THESIS IIIIIIIIIIIIIIZIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII _ ~ _ » 31049 1 9646 . 63153"? . This is to certify that the thesis entitled CONGRUENCE OF PERCEPTION ON CLIENT SYMPTOMS AND THERAPY OUTCOME presented by JOHN PILAK has been accepted towards fulfillment of the requirements for MA degree in 6)? [O/C’jfy Major professor Date 1—,[7-7/[27/ 0-7639 MSU RETURNING MATERIALS: Place in book drop to remove this checkout from LIBRARIES .—::—. your record. FINES will be charged if book is returned after the date stamped below. . ‘39 l 6 mg rfngLLJljfEQ (d IE“ \ \ E {3 in CONGRUENCE OF PERCEPTION ON CLIENT SYMPTOMS AND THERAPY OUTCOME BY John Filak A THESIS Submitted to Michigan State University partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Psychology 1982 ABSTRACT CONGRUENCE OF PERCEPTION ON CLIENT SYMPTOMS AND THERAPY OUTCOME By John Filak This study examined congruence of perception (clini- cian's judgment and client self-report) on client symptoma- tology and its relationship to outcome ratings of psycho- therapy. The construct, congruence of perception, was defined as the degree of agreement between therapist and client through independent ratings based on the Symptom Distress Check List's (SCL-9OR) ten symptom dimensions. The major hypothesis stated that: "congruence of perception on client symptoms at posttherapy is related to positive out- come as rated by therapist and client." This hypothesis was supported. Additional analyses on the congruence measure are reported, as well as analyses on clinicians' use of the SCL-90R. Findings suggest that therapists' accuracy or congruence to clients' symptomatic status may be an impor- tant factor in psychotherapy. Future research in this area is suggested. To Laura and to my parents ii ACKNOWLEDGEMENTS I would like to express my warm appreciation to my thesis committee. I would like to thank Dr. Norman Abeles, Chairman of my thesis committee. for the way he has been with.me. His guidance, competence, flexibility, and Openness has contri- buted greatly to the develOpment of this project. The chance for me to see him as a teacher. and to see his acceptance of his students, has been very important to me. I would like to thank Dr. John Hurley, for his considerable contributions to this thesis. His work and energy into this project is heartfelt. His understanding of the deve10pment of clinical psychology through its inter- personal influences. his individuality, and his confrontative and caring approach with me are greatly valued. I would like to thank Dr. Donald Grumman, for his con- tributions to this project. His impact on my deveIOpment, present and future, as a psychotherapist is profound. Thanks also to department secretaries, Marjorie Curtis and Loismarie Williams. for their c00peration with the typing of earlier drafts, and to Judie Shepherd, a professional typist, for her work on the final draft. iii TABLE OF CONTENTS LIST OF TABLES . . INTRODUCTION . . . . . . REVIEW OF THE LITERATURE . . . . . . . Person Perception Research. . . Judgment Process -- Person Factors. . Relevant Social Psychological Research. Clinical Studies. . . . . . . Methodological Considerations . Physician- -Patient Congruence. HYPOTHESES . METHOD . . . . . . . . . Data. . . . Clients . . Therapists. Instruments A. Symptom Checklists . B. Outcome Measures . C. Congruence of Perception Measures: Procedure . . . . . RESULTS. . . . . . . Preliminary Comments. . . . . . . . A. Change from two congruence measures to one. . . . B. Interpretation of congruence measure. . C. Pooling of male and female data. Hypothesis I. . . . . . . . . . . . Hypothesis II . . . . . . . . . . Hypothesis III. . . Hypothesis IV . Hypothesis V. iv Pa e v 37 38 38 38 41 42 42 44 Hypothesis VI . . . . . . . . . . . . . . . Hypothesis VII. . . . . . . . . . . Additional Data Analyses. . . DISCUSSION . . . . . . . . . . . . . . . . . . . . Congruence of Perception Findings . . . . . . . . A. Therapy Outcome. . . . B. Congruence of Perceptions - Person factors . . . . C. Congruence of Perception and Ear Termination. . . . . . D. Congruence of Perception - Intake vs. Posttherapy . . . . . . . E. Implication for Future Research. . . . . The Use of the SCI-90R in a Psychotherapy Setting . . . . . . . . . . . . . SUMMARY. . . . . . . . . . . . . . . . . . . . . . . LIST OF REFERENCES ... . . . . . . . . APPENDICES A SYMPTOM DISTRESS CHECKLIST (SCI-90R) CLIENT FORM. . . . . . . . B SYMPTOM DISTRESS CHECKLIST (SCL-9OA) CLINICIAN POEM . . . . C POSTTHERAPY CLIENT QUESTIONNAIRE . D POSTTHERAPY THERAPIST QUESTIONNAIRE. . E STANDARD (T-SCORE) NORMS FOR MICHIGAN A. Discrepancies between clinicians and clients on the SCL- 90R' 3 Ten Scales. B. Correlations between clinicians' global distress ratings and separate scale discrepancies. . . . C. Correlations between global distress ratings and separate scale ratings . . D. Summary and Comparison with Derogatis et al' 8 (1976) Study . . . . . . . . . . STATE PSYCHOLOGICAL CLINIC CLINICIAN RATINGS OF MALE AND FEMALE OUTPATIENTS ON THE SCL-9OR'S SYMPTOM DIMENSIONS AND GLOBAL PATHOLOGY INDEX . . . . Page 45 46 49 49 49 53 55 56 57 59 59 61 63 64 67 68 71 77 78 84 88 Table 10. LIST OF TABLES Correlations between congruence of percep- tion scores and clients' and therapists' posttherapy ratings . . . . . Comparison of early terminators and continuers' congruence of perception scores at intake . Comparison of congruence of perception scores at intake and at posttherapy for clients who had the same person for intake worker and therapist . . . . Intake workers' mean congruence of perception scores at intake . Analysis of variance of intake workers' mean congruence of perception scores . Therapists' mean congruence of perception at posttherapy. . . . . . . . . . . . . Analysis of variance of therapists mean congruence of perception scores at posttherapy . . . . . . . . Mean T- Score points discrepancies between clinician and client for each of the ten dimensions. Correlations between clinicians' global distress ratings of clients' symptomatology and discrepancy scores for individual symptom.dimensions. . . . . Correlations between clinicians' global distress ratings and their ratings of clients on the separate scales. . . . . . . . . vi Page . 