THE RELATIONSHW BETWEEN CLEENT mvowmam m ¥ ‘ PSYCHOTHERAPY EVALUATION AND , EFFECTNENESS 0F SHORMERM ~ PSYCHDTHERAPY: A PILOT STUDY r ‘ Sissez‘tation far the Degg'eezof Ph‘ D. V 5 , meme»; sma UNIVERSITY- PfiWICEA RESEK UPDYKE ' ‘ I 9 75 ABSTRACT THE RELATIONSHIP BETWEEN CLIENT INVOLVEMENT IN PSYCHOTHERAPY EVALUATION AND EFFECTIVENESS OF SHORT-TERM PSYCHOTHERAPY: A PILOT STUDY The purpose of the study was to examine the relationship of client involvement in the evaluation of psychotherapy sessions to the effectiveness of short-term psychotherapy. Very few studies of psycho- therapy effectiveness have focused on the client's evaluation. Even fewer studies have been conducted in a community mental health setting. All types of public agencies are increasingly being confronted with the need to provide data which verify that effective services are being delivered. The further impact of the consumers' rights movement has added impetus to the current thrust toward accountability. Be- cause many public facilities espouse a short-term psychotherapy focus. sometimes out of necessity, and sometimes out of a firm belief in the phleSOphy of short-term treatment, a need exists to evaluate the effectiveness of such a focus. Little research has been directed toward evaluating short-term psychotherapy, yet increasing numbers of clinicians and agencies are adopting this treatment orientation. The additional factor of involving clients in the evaluation of their own treatment has many implications, for both the treatment itself and the accountability demands. It was with these issues of client involve- ment in evaluation and effectiveness of short-term psychotherapy in mind that the study was undertaken. Patricia Resek Updyke Thirty-eight outpatient clients seen for individual short- term psychotherapy at the Ingham community Mental Health Center in Lansing, Michigan comprised the sample of the study. One Treatment Group (n = l2) included those clients "involved" in the evaluation of their psychotherapy sessions by completion of the Therapy Session Re- port after each of four psychotherapy sessions. The experimental group was compared with two control groups (n = ll; n = l5) on meas- ures of psychotherapy effectiveness. Psychotherapy effectiveness was defined by three components. Two of the components consisted of positive changes between pretesting and posttesting measures of self concept, as measured by the Tennessee Self Concept Scale (TSCS) and number of self-identified problems, as measured by the Mooney Problem Checklist. The third component was a posttest measure of satisfaction with psychotherapy (Therapy Session Report-SATS). Clients were randomly assigned to therapists and treatment groups, and their scores on the three measures of psychotherapy effec- tiveness were compared simultaneously by multivariate analysis of covariance. Helmert Contrasts were used in the analysis to combine the control group means for comparison with the Treatment Group means. The primary research hypothesis was that at the end of short- term psychotherapy, clients who evaluated their own psychotherapy would score higher on measures of self concept and satisfaction with psychotherapy, and report fewer self-identified problems, than clients in two control groups who did not evaluate their own psychotherapy. The null hypothesis was ngt_rejected. Patricia Resek Updyke Because the research hypothesis was not rejected, post hoc analyses were conducted to help identify possible reasons for the findings. Ten questions were raised about the research design and problems encountered during the implementation of the study. The analyses indicated that the TSCS was a stable instrument from pre- testing to posttesting, but the MPC was highly unstable. This finding implied that the semantic meanings of the MPC questions changed from pretesting to posttesting. A high drop-out rate did not appear to be related to dissat- isfaction with services. Possible reasons for the high drop-out rate were conjectured regarding annoyance or anxiety factors associated with the completion of the checklist (TSR). Evidence was cited which supported the appropriateness of the instruments used in the study and the number of sessions chosen to define short-term psychotherapy. Theoretical issues regarding effectiveness of the treatment, client "involvement," and appropriate- ness of the sample for short-term psychotherapy were explored which led to suggestions for refinements in the design. Because the research hypothesis was not rejected; the major conclusion was that client involvement in evaluation of short—term psychotherapy did not significantly affect psychotherapy effectiveness. The recommendation was made that the MPC means and standard deviations be used as norms for populations similar to the mental health center adult sample. THE RELATIONSHIP BETWEEN CLIENT INVOLVEMENT IN PSYCHOTHERAPY EVALUATION AND EFFECTIVENESS OF SHORT-TERM PSYCHOTHERAPY: A PILOT STUDY By Patricia Resek Updyke A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Counseling, Personnel Services and Educational Psychology 1975 ACKNOWLEDGMENTS I wish to express sincere appreciation to the many people whose support was invaluable during the completion of this dissertation. First, I wish to thank Eric for his enduring patience. To Dr. Alex Cade, my adviser and dissertation chairman, I wish to express special thanks. Through our relationship, I have grown both as a person and as a professional. Hera istnlmve has 86\ I also wish to pay special tribute to Dr. William Farquhar, who gave so freely of his time and efforts to facilitate the completion of this dissertation. Through his support I was able to overcome seemingly insurmountable obstacles throughout the final stages of my doctoral program. To my other committee members, Dr. Teresa Bernardez- Bonesatti, Dr. John Hurley, and Dr. John Jordan, I extend sincere appreciation for their guidance, interest, and constructive contribu- tions. Appreciation and affection is extended to Dr. Romualdas Kriauciunas, for his inspiration, encouragement, and advice throughout my entire doctoral program. Our relationship has taught me the true value of friendship. A special thank you is extended to Bob Moreas, whose under- standing and support were always offered when needed the most. 11 The Ingham Community Mental Health Center staff, who toler- ated many inconveniences during the implementation of this study, deserve special recognition. A note of appreciation is also due to those whose assistance was vital: Jean Emerick, Jim Jenkins, Connie Pease, Nancy Thelen, and Helen Therrien; and to the therapists who helped collect the data: Mary Alice Collins, Jan Conti, Dave Harley, Jean McIntosh, Christy Minning, Bob Moreas, Bob Neville, Kathy Nouveau-ne, Lynne Rich, Muriel Rokeach, Jim Stueland, and Betsy VanNesten. The administrative support of Dr. Eugene Friesen and Mr. Edward Oxer, and the Tri-County Mental Health Board-Clinton, Eaton, Ingham, was invaluable. Appreciation is also extended to Dr. David Orlinsky for his permission to use the Therapy Session Report. And finally, special thanks are extended to my parents for their encouragement throughout my academic endeavors. Through their efforts, I have acquired a love of learning and creativity. Use of the Michigan State University computing facilities was made possible through support, in part, from the National Science Foundation. TABLE OF CONTENTS CHAPTER I. THE PROBLEM ....................... Statement of the Problem ............. . . Purpose of the Study ................. Definitions. .................. . . . . Overview ....................... II. REVIEW OF THE LITERATURE ................ Short-Term Psychotherapy ............... Historical Perspective ........... . . . . Application of Short-Term Psychotherapy and Related Research ................. Summary of.Short-Term Psychotherapy Research . Phenomenological Self Theories ............ Historical Perspective .............. Definition and Dimensions of Self Concept . . . . . . Application of Self Concept Theory and i Related Research ................. Self Concept Change ................ Creative Client Involvement in Psychotherapy Process .............. Client Involvement in Evaluation of Psychotherapy .................. Summary of Self Concept Research .......... III. DESIGN OF THE STUDY ................... Sample ........................ Instruments. ...................... Tennessee Self Concept Scale (TSCS) ......... Introduction and Test Development ......... Administration ................... Reliability and Validity ............. Advantages and Criticisms of the TSCS ....... Mooney Problem Checklist (MPC) - Form A ....... Introduction and Test Development . . ....... Design and.Administration ........... . . MPC Scales .................... Reliability and Validity ............. Therapy Session Report — Form P ........... Introduction and Test Development ......... iv Administration .................. 54 Reliability and Validity ............. 54 How the TSR Was Used in the Study ......... 55 Instrument Correlations ................ 57 Design . . ................. 58 Testable Hypothesis ................. 60 Analysis ...................... 60 Summary ........................ 61 IV. ANALYSIS AND DISCUSSION OF RESULTS ........... 63 Results ............... . ......... 63 Hypothesis Testing ................. 63 Means and Standard Deviations of the Sample ..... 64 Exploratory Analyses ................. 64 Post Hoc Factor Analyses .............. 68 Checklist Questions ................. 73 Satisfactioneby-Therapist Analysis ......... 77 Discussion ...................... 78 High Drop-Out Rate ....... . . ........ 79 Treatment Effectiveness ............... 82 Adequacy of Theory and Instruments ......... 83 Timing of Posttesting ................ 84 Instrument Appropriateness ............. 86 Therapist- by-Client Interaction Effect ....... 88 Number of Short- Term Psychotherapy Sessions ..... 89 Summary .............. . . . ...... 90 V SUMMARY AND CONCLUSIONS ............ . . . . . 93 Summary ........................ 93 Conclusions ...................... 95 Recommendations for Future Research .......... 95 APPENDICES A. THERAPY SESSION REPORT: FORM P (TSR) .......... 98 B. SATISFACTION QUESTIONNAIRE (SATS) ............ ll3 C. THERAPIST DEMOGRAPHIC DATA ................ ll4 D. CLIENT CONSENT FORMS . . . ................ ll5 E. RELIABILITY OF THE TENNESSEE SELF CONCEPT SCALE (TSCS) IDENTITY SELF. JUDGING SELF. AND BEHAVIORAL SELF FOR THE SAMPLE ..................... 117 F. MEANS AND STANDARD DEVIATIONS OF THE TENNESSEE SELF CONCEPT SCALE (TSCS) IDENTITY SELF, JUDGING SELF AND BEHAVIORAL SELF FOR THE SAMPLE PRETEST ............. . ...... G. MEANS AND STANDARD DEVIATIONS OF THE TENNESSEE SELF CONCEPT SCALE (TSCS) IDENTITY SELF. JUDGING SELF,.AND BEHAVIORAL SELF FOR THE SAMPLE POSTTEST ................... BIBLIOGRAPHY ......................... vi Page 118 119 120 Table 3.1 3.2 3.3 3.4 LIST OF TABLES Client Assignment and Therapy Completion Rates for the Sample of the Study ......... Mean Ages of Clients by Treatment Group for the Sample of the Study ............ Percent Age Distribution of Sample and Clients of the Ingham Community Mental Health Center (ICMHC) ...................... Reliability of the Tennessee Self Concept Scale (TSC) .................... Comparison of Tennessee Self Concept Scale Sample Means with Norms and Related Studies . . . . Comparison of.Mooney.Problem Checklist Sample Means and Standard Deviations with Writing Therapy Sample .................. Intercorrelation Matrix of the Dependent Variables Range of Significant Correlations of the Inter- correlations of the Dependent Variables ...... . . Number of Significant Correlations Between the Variables in the Study .............. Means and Standard Deviations of All Measures for the Treatment Group of the Study ....... Means and Standard Deviations of All Measures for Control Group I of the Study ......... Means and Standard Deviations of All Measures for Control Group II of the Study ......... Varimax Rotation Analysis of Sample Pretests . . . . Varimax Rotation Analysis of Sample Posttests . . . . vii Page 38 4O 4O 45 47 51 56 57 67 69 7O Page 4.6 Mean Satisfaction.Scores (SATS) and Spot-Check Question Scores of the Treatment Group . . ...... 74 4.7 Mean Satisfaction Scores (SATS) of the Sample by Therapist . . . . . . . . ............. 77 viii LIST OF FIGURES Figure Page 4.1 Factoral Structure of the Varimax Rotation Analysis of Tennessee Self Concept Scale (TSCS) and Mooney Problem Checklist (MPC) Pretest and Posttest Scores of the Sample ....... 72 ix CHAPTER I THE PROBLEM In recent years, there has been a growing effort to devise and implement more efficient mental health services to an increasingly aware public. As consumers demand more effective services, health care professionals are faced with a manpower shortage. It is highly unlikely that demand for services will ever decrease. Our society "is becoming increasingly insistent upon the dignity, the well-being and good health of all its citizens" (Small, l97l). Therefore, the effects of this shortage on mental health programs must be reduced, either by dealing with the shortage itself and increasing the supply of trained professionals and paraprofessionals who provide the services, or by innovative changes in treatment itself. Various methods of increasing manpower supply have been proposed and imple- mented. The current paraprofessional revolution is one effort toward a solution. Another solution is geared toward the more efficient utilization of current manpower resources in treatment. Given a limited supply of manpower, how can these resources be distributed for maximal benefit? One answer is the development of short-term psychotherapeutic.techniques which will allow the professional and paraprofessional to meet the immediate mental health needs of more individuals than could ever be accommodated by the traditional long- term psychotherapist. Statement of the Problem Two focal issues generating concern among health care pro- fessionals have surfaced in recent years. The value and effectiveness of short-term psychotherapy is one issue. The second issue is the ever-increasing pressure to deliver more services, without jeopardizing effectiveness, and at the same time, to provide accountability for those services to funding sources and consumers. Before investigating the issue of effectiveness of short- term psychotherapy, the terminology must be defined. For purposes of this study, psychotherapy will be defined as an educative process in- volving a relationship between at least two persons, one of whom espouses a theoretical framework of the change process and accepts responsibility for helping the other to change. Regarding the number of sessions, how short is "short-term?" In their review of the liter- ature, Phillips and Wiener (I966) cite numerous studies of short-term psychotherapy which define short-term as anywhere from l to 100 sessions. Small (1971) quotes many studies with short-term ranging from l to 217 sessions. Both authors agree that the number of hours cannot be set, and issues such as length of sessions and time span between sessions are additional factors which need to be investigated. Bellak and Small (l965) expect brief psychotherapy to be accomplished in the short range of l to 6 therapeutic sessions of customary duration (45 to 60 minutes). Avnet (Wolberg, 1965) defines short-term as a maximum of 15 sessions. Parad (1969) defines "brief" as 10 to 12 interviews in a period of up to three months. For purposes of the study, short-term is defined as 4 psychotherapy sessions, each spaced one week apart. The Ingham Com- munity Mental Health Center, located in Lansing, Michigan, has an outpatient program which operates within a short-term psychotherapy model. Clients of the Outpatient Service come for an average of 4.06 sessions. The second issue, that of accountability, generates concern about "who" should do the evaluating and provide the data used for accountability purposes. Most research related to the process or outcome of psychotherapy has depended primarily on the reports of therapists or independent observers. A dearth of literature is avail- able using client self-reporting of progress or effectiveness of psy- chotherapy. The scarcity has also been influenced by the traditional belief that the therapists are the experts, and the clients, since they are the ones seeking help, are the "sick" ones. This belief follows the traditional medical model regarding treatment and its evaluation and implies impaired perception and judgment of the clients. An additional factor influencing the dearth of literature has been the scientific approach to research. In the scientific framework, subjec- tively has been seen as unreliable. The only legitimate data were based on behavior, objective measurement, and observation (Glick, 1971). It has only been in the last two decades that the phenomeno- logical approach to research has been accepted as valid in some circles of the scientific community. The phenomenological viewpoint asserts that ". . . all behavior is both determined by, and related to, the behaving organism's phenomenal field . . . which includes everything of which a person is aware at the moment of action. Though the degree of awareness may vary, an individual's behavior is the result of his perceptions of reality, not the result of the physical reality itself" (Fitts, 1971). Many contemporary theorists and practitioners hold that the subjective experience of the individual is most relevant (Rogers, 1951; Berne, 1961; Kelly, 1963; Perls, 1971; and Ellis, 1971). Today, many therapists regard the client as more than a recipient of services. Mann (l973)_points out that there is a current "anti-intellectual" atmosphere in our country which is putting pressure on therapists to be more human and less clinical. Clients are also perceived as integral participants contributing to a mutually-enhancing and growth-producing relationship (Kell and Mueller, l966). Steinzor (1967), in describing therapists as "value-makers," adds that the "therapist's stake in the immediate relationship between himself and the patient is as great as the patient's stake." He believes ". . . if I am to reach him, he must also feel I have been affected by him." O'Connell (1970) also recognizes the partnership that psychotherapy involves. It is important that the client's report of the collabora— tive therapeutic experience be seen as valid, not only by therapists, but by researchers as well. Unfortunately, the literature is not replete with studies using self-reporting methods. The majority of literature evaluating psychotherapy still depends on trained clinician: to provide data defining client progress. Feedback from the client- participant is sadly neglected and goes untapped as a resource for learning more about the psychotherapeutic process. For these reasons, this