some things that seem to be bothering her. She moves quickly into important problem areas. The counselor seems to be slowing down the process, not knowing where to take the interview. He moves over previously- covered territory. The client is hurt but doesn't approach it. The counselor does ask some questions, but they don't seem to be relevant. The client is moving into her feelings of being hurt, but the counselor is not handling it very well. He seems to be missing her reactions and consequently slows her down. Counselors Rated Neither High Nor Low-- Mixed Degree of Client Movement (121.3) Judge #1: The counselor in this session retards any movement that might take place. He focuses on a tangential question, makes Judge #2: Judge #3: (127.7) Judge #1: Judge #2: Judge #3: (128.2) Judge #1: 152 interpretations quickly, and generally so controls the session at a point different from where the client is headed, so as to forestall anything important taking place. The client appears to show very little understanding of people in general or her- self in particular; she is quite naive. The counselor is quite punitive and is punishing her for her naive views. The client begins the session with much unfocused anxiety. The counselor picks up one aspect of her communication and lectures to her on the meaning of patri- otism. The client appears to be totally confused as to what is really going on. The counselor is not picking up anything and the client is getting very angry. The client is not moving; the counselor is not picking up her affect. He seems to be defeating the client. Bad interview. The counselor is very bad. This session is very affect-loaded with the client get- ting very angry. The counselor is just not hearing her and is leading her away from every important area. The client is going ahead and exploring. The counselor is going backwards. He is protective of the school policy, jus- tifies certain actions and over all, tunes her out. He seems to misinterpret everything she says. The counselor slows this client down. He isn't focusing on the problem at hand and pulls the client away from her feeling. Judge #2: Judge #3: (128.u) Judge #1: Judge #2: Judge #3: (135.3) Judge #1: 153 The client wants to discuss her feelings but is being directed otherwise by the counselor. The client is feeling depres- sed and would explore this is she could. The counselor seems helpful sometimes and not helpful at other times. The counselor is problem oriented and doesn't pick up the feelings the client is trying to express. The client makes some movement in spite of the counselor. The counselor tends to lead the client astray here. The counselor doesn't attempt to react to or to clarify the client's feelings. The client spends most of the session giving examples of what he means in order to make the counselor understand. The counselor is slowing the client down a bit. She is missing a lot of what is going on, but does help to clarify the client's statements. It is hard to judge the client's involvement with the session, but he is expressing some hostile feel- ings and is working with them pretty well. As the counselor began to help the client focus on his problem, the client became somewhat more resistant. The counselor consequently missed some responses, and the session then became problem oriented, not dealing with the feelings which greatly affected their interaction. The counselor pulls the client away from her feeling. He seems to hit once in a while at certain important points but doesn't stay with it, coming back to more cognitive material. No movement seems to be taking place. Judge #2: Judge #3: (136.6) Judge #1: Judge #2: Judge #3: (127.8) Judge #1: 15“ The client begins to get into the family situation, but the counselor draws her away from it and into the vocational area. It is hard to know where the client is going or whether the vocational area is relevant at all. This makes the session difficult to judge. The client is on the cognitive level, and so is the counselor. The client is not likely to move without any counselor initiative, and he is not making any. The client flirts with the tendency to explore but the counselor reinforces her problem-solving behavior. The counselor is supportive and reflec- tive, but seems to move first toward and then away from the client's concerns. He pulls the client off the feeling level and works into irrelevant areas. When a feeling does come out, he isn't able to deal with it. The client opens up the session with strong feelings of doubt about himself and his work. The counselor turned it off and fell into the trap of supporting the client's laziness in school. The counselor is not focusing well and jumps all over the place. The client seems willing to move, and makes a little move- ment, but simply doesn't know how to talk about it. The counselor is fishing for ideas ini— tially and not following through on what he does turn up. The client is somewhat sullen. The counselor is using a shotgun approach with little in the way of results. This session is informational, with a lot of social chitchat. The client is extrem— ely verbal and seems to overwhelm the counselor with it; there is no movement taking place. Judge #2: Judge #3: (121.1) Judge #1: Judge #2: Judge #3: 155 The client is talking endlessly on and on; the counselor doesn't cut in. This session is very poor. The client is talking endlessly; the counselor cannot or does not try to get a word in. Nothing is going on here except for a long trail of verbal garbage. There is a decided lack of movement in this session. The counselor seems a bit hostile and out of touch with the client. He seems to have a few standard ques- tions that he is going to ask regardless of whether they are appropriate. The counselor and client are moving around in the area of relations with parents. They are not moving ahead rapidly, but do manage to hit some sig- nificant areas. The counselor, however, tends to move away from the subject at hand. The client produces some gain in this session, but it is on her own. The coun- selor seems to move in and out, hitting a chord but then not dealing with it. The client seems willing to deal with her problems. "711111111111111')?