4O . 41 . 42 . 43 . 43 . 44 . 4S . 46 . 47 . 48 INTRODUCTION This study examines congruence of perception (clinician's judgment and client self-report) on client symptomatology and its relationship to process and outcome ratings of psycho- therapy. The degree to which a clinician perceives client symptomatology in congruence with client self-report is tested as a therapist variable in the process of therapy. Congruence of perception is also examined and compared at specific points in the course of clinician-client contact and is examined in relation to early termination. There are two major objectives of this study. One is to determine whether congruence of perception between therapist and client on client symptoms and level of distress is an important factor in psychotherapy especially as it relates to variables such as the client per- ceived satisfaction of the therapy and therapist rating of client benefit. AnothEr objective of this study is to compare the degree of congruence of perception between mental health professionals and their clients across symptom.dflmensions and a global distress index. REVIEW OF THE LITERATURE Is the therapist's possession of an accurate under- standing or impression of his or her client's distress an important factor in psychotherapy? Is it important for a therapist to have a capacity to graSp the degree of particu- lar symptomatology present as well as the overall sense of symptom configuration from interviewing the client? While such therapist skills may be taken for granted as being an important part of clinical performance and training, their value as therapeutic skills in relation to the effectiveness or outcome of therapy is not really known. In fact, the issue of the clinician's ability to judge peOple in general on social and psychological dimensions has had a long and arduous history which is still in the process of being understood conceptually and clarified. This review begins with the social psychological research in the person percep- tion area with a focus on relevant conceptual advances in the area. From this base, clinical studies are reviewed in which the majority of these studies have found that a clinician's capacity to be a good judge of certain attributes in peOple is important. Following this, relevant work on the perception of symptoms is presented as well as a ration- ale for some of the methodology of this study. Person Perception Research Considerable research in psychology has focused on the capacity of one person to ascertain the behavior, emotions, 2 3 attitudes, or traits of another person. This skill or capacity for one person to conceptualize and often to quan- tify specific aSpects of another person on social or psycho- logical dimensions has been labeled variously as psychologi- cal understanding, judgment, empathy, accuracy of perception, intuition, interpersonal sensitivity, etc. Early investigators assumed that the ability to judge people accurately was a skill that mental health professionals were most likely to possess because of the nature of clinical work and training (Weiss, 1963). In fact, one of the most widely employed ways for these studies to proceed was to use the "expert" Opinion of psychologists and psychiatrists on case material to set up the criteria of "reality" or "objec- tive standard" against which to rate the accuracy of judges' perceptions of subjects (Taft, 1955). However, when mental health professionals as a group were put in the role of judges and compared to other groups in assessing individuals, they did not fare as expected. In tasks involving predicting how subjects would answer ques- tionnaires or personality inventories, or even in predicting behavior from case abstracts, research findings found that on the average nonpsychologists fared better than psycholo- gists on these kinds of tasks (see Taft, 1955 for a review). This finding about psychologists was fairly common yet perplexing (Crow, 1957; Cutler, Bordin, Williams & Rigler, 1958; Weiss, 1963), and cast doubt on the importance of the clinician to be a good judge of peOple since, as a group, mental health professionals could outperform neither physical 4 scientists nor undergraduates on person perception tasks. Further, it was found that the more psychological training a judge had, the worse in general he or she did on these psychological judgment tasks (Taft, 1955). In his review Taft entertained three hypotheses regarding these findings: 1) training in psychology blunts a judges' ability, 2) being a good judge of others is incompatible with the social skills necessary to be a clinical psychologist, and 3) many profes- sors and clinicians tend to live in isolation from the general life experiences of the peOple whom they are endeavored to understand. Another hypothesis was that clinical training over- sensitized psychologists for these tasks to individual differences (Crow, 1957), or similarly that training, at least psychoanalytic training, may lead to a somewhat greater