study's definition of effectiveness is based on three components of client self-reporting: (l) a more positive self concept as meas- ured by the Tennessee Self Concept Scale (Fitts, 1965); (2) fewer self-identified problems as measured by the Mooney Problem Checklist (Gordon and Mooney, 1950); and (3) satisfaction with psychotherapy as measured by a section of the Therapy Session Report (Psychotherapy Session Project, 1966 [See Appendix A]). Agencies of all types, whether they receive funding from public or private sectors, are constantly confronted with the necessity of providing data to verify that effective services were, in fact, provided to identified populations. This study, which assesses one community mental health center, looks at an agency not unlike other social service organizations and counseling centers. As an example, the center investigated is accountable to county, state, and federal governments, as well as to third party insurance carriers such as Blue Cross, Blue Shield, Medicare and Medicaid. With the current interest in consumers' rights, accountability is under the close scru- tiny of consumers themselves. In this study, the consumers are the clients of the Ingham Community Mental Health Center. A movement in the direction of consumerism is already underway through the efforts of a Citizens' Advisory Council. As consumers become more educated about quality of care, they will be in a better position to "shop around" for mental health services to meet their individual needs. A basic underlying premise of accountability is that cost for services must be justified in terms of effectiveness. In a community setting such as that investigated in this study, focus could be on any one of numerous agents: the therapist, middle management, the agency, the Tri-County Mental Health Board, the State of Michigan Department of Mental Health, or, at the federal level, the National Institute of Mental Health. Needless to say, every level is accountable to some-. one, be it the consumer or some level of the governmental bureaucracy. Accountability data obtained from the client-recipient is a relevant beginning. It was with these issues in mind that the study was under- taken. Purpose of the Study The purpose of the study is to investigate the relationship between client involvement in psychotherapy evaluation and effective- ness of short-term psychotherapy. The study investigated the following research hypothesis: At the end of short-term psychotherapy, clients who are involved in the evaluation of their own psychotherapy will score higher on measures of self concept and satisfaction with treatment, and report fewer self identified problems than clients in two control groups (which do not differ) who do not evaluate their own psy- chotherapy. Definitions For purposes of this study, the following definitions were used: CLIENT INVOLVEMENT: the process by which the clients themselves evalu- ated their psychotherapy sessions by completing a questionnaire (the Therapy Session Report, Form P) after eaCh session. (See Appendix A) EFFECTIVENESS consists of three components. Two components involve positive changes between the time immediately prior to the first psychotherapy session and the time immediately after the fourth session, on two personal phenomena: (1) self concept, as measured by the Tennessee Self Concept Scale, and (2) number of self- identified problems, as measured by the Mooney Problem Checklist. A third component, satisfaction, was measured by the Therapy Ses- sion Report (See Appendix B) after the fourth psychotherapy session. SHORT-TERM PSYCHOTHERAPY: four sessions of an educative process in- volving a relationship between at least two persons, one of whom espouses a theoretical framework of the change process and accepts responsibility for helping the other to change. Overview Chapter I includes a statement of the problem, delineates the purpose of the study, provides a brief theoretical overview, identifies a general research hypothesis, and defines the terminology. Chapter II is comprised of a more extensive review of the literature relating to short-term psychotherapy and the phenomenological self theories. The design of the study is described in Chapter III. The presentation, analysis, and discussion of the results are reported in Chapter IV. A summary, the conclusions, and the directions for future research are presented in Chapter V. CHAPTER II REVIEW OF THE LITERATURE A review of the literature has been summarized in relation to two theories. First, short-term psychotherapy theory and related research are reviewed. Second, phenomenological self theories and related research are summarized, with special emphasis on the client's perspective of psychotherapy. Short-Term Psychotherapy Historical Perspective "Brief psychotherapy is at least as old as Freud's efforts to find a cure for the neuroses, . . . ." Thus began Small's (1971) compendium of the briefer psychotherapies. Even Freud searched for a "quick cure," with emphasis on quick diagnosis of psychodynamics and their undoing through active interpretation. The concepts of brief vs. long-term therapies did not evolve until long after psy- chotherapy and psychoanalysis had developed into lengthy processes. Phillips and Wiener (1966) also looked at the question of how to shorten psychotherapy, viewing it as a question again dating back to the days of Freud. Interest in psychotherapy has grown at a rapid rate since the mid-1940's, a time when concern about mental health also became widespread. Around that same time, Alexander and French (1946) introduced their ideas about "psychoanalytic therapy," a briefer 8 form of psychoanalysis which placed more emphasis on the activity of the therapist and individualized treatment. According to Wolberg (1965), investigation of the methodologies of short-term psychotherapy have been hampered by the traditional view that long-term psychotherapy is the most effective approach to solving emotional problems. This view was further supported by historic, philosophic and economic factors. Wolberg (1965) further stated that time spent in formal therapy was not the only variable involved in therapeutic gain, nor is it necessarily the most important ingredient. Also contributing to increased interest in shortening psy- chotherapy has been the development of health insurance programs, many of which now finance a limited number of psychotherapeutic ses- sions (Wolbert, 1965). Short-term psychotherapy is not a poor sub- stitute for longer-term varieties. Short-term psychotherapy is often a treatment of choice (for reasons which will be delineated in the next section) and is not, as many believe, a treatment approach for the poor, the naive, the unmotivated, the lower class, or those who are not verbally fluent. Phillips and Wiener (1966) reported that structured short-term therapy is better than long-term, conventional therapy because it is "structured to solve specific problems, regard- less of whether they are chronic and serious or only mildly disabling." Small (1971), in reviewing the briefer psychotherapeutic techniques, pointed out that little was reported during the 1940's or during the next decade, to reflect the actual rapid growth of brief psychotherapies. He noted that it was Lindemann's (1944) classic investigation of the survivors of the Coconut Grove nightclub fire in 10 Boston, and his enunciation of immediate intervention procedures, which almost solely advanced the concept of emergency psychotherapy. Mean- while, other factors were influencing momentum of growth of shorter psychotherapeutic techniques: growing impatience with psychoanalysis; a short supply of trained therapists; the necessity to begin providing more adequate mental health services to the poor; an increasing public awareness of availability and effectiveness of psychotherapy; govern— mental influence and support of mental health services; and various theories which focused on the problems of dealing with life crises. Fenichel (1954) prophetically spoke of brief psychotherapy as ". . . the child of bitter practical necessity." Steiper and Wiener (1965), commenting on the therapist as a "prime mover" who must "abandon favorite illusions of non—commitment and non-involvement" have described innovative short-term practices, such as time limits, homework, bibliotherapy, goal setting, and propagandizing as attempts to facilitate the greatest possible impact in the shortest possible time. They pointed out that over three de- cades ago Rank thought of the idea of time-limiting psychotherapy, but the idea has only recently been exploited. Lewin (1970), in describing the urgent need for effective brief psychotherapy in our "anxiety-ridden" age, stated that current treatment methods cannot possibly fill ever-increasing demands for psychiatric care. "If psychotherapy could be shortened and still re- main effective, many of the problems of community health care would be ameliorated. Regardless of any other benefits, the number of patients receiving therapy could be doubled or tripled." Lewin failed 11 to put much emphasis on the preventive aspects of brief psychotherapy, and this preventive aspect is foremost in importance for many propoe nents of brief psychotherapeutic techniques (Jacobson gt_al,, 1965; Wayne and Koegler, 1966; Farberow, 1968). The preventive role of brief psychotherapy is rapidly being recognized and accepted by mental health professionals. Historically, the theoretical and practical evolution of the crisis intervention model of short-term psychotherapy took place con- currently with the development of the Community Mental Health move- ment. One of the major findings documented by the U.S. Commission on Mental Illness and Mental Health in the United States (reported to Congress in the 1962 publication, Action for Mental Health) was the lack of immediate mental health services in the local communities of America. The Joint Commission recommended five basic components for comprehensive mental health services: (1) community consultation and education, (2) emergency service, (3) outpatient service, (4) partial- hospitalization and (5) inpatient service. The components of compre- hensive mental health services drew heavily on the preventive psychiatry ideas formulated by Caplan (1964). In a report on psychiatry's new approach, Time magazine (May 9, 1969) reported that many practitioners heralded the mental health movement as the third revolution in mental health. The first revolution was the medical discovery that the insane were neither criminals nor possessed by demons, but sick people. The second was Freud's insights into the emotional topography of the mind. The article concluded that the crisis intervention model of short-term psychotherapy was the most successful technique developed so far by the rapidly expanding Community Mental Health movement. 12 Application of Short-Term Psychotherapy and Related"Research Small (1971) was accurate in his assessment that "little material is available that analyzes process in brief psychotherapy; there is much theory but little research." The exceptions he listed were studies by Malan (1963) and Bellak (1965). Both studies were psychoanalytically-oriented. The conclusion Small reached was that “independent, well-trained clinicians are able, to a highly signifi- cant degree, to agree on the formulations of psychodynamics, a concise treatment plan and upon the actual process of psychotherapy . . ." furthen,they show that brief,we11-conceptualized, psychoanalytically- oriented therapy has both a demonstrable rationale and a success, and therefore merits a place in comprehensive mental health programs." Length of psychotherapy is one aspect of treatment which has been somewhat researched. In comparing the effects of length of therapy, two groups of patients, one seen six to ten times, and the other more than twenty-one times, Errera gt_al, (1967) found no sig- nificant differences in improvement rates, either as recorded by the therapist or as evaluated by independent raters. In their investigation of the "interminability“ of outpatient psychotherapy, Steiper and Wiener, (1959) found no correlation between length of time in therapy and the extent of the patient's improvement. They interpreted this finding as reflective of the failure of long- term therapists to formulate goals. After reviewing studies of long vs. short-term psychotherapy, Phillips and Wiener (1966) concluded that ". . . long-termness or interminability . . . seems most likely 13 to be determined by the dependency and conceptual needs of the patient and by the personality and theory of the therapist, and it apparently has little direct relationship to improvement in treatment." They reported in depth on four systematic follow-up studies (Batrawi, 1964; Test, 1964; Muench, 1964; Lorr gt_gl,, 1966). One study, conducted by Batrawi (1964), which compared structured and unstructured therapies, asked, ". . . if manipulation is a necessity (or if it is more evident in producing change), then why not introduce it openly and define the aspects of behavior which it can economically and more efficiently change?" Structured therapy produced definite, directly observable changes in behavior, and unstructured permissive therapy yielded more nebulous changes. In another study '(Test, 1964), which compared three types of structured short-term therapy (group therapy, individual therapy, and writing therapy, and a fourth group which declined help), each client received ten sessions of psychotherapy. The writing therapy group showed the largest number of changes, and the most significant ones, on measures of the MMPI, the Edwards Personal Preference Schedule, and the Butler-Haigh Q-sort. Test con- cluded that although ". . . each method produced some personality and behavior change, the results warrant further systematic study of such short-term therapy methods." In both these studies (Batrawi, 1964; Test, 1964), structured therapies involved changes in attitudes as well as parallel overt behavior change outside of therapy. A college study involving students (Muench, 1964) compared short-term (three to seven interviews), long-term (twenty-plus interviews) and time- limited (eight to nineteen interviews) therapy over a five year period, 14 and found significant changes on both the Rotter and the Maslow in- ventories for the short-term and the time-limited therapy groups, but not for the long-term therapy group. Lorr gt_gl, (1966) compared the therapeutic effects of time-limited and time-unlimited individual psychotherapy in a mental hygiene clinic. Over an eighteen week period that involved three separate examination times, the time-limited patients improved significantly with respect to somatic distress, tension, depression, bewilderment, and fatigue, but the time-unlimited cases made no improvement on these same measures during the same eighteen week period. These gains reported for the time-limited group had not decreased at the time of a forty week follow-up. Avnet's study (1965) of the effectiveness of short-term therapy reported cure or improvement of 76 percent of the clients at the end of the limited treatment period (maximum of fifteen sessions), despite the fact that most of the 1200 participating psychiatrists involved were analytically and long—term oriented and were skeptical of short-cuts in psychotherapy. A follow-up study two and one half years after termination showed that 81 percent of the patients re- ported sustained recovery or improvement. Avnet (1965) concluded that, in spite of the lack of scientific instruments to permit the objective measurement of these gains, there was impressive agreement by both the patients and the psychiatrists regarding the recovery or improvement with short-term methods, and that limited therapy "must" be effective for the majority of ambulatory patients. Shlien gt_gl, (1962), in comparing the effects of time limits in two different therapies (client centered and Adlerian), also 15 addressed the issue of deliberately brief therapy, defined as a maxi- mum of twenty sessions. The unlimited therapy group, with an average of thirty-seven interviews, showed significant results between pre- and posttesting and follow-up, indicating that “. . . unlimited therapy is effective." The time-limited client centered group, with an ave- rage of eighteen sessions, demonstrated that time-limited therapy is not only effective, but ". . . twice as efficient.