ability to entertain multiple alternative hypotheses, eSpe- cially in the face of increasing amounts of contextual information (Cutler et al, 1958). The effect of psychologi- cal training then was to increase the judges variability in responding to subjects when in general such variability or diacriminations led to a decrease in accuracy scores (Crow, 1957). Several studies supported the idea that clinicians are more attuned to individual differences or that they may entertain too many hypotheses about a person as information increases for performing well on many kinds of person perception tasks. Fer instance, Weiss (1963) found that with only minimal information available, clinical psychologists 5 outperform physical scientists on prediction tasks involving peOple, but that as the amount of information increased, physical scientists began to do better than psychologists. A.number of studies in another review by Taft (1959) found that increasing the amount of case material past a rapidly reached ceiling does not generate an increase in accurate assessment by clinicians and may even hamper judgment. Cline and Richards (1962) found that stereotypic prediction is often the most accurate in making judgments and that psychologists do tend to overrely on individual differences. These findings were consistent with Cronbach's (1955, 1958) critiques of much of the previous research in this area for the blending of different kinds of accuracy. Cronbach convincingly demonstrated that many studies, but especially those in which judges had to predict how subjects would answer items on personality questionnaires or how subjects would react in certain situations, confounded dis- criminative accuracy or interpersonal sensitivity with judges' knowledge of stereotypic responding and the degree of similarity between judges and subjects. Cronbach's distinction between stereotypic accuracy and discriminative accuracy was an important conceptual advance. Bronfenbrenner, Harding & Gallwey (1958) arrived at the same conceptual distinction with the concepts sen- sitivity to the generalized other" and "sensitivity to individual differences" or "interpersonal sensitivity." While the distinction is not crucial to the present metho- dologY. it is again referred to in the discussion of clinical studies. A number of inferences may be made from the research cited thus far. Research suggests that while psychologists ‘might not do well on tasks that require a knowledge of certain stereotypes, the ability to detect and judge indi- vidual differences may be an important clinical skill. It may be that clinicians, because of their psychological training or the nature of their work, hold a set of different stereotypes from those whom they judge in these studies. A broad conclusion from these studies is that the nature of the judgment task seems to be a fundamental consideration. A task may vary, for instance, on the relative weight Of stereotypic and discriminate accuracy as well as level of complexity and information available. It would seem impor- tant in person perception studies when assessing a certain skill to match the prOper experimental task, or similarly, in using a certain task to understand what kind Of skill may be measured. Fancher (1967) made an important methodological dis- tinction regarding the nature Of the judgement task and the correSponding abilities required of judges. He made the distinction between the ability to make behavioral predic- tions which many, if not most of the previous research investigated, and the ability to formulate a valid conceptu- alization Of personality. These abilities, first distin- guished by Wallins (1941), while appearing to be highly related in that both require an "understanding" of a person, called for qualitatively different cognitive Operations. 7 According to these authors, the prediction of individual behavior appears to involve an empathic, "nonanalytic" ap- proach to case material while the formulation of personality descriptions or conceptualizations appears to call for an inferential "analytic" approach. Fancher considered the latter ability more relevant to clinical psychology than the former, as in "insight psychotherapy" which appears to call primarily for the conceptualization and clarification of personality data, rather than for explicit predictions. In his study, predictive ability was Operationalized as the ability to predict true life events from case history material under a multiple-choice format. Valid conceptualization was Operationalized as the ability to write a conceptualization of personality on a case from which another group of judges could use as an aid in predicting life events of the case material. In other words, the validity of the conceptuali- zations was assessed by how well other judges could do using such information. Fancher hypothesized and found a negative correlation between these two abilities in undergraduate students. In a further post hoc analysis, Fancher found that these students' course grades in abnormal psychology correlated negatively though insignificantly to predictive ability (3.