“ The third (Adlerian) group, also time-limited, supported the effectiveness and efficiency of that approach with a sharp change after only seven . interviews. The authors concluded that the ". . . structure of time limits will promote certain similar effects even where the therapists are distinctly different in their behavior." Summary of Short-Term Psychotherapy Research ‘ Although not much research has been done which relates specifically to short-term psychotherapy, of those studies reviewed, all present evidence which supports the value of a short-term focus. Special emphasis on the importance of goal setting was provided by Small (1971). Studies by Errera pipe}, (1967) and Steiper and Wiener (1959) could find no significant differences on improvement rates be- tween short and long-term psychotherapy clients. In addition to short-term psychotherapy being more efficient, the advantage of the concept of structure introduced into the psycho- therapeutic process has warranted much attention (Batrawi, 1964; Shlien et al., 1962; Test, 1964). Goal setting not only facilitated outcome 16 evaluation, but influenced attitudes and behaviors outside the therapy setting. Even traditional long-term oriented psychotherapists effected positive changes with their clients in studies requiring short-term focus, with gains persisting long after psychotherapy had ended (Avnet, 1965; Bellak, 1965; Lorr gt_a1,, 1966; Malan, 1963; Muench, 1964). Phenomenological Self Theories Historical Perspective Although there is a current resurgence among psychologists_ to focus on self concept and phenomenological components of psycho- therapy, William James, as long ago as 1890, "set the stage . . . for much of what is written today about the self and the ego . . . ." (Hall and Lindzey, 1957). James""empirica1 me" was the sum total of all that a person called 'his,' including body, traits, abilities, possessions, family and friends, vocation and avocation. Current usage of the concept (Hall and Lindzey, 1957) has two distinct con- notations, the 'self-as-object' (person's own attitudes, feelings, perceptions and evaluations, and 'self-as-process') and self as doer (processes of thinking, remembering, perceiving, which govern behavior). Since the time of James, social scientists of every genre have gradu- ally come to view the self concept as a central construct for under- standing people and their behavior (Fitts gt_al,, 1971). Early in- vestigators looked at many different aspects of the self. C. H. Cooley (1902) was concerned primarily with how the self grows as a 17 consequence of interpersonal interactions, and coined the term "the looking glass self," to refer to the social self. G. H. Mead (1934) perceived the self as an object of awareness, and his ”socially formed self" suggested that a person can have as many selves as there are variety of groups to which he belongs. H. S. Sullivan's (1953) “reflected appraisals“ were what the child assimilates to develop ex- pectations and attitudes toward himself as an individual. Other theorists have included Adler, Horney, Goldstein, F. Allport, G. Allport, Murphy, Kelly, Grinker, Combs and Snygg, Lecky, Wylie, Perls, and Berne (Hall and Lindzey, 1957; Hamachek, 1971; Fitts, 1971; Raimy, 1971; Patterson, 1973). Theorists and practitioners who have had the strongest in- fluence on the development and acceptance of self theories have been Rogers (1951) and Maslow (1954). Maslow's unique contribution was the focus on the healthy personality and motivations underlying a need hierarchy, while Rogers' contribution was not only theoretical, but also encompassed research studies to support his theoretical assumptions. Rogers' work added to the earlier works of Snugg and Combs (1949) which stated that all behavior is determined by and re- lated to the organism's phenomenal field, i.e. perceptions of reality, not physical reality itself, and man's basic need for "adequacy." To them, the degree to which behavioral adequacy is achieved is largely a function of the individual's self perceptions. The adequate personality is characterized by positive self-regard, openness to experience, and the ability to identify with a variety of persons, roles, and institutions. Rogers later emphasized the significance of 18 the self concept in determining behavior and psychological adjustment. His "fully functioning person" was very similar to Snygg and Combs' adequate personality, adding the dimensions of self-direction, differ- entiation and complex self-structure, and flexibility to assimilate new experiences (Fitts et al., 1971). Definition and Dimensions of Self Concept The major phenomenological assumption is that each person is influenced by his own phenomenal field. Each individual perceives events (phenomena) in terms of self perceptions of the world which may, or may not, correspond with physical reality. Reality is highly individualized because it is governed by the person's perceptions, both internal and external, which are based on past experiences in dealing with the environment. According to Brammer and Shostrom (1960), this field grew from the Gestalt psychologists' concept of the "self" which involves the way the individual sees himself as determined by his organization of values, goals, concepts and ideals into patterns of behavior. In short, the "self" is the way the in- dividual defines "I" or "me." This basic philosophy has influenced phenomenological approaches to psychotherapy by emphasizing the therapist's ability to understand the client's internal frame of reference and concept of reality in order to facilitate change (Kolt and Trotter, 1969). The "self" has been defined many ways. The early definition of William James' (1890) "empirical self" included the subsystems of 19 "special self," "spiritual self," and "physical self." Symonds (Hall and Lindzey, 1957) perceived the self as the "ways in which the in- dividual reacts to himself," consisting of four aspects: how a person perceives himself, what he thinks of himself, how he values himself, and how he attempts to enhance or defend himself. Snygg and Combs' (1949) "phenomenological self" included "all those parts of the phe- nomenal field which the individual experiences as part or characteris- tic of himself." Lundholm (Hall and Lindzey, 1957) made a distinction between a 'subjective self' ("what I think of myself“) and an "objec- tive self' ("what others think of me"). Sherif and Cantril's (Hall and Lindzey, 1957) "ego" was defined as a constellation of attitudes, "what I think of myself, what I value, what is mine and what I identify with." Sarbin (Hall and Lindzey, 1957) viewed the self as a cognitive structure which consists of one's ideas about various aspects of his being: "somatic self," "receptor-effector self," and "social self." Stephenson (Hall and Lindzey, 1957) was the first to introduce the concept of an ideal self in relation to self concept. He referred to a person's "self-reflections," which he measured with an instrument called the Q-methodology. In reporting Secord and Jourard's (1953) findings that the feelings a person had about his body were commen- surate with the feelings he had about himself as a person, Hamachek (1971) pointed out that "self-image is first and above all a body image." Raimy (1971) described the self concept as a "learned per- ceptual system which not only influences behavior, but is itself altered and restructured by behavior and unsatisfied needs, and may 20 have little or no relationship to external reality." This definition is still widely accepted by current theorists and researchers, even though it was originally written in 1943. Raimy saw the self concept as both a product and process learned from experience, and recognized it as an intricate system with many sub-systems. He also asserted that the self observations, with their attendant values, were organized into a hierarchy, with some aspects of the self having greater import- ance or significance than others. This hierarchy, too, varies from individual to individual. He pointed out that perceived negative evaluation of one aspect of the self concept may be compensated for by positive valuing of other parts. The final evaluation of the total self is dependent on the balancing of the many components of the self concept. He described the functions of the self concept as: provid- ing a frame of reference for behavior which has social significance, regulating and controlling behavior over long periods of time, defin- ing the individual's status and function in society, forming the cri- terion against which choices of behavior are made, and projecting into the future imaginary self concepts (later called 'ideal self' by others). According to Rogers (1951), the self concept is "an organized, configuration of perceptions of the self which are admissable to aware- ness." It is composed of one's perceptions of characteristics and abilities, self in relation to others and the environment, values, goals and ideals. Rogers viewed behavior disorganization as the result of incongruence between the self concept and perceived experience. 21 "Freedom from inner tension, or psychological adjustment, exists when the concept of the self is . . . congruent with all the experiences of the organism" (Rogers, 1951). Toffler's (1970) assessment of the future spoke to the issue of "serial selves," a series of different selves which are developed and discarded in adaptation to a rapidly changing world. This, too, was a new idea in the area of self concept, and has much potential for future research. Purinton §t_al, (1974) viewed the self as composed of many different layers, using the analogy of an onion, with outer layers being superficial and inner layers being private and spiritual. The outermost layer is described as the "positive but false self," which contains the images and facades the person presents to the world in an attempt to gain respect and approval from other people. The second layer is the "negative but false self," which includes feelings of weakness, worthlessness, anger, hate, loneliness and sadness. These images, like the first layer, are also false. The third layer, the true inner core of the individual, is the "source of dignity and humility," beyond the manipulative roles and feelings. Purinton's (1974) viewpoint is consistent with contemporary Gestalt therapy views, and akin to the layers of self identified by Perls (Fagan and Shepherd, 1970): phony, phobic, impasse, implosive, explosive, and self. Fitts §t_al, (1971) pointed out that the self has played a prominent role in theory and research on social control, economic be- havior, social deviance, personal aspirations, psychological develop- ment, interpersonal attraction, social influence, psychopathology, 22 and psychotherapy. Fitts' (1965) conceptualization of the self was composed of an internal and an external dimension, with each dimension consisting of several parts or subselves. The parts of the internal dimension were self-as-object (Identity Self), self-as-doer (Behavioral Self), and self-as-observer and judge (Judging Self). While self theorists have emphasized the Identity Self, and Behaviorism has con- centrated on the Behavioral Self, Fitts' position was that both are equally important, each influencing the other, and true integration of the self requires their free, continual and accurate interaction. Looking toward the external dimensions of the self, Fitts (1965) identified five subselves: Physical Self, Moral-Ethical Self, Personal Self, Family Self, and Social Self. Complete understanding or measurement of any person's self concept necessitates consideration of all these unique sets of subselves, and it was on this premise that Fitts developed the Tennessee Self Concept Scale. This instru- ment yields a variety of scores, each measuring the subselves which comprise the complexity of the concept known as the "self." Since the self is indeed complex, it cannot be adequately described along a single continuum or by a single score or label. Most self theorists would agree with this viewpoint. Application of Self Concept Theory and Related Research In relating the self concept to psychotherapy literature, diverse approaches have been used. Some research has focused on self concept change. Other studies have involved clients in the process 23 of their psychotherapy in creative ways. Still other studies have investigated client evaluations of the psychotherapy experience. These three aspects of applied self theory are summarized. Self Concept Change.--Self concept change has been investi- gated by many researchers, including Raimy (1948, 1971), Butler and Haigh (1954), Rogers and Dymond (1954), Taylor (1965) and Fitts (1965; 1971; 1972). Originally in 1943, Raimy (1971) studied self concept change. The basic hypothesis of personality reorganization was ". . In formal psychotherapeutic situations which result in reor- ganization of the client's personality, essential structural relation- ships within the Client's self-concept will be altered. Where reorganization of personality does not occur, the essential structural relationships within the self concept will remain unchanged." Raimy found that in cases of successful therapy, the client's revealed estimation of self underwent a reversal from predominance of negative self-references at the beginning of therapy to a heavy weighting of positive self-references at the conclusion of therapy. In unsuccess- ful cases, a reversal in self-esteem did not occur. A later study by Raimy (1948) supported his earlier work. He found that at the start of therapy, clients gave a preponderance of disapproving or ambivalent self-references. As therapy progressed, fluctuations in self-approval occurred with mounting ambivalence. At the conclusion of therapy, those clients who were judged to be improved were making a preponderant number of self-approving statements, while those clients who had not improved were still being ambivalent and disapprov- ing of themselves. 24 Butler and Haigh (1954) studied self concept change using the Stephenson Q-technique. Their study showed a significant change from pre-therapy to post-therapy self-ideal correlations. They con- cluded that one of the changes associated with client-centered therapy is that self-perception is altered in a direction which makes the self more highly valued. This finding was not transient, but persisted long after therapy had terminated. Research conducted by Rogers and Dymond (1954) demonstrated that primarily the self concept, not the ideal self, changes in therapy. The ideal self does tend to change, but less so than the self concept. Ideal self change is in the direc- tion of becoming less demanding and more achievable. Similar results were reported by Rudikoff (Hall and Lindzey, 1957), who also concluded that the ideal self was somewhat lowered in the direction of the self image during psychotherapy. Other changes in the self concept were reported by Taylor (1965). Alterations in the self concept accompanied "improvement," as rated by therapists. The self concept became more positive, more congruent with the ideal self, and more consistent. These changes were identified as criteria for successful therapy. Taylor (1965) then investigated changes in self concept without psychotherapy, and found trends similar to those reported for successful psychotherapy clients. This finding was explained by conjecturing that the inten- sive self-introspection, without psychotherapy, appeared to be ad- companied by increased positive attitudes toward the self. Repeated descriptions of the self and ideal self were also accompanied by an increased positive relationship between the two concepts of self 25 and ideal self, though in smaller increments than those reported for successful psychotherapy clients. Because increased consistency of self concept was achieved so readily by self-descriptions without psychotherapy, Taylor (1965) warned against using this criteria for evaluating pSychotherapy. Self concept change in psychotherapy has probably been in- vestigated most extensively by Fitts (1965; 1971; 1972) and his students. The development of the Tennessee Self Concept Scale was the result of his early work. Subsequent studies using this instrument to assess self concept change have been prolific. Ashcraft and Fitts (1964) reported that the underlying assumption of their study was "any changes in self concept which are produced by psychotherapy will be more adequately portrayed by an instrument which measures many different aspects of the self concept, i.e. a multiple factor concept." Fitts (1971, III) also reported on several studies which supported the hypothesis that self concept is an index of self actualization or personality integration. He summarized by pointing out that self concept is an adequate index of self actualization in that the self concept shows a consistent relationship to behavioral competence and effective adjustment. Fitts (1972, VI) also summarized 400 studies which supported the conclusion that interpersonal behavior is predict- able from the self concept. In many ways, the studies conducted by Fitts operationalized the theory Rogers (1954) spoke of many years earlier. Creative Client Involvement in Psychotherapy Process.--Some creative approaches to involving clients in the process of their 26 psychotherapy have included variations of written or verbal communica- tion techniques: written responses to questions (Lipkin, in Click, 1971), autoelaboration (Messinger, 1952), written therapy sessions (Farber, 1953; Phillips and Wiener, 1966; Bastien and Jacobs, 1974), directed writing therapy (Widroe and Davidson, 1961), graffiti therapy (Shulman et_al,, 1973), bibliotherapy and poetry therapy (Leedy,”l965; 1973), recorder self therapy (Shor, 1955), Interpersonal Process Recall (Kagan gt_al,, 1963) and the Therapy Session Report (Howard, Orlinsky and Hill, 1970). Use of these techniques incorporates an underlying value of the importance of client involvement and feedback in the psychotherapeutic process. As early as 1948, Lipkin (Glick, 1971) reported a study which involved clients in the evaluation of their short-term psychotherapy by having them respond in writing to three unstructured questions concerning pre-therapy and post-therapy adjustment, and views about what took place during their therapy sessions. This was one of the earliest studies to employ a freearesponse method of soliciting the client's subjective experience of therapy. Lipkin concluded that the clients' accounts independently corroborated the descriptions of those who write and theorize about non-directive therapy. Messinger (1952) wrote of a technique he called "autoelabor- ation," a homework procedure whereby the client free-associated to the therapists' notes of previous psychotherapy sessions by copying them, elaborating on them, and then discussing them at length during sub- sequent therapy sessions. 27 Farber (1953), a deaf psychotherapist, reported a modified treatment technique whereby clients communicated with him during their sessions via writing or typing. Farber, himself, verbalized since his deafness was incurred after adulthood while he was a practicing psychotherapist. Advantages of this technique were identified as slowed tempo, which facilitated more evaluation, decreased wasted motion and excess verbiage, increased memory of what transpired during sessions, and resulted in a more effective sense of participation on the part of the clients. Similar written communication techniques as a form of psychotherapy were reported by Bastien and Jacobs (1974) who conducted a study of clients who received only written therapy through the mail, and by Phillips and Wiener (1966) who studied written therapy techniques in a counseling center where students came to an isolated room for writing therapy sessions by appointment. The use of "directed writing therapy“ was reported by Wildroe Davidson (1961), who found that writing outside the therapy hour aided verbal therapy ". . . so significantly as to alter the course of treatment." This technique consisted of having inpatients of a psy— chiatric ward write accounts of their daily activities. Those accounts were discussed during therapy hours. A serendipitous result of the project was that the writing itself became an important group activity. It was a way for patients to communicate with one another. A similar finding was reported by Shulman gt_gl, (1973), who discovered “Graffiti Therapy" as an effective means of promoting communication between patients and staff of an urban mental health unit. The graf- fiti analysis aided diagnostic insights and clues about ways of relat- ing to the patients. 28 Verbal methods of involving clients in the process of psy- chotherapy have also yielded interesting results. Recorder self- therapy, reported by Shor (1955), is a technique of self-therapy using a tape recorder. The technique is often supplemented with nonverbal cues by having the client face a mirror while recording and listening to tapes. In view of limited availability of skilled practitioners and limited financial resources of many persons needing psychotherapy, the major advantage of this method was the provision of help for more patients. The omission of the interpersonal relationship, transference, and objective interpretive functions of the therapist could be viewed as disadvantageous, but the authors pointed out that not all therapists deem these aspects of psychotherapy essential for effective results. They further pointed out the usefulness of the technique as an adjunct to traditional therapy and as a valuable training tool. Interpersonal Process Recall (Kagan gt_al,, 1963) has shown much potential for investigating the subjective experiencing of psychotherapy. At the conclusion of interviews, participants viewed videotape playbacks of their interviews, and an interrogator aided the elicitation of feelings, thoughts, and reactions of the clients to the events in the replay. The playback allowed subjective experi- ence to be recalled and recorded, and provided new ways of examining the interpersonal event of psychotherapy for the participants. The use of videotape as a therapeutic adjunct is a growing interest area in the helping professions. Client Involvement in Evaluation of Psychotherapy,--Surpris- ingly few studies have been reported which used clients to evaluate 29 psychotherapy. Grigg (1957) recognized the value of clients as judges of therapist's performance. He viewed clients as a pool of independ- ent observers of a fairly well-delineated job performance. Although he admitted that clients are not unbiased, he stated that any criteria of counselor performance ". . . must include some client-observed and client-reported variable." Grigg (1957) viewed the successful prac- titioner as one who elicits favorable reactions from the recipients of his services, and quoted studies which showed evidence that client evaluations might be important therapeutic variables. The major find- ings of Grigg's study were that therapists varied in kind of technique and in consistency of their modal technique, clients tended to report they obtained what they wanted from therapy if therapists played an active and directive role in the process, clients who reported favor- able attitudes toward therapy outcome also reported favorably on feelings while undergoing therapy, clients who felt their counselor took an active interest in them reported greater satisfaction with therapy, no significant correlation was found between length of time in treatment and attitude of clients about the outcome. Simultaneous therapist and client evaluations occupy most of the literature relating to client involvement in evaluation of psychotherapy. Feifel and Eells (1963) felt that the participating parties in psychotherapy were in the most favored position to provide promising leads concerning what takes p1ace.in the sessions. Their study found interesting differences between client and therapist per- ceptions. Clients accented insight changes, while therapists accented symptom relief and behaviors. Another difference was that clients 3O favored the opportunity to discuss their problems and found the human characteristics of the therapists helpful, while the therapists high- lighted therapeutic technique and support. Other studies have also found contrasting therapist-client impressions. Tucker's (Glick, 1971) research suggested that the client and therapist were evaluating therapy from different perspectives and by different criteria. The client used an everyday operational system of values, and the thera- pist used a psychological value system. Blaine and McArthur (Glick, 1971) also found therapist-client discrepancies regarding psycho- therapy process. In interviews with post-therapy clients, they dis- covered that clients focused on better understanding and acceptance of themselves, and therapists focused on theoretical and historical aspects of psychotherapy. There was mutual agreement about the im- portance of the client-therapist relationship. One other study (Strupp, Fox and Lessler, 1969) compared client and therapist assess- ments of change and found that perspectives of therapy outcome were clearly different. Clients valued the opportunity to discuss their problems, and therapists, again, focused on theory and technique. Therapist assessments of gain were more conservative than client assessments. The above studies point to the importance of therapist awareness of the client's perspective of the psychotherapy process. Clarification of expectations and mutual feedback during the process can facilitate better comunication in the therapeutic relationship. Extensive research of psychotherapy sessions, using both client and therapist feedback, has been conducted in Chicago by Hill (1969), Howard, Orlinsky, and Hill (1968; 1969; 1970) and Orlinsky 31 and Howard (1966; 1967; 1968). This group of researchers devised an instrument entitled the "Therapy Session Report" which was used to assess many aspects of psychotherapy sessions. They found that self ratings which are descriptive of behavior and feelings in a specific situation reflect the subject's actual performance in that situation. The "good therapy hour" definitions from clients and therapists had much the same character. The manner of therapeutic work which clients and therapists both found most valuable was actively collaborative, genuinely warm, affectively expressive, and humanly involving. These studies were the only ones reviewed which found high client-therapist correlations regarding the psychotherapy process. A third dimension to psychotherapy evaluation was added by Horenstein et_al, (1973). Clients, therapists, and independent judges all evaluated the same psychotherapy sessions. Clients"evaluations were unrelated to their therapists' evaluations, but were highly re- lated to the evaluations made by the independent judges. This finding suggested that clients might be ". . . good, or at least better than their therapists, at evaluating their therapy progress." In light of current trends toward consumer rights, the authors also recognized that client-therapist disagreement over the progress of therapy might be a moot issue. Because clients know why they came for therapy, they are in a better position to determine whether or not the reasons for coming still exist. "If the client honestly believes that the reason for which he came to therapy no longer exists, or has been substantially reduced, therapy has been successful" (Horenstein et al., 1973). In 32 summary, if the goal of therapy revolves around the comfort of the client, then the therapist can rely on the client's report of thera- peutic progress. Summary of Self Concept Research Self concept change has been the focus of numerous studies under a variety of settings (Raimy, 1948; 1971; Butler and Haigh, 1954; Rogers and Dymond, 1954; Taylor 1965; and Fitts, 1965; 1971; 1972). There is some consensus that in psychotherapy, the self con- cept changes more than the ideal self, and psychotherapy facilitates a more positive and more consistent self concept. Studies involving clients in the evaluation of their own psychotherapy are few. Innovative therapeutic techniques designed to involve clients more have included written, verbal, reading, and re- call assignments. These techniques incorporate an underlying value of the importance of client involvement and feedback in psychotherapy. Even fewer studies used client-therapist comparisons. In general, the studies have shown that therapists and clients view psychotherapy somewhat differently. Therapists focused on symptom relief, behavior, theory, techniques, and a psychological value system. Clients focused on insight changes, therapist characteristics, self acceptance, and an everyday operational value system (Glick, 1971; Feifel and Eells, 1963). Horenstein §t_gl, (1973) found that client evaluations were highly related to independent judges' ratings but unrelated to thera- pists' evaluations. Orlinsky and Howard's (1967) study was the only one reviewed which showed congruence between therapist and client evaluations. 33 Of the literature reviewed in the preceding sections regard- ing the clients' evaluation of psychotherapy, only one group of studies was conducted in a community mental health setting (Hill, 1969; Howard, Orlinsky and Hill, 1968; 1969; 1970; and Orlinsky and Howard, 1966; 1967; 1968). An area of challenging research is open for in- vestigation of psychotherapy evaluation by clients in short-term psy— chotherapy. It is toward this end that the present study turned its attention. CHAPTER III DESIGN OF THE STUDY The purpose of this chapter is to provide a detailed descrip- tion of the methodology of the study. Within the chapter may be found the sample, the instruments, the design and procedures followed, the testable hypothesis, and the method of statistical analysis. Sample The population from which the sample was drawn was composed of adults who sought outpatient psychotherapy at the Ingham Community Mental Health Center in Catchment Area II of Lansing, Michigan. A catchment area is a geographical region delineated by federal guide— lines, and is designed to provide boundaries for service areas. Each area includes a population of 75,000 to 200,000 residents. The mental health center presiding over the area is expected to provide comprehensive mental health services to its residents. The study was conducted between May and November of 1974 at the Ingham Community Mental Health Center which serves only Catchment Area 11 residents in Lansing, Michigan. Lansing is the capital city of Michigan, with a population of 145,000. The economy of the city is centered around state government, the automotive industry, and nearby Michigan State University. The residents of Catchment Area II reside primarily in the southern portion of the city of Lansing, with 34 35 the remainder of the residents inhabiting suburban and rural areas in the adjoining Eaton County. The Ingham Community Mental Health Center has two satellite centers designed to serve the rural populations of the Catchment Area. The sample of the study was drawn from the main Center which is used primarily by urban residents. Because the pattern of service utilization is quite stable, it was assumed that the clients who requested mental health services during the time the study was conducted were representative of all clients who request services from the Outpatient Unit at the Center. The Outpatient Unit of the Ingham Community Mental Health Center focuses its services on a crisis intervention model. Clients who have a his- tory of psychiatric hospitalization and/or chronic impairment (usually with a diagnosis of one of the psychoses), are typically excluded from the Outpatient Unit and receive services from the Aftercare Unit at the Center. Only persons 18 years of age and over are served by the Ingham Community Mental Health Center. ' The sample was obtained in the following manner. A11 17 therapists actually engaged in seeing new clients in the Outpatient Service of the Center were asked to participate in the study. Two therapists declined to participate in the study, one newcomer who was unfamiliar with routines, and one who did not state reasons. All other therapists agreed to participate. Of these 15, thirteen were selected by supervisory consensus to participate in the study. From the thirteen selected, ten yielded usable data. Therapist demographic data is presented in Appendix C. The three therapists whose clients' 36 data were not usable included two therapists who left the Outpatient Service during the time of the study, and a third therapist who con- sistently failed to notify the researcher of client appointment times. Clients were randomly assigned to one of three groups. Clients assigned to the Treatment Group participated in 45-minute psychotherapy sessions followed by a lS-minute checklist completion session. Control Group I clients participated in 60-minute psycho- therapy sessions. Control Group II clients participated in 45-minute psychotherapy sessions. All therapists were expected to use the Treatment and the two control methods. In order to randomly assign clients to therapists, the following procedure was employed: 1. All therapists were assigned a code number between one and 515F15§3 (later referred to by letters A through M for reading 2. The therapist's code number was located in a table of random numbers by flipping the pages of the table open to any section. 3. The table of random numbers was searched, moving down the column, until any combination of the numbers 1, 2, 3 was found. The resulting sequence became the order followed in assigning clients to the therapist. A11 clients who called the Center for an initial appointment were first assigned to a therapist by alphabetical rotation based on the therapist's last name. Thereafter, assignment to a treatment group was based on that particular therapist's sequence code. Al- though approximately equal numbers of clients might be expected to be assigned to each treatment group, there was no way to anticipate which clients would keep their initial appointments and which would not. The result was that equal sample size was not obtained for all treat- ments and therapists. 37 A specific breakdown of client and therapist assignments is presented in Table 3.1. The discrepancy in number of clients initially assigned to each therapist is accounted for by the fact that not all therapists were assigned the same amount of time to the Outpatient Service. The range of time assignments was from one to three days per week. Of those clients who kept their initial appointments, 51 had been assigned to the Treatment Group (45-minute psychotherapy sessions plus a lS-minute checklist completion session), 34 had been assigned to Control Group I (60-minute psychotherapy sessions), and 34 had been assigned to Control Group II (45-minute psychotherapy sessions). Of the 120 clients who were approached about voluntary participation, all but one consented to participate in the study. From the 119 clients who volunteered, 38 sets of usable data were obtained: i.e., 38 clients (34.4 percent) completed four psychotherapy sessions and had complete pretests and posttests. In the Treatment Group, 14 clients (27.45 percent) completed four psychotherapy sessions. Control Group I had 12 clients (35.29 percent) completing four psy- chotherapy sessions, and Control Group II had 15 clients (44.11 per- cent) completing four psychotherapy sessions. Previous studies of the Outpatient Service's clientele of the Ingham Community Mental Health Center indicated that only 52 per- cent of that clientele can be expected to complete four or more psychotherapy sessions. This yields a drop-out or termination rate of 48 percent. The drop-out rate for th e study was 65.55 percent. It was not possible to determine how many clients in the study self- terminated and how many were referred elsewhere for more appropriate 38 Acowmmmm Amcowmmmm hamgmggoguxma mpzcwsumev HH azogw pogucoo "N Amcopmmmm hamgmspoguxma muscws omv H qsogu pocpcou u. copum—asou umwpxomgo muzcwsump Napa NcoNNNmN zgmngpoguamq muzcwsumev azogu pcosummgp u u u H cmumpov NNNe mumpasoucp mo pmm mco « NN NN NN NN Nmm acmvumapso Nam umggmwog cmcvppzo nmncaumpn appmcowgoem mo mpcmcmav Ammmpv :NEeo: .e Amemcpxm agupmm; xppmowmo—ozuxma .m.w .cowumcmmpcw apwpmcomgma mmmp Nuuwu .m NNNNNNNNN UNLNNNNNNNN NNNN NNNNN .N ANNNN NNN.N ._ Ne.~o oo.eo o.ON m.¢m N¢.eN o.mm ep.mm epmm _N_uom oo.mm mN.mm _.mN m.mm me.NN m.¢m mm.oN Npmm NFNENN mo.mm mm.~m m.No o.Pm mN.FN m.om mm.¢m epmm chomcmm oo.~o m~._m e.NN N.¢N mN.mN N.mo mm.ON Npmm —Nuvsum-pmcoz mo.~o m~.~m m.mm m.¢o mN.NN m.NN mN.FN Npmm quwmxza mm.wm~ mm.mm~ ~.mmm N.N~m Po.NNm o.m~m Nm.mem m>NpNNoa FNuoh oo.om mm.um N.om m.mm mm.mm o.om em.mm Emwowprgo epmm wmuz mmnz omuz omuz munz mmmuz ommuz ammuumoa umwumcm mpaENm «Fasmm muwcwpouzoz Nowcwpo mHm Npcmwpmm FNENoz mpoum Nw_u=pm umpmpmm ucm NEL.oz saw: Ncmmz m_asmm mpmom pamucoo mpmm mmNchcmh mo :oNNNNano m.m mpnmh 48 Mooney Problem Checklist--Form A (MPC) Because persons seeking psychotherapy typically view them- selves as having problems, and the concept of "presenting problem" is widespread among psychotherapists, one criteria of psychotherapy effectiveness is relief from the problems which necessitated a person's initiating treatment. It was assumed that clients seeking psychotherapy would want to communicate their concerns, and the MPC is one instrument specifi— cally designed to facilitate such communication. Because the clients were assured of anonymity, it was further assumed that they would be truthful in completing their checklists, and not bypass concerns which might, at that point in time, be embarrassing or inducive of defensiveness. Introduction and Test Development The authors, Gordon and Mooney (1950) state that the ". . . essential purpose of the MPC is to help individuals express their personal problems." The MPC was developed for use with late adols- cents and adults who were principally of non-student status. The items were developed from ". . . original problem literature, write- in statements made by . . . students on the College Form, and thousands of problem items accumulated in the development of the other forms of the MPC series" (Gordon and Mooney, 1950). The authors emphasize that the MPC is not a "test" and does not yield scores on traits. Neither does the MPC allow for the adjustment status of the respondent. The 49 MPC is solely intended as a form of communication designed to accel- erate the understanding of the respondents and their problems. Desigp_and Administration The MPC is a self-administered checklist consisting of 288 statements. The respondents underline the problems which are of con- cern to them, circle the ones of prime concern, and write a summary in their own words. Completion time is estimated to be 20 to 30 minutes. The list is constructed so problem areas run horizontally across the page in groupings of six items, but most respondents do not discover this grouping. This format facilitates avoidance of entire category sections being skipped by respondents who might avoid areas of low social acceptability (e.g. personality and sex) or sec- tions inappropriate to their life situations. In this study, the written summary in the client's own words was eliminated. Only the tallies of numbers of problems underlined were used as indicators of number of problems identified in each of the nine problem areas covered by the MPC. MPC Scales The nine areas surveyed by the MPC include: (1) Health-36 items, (2) Economic Security-36 items, (3) Self Improvement-36 items, (4) Personality-72 items, (5) Home and Family-36 items, (6) Courtship- 18 items, (7) Sex-18 items, (8) Religion-18 items, and (9) Occupation- 18 items. 50 Reliability and Validity In the Manual for the Adult Form (Gordon and Mooney, 1950), the authors do not present any reliability figures. Their reason for this was that the MPC was designed to reflect problems at a specific point in time only. Since there are "technically" no scores, no reliability estimates can be calculated. Again, because the MPC was not designed to be a "test," validity indices were also deemed meaningless. According to the authors (Gordon and Mooney, 1950), data obtained from the MPC should be studied in relation to the people in their specific situations. Therefore, local norms were recommended. Local norms were not available, but the sample means were compared with means obtained in a writing therapy study conducted by Bastien (1974) using college students in psychology courses. These figures are presented in Table 3.6. Bastien investigated three modes of writing therapy: (1) written therapeutic responses to the “writ- ing therapy" group, (2) reassurance letters to the "reassurance" group, and (3) requests for frequency of occurrence of problem be- haviors on a mailed form for "base rating." The findings of Bastien's (1974) study showed that §§_treated by means of writing therapy re- ported significantly greater decrease in number of personal problems than §§_treated with reassurance or those who base rated their behavior. Bastien (1974) also reported that the mean number of problems on the MPC for college students is 22, while her study's pretest mean number of problems was 45.27, over double that of the average college student, 51 AcoNNNmN cowumPQEou pmwpxuwgu muscwsump N=_a :oFNNmN maNnguosuxma mu::NE-m¢v azogu acmEuNoNp azocm mumgummmm n mm aaogm mucmgzmmmmm u a azogm anaemza mcwpwez u 3 pmuoh u no» AcoNNNmN xamcmzuosuzma mpacpslmev HH azoew poNucou Nu AcoNNNmN zaNngpozoama mpzcwsuomv H asogu Focucoo _ II 0 H emmp cmwpmmme v.03 ~.NF m.- om.mm w.m¢ —.m~ N.nm em.pm mo.mm mw._m mo.m~ op.¢¢ mm.m¢ eo.pe up.mm ummu -pNom III. N.Nm ¢.- ¢._m N~.m¢ m.w¢ N.~m o.mm mm.mm NN.mm ¢¢.Nm om.~m N¢.¢m oo.NN mm.om mm.me “New .uoh mm a 3 punch mm m 3 .ao» No » .Nop No No b 1mg; .u.N :Nmz .u.N cam: «mamgwmhummpu_g3 mmammm «mPaENm xamsmgp mcwuwg3 sum: Ncowumw>ma NNNucmum New Name: NPQENN umwpxumsu EanoNN xucooz mo coNFNquoo m.m mpnmh 52 "clearly a more troubled sample than the average college student." Inspection of Table 3.6 shows that the total sample of the current study had a greater number of self-identified problems both at pre- testing and posttesting than the total sample reported in Bastien and Jacobs' (1974) study. Therapy Session Report--Form P (TSR) The TSR (Appendix A) was the instrument chosen to measure the satisfaction with psychotherapy of the clients in the sample. Again, referring back to the self theories, a client's experience in psychotherapy should be one which the client finds satisfying and which is conducive to a more positive and integrated self concept. As Grigg (1957) observed, clients' evaluations can be important therapeutic variables, and favorable attitudes toward counseling out- come coincided with favorable feelings while experiencing the psycho- therapeutic process. Horenstein edged, (1973) also theorized the value of client satisfaction in psychotherapy, finding that clients were at least as good as, and often better than, their therapists in evaluating psychotherapy progress. They also recognized that any client-therapist disagreement over psychotherapy progress might be a moot issue, in light of the fact that since clients know why they came for psychotherapy, they are in a position to determine whether or not their reasons for seeking treatment still exist. 53 Introduction and Test Development The TSR is a questionnaire developed as part of the Therapy Session Project at the Institute for Juvenile Research (Orlinsky and Howard, 1966). The instrument has primarily been used with female outpatients of the Katherine Wright Mental Health Clinic in Chicago. The project's intent was to focus on the experiences of patients and therapists by the use of two parallel structured-response question- naires. The methodology involves the systematic and quantitative use of structured participant-observation and requires the patient and therapist to each complete a questionnaire immediately following a psychotherapy session. Parallel forms survey comparable aspects of the sessions. The Patient Form of the TSR is comprised of 168 items, and the Therapist Form has 167 items. All items are either of a checklist or rating scale format. Composition of the item pool was accomplished through extensive pilot testing and consultation. The entire questionnaire can be completed in five to ten minutes. All items are designed to avoid the terminology of any special theoretical orientation, but to include issues which are meaningful to most all orientations. Five general facets of the therapy experience are investi- gated by use of the TSR. First, the topical content of "dialogue," as established by the patient, is reported. Second, the nature of the interpersonal "relationship" between patient and therapist, as re- flected in manner of relating to each other is included. Third, the "affective processes" are assessed through reported feelings 54 experienced. Fourth, the "exchange process" of patient's wants and therapist's goals is reviewed. Last, the “development of the session" is explored, including the motivation, implementation, and consummation of goals. Administration The TSR is self-administered and consists of simply indicat- ing one's opinion about various aspects of the psychotherapy session being reviewed. Patient and therapist can independently report on their experiences in the session by filling out their respective forms of the TSR. Reliability and Validity The extent of patient-therapist agreement about content of "dialogue" was analyzed using an adaptation of the Campbell and Fiske multitrait-multimethod matrix. Correlations ranging from .40 to .80 were reported (Howard, Orlinsky and Hill, 1969). Orlinsky and Howard (1967) reported that the " . . . validity of the instruments is, at least, in a minimal sense, established by the consensus so often noted between patient and therapist raters, and by the inter-questionnaire correlations which were obtained." No specific reliability figures were given, however, for the questionnaire in its entirety. 55 How the TSR Was Used in the Study The TSR was used in two ways in the study. Client "involve- ment" in evaluating psychotherapy sessions was defined as completing the TSR--Form P, in its entirety after each of four sessions of short- term psychotherapy. The completed questionnaires were not used for any purpose other than to ensure that the clients in the Treatment Group took a specified amount of time, i.e., 15 minutes immediately follow- ing their psychotherapy sessions, to evaluate what transpired during that session. Therefore, "involvement" was comprised solely of the act of completing the questionnaire. To ensure confidentiality and promote honesty of respnses, the clients were assured that their therapists would not have access to the completed questionnaires. The second way the TSR was used was to assess satisfaction with psychotherapy in general. The section of the TSR which dealt with Satisfaction was printed on a separate sheet, which was then used at posttesting only (Appendix B). The Satisfaction measure con- sisted of 13 statements relating to patient satisfaction, each of which could be responded to on a scale from O to 2. Space was provided for client comments, although these comments were not scored. The Satisfaction score consisted of the total of the response ratings for the 13 scaled statements. The .84 estimated reliability for the present sample was calculated by the Coefficient Alpha variation of the Kuder Richardson (Mehrens and Ebel, 1967). No reliability estimates were reported for this form by the authors (Orlinsky and Howard, 1966). 56 cowaoaauuo um: u o.: NNNNN_NN Na: a N-z xmm um: n m1: qwsmugaou um: u 9:: upmm papuom mumh u mmuh hpwemm new use: on: n uznz upwm prEam womb H mm:# qu—acomgmm on: u an: m—om pacomgma womb n wank Auuowga>ouv —mu0h on: u UHOHII acmem>ogaeu wpmm on: u mmuz mpmm —mu_:umupugo: mumh u mznh Amummgm>ouv m>wuwmoa —muo# mom» a UFOPIH qussuom uweocouw on: u mun: w—wm poumeza mumk u mzanh cowuuammwumm amp * mhmx po. v 3 mm mo. v m m oo.p mmvm.u mp.u mmpn. mmoo. mmmo. mmMN. moe. mmmm. mmmw. mmew. mmmo. mmmm.u mem.n mm~e.: w~.u mmm¢.1 mmme. UPOPit oo.— m—. mom.u ¢N.1 9—.1 NN.1 ~—.n mmeo.1 mmmv.a mva.1 m—.1 mmvm. mmON. mmmm. mmmm. mmoo. mN.1 chowuh oo.— mmm.1 —p.1 —o.1 no. mm~¢.1 mN.1 up.1 39.1 mp.1 men. MN. mmme. no.1 mmm. m—.n mp “conceaoo we» mo xNNuN: coves—occoucuucm 57 Instrument Correlations An intercorrelation matrix of the 16 dependent variables is presented in Table 3.7. There were 68 significant correlations. Of these significant correlations, 25 correlations (37 percent) were p < .05, and 43 correlations (63 percent) were p < .01. The range of significant correlations varied greatly and is summarized in Table 3.8. Table 3.8 Range of Significant Correlations of the Intercorrelations of the Dependent Variables. Correlation range Number of significant correlations .32-.4O (p < .05) 25 .41-.50 (p < .01) 16 .51-.6O 14 .61-.7O 9 .7l-.80 3 .81-.9O O .91-.99 1 The highest correlation (r = .92) was between MPC-Person- ality and MPC-Self Improvement, an indication that these two variables were closely measuring the same phenomenon. However, because there was no theoretical base for doing so, these scales were not collapsed. Three of the significant correlations were in the .70's: MPC-Person- ality and MPC-Occupation (r = .78); TSCS-Personal Self and TSCS-Social 58 Self (r = .75); and MPC-Occupation and MPC-Self Improvement (r = .73). The MPC-Personality, MPC-Self-Improvement, and MPC-Occupation all ap- pear twice, each appearing with each other variable and thus comprising three of the four highest correlations of the entire matrix. It seems plausible that these three variables are closely related to each other in the frame of reference of the respondent, and problems in one area WBuld imply high probability of problems in the other two areas. The remainder of the significant correlations were below .70. In every instance, an inverse relationship was found between the self concept measure (TSCS) and the number of reported problems (MPC). This finding implies that greater numbers of problems are associated with lower self concepts, and less problems are associated with higher self concepts. There appeared to be no pattern with the SATS correlations. All of the measures correlated significantly with at least four other variables. A sunmary of the number of the significant correlations by variable is presented in Table 3.9. Design All clients who agreed to participate in the study were randomly assigned to a treatment group by the procedures described earlier. Each participating therapist was assigned clients for the three treatment groups: (1) Treatment: 45-minute psychotherapy ses- sion plus 15-minute checklists-completion session, (2) Control Group I: 60-minute psychotherapy session, and (3) Control Group II: 45-minute psychotherapy session. 59 Table 3.9 Number of Significant Correlations between the Variables in the Study. Variable T°tca01rr5e119antwncsa"t P < .05 P < .01 MPC-Self Improvement 13 2 11 MPC-Personality l3 2 ll MPC-Occupation 13 5 8 MPC-Economic Security 11 5 6 TSCS-Physical Self 10 4 6 MPC-Health 9 3 6 TSCS—Moral-Ethical Self 9 3 6 TSCS-Personal Self 9 2 7 TSCS-Family Self 9 3 6 TSCS-Social Self 8 3 5 MPC-Sex 7 3 4 MPC-Religion 6 2 4 TSCS-Self Criticism 5 5 0 MPC-Courtship 5 4 1 SATS-Satisfaction 5 3 2 MPC-Home and Family 4 1 3 The participating therapists' responsibilities included strictly abiding to the time limitations imposed by the treatment assignment, and notifying the researcher of client appointment times in advance. Therapy was to proceed in the typical style of the ther- apist for four individual sessions. Thereafter, the decision for continuing with psychotherapy or for terminating treatment was left to the judgment of the client and the therapist. 60 A pretesting session immediately prior to seeing the therapist for the initial interview consisted of having the clients complete the the TSCS and the MPC, and sign consent to participation forms (Appendix D). The posttesting session, consisting of completion of the TSCS, the MPC, and the TSR Satisfaction measure, took place immediately after the fourth psychotherapy session. All testing took place individually in a secluded room which was separate from the therapist's office. Testable Hypothesis To answer the primary question of interest about the popu- lation, the following hypothesis was tested: Null hypothesis: No differences will be found on measures of psychotherapy effectiveness between clients "involved" in the evaluation of their own psychotherapy and clients who do not evaluate their own psychotherapy. No differences are anticipated between the two control groups. Alternate hypothesis: The mean score on measures of psychother- apy effectiveness of clients who evaluate their own psychotherapy will exceed that of clients who do not evaluate their own psycho- therapy. No differences are anticipated between the two control groups. Analysis Because population means are unknown, the sample means (M) were tested against the multivariate analysis of covariance. The two covariates were the TSCS Total Positive score and the MPC Total score. 61 The 16 dependent variables were TSR Satisfaction, TSCS-Self Criticism, TSCS-Physical Self, TSCS-Moral-Ethical Self, TSCS-Personal Self, TSCS- Family Self, TSCS-Social Self, MPC-Health, MPC-Economic Security, MPC- Self Improvement, MPC-Personality, MPC-Home and Family, MPC-Courtship, MPC~Sex, MPC-Religion, and MPC-Occupation. For a client's data to be usable in the analysis, scores for both covariates (pretest scores on the TSCS-Total Positive and the MPC-Total), and scores for all 16 dependent variables at posttesting, listed above, were mandatory. A total of 38 clients"data met these criteria. The analysis was multivariate analysis of covariance, with posttest scores on the 16 dependent variables covaried against the two pretest scores. The analysis was performed using Helmert Contrasts, a system which compares the Treatment Group with both control groups simultaneously. Helmert Contrasts involves contrasting the mean of the Treatment Group with the average of the means of the two control groups. In addition, the means of the two control groups were con- trasted with each other. The analyses were based on the means of vectors which consisted of the 16 dependent variables' scores consid- ered simultaneously. Summary The study was conducted using a sample of 38 outpatient clients of an urban Community Mental Health Center whose services focus on short-term psychotherapy. A multivariate analysis of covari- ance, simultaneously analyzing 16 variables from the Tennessee Self Concept Scale, the Mooney Problem Checklist, and the Therapy Session 62 Report Satisfaction scale, contrasted the performance of one Treatment Group (defined as those clients "involved" in the evaluation of their psychotherapy) with the performance of two control groups. A further analysis compared the two control groups with each other to assess whether significant differences existed between these two groups. CHAPTER IV ANALYSIS AND DISCUSSION OF RESULTS The purpose of this chapter is to report the results of the study and discuss the implications of the findings. Results Hypothesis Testing. The major hypothesis investigated was that no differences would be found on measures of psychotherapy effectiveness between clients "involved" in the evaluation of their own psychotherapy and clients not involved in evaluation. The multivariate analysis of variance F-ratio for this hypothesis, testing the equality of mean vectors of the 16 dependent variables described earlier, was 1.5313 ( p = .1822). The value was larger than the g_p§ig§j_alpha of .05, and the hypothesis was not rejected. The F-ratio for the multivariate analysis of variance, testing the equality of mean vectors of the 16 dependent variables of the two control groUps was 1.1503 (p = .3786). This value was also larger than the g_pgigri_alpha of .05, and this part of the hypothesis was not rejected. The multivariate analysis of covariance resulted in an F- ratio of 1.2275 (p = .3352), which was larger than the g_priori alpha 63 64 of .05, and the hypothesis was not rejected. The multivariate analysis of covariance F-ratio for the two control groups was .9513 (p = .5366), and this part of the hypothesis was also not rejected again. Multi- variate analysis of covariance showed that the two covariates (TSCS- Total Positive and MPC-Total) accounted for only 8.93 percent of the total variation (based on pooled within cell correlations). Means and Standard Deviations of the Sample The mean scores of all variables for the three sample groups did not significantly differ. Pretest and posttest means and standard deviations of all measures for the Treatment Group are summarized in Table 4.1. Similar data for Control Group I is presented in Table 4.2. For Control Group II, the data is presented in Table 4.3. The means and standard deviations for the TSCS Identity Self, TSCS Judging Self, and the TSCS Behavioral Self, which could not be used in the Multi— variate analyses, as described earlier, are summarized in Appendix F and Appendix G. Exploratory_Analyses Because the major hypothesis, that no differences would be found on measures of psychotherapy effectiveness between clients "in- volved" in the evaluation of their own psychotherapy and two control groups of clients who did not evaluate their own psychotherapy, was not rejected, three avenues of exploratory analysis were pursued to help identify factors possibly influencing the finding. The explora- tory analysis consisted of: (1) pg§EDhgg_factor analysis to 65 Table 4.1 Means and Standard Deviations of All Measures for the Treatment Group* of the Study. Treatment Group n = 12 Pretest Posttest . 1T so if so S’JTscs Self Criticism 37.58 4.74 35.00 5.12 **TSCS Total Positive 298.83 36.43 298.58 35.83 TSCS Physical Self 62.25 7.70 62.08 8.47 TSCS Moral-Ethical-Self 61.25 8.23 62.00 10.78 TSCS Personal Self 52.58 12.06 53.08 8.51 TSCS Family Self 58.75 7.88 59.00 8.15 TSCS Social Self 64.00 10.85 62.42 9.16 MPC Health 4.17 4.26 3.25 2.83 MPC Economic Security 3.67 2.57 3.50 2.20 MPC Self Improvement 6.00 4.57 6.50 6.83 MPC Personality 16.00 10.75 13.83 14.24 MPC Home and Family 5.83 5.80 5.50 6.02 MPC Courtship 1.33 1.16 1.83 1.34 MPC Sex 1.00 1.13 1.08 1.17 MPC Religion 2.75 2.01 1.50 1.51 MPC Occupation 1.83 2.62 2.17 4.02 **MPC Total 42.58 22.80 39.17 29.07 SATS Satisfaction --- —-- 15.75 6.55 *45-minute psychotherapy session plus a 15-minute checklist completion session **Covariates 66 Tab1e 4.2 Means and Standard Deviations of All Measures for Controerroup I* of the Study. Control Group I n = 11 Pretest Posttest 7' SD 7’ SD TSCS Self Criticism 37.36 6.61 38.64 4.80 **TSCS Total Positive 279.46 49.15 292.36 41.02 TSCS Physical Self 56.91 9.12 59.00 8.34 TSCS Moral-Ethical Self 59.64 12.13 62.00 11.47 TSCS Personal Self 49.73 11.73 52.27 9.18 TSCS Family Self 56.64 11.62 59.73 9.31 TSCS Social Self 56.55 14.47 59.36 11.44 MPC Health 3.73 1.85 2.91 2.66 MPC Economic Security 5.27 3.55 4.73 4.22 MPC Self Improvement 8.27 6.53 7.18 8.47 MPC Personality 21.55 16.78 16.55 14.36 MPC Home and Family 4.55 2.95 3.91 3.15 MPC Courtship 2.00 2.24 1.73 1.62 MPC Sex 1.00 1.55 0.73 1.10 MPC Religion 1.82 1.60 1.82 2.32 MPC Occupation 2.18 3.13 2.09 2.43 **MPC Total 50.36 32.44 41.64 31.83 SATA Satisfaction --- --- 15.27 4.61 * 60-minute psychotherapy session ** Covariates 67 Table 4.3 Means and Standard Deviations of All Measures for Control Group II* of the Study. Far Control Group II n — 15 Pretest Posttest 11' so 1T so TSCS Self Criticism 37.47 6.021 35.87 4.50 **TSCS Total Positive 287.87 31.93 301.33 34.20 TSCS Physical Self 60.33 8.00 64.93 10.24 TSCS Moral-Ethical Self 62.80 8.30 64.40 6.72 TSCS Personal Self 48.73 9.61 53.13 10.08 TSCS Family Self 60.80 7.25 61.47 9.24 TSCS Social Self 54.47 10.13 57.40 9.33 MPC Health 5.00 3.16 3.73 2.71 MPC Economic Security 6.40 4.95 4.40 3.74 MPC Self Improvement 11.67 7.76 8.80 6.74 MPC Personality 27.07 17.32 21.00 14.40 MPC Home and Family 6.13 3.82 4.13 3.89 MPC Courtship 4.00 3.78 2.73 2.05 MPC Sex 2.13 2.62 1.27 2.22 MPC Religion 2.27 2.52 2.07 2.12 MPC Occupation 2.47 2.30 1.73 2.69 **MPC Total 67.00 39.72 49.87 35.09 SATS Satisfaction --- --- 16.07 4.73 * 45-minute psychotherapy session ** Covariates 68 investigate the stability of the instruments used in the study; (2) examining the checklists 0f the Treatment Group sample on three rele- vant questions concerning current level of functioning and satisfaction with the therapy sessions to glean clues regarding the high drop-out rate of the sample; and (3) inspecting the SATS scores of clients-by- therapists to perhaps find an indication of a client-by-therapist interaction effect. Post-Hoc Factor Analysis One conjecture regarding the lack of significant findings was that perhaps the instruments used in the study were unstable. To further investigate this idea, three factor analyses were conducted. The first factor analysis included the pretest scores for all the TSCS and MPC variables. The Varimax rotation analysis of the sample pretest scores is presented in Table 4.4. The TSCS-Self Criticism score, consisting of ten MMPI L-Scale items, comprised one of the four resulting factors. A second factor consisted of the five TSCS variables (Physical Self, Moral-Ethical Self, Personal Self, Family Self and Social Self). A third factor was comprised of eight of the nine MPC variables (Health, Economic Security, Self Improvement, Personality, Courtship, Sex, Re- ligion and Occupation). A fourth factor was the MPC-Home and Family variable. The analysis indicates that each of the instruments at pretesting was measuring different phenomena. 69 Table 4.4 Varimax Rotation Analysis of Sample Pretests. Factor 1 Factor 2 Factor 3 Factor 4 TSCS-Self Criticism .2498 .1402 .1658 48618? TSCS-Physical Self -.3468 -;§4§§f .0681 .1959 TSCS-Moral-Ethical Self .0425 -;§1lgf —.O703 -.1367 TSCS-Personal Self -.1641 -;§§l§f -.O720 -.1880 TSCS-Family Self .0783 -;§§lgf -.6079 .0167 TSCS-Social Self —.2975 -;Z§12f .0229 -.1586 MPC-Health gggzgr .0061 .4732 -.3105 MPC-Economic Security 44316? .3730 .4088 -.l750 MPC-Self Improvement 48464? .3234 -.0180 .1708 MPC-Personality 42265? .4947 .1350 .1752 MPC-Home and Family .1307 «.0709 48239? .2746 MPC-Courtship 48953? .0545 .0574 .0376 MPC-Sex ngzgr -.0332 .0956 .0648 MPC-Religion 45541? .0591 .4294 .0425 MPC-Occupation :1132f .2313 .0541 .1124 * Highest correlation for this variable. 7D The second factor analysis included the posttest scores for all the TSCS and MPC variables. The Varimax rotation analysis of the sample posttest scores is presented in Table 4.5. By inspection, it can be seen that, again, Table 4.5 Varimax Rotation Analysis of Sample Posttests Factor 2 Factor 1 Factor 3 Factor 4 TSCS-Self Criticism 46432? -.0846 -.1289 .0657 TSCS-Physical Self -;ZQQ§f -.1824 -.3143 -.0597 TSCS-Moral-Ethical Self -;§§§Qf -.3629 .2413 .5406 TSCS-Personal Self jgzzggf -.1311 -.3179 .1526 TSCS-Family Self -.3633 -.3306 -.2795 gggggr TSCS—Social Self -;§2§§f -.0551 -.l655 .0449 MPC-Health .0701 gzgggf .4157 .2054 MPC-Economic Security' .2999 L195]? .1428 .0086 MPC-Self Improvement 42361? .5874 -.0224 -.0730 MPC-Personality gzgggf .5806 .1322 -.1157 MPC-Home and Family -.0705 .4588 |gzg§gf .0703 MPC-Courtship .2953 .4597 -.O753 ;§§§Zf MPC-Sex .1181 gzgggf -.0634 .0224 MPC-Religion -.0247 gzgggr .0639 -.1728 MPC-Occupation $5641? .5102 .0491 -.2989 SATS —.3157 .1041 -g§Q§Zf .0749 *Highest correlation for this variable. 71 four factors emerged, but variables comprising these four factors are quite different than those variables comprising the four factors of the pretest scores. On the posttest factor analysis, one factor con- sists of eight variables which cut across instrument lines. This factor includes five variables from the TSCS (Self Criticism, Physical Self, Moral-Ethical Self, Personal Self, and Social Self) and three variables from the MPC (Self Improvement, Personality, and Occupation). A second factor is composed of four MPC variables (Health, Economic Security, Sex and Religion). A third factor consists of one MPC vari- able (Home and Family) and the SATS (Satisfaction). The fourth factor consists of one TSCS variable (Family Self) and one MPC variable (Courtship). A third Varimax rotation analysis was computed using all of the TSCS and MPC pretests and posttests (only scale thereby excluded was the SATS). The third analysis resulted in eight factors which are diagrammed in Figure 4.1. The semantic instability of the MPC is graphically demonstrated in Figure 4.1. It can be seen that all six of the TSCS variables remained stable from pretesting to posttesting, but only three of the MPC variables (Courtship, Home and Family, and Occupation) were included within the same factor from time of pre- testing to time of posttesting. Six of the MPC variables were located in one factor at pre- testing, and in a different factor at posttesting. These six MPC variables were Health, Economic Security, Self Improvement, Personality, Sex and Religion. 72 F.e mesmem .epeeem egN No NeNoom NNNNNNON use NNeNeNm Auezv NNN_xue3u ENNNONN ameooz use Amumhv epeem Nneocou epem eeNNeceeh mo NNN3NOe< caepeuox xeeNNe> eeN we ONONOONNN Nespuee . NNNNNNoa op pneuega seem mpaeNNe> mo xuNFNaeNN eNeoNecw Nepnewee> em—NNNum umeeumon oN NNoNeNn sage NoNuem mo 3uNFNneNN eNeuNecN Neege eeeegmmm NNON NONOON NNNN u NN-N NNNEON ONO ONO: NO: u Nz-z NNON NNNEON NNNN n NN-N OONNOOOOOO Ne: u 0-: NNNNOOONNOO ON: u O-= NNON NOOONNON NNNN u NN-N OONNNNON Na: u N-z NOOEO>OLNEN NNON NO: u NN-z NNON NOONNNN-NOLOz NNNN u N2-N NON ON: u N-z NNNNOOON ONEOOOON Ne: u NN-z . NNON NOONNNNN NNNN u NIN-N ONNNNNOON ON: n N-z NNNOOI Ne: u 2.: N. ENNONNNLN NNON NNNN u NN-N w NNON ~72 NN.- ON NN.- No 7 ONN-N=N-N NNON N-z ON.. ON NN.- be... 1-: _ 8.- ON 3.. , 7 ON.. ON ON.- ON ON N_.- NF. ON ON NN. ON ON. we Ne”: NOON 3-2 NN. ON 8. NON N z \u NN. ON ON. m NoNueN . N wepuee emu Nepoem «mm NoNueN e NoNueN . m NoNueN N gouge; _ F NoNueN 73 Only two factors remained "pure" from pretesting to post- testing. Factor 5 was composed solely of the TSCS Self Criticism variable both at pretesting and at posttesting. Factor 6 was comprised solely of the TSCS Moral-Ethical Self and the TSCS Family Self both at pretesting and posttesting. Checklist Questions In an effort to identify reasons for the high drop-out rate, the checklists (TSR-Form P), which had been done only by the Treat- ment Group sample, were investigated. The Satisfaction (SATS) measure and three questions which logically related to factors which typically influence drop-outs or terminations in psychotherapy were chosen. Since the Treatment Group had some feedback, via the completed TSR's, it was hoped that some clues to the reasons for dropping out of treat- ment would result from perusing the answers. Of special interest, also, was the fact that the Treatment Group had the highest drop-out rate (72.55 percent) of the three sample groups. The three questions chosen for review were: (1) How help- ful do you feel your therapist was to you this session?; (2) How well do you feel that you are getting along, emotionally and psychologi- cally, at this time?; and (3) To what extent are you looking forward to your next session? A summary of the mean scores for these questions, and for the SATS, are presented in Table 4.6. The first two questions had six response choices, and the third question had five response choices. The number of the response circled by the respondent was the score for that item. For the three questions, a low score 1.1111 i 74 Table 4.6 Mean Satisfaction Scores (SATS) and Spot-Check Question Scores of the Treatment Group COmpleted 1 1 ’ 2 3 Checklists n SATS Question 1 Question 2 Question 3 0 ll -- -- -— ~- 1 12 12.41 3.17 3.79 3.13 2 9 12.72 2.94 4.33 2.89 3 5 11.53 2.66 3.60 2.86 4 14 15.74 2.69 3.04 2.71 51 1. HOW HELPFUL DO YOU FEEL YOUR THERAPIST WAS TO YOU THIS SESSION? 1. 2. 3. 4. 5. 6. 2. H P 1. 2. 3. 4. 5. 6. 3. mthd o o o o 0 Completely helpful Very helpful Pretty helpful Somewhat helpful Slightly helpful Not at all helpful 0W WELL DO YOU FEEL THAT YOU ARE GETTING ALONG, EMOTIONALLY AND' SYCHOLOGICALLY, AT THIS TIME? Very well; much the way I would like to. Quite well; no important complaints. Fairly well; have my ups and downs. So-so; manage to keep going with some effort. Fairly poorly; life gets pretty tough for me at times. Quite poorly; can barely manage to deal with things. T0 WHAT EXTENT ARE YOU LOOKING FORWARD TO YOUR NEXT SESSION? Intensely; wish it were much sooner. Very much; wish it were sooner. Pretty much;.will be pleased when the time comes. Moderately; it is scheduled and I guess I'll be there. Very little; I'm not too sure I will want to come. 75 indicates higher satisfaction with either the session or self than a high score does. With the SATS form, however, a higher score indi- cates greater satisfaction with the session than a low score does. 0f the 51 clients initially assigned to the Treatment Group, 11 clients had no completed checklists, five clients had three com- pleted checklists, and 14 clients had four completed checklists. The mean SATS score for clients who completed two checklists (7': 12.72) was only slightly higher than the mean score of those clients who completed one checklist (7'= 12.41). Clients who com— pleted three checklists had a lower mean SATS score (R'= 11.53) than either the one-checklist or the two-checklist clients. The clients who completed all four checklists had the highest mean SATS score (7'= 15.74) of all the checklist clients. . For the first question, “How helpful do you feel your therapist was to you this session?", the mean reflecting the least helpfulness was for the clients who completed only one checklist (7'= 3.17). This mean score reflected a response between "pretty helpful" and "somewhat helpful." The clients who completed two check— lists perceived their therapists as slightly more helpful than the clients who completed only one checklist. The mean for the two- checklist clients was 2.94, indicating a response between "very help- ful" and "pretty helpful." The clients who completed three checklists had a mean score of 2.66 for question one, and the clients who com- pleted four checklists had a mean score on this item of 2.69. Both of these mean scores also fall between "very helpful" and "pretty he1pfu1." 76 For the second question, "How well do you feel that you are getting along, emotionally and psychologically, at this time?", the clients who completed two checklists had the highest mean (7'= 4.33), reflecting the lowest feeling of well-being of the checklist clients. This mean score fell between "So-so; manage to keep going with some effort" and "Fairly poorly; life gets pretty tough for me at times." The mean score for this question was 3.79 for the clients who com- pleted one checklist, a response between "Fairly well; have my ups and downs" and "So-so; manage to keep going with some effort." Also within this same range of responses were the mean scores of the three- checklist clients (7'= 3.60) and the mean scores of the four-checklist clients (Y'= 3.04). The third question, "To what extent are you looking forward to your next session?", was the only question of those investigated which exhibited a consistent trend from the one-checklist clients through the four-checklist clients, although the range of the means was quite small (2.71 to 3.13). In fact, this range was the smallest range of all questions examined. The clients who completed one check- list had a mean score of 3.13, a score which falls between "Pretty much; will be pleased when the time comes" and "Moderately; it is scheduled and I guess I'll be there." The two-checklist clients' mean score (7'= 2.89), the three-checklist clients' mean score (7'= 2.86), and the four-checklist clients' mean score (7'= 2.71), all fall between the responses "Very much; wish it were sooner" and "Pretty much; will be pleased when the time comes." 77 Satisfaction-by-Therapist Analysis A second conjecture regarding possible factors influencing the lack of significant findings for the major hypothesis concerned the possibility of a therapist-by-client interaction effect. 0f the ten therapists whose clients had usable sets of data, it was observed that the percentage of clients who completed four psychotherapy ses- sions varied greatly between therapists (see Table 3.1). The SATS scores of the 38 clients of the sample were extracted from the data and sorted by therapist. A mean SATS score was then calculated for each therapist. A summary of the findings is presented in Table 4.7. Table 4.7 Mean Satisfaction Scores (SATS) 0f the Sample by Therapist v—v Therapist n Range of SATS Mean SATS Percent of assigned Scores Score clients who finished four sessions A 2 6 - 18 12.00 17.64 B 5 2 - 21 10.80 40.00 C 6 10 - 21 16.33 26.08 D 3 15 - 20 17.33 21.42 E 7 10 - 18 15.42 80.00 F 3 15 - 22 19.66 75.00 G 2 l4 - 21 17.50 40.00 H 2 10 - 15 12.50 40.00 I 3 18 - 19 18.33 42.85 J 5 10 - 23 17.60 38.46 Total N = 38 2 - 23 15.74 34.45 78 Among the ten therapists, the number of clients who completed the study ranged from two to seven. The range of the SATS scores for the entire sample was from two to 23. The mean SATS scores by ther- apists ranged from a low of 10.80 to a high of 19.66. The percent of assigned clients who finished four psychotherapy sessions ranged be- tween therapists from a low of 17.64 percent to a high of 80.00 percent. Therapist B, with the lowest client SATS mean score (10.80) was also the therapist with the widest range (2-21) of client SATS scores. Therapist E, with the highest number of sample clients ( n = 7) also had the highest percentage of clients who completed the study (80 percent). Therapist F had the highest client mean SATS score (Y'= 19.66, n = 3), and also had the second highest percentage (75 percent) of clients complete the study. Discussion A positive finding of the study was that the two control groups were not significantly different. Although this finding can- not be interpreted to mean that the two control groups were the same, they were not significantly different to be considered two different sample groups. Therefore, the possibility that the time element alone, i.e., 45 minutes vs. 60 minutes, which could account for any differ- ences in results, was controlled. Because the major hypothesis was not rejected, a number of reasons for such a finding can be entertained. At least ten questions might be raised about the lack of significant findings: (1) How did the high drop-out rate affect the results? (2) Does lack of 79 significant findings imply psychotherapy was ineffective? (3) Was the theoretical base faulty? (4) Were the instruments sufficiently stable? (5) How did the timing of posttesting affect the results? (6) Were the clients in the Treatment Group, in fact, "involved" in the evaluation of their psychotherapy? (7) How appropriate were the instruments for use with a short-term psychotherapy focus? (8) Could there have been a therapist-by-client interaction effect which in- fluenced the findings? (9) Were the sample clients appropriate short- term psychotherapy candidates? and (10) Were four sessions too short for short-term psychotherapy? Each of these questions is explored, in turn, in the dis- cussion. Some are easier to answer than others, and some provide direction for refinements of the design for future studies. Hopefully, some will stimulate future research'gfideavors. High Drop-Out Rate How did the high drop-out rate affect the results? This question only can be answered speculatively. The drop-out rate for the Treatment Group (72.55 percent) was higher than the drop-out rate for both Control Group I (64.71 percent) and Control Group 11 (55.89 percent). All three sample groups had higher drop—out rates than normally expected at the Center. As reported earlier, only 52 percent of the Outpatient Services clientele at the Ingham Community Mental Health Center can be expected to complete four or more psy- chotherapy sessions, leaving a drop-out or therapy termination rate 80 of 48 percent. The 48 percent figure excludes post-hospitalization and chronically impaired clients requiring intensive services from the Aftercare unit at the Center. Because the Outpatient Services focus on a short-term crisis model, some clients can reasonably be expected to terminate treatment before completing four sessions. Exact figures were not obtainable from the participating therapists regarding how many clients self- terminated treatment and how many were referred elsewhere. A 2933. ‘hgg_examination of client case records did indicate, however, that most of the clients did self—terminate treatment. A few clients moved, at least one client was hospitalized, and some needed only medication reviews. In general, however, all clients and therapists entered the study with the expectation that four psychotherapy ses- sions would be a minimum requirement. As mentioned earlier, staff morale was low at the time of the study. Some reasons for low morale included poor salaries, re- cent acceptance of an employees' union, large caseloads, and hiring freezes. It does not seem unreasonable to conjecture that the added responsibility of informing the research team of client appointment times, cancellations, and checklist times, was seen as just another burden which was often forgotten among other daily demands. Examina- tion of client records did reveal that some clients had four or more psychotherapy sessions, but the appointment times had not been given to the research team. As a consequence, posttesting of these clients was missed. Because some of those who missed posttestings were 81 clients in the Treatment Group, the criterion checklists had not been completed either. Therefore, the problem was compounded because posttesting of these clients would not have been appropriate. As reported in the results section, the questions examined on the checklists of the completed Treatment Group sample did not indicate that the drop-outs were dissatisfied with the service they received. Their therapists were perceived as quite helpful, whether the clients came for one or four psychotherapy sessions. At the time of the treatment sessions reported on via the checklists, the clients also felt they were getting along moderately well emotionally and psychologically, and yet were looking forward to their next sessions, ,increasingly so as the number of sessions increased. Based on the responses gleaned from the checklists regarding satisfaction (SATS), it appeared as though the clients were somewhat satisfied with their sessions. Why, then did so many drop out? Comparable data was not available from the drop-outs of the two control groups, so it could only be assumed that they were not significantly different from the Treatment Group. Two possible explanations for the Treatment Group having the highest drop-out rate can be conjectured. The first reason could be an annoyance factor. Many clients had not been properly advised that pretesting would take approximately one hour prior to appointment times with their therapists for intake. These clients were not only upset with the pretesting procedure, but were further critical of their being asked to again do some paperwork task after their session that same day. The unanticipated time, plus the task 82 itself, could have caused sufficient resistance to increase the Treat- ment Group sample loss. Clients in the two control groups had only to contend with the pretesting annoyance. A second possible explanation could be increased anxiety generated by having to focus on 168 questions specific to the therapy sessions, and to further commit oneself to an answer, in writing, to each of the questions. If too many stimuli were too compacted, per- haps an inordinate amount of anxiety was generated and contributed to high drop-oat rates. .The taSk of completing a checklist after each session might have been so adverse that it was actively avoided. Was the checklist, therefore, a deterrent to continuation of treatment? Only a study designed to incorporate client feedback regarding the task of checklist completion would accurately answer such questions. Q Treatment Effectiveness A second interpretation of lack of significant findings is the most obvious, i.e., a Type II error was committed in failing to reject the null hypothesis. If this were so, only replication of the study could lend support to this conclusion. Because so many questions were raised by the design and the complications encountered during the implementation, a study designed to control for many of the conjectured invalidating factors would not only be desirable, but highly recom- mended. Two other studies yielded results contradictory to the findings of this study. Test (1964) compared three types of structured short-term therapy (group therapy, individual therapy, and writing 83 therapy), with each group receiving ten psychotherapy sessions. He found the writing group to have the largest number of changes, and the most significant changes, on measures of the MMPI, The Edwards Personal Preference Schedule, and the Butler-Haigh Q-Sort. Bastien (1974) also found a significant difference between §§_in a writing therapy group and two control groups on the Mooney Problem Checklist and the Semantic Differential. These two studies, which also used some aspect of written response from the clients, found results which contradicted the lack of significant findings of the current study. Adequacy of Theory and Instruments Because the self construct is an abstraction, it is diffi- cult to operationalize, and therefore difficult to measure. The utility of self concept theory in predicting outcome of psychotherapy is still a moot issue among theorists. For pusposes of the study, the measure chosen to operationalize the self concept was the Tennes- see Self Concept Scale. Expecting self concept to change significantly in a period of approximately four weeks might be unrealistic. However, the TSCS was not the only measure used in the study. The other two measures, the Mooney Problem Checklist and the TSR Satisfaction form, were combined with the TSCS to yield a mean vector of all measures for each client. The mean vectors were used for hypothesis testing. The TSCS, the MPC, and the TSR Satisfaction measures were factor analYZEd- The self concept measure (TSCS) was stable from pre to posttesting. The MPC appeared to be a highly unstable instrument in the way it 84 loaded on pre to posttest factors. ‘To further c0mplicate'matters, a high intercorrelation between many of the measures was found which increased the probability of making a Type II error. Complications might also have arisen because previous studies (Secord and Jourard, 1953; Fisher, 1973) have demonstrated a time lag between actual change and self-reported change. Out-of-date concepts might therefore, exist for a period of time after certain stages of psychotherapy, until new experiences become integrated into the self concept. Even though the semantic stability of the TSCS was demonstrated in the study, it is possible that the self concepts of the clients could have changed more than was measurable at posttesting. However, even then, it is not possible to predict which direction self concept might have changed more. Timing of Posttesting_ The above discussion suggests either postponed and/or multiple posttesting sessions. Such a refinement would control for the time lag for integrating self concept changes, and would also provide data on stability of therapeutic interventions. Satisfaction with psychotherapy might more appropriately be assessed sometime after psychotherapy has terminated and affect-laden responses are minimized. The timing of posttesting could also have been affected by stage of treatment. -For clients who were to be involved in continuing treat- ment after the fourth session posttesting, it is reasonable to assume that these clients would be in a different stage of the psychothera- peutic process than clients who were to terminate treatment at or nearer to the fourth psychotherapy session. 85 In considering the amount of stimulus input, Control Group II (45-minute sessions) had a total of three hours of stimuli across four psychotherapy sessions, Control Group I (60-minute sessions) had a total of four hours of stimuli across four psychotherapy sessions, and the Treatment Group (45-minute sessions plus 15-minute checklist completion sessions) had a total of four hours of stimuli across four psychotherapy sessions. Further, for the Treatment Group, the last 15 minutes (checklist completion) supposedly had more stimuli for the clients to deal with than would be expected in a 15-minute verbal exchange, considering that the TSR had a total of 168 questions and/or ratings which required a response in those 15 minutes. The issue of stage of psychotherapy process has important implications for the study. Many theorists (Rogers, 1951; 1961; Blocher, 1966; Kell and Mueller, 1966; Kell and Burrow, 1970; Raimy, 1971) have demonstrated that the process of psychotherapy follows a particular developmental sequence. The early stage focuses on orien- tation, ventilation, catharsis, disapproving self-references, and is, in general, an unfolding process leading to increased self disclosure and self awareness. The initial stage is characterized by denial of unacceptable parts of the self (Rogers, 1951). Examining, reorganiz- ing and beginning to integrate new awarenesses accompany a gradual shift from "symptoms" to "self" (Rogers, 1951). "As therapy proceeds . . . (the client) . . . often feels even more discouraged about and critical of himself," . . . but as therapy approaches termination, . . . (the client) experiences . . . "greater acceptance of the self" (Rogers, 1951). For the current study, if the checklists were, in 86 fact, a mode of "involvement", is it not possible that clients in the Treatment Group were in a more advanced stage of psychotherapy than the clients who were not exposed to the checklist completion task? Although this idea is highly speculative, it could be investigated in future studies by controlling for assessing the process of the psycho- therapy sesSions, as well as investigating the outcomes, as measured in the current study. Some method of judging the clients' stages of psychotherapy would facilitate equalizing the sample groups so all would be posttested at comparable stages. Instrument Appropriateness Were the instruments appropriate for use with a short—term psychotherapy focus? Again, no definitive answer is possible, but support for the instruments used in the study could be logically made by examining similar studies. Unfortunately, a review of the litera- ture did not reveal any short-term psychotherapy oriented studies which used the TSCS, therefore an exact comparison could not be made. The three studies reported (Test, 1964; Phillips and Wiener, 1966; Bastien, 1974) which used some writing therapy technique used instruments similar to those used in the current study. The Butler- Haigh Q-Sort, an instrument designed to measure self concept, was used by Test (1964) and Phillips and Wiener (1966). The Mooney Prob- lem Checklist was used by Bastien (1974). None of these studies used both instruments simultaneously. Only in the writing therapy group of the study conducted by Test (1964) were there significant changes on the Q-Sort (p = .04). In Bastien and Jacobs'(l974) study, . 87 significant differences (p < .025) were found on the MPC for the writing therapy group. Phillips and Wiener (1966), in describing a writing therapy case study, reported a figure “equal to the one re- ported by Rogers and Dymond (1954)" on pre-therapy to post-therapy Q-Sorts. The figures they reported were a change from -.08 before therapy to +.85 after therapy on the self-ideal Q-Sort. They con- cluded that, even though the writing therapy had a behavioral emphasis, ". . cognitive change may actually represent a shift in self- references brought about by successful, overt, behavioral changes." The instruments used in the study, therefore, do not seem inappropriate. Other studies have used comparable instruments and have found significant results. Phillips and Wiener (1966) stressed the importance of structure and goal setting in short-term psycho- therapy. At the time of this study, specific methods for structuring goal setting activities were not a part of the routine for therapists seeing short-term psychotherapy clients. The procedure has since changed at the Center where the study was conducted, and replication might reflect this added dimension of structured goal-setting to the short-term psychotherapy process. A few additional comments are warranted concerning the scales of the instruments used. As was reported in Table 3.8, only one scale correlation (MPC-Personality and MPC Self Improvement, r = .92) was in the .90's. This finding indicates that these two scales might have been collapsed into one scale, but because there was no theoretical base for such a decision, both scales were included in the analysis. Future studies might consider using only the MPC-Total in lieu of all sub-scores. 88 Therapist-by-Client Interaction Effect The question of a possible therapist-by-client interaction effect is more easily answered than many of the other questions raised by the lack of significant findings in the study. As was shown in Table 4.7, when SATS means were calculated for each therapist, based on the SATS scores of the therapist's clients who completed the study, a wide range (X'= 10.80 to 7'= 19.66) was observed. Whether a true interaction effect existed or not cannot be determined, except through replication. The evidence shown in Table 4.7 points very strongly to such an effect, at least with respect to satisfaction with psycho- therapy. Some therapists might produce negative changes in clients, thus neutralizing the positive influence of the more effective thera- pists (Bergin, 1963). A caveat is offered regarding satisfaction with psychotherapy. Satisfaction should not be equated with effectiveness per se, nor should it necessarily be seen as a reflection of better treatment. Most clinicians are aware that satisfaction with treatment per se does not mean that effective treatment is ensuing. In fact, often the opposite is true, because part of a therapist's role is to con- front the client on behaviors which are inappropriate or destructive. Future research dealing with optimal levels of satisfaction with psychotherapy could shed light on the questions raised regarding this issue. A further question deals with appropriate screening of clients for short-term psychotherapy. At the Ingham Community Mental Health Center, the only screening process which determined whether or 89 not a client was assigned to a therapist in the Outpatient Service was previous history of long-term psychiatric hospitalization. Clients with a lengthy psychiatric hospitalization history were assigned to the Aftercare Services Unit at the Center. All other clients were assigned to the Outpatient Services unit. Beyond this point, it was assumed that all Outpatient Service clients would receive short-term crisis-oriented services. The expectation might be realistic in terms of service demands, but unrealistic in terms of client needs. Occasionally, therapists did continue to see a client beyond the. short-term limit of the crisis model. However, there were limits re- garding how many clients a therapist could realistically handle in this manner without suffering from "caseload overload." It is pos- sible that all of the clients in the sample were not appropriate short-term psychotherapy candidates. Number of Short-Term Psychotherapy Sessions A last question centers on the number of psychotherapy ses- sions chosen for the study. Phillips and Wiener (1966), in their book on short-term psychotherapy, reviewed numerous studies which defined "short-term" as anywhere from one to one hundred sessions. Four ses- sions were chosen for the current study because 1974 figures for the Ingham Community Mental Health Center indicated that Outpatient Service clients come for an average of 4.05 sessions. Most of the studies in the literature, reviewed earlier, which reported on short-term psy- chotherapy, compared short-term with longer-term psychotherapy. No studies were found which elaborated on different time-limits of 9O short-term psychotherapy. Clarification of the definition of short- term psychotherapy would, obviously, be a prosperous area for future research. The four-session limit of the present study does not seem out of line with other studies in the area of short-term psychotherapy. What might be asked is, "Was this psychotherapy or was it something else?" By the criteria of the study's definition of psychotherapy, it was psychotherapy: an educative process involving a relationship between at least two persons, one of whom espouses a theoretical framework of the change process and accepts responsibility for helping the other to change. Summar The major hypothesis of the study, that outpatient clients in short-term psychotherapy who evaluated their own psychotherapy would score higher on measures of self concept and satisfaction with treatment, and report fewer self-identified problems than clients who did not evaluate their own psychotherapy, was not rejected. Ten questions raised by the design and rsults of the study were discussed: (1) How did the high drop-out rate affect the results? (2) Does the lack of significant findings imply psychotherapy was in- effective? (3) Was the theoretical base faulty? (4) Were the instru- ments used sufficiently stable? (5) How did the timing of posttesting affect the results? (6) Were the clients in the Treatment Group, in fact, "involved“ in the evaluation of their psychotherapy? (7) How appropriate were the instruments for use with a short4term psycho- therapy focus? (8) Could there have been a therapist-by-client 91 interaction effect which influenced the findings? (9) Were the sample clients appropriate short-term psychotherapy candidates? and (10) Were four sessions too short for short-term psychotherapy? Egg; hgg_exploratory analyses, investigating possible reasons for lack of significant findings, were reported. These analy- ses included factor analyses, examination of the TSR checklists, and inspection of SATS scores by therapist. The exploratory factor analyses indicated that the TSCS was a stable instrument from pretesting to posttesting, but the MPC was highly unstable. The finding implies that the semantic meanings of the MPC questions changed from pretesting to posttesting. The examination of key questions on the TSR checklists of the Treatment Group clients was conducted to identify possible reasons for the high drop-out rate. The clients indicated they were relatively satisfied with the services they received prior to dropping out, they perceived their therapists as quite helpful, and felt they were getting along moderately well, even though they looked forward to their next sessions. Therefore, reported dissatisfaction with treatment did not seem to be a factor in the high drop-out rate. When satisfaction with psychotherapy (SATS) was assessed for the entire sample by inspection of their SATS posttests only. widely discrepant scores by therapist indicated a possible therapist- by-client interaction effect. In relating the ten questions raised to the literature, support was lent to the theoretical base of the study. Stage of the psychotherapy process might have affected the posttesting results. .Q I‘ 92 Evidence was cited which supported the appropriateness of the instru- ments used in the study and the number of sessions chosen to define short-term. Theoretical issues regarding effectiveness of the treat- ment, client “involvement" and appropriateness of the sample for short-term psychotherapy were explored which led to suggestions for refinements in the design. CHAPTER V SUMMARY AND CONCLUSIONS In the final chapter the study is summarized, conclusions are discussed, and recommendations for future research are offered. Summary The purpose of the study was to examine the relationship of client involvement in the evaluation of psychotherapy to effectiveness of short-term psychotherapy. The majority of literature reporting on psychotherapy evaluation still depends on data obtained from clinicians, independent judges, or, even more rarely, therapist-client correla- tions. If psychotherapy is viewed as a special relationship which facilitates change, then this collaborative experience is lacking crucial feedback by ignoring the client as a valuable and necessary source of information about the change process. The sample consisted of 38 outpatient clients of the Ingham Community Mental Health Center in Lansing, Michigan. The Outpatient Service of this Center focuses on short-term psychotherapy. Immedi- ately prior to intake, each voluntary client completed a Tennessee Self Concept Scale (TSCS) and a Mooney Problem Checklist (MPC). After the fourth psychotherapy session, these two measures, plus the Therapy Session Report satisfaction form (SATS), were completed. Clients were randomly assigned to one of three groups: (1) Treatment Group (45- 93 94 minute psychotherapy sessions, plus 15-minute checklist-completion session; n = 12), (2) Control Group I (GO-minute psychotherapy ses- sions; n = 11), and (3) Control Group II (45-minute psychotherapy sessions; n = 15). The three groups were compared by Multivariate Analysis of Covariance on 16 variables simultaneously (TSCS-Self Criticism, TSCS-Physical Self, TSCS-Moral-Ethical Self, TSCS-Personal Self, TSCS-Family Self, TSCS-Social Self, MPC-Health, MPC—Economic Security, MPC-Self Improvement, MPC-Personality, MPC-Home and Family, MPC-Courtship, MPC-Sex, MPC-Religion, MPC-Occupation, and SATS). Helmert Contrasts were used in the analysis to combine control group means for comparison with the Treatment Group means. The research hypothesis, that outpatient clients in short- term psychotherapy who evaluated their own psychotherapy would score higher on measures of self concept and satisfaction with treatment, and report fewer self-identified problems, than clients who did not evaluate their own psychotherapy, was not rejected. Ten questions were raised by the design and results of the study, and were discussed in depth: (1) How did the high drop-out rate affect the results? (2) Does the lack of significant findings imply psychotherapy was ineffective? (3) Was the theoretical base faulty? (4) Were the instruments used sufficiently stable? (5) How did the timing of posttesting affect the results? (6) Were the clients in the Treatment Group, in fact, "involved" in the evaluation of their psychotherapy? (7) How appropriate were the instruments for use with a short-term psychotherapy focus? (8) Could there have been 95 a therapist-by-client interaction effect which influenced the findings? (9) Were the sample clients appropriate short-term psychotherapy candidates? and (10) Were four sessions too short for short-term psy- chotherapy? Conclusions The major conclusion of the study was that involving clients in the evaluation of their own psychotherapy did ggt_significantly affect psychotherapy effectiveness, defined as a more positive self concept, fewer self-identified problems, and greater satisfaction with treatment. Client completion of a checklist following each psycho- therapy session ("involvement") did not significantly affect psycho- therapy effectiveness. Exploratory post hoc analyses investigating possible reasons for lack of significant findings and research design complications resulted in two additional conclusions: (1) The Mooney Problem Checklist was highly unstable from pretesting to posttesting, indi- cating that semantic meanings of the problems listed on the MPC changed over time; and (2) The design of the study should be refined to control for factors which complicated the implementation and possibly influenced the failure to find significant results. Recommendations for Future Research One criterion for usefulness of research is heuristic value. Recommendations are offered regarding the instrumentation used in the study, and refinements in the design are enumerated to facilitate future efforts in short-term psychotherapy research. 96 Recommendations regarding instrumentation are twofold. First is the suggestion that the MPC means and standard deviations be used as norms for client populations similar to the outpatient sample reported in the study. This could be a starting point for future research investigating clinic populations consisting primarily of non-student adults. Second, because the covariates in the study accounted for less than 9 percent of the total variance, the two covariates should be eliminated in future research.‘ Instead, a model which incorporates measurement of the treatment process, the treatment outcome, and satisfaction with treatment, would be ideal. Several recommendations are also offered to increase the precision and validity of the deSign. Replication efforts might con- sider the following: (1) Clients should be screened for appropriate- ness for short-term psychotherapy by some method using severity of impairment, diagnosis, and/or service expectations; (2) The design should incorporate controls for testing a therapist-by-client inter- action effect; (3) Close monitoring during the implementation should include better client preparation regarding additional time involved, rigorous orientation of the participating therapists, structured therapeutic goal-setting and control for the process and timing of the psychotherapy sessions; (4) Delayed, and possibly multiple, post- testing sessions should be incorporated into the design; and (5) Client reactions to giving feedback regarding their psychotherapy sessions should be solicited. 97 Beyond these refinements in the design, future research efforts might consider two fertile areas of interest. Little informa- tion is reported in the literature about short-term psychotherapy research. Of the studies reviewed, none focused on different time limits of short-term psychotherapy. Each study defined short-term psychotherapy in its own way. To arrive at an operational definition of short—term psychotherapy, staggered numbers of sessions would be recommended for studies investigating the optimal number of sessions necessary and sufficient for "short-term psychotherapy." A last recommendation expands the concept of "involvement" as it relates to psychotherapy effectiveness. Future studies might investigate this concept further by hypothesizing varying degrees or kinds of "involvement." Involvement could progress from "no involve- ment" (doing nothing after psychotherapy sessions), to "low involve- ment" (client-completion of a checklist after each psychotherapy session), to "moderate involvement“ (client-dictated process summary after each psychotherapy session), to "high involvement" (client- written or videotaped process summary after each psychotherapy session). There can be many ways to define involvement. Just as individuals differ in the ways they learn, might they not also differ in the ways they feel "involved?" How various types of involvement might affect the psychotherapy relationship remains to be discovered. APPENDICES APPENDIX A APPENDIX A (INSTRUCTIONS TO TREATMENT GROUP--CHECKLIST) DEAR CLIENT: WE'D LIKE YOU TO TAKE A FEW MINUTES TO ANSWER SOME QUESTIONS ABOUT THE THERAPY SESSION YOU HAVE JUST FINISHED. PLEASE ANSWER ALL THE QUESTIONS AS HONESTLY AS POSSIBLE. YOUR THERAPIST WILL NQI_SEE YOUR COMPLETED CHECKLIST. YOU NOW HAVE 15 MINUTES TO COMPLETE THE ATTACHED CHECKLIST. WHEN 15 MINUTES ARE UP YOU WILL BE NOTIFIED. IF YOU FINISH BEFORE 15 MINUTES. PLEASE RETURN THE CHECKLIST TO THE MANILA ENVELOPE ON THE WALL BY ROOM 229. THANK YOU! 98 99 (P) THERAPY SESSION REPORT This booklet contains a series of questions about the therapy session which you have just completed. These questions have been designed to make the description of your experiences in the session simple and quick. There are two types of questions. One type of question is followed by a series of numbers on the right— hand side of the page. After you read each of the questions, you should circle the number "0" if your answer is "no"; circle the number "1" if your answer is ”some"; etc. The other questions have a series of numbered statements under them. You should read each of these statements and select the one which comes closest to describing your answer to that question. Then Ei—17c1e the number in front of your answer. Once you have become familiar with the questions through regular use, answering them should take only a few minutes. Please feel free to write additional comments on a page when you want to say things not easily put into the categories provided. BE SURE TO ANSWER EACH QUESTION. Identification Date of Ses sion Copyright by Psychotherapy Session Project, 1966. All prOperty rights reserved by the Psychotherapy Session Project, 907 South Wolcott Avenue, Chicago, Illinois, U.S.A. 100 HOW DO YOU FEEL ABOUT THE SESSION WHICH YOU HAVE JUST COMPLETED? (Circle the one answer which best applies.) THIS SESSION WAS: 1. Perfect. 2. Excellent. 3. Very good. 4. Pretty good. 5. Fair. 6. Pretty poor. 7. Very poor. 10] WHAT SUBJECTS DID YOU TALK ABOUT DURING THIS SESSION? (For each subject, circle the answer which best applies.) DURING THIS SESSION I TALKED ABOUT: Z. 10. ll. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. MY MOTHER. MY FATHER. MY BROTHERS OR SISTERS. MY CHILDHOOD. MY ADOLESCENCE. RELIGIOUS FEELINGS, ACTIVITIES OR EXPERIENCES. WORK, CAREER OR EDUCATION. RELATIONS WITH OTHERS OF THE SAME SEX. RELATIONS WITH THE OPPOSITE SEX. FINANCIAL RESOURCES OR PROBLEMS WITH MONEY. FEELINGS ABOUT SPOUSE OR ABOUT BEING MARRIED. HOUSEHOLD RESPONSIBILITIES OR ACTIVITIES. FEELINGS ABOUT CHILDREN OR BEING A PARENT. BODY FUNCTIONS, SYMPTOMS, OR APPEARANCE. STRANGE OR UNUSUAL IDEAS AND EXPERIENCES. HOPES OR FEARS ABOUT THE FUTURE. DREAMS OR FANTASIES. ATTITUDES OR FEELINGS TOWARD MY THERAPIST. THERAPY: FEELINGS AND PROGRESS AS A PATIENT. OTHER: NO SOME A LOT 0 1 2 0 l 2 0 1 Z O l 2 0 l 2 0 l 2 0 l 2 O 1 Z 0 1 2 0 l 2 0 1 2 0 l 2 O 1 2 O 1 2 0 l 2 0 l 2 0 1 2 0 l 2 O l 2 1 2 BE SURE THAT YOU HAVE CHECKED EVERY ITEM. 102 WHAT DID YOU WANT OR HOPE TO GET OUT OF THIS SESSION? (For each item, circle the answer which best applies.) THIS SESSION I HOPED OR WANTED TO: NO SOME A LOT 22. GET A CHANCE TO LET GO AND GET THINGS O 1 2 OFF MY CHEST. 23. LEARN MORE ABOUT WHAT TO DO IN THERAPY, O 1 2 AND WHAT TO EXPECT FROM IT. 24. GET HELP IN TALKING ABOUT WHAT IS REALLY 0 l 2 TROUBLING ME. 25. GET RELIEF FROM TENSIONS OR UNPLEASANT 0 l 2 FEELINGS. 26. UNDERSTAND THE REASONS BEHIND MY 0 l 2 FEELINGS AND BEHAVIOR. 27. GET SOME REASSURANCE ABOUT HOW I'M DOING. 0 l 2 28. GET CONFIDENCE TO TRY NEW THINGS, TO BE 0 l 2 A DIFFERENT KIND OF PERSON. 29. FIND OUT WHAT MY FEELINGS REALLY ARE, 0 l 2 AND WHAT I REALLY WANT. 30. GET ADVICE ON HOW TO DEAL WITH MY LIFE 0 l 2 AND WITH OTHER PEOPLE. 31. HAVE MY THERAPIST RESPOND TO ME ON A O l 2 PERSON- TO-PERSON BASIS. 32. GET BETTER SELF CONTROL. 0 l 2 33. GET STRAIGHT ON WHICH THINGS I THINK AND 0 l 2 FEEL ARE REAL AND WHICH ARE MOSTLY IN MY MND. 34. WORK OUT A PARTICULAR PROBLEM THAT'S O 1 2 BEEN BOTHERING ME. 35. GET MY THERAPIST TO SAY WHAT HE (SHE) 0 l 2 REALLY THINKS. 36. OTHER: 1 2 BE SURE THAT YOU HAVE CHECKED EVERY ITEM. 103 WHAT PROBLEMS OR FEELINGS WERE YOU CONCERNED ABOUT THIS SESSION? (For each item, circle the answer which best applies.) DURING THIS SESSION I WAS CONCERNED ABOUT: NO SOME A LOT 37. BEING DEPENDENT ON OTHERS. 0 l 2 38. MEETING MY OBLIGATIONS AND O 1 2 RESPONSIBILITIES. 39. BEING ASSERTIVE OR COMPETITIVE. O l 2 40. LIVING UP TO MY CONSCIENCE: SHAMEFUL 0 l 2 OR GUILTY FEELINGS. 41. BEING LONELY OR ISOLATED. 0 l 2 42. SEXUAL FEELINGS AND EXPERIENCES. O l 2 43. EXPRESSING OR EXPOSING MYSELF TO OTHERS. O 1 2 44. LOVING: BEING ABLE TO GIVE OF MYSELF. O 1 2 45. ANGRY FEELINGS OR BEHAVIOR. 0 l 2 46. WHO I AM AND WHAT I WANT. 0 l 2 47. FEARFUL OR PANICKY EXPERIENCES. 0 1 2 48. MEANING LITTLE OR NOTHING TO OTHERS: 0 l 2 BEING WORTHLESS OR UNLOVABLE. 49. OTHER: 1 2 PLEASE DO NOT WRITE BELOW THIS LINE. 57_ 60_ 65__ 7o_ 72__ 75_ 79_ 58__ 61__ 66_l_ 71_ 73__ 76_ 80__ 59_ 62__ 67_ 74__ 77__ 63 68 _; 78_ 64 69__1 104 WHAT WERE YOUR FEELINGS DURING THIS SESSION? (For each feeling, Circle the answer which best applies.) DURING THIS SESSION I FELT: 10. 11. 12. 13. 14. 15. 16. 17. . CONFI DENT . EMBARRASSED . RELAXED WITHDRAWN . HE LP LESS DETERMINED . GRATEFUL . RELIEVED TEARFUL C LOSE IMPATIENT GUILTY STRANGE INADEQUATE LIKEAB LE HURT DEPRESSED BE SURE NO SOME A LOT 0 1 2 0 l 2 0 1 2 0 1 2 0 l 2 0 1 2 0 l 2 0 l 2 0 1 2 0 l 2 0 1 2 0 1 2 0 1 2 0 l 2 0 1 2 0 1 2 0 1 2 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. AFFEC TIONATE SERIOUS ANXIO US ANGRY PLEASED INHIBITED CONFUSED DISCOURAGED ACCEPTED CAUTIOUS FRUSTRATED HOPEFUL TIRED ILL THIRSTY SEXUALLY ATTRACTED OTHER:____ NO SOME A LOT 0 1 2 0 l 2 0 1 2 0 1 2 0 1 2 0 l 2 0 1 2 0 1 2 0 1 2 0 1 2 0 l 2 0 1 2 0 1 Z O l 2 0 1 2 0 l 2 THAT YOU HAVE CHECKED EVERY ITEM. 105 Slightly or not Pretty Very DURING THIS SESSION, HOW MUCH: at all Some Much Much 35. DID YOU TALK? 0 l 2 3 36. WERE YOU ABLE TO FOCUS ON WHAT 0 1 2 3 WAS OF REAL CONCERN TO YOU? 37. DID YOU TAKE INITIATIVE IN BRINGING UP 0 1 Z 3 THE SUBJECTS THAT WERE TALKED ABOUT? 38. WERE YOU LOGICAL AND ORGANIZED 0 1 2 3 IN EXPRESSING YOURSELF? 39. WERE YOUR EMOTIONS OR FEELINGS 0 l 2 3 STIRRED UP? 40. DID YOU TALK ABOUT WHAT YOU WERE 0 l 2 3 FEELING? 41. WERE YOU ANGRY OR CRITICAL TOWARDS 0 l 2 3 YOURSELF? 42. DID YOU HAVE DIFFICULTY THINKING OF 0 1 2 3 THINGS TO TALK ABOUT? 106 DURING THIS SESSION, HOW MUCH: 43. 44. 45. 46. 47. 48. 49. 50. FRIENDLINESS OR RESPECT DID YOU SHOW TOWARDS YOUR THERAPIST? WERE YOU FREE AND SPONTANEOUS IN EXPRESSING YOURSELF? DID YOU TRY TO PERSUADE YOUR THERAPIST TO SEE THINGS YOUR WAY? WERE YOU ATTENTIVE TO WHAT YOUR THERAPIST WAS TRYING TO GET ACROSS TO YOU? DID YOU TEND TO ACCEPT OR AGREE WITH WHAT YOUR THERAPIST SAID? DID YOU HAVE A SENSE OF CONTROL OVER YOUR FEELINGS AND BEHAVIOR? WERE YOU NEGATIVE OR CRITICAL TOWARDS YOUR THERAPIST? WERE YOU SATISFIED OR PLEASED WITH YOUR OWN BEHAVIOR? Slightly or not at all Some Pretty Very Much Much 2 3 2 3 2 3 2 3 2 3 2 3 51. 52. HOW DID YOU FEEL ABOUT COMING TO THERAPY THIS SESSION? 1. HOW MUCH PROGRESS DO YOU FEEL YOU MADE IN DEALING WITH 107 Eager; could hardly wait to get here. Very much looking forward to coming. Somewhat looking forward to coming. Neutral about coming. Somewhat reluctant to come. Unwilling; felt I didn't want to come at all. YOUR PROBLEMS THIS SESSION? A great deal of progress. Considerable progress. Moderate progress. Some progress. Didn't get anywhere this session. In some ways my problems seem to have gotten worse this session. 108 53. HOW WELL DO YOU FEEL THAT YOU ARE GETTING ALONG, EMO- TIONALLY AND PSYCHOLOGICALLY, AT THIS TIME? I AM GETTING ALONG: 1. Very well; much the way I would like to. 2. Quite well; no important complaints. 3. Fairly well; have my ups and downs. 4. So-so; manage to keep going with some effort. 5. Fairly poorly; life gets pretty tough for me at times. 6. Quite poorly; can barely manage to deal with things. 54. TO WHAT EXTENT ARE YOU LOOKING FORWARD TO YOUR NEXT SESSION? 1. Intensely; wish it were much sooner. 2. Very much; wish it were sooner. 3. Pretty much; will be pleased when the time comes. 4. Moderately; it is scheduled and I guess I'll be there. 5. Very little; I'm not too sure I will want to come. PLEASE DO NOT WRITE BELOW THIS LINE. 57__ 6o_ 65_ 7o_ 72__ 75_ 79__ 58__ 6I_ 66_I 71_ 73___ 76__ so_ 59__ 62___ 67___ 74___ 77__ 63 68 ___; 73_ 64 69.2. 109 WHAT DO YOU FEEL THAT YOU GOT OUT OF THIS SESSION? (For each item, circle the answer which best applies.) I FEEL THAT I GOT: 10. 11. 12. 13. 14-18. A CHANCE TO LET GO AND GET THINGS OFF MY CHEST. HOPE: A FEELING THAT THINGS CAN WORK OUT FOR ME. HELP IN TALKING ABOUT WHAT WAS REALLY TROUBLING ME. RELIEF FROM TENSIONS OR UNPLEASANT FEELINGS. MORE UNDERSTANDING OF THE REASONS BEHIND MY BEHAVIOR AND FEELINGS. REASSURANCE AND ENCOURAGEMENT ABOUT HOW I'M DOING. CONFIDENCE TO TRY TO DO THINGS DIFFERENTLY. MORE ABILITY TO FEEL MY FEELINGS, TO KNOW WHAT I REALLY WANT. IDEAS FOR BETTER WAYS OF DEALING WITH PEOPLE AND PROBLEMS. MORE OF A PERSON-TO—PERSON RELATION- SHIP WITH MY THERAPIST. BETTER SELF CONTROL OVER MY MOODS AND ACTIONS. A MORE REALISTIC EVALUATION OF MY THOUGHTS AND FEELINGS. NOTHING IN PARTICULAR: I FEEL THE SAME AS I DID BEFORE THE SESSION. OTHER: NO SOME A LOT 0 l 2 0 1 2 0 l 2 0 1 2 0 1 2 0 l 2 0 l 2 0 1 2 0 1 2 0 1 2 0 l 2 0 1 2 0 l 2 19. 20. 110 HOW WELL DID YOUR THERAPIST SEEM TO UNDERSTAND WHAT YOU WERE FEELING AND THINKING THIS SESSION? MY THERAPIST: l. Understood exactly how I thought and felt. 2. Understood very well how I thought and felt. 3. Understood pretty well, but there were some things he (she) didn't seem to grasp. 4. Didn't understand too well how I thought and felt. 5. Misunderstood how I thought and felt. HOW HELPFUL DO YOU FEEL YOUR THERAPIST WAS TO YOU THIS SESSION? 1. Completely helpful. 2 . Ve ry helpful. 3. Pretty helpful. 4. Somewhat helpful. 5. Slightly helpful. 6. Not at all helpful. 111 DURING THIS SESSION, HOW MUCH: 21. 22. 23. 24. 25. 26. 27. 28. DID YOUR THERAPIST TALK? WAS YOUR THERAPIST ATTENTIVE TO WHAT YOU WERE TRYING TO GET ACROSS? DID YOUR THERAPIST TEND TO ACCEPT OR AGREE WITH YOUR IDEAS AND POINT OF VIEW? WAS YOUR THERAPIST NEGATIVE OR CRITICAL TOWARDS YOU? DID YOUR THERAPIST TAKE INITIATIVE IN BRINGING UP THINGS TO TALK ABOUT? DID YOUR THERAPIST TRY TO GET YOU TO CHANGE YOUR POINT OF VIEW OR WAY OF DOING THINGS? WAS YOUR THERAPIST FRIENDLY AND WARM TOWARDS YOU? DID YOUR THERAPIST SHOW FEELING? Slightly or not at all Some Pretty Very Much Much 2 3 2 3 2 3 2 3 2 3 2 3 2 3 2 3 112 HOW DID YOUR THERAPIST SEEM TO FEEL DURING THIS SESSION? (For each item, circle the answer which best applies.) MY THERAPIST SEEMED: No SOME A LOT NO SOME A LOT 29. PLEASED o l 2 43. ATTRACTED o 1 2 3o. THOUGHTFUL o I 2 44. CONFIDENT o 1 2 31. ANNOYED o 1 2 45. RELAXED o l 2 32. BORED o l 2 46. INTERESTED o 1 2 33. SYMPATHETIC o 1 2 47. UNSURE o 1 2 34. CHEERFUL o l 2 48. OPTIMISTIC o l 2 35. FRUSTRATED o l 2 49. DISTRACTED o 1 2 36. INVOLVED o 1 2 so. AFFECTIONATE o l 2 37. PLAYFUL o 1 2 51. ALERT o 1 2 38. DEMANDING o I 2 52. CLOSE o 1 2 39. APPREHENSIVE o 1 2 53. TIRED o 1 2 4o. EFFECTIVE o 1 2 54. OTHER: __ l 2 41. PERPLEXED o 1 2 42. DETACHED o 1 2 BE SURE THAT YOU HAVE CHECKED EVERY ITEM. APPENDIX B SATISFACTION (SATS) APPENDIX B SATISFACTION (SATS) WHAT DO YOU FEEL THAT YOU GOT OUT OF THESE SESSIONS? (For each item, circle the answer which best applies.) I FEEL THAT I GOT: 1. A CHANCE TO LET GO AND GET THINGS OFF NO SOME A LOT MY CHEST. O 1 2 2. HOPE: A FEELING THAT THINGS CAN WORK OUT FOR ME. 0 1 2 3. HELP IN TALKING ABOUT WHAT WAS REALLY TROUBLING ME. 0 1 2 4. RELIEF FROM TENSIONS 0R UNPLEASANT FEELINGS. 0 1 2 5. MORE UNDERSTANDING OF THE REASONS BEHIND MY BEHAVIOR AND FEELINGS. O 1 2 6. REASSURANCE AND ENCOURAGEMENT ABOUT HOW I'M DOING. O 1 2 7. CONFIDENCE TO TRY TO DO THINGS DIFFERENTLY. O 1 2 8. MORE ABILITY TO FEEL MY FEELINGS, TO KNOW WHAT I REALLY WANT. O 1 2 9. IDEAS FOR BETTER WAYS OF DEALING WITH PEOPLE AND PROBLEMS. O 1 2 10. MORE OF A PERSON-TO-PERSON RELATIONSHIP WITH MY THERAPIST. O 1 2 11. BETTER SELF CONTROL OVER MY MOODS AND ACTIONS. O 1 2 12. A MORE REALISTIC EVALUATION OF MY THOUGHTS AND FEELINGS. O 1 2 13. NOTHING IN PARTICULAR: I FEEL THE SAME AS I DID BEFORE THE SESSIONS. O 1 2 113 APPENDIX C THERAPIST DEMOGRAPHIC DATA APPENDIX C THERAPIST DEMOGRAPHIC DATA Therapist Sex Age Years experience in NH setting A F 26 3.00 B M 30 5.00 C F 27 4.50 D F 27 3.00 E M 28 2.50 F M 32 2.00 G F 26 2.75 H F 26 3.50 I M 28 5.00 J F 29 4.00 6 females X = 27.90 4 males >