- -.31, p_<.15) but positively and significantly ‘with valid conceptualizing ability (g'- .40, p¢<.05). Fancher concluded that clinicians may yet demonstrate that their greater competence lies in the conceptualization of person- ality. A possible extrapolation from.such a finding would be that the better analytic or conceptual thinking of a 8 therapist, the better service he or she may be able to pro- vide a client. While Fancher's study appears easy to assimilate in that nonanalytic predictive ability appears to be highly related to stereotypic accuracy or sensitivity to the gen- eralized other, and analytic conceptual ability appears to be highly related to differential accuracy or sensitivity to individual differences, other findings in his study sug- gest that such relationships are by no means isomorphic. FOr instance, the undergraduate judges‘ predictive ability was positively correlated with an emphasis on the subjective frames of reference of the individuals in the presented cases and with a tendency to use Objective and nonevaluative con- structs on the Kelly Rep Test. Judges' valid conceptual ability was positively related to a tendency to be evalua- tive and nonobjective on the Rep Test with an implication that they treated the presented cases as Objects to be evaluated and categorized. The latter finding is a trend similar to Abeles‘ (1967) finding that the higher discrimin- ative ability of the judges, the more evaluative they tended to be. Abeles' study had the additional finding that these judges also had lower ratings of "liking" of the clients they rated. The possible inverse relationship between analytic conceptual ability and the capacity to like others and to be empathic suggests a perspective contrary to Fancher's suggestion that clinicians may yet demonstrate that their greatest competence lies in conceptualization: at least as 9 far as the qualities of a therapist may be concerned. Fancher makes the assumption in develOping the implications of his study for clinical psychology that the higher a student's course grade in abnormal psychology, the more likely that person is to have the capacity to be a thera- pist. However, the importance of a therapist's capacity to conceptualize, as well as what social or psychological dimensions conceptualizations would be advantageous, is uncertain. Yet, Fancher's study does provide another ex- planatory hypothesis for clinicians"weak performance on traditional person perception tasks and suggests that further research on conceptual ability, eSpecially in a clinical setting, may be worthwhile. Judgment Process -- Person Factors Estes (1938) initially Observed that one significant variable in the judges' accuracy of perception of subjects was a subject's ability to be judged. Gottheil, Exline, and Winkelmayer (1979) found this to be the case, with normal peOple being the easiest to judge. These authors concluded from their earlier research that there is less congruence between the verbal and nonverbal expressions of schizophrenic individuals than of normal subjects and that schizophrenics' nonverbal emotional communications were harder to identify than were those of normals' (Gottheil et a1, 1979). It could be expected in a person perception study that the capacity to be judged accurately would emerge as a significant factor. 10 The ability to judge peOple could also be expected to emerge as a significant factor in person perception research (Taft, 1955; Cline and Richards, 1962). Furthermore, the clinician's cognitive-perceptual style would be eXpected to have a certain overall bias across clients (Grosz and Grossman, 1968). The differences between groups of judges such as physical scientists versus psychologists indicates such biases in the rating of the individual judges within each group. Katz, Cole and Lowery (1969), found that diag- nostic disagreements among clinicians appeared partially to be based upon actual differences in their perceptions of certain kinds of pathology rather than being solely due to semantic differences. Katz et al (1969) found group differ- ences between American and British clinicians in their interpretation of the level of affect, particularly apathy, and in their threshhold for perceiving and reporting dis- tortions in the perception of reality in patients when making diagnoses such as psychotic, psychoneurotic, or personality disorder conditions. Relevant Social Psychological Research Before reviewing pertinent clinical studies, some re- cent social psychological research appears highly illuminating and relevant to this area. The research on field dependence- independence advanced by Witkin, Dyk, Paterson, Goodenough, and Karip (1962-1974), as reported in a research study by Widiger, Knudson, Rorer, (1980), while essentially appearing quite independent of person perception research, is 11 strikingly similar to suggesting that the variables under investigation have some fundamental nature to them. The basic dimension underlying the phenomena of field dependence- independence is an analytic versus global one (Widiger et al, 1980). While originally conceptualized as cognitive "styles," there is good evidence to consider these variables, analytic and global, as well as the corresponding variables of field independence and field dependence as "abilities" (Widiger et a1, 1980). Analytic ability refers to the tendency to break.stimuli into parts: to perceive and concentrate on the details. The global ability is defined as the absence of the tendency to either analyze wholes into parts or_syn- thesize parts into wholes (Widiger et a1, 1980). There is also an interesting stylistic interpretation of field dependence-independence (FDI). The FBI dimension is characterized as the tendency to rely on internal vs. external referents. An internal frame of reference is supposed to result in a greater ability on cognitive tasks, since many such tasks require some form of cognitive re- structuring that theoretically is based on an internal frame of reference (Widiger et a1, 1980). An external frame of reference, on the other hand, results in social effective- ness because such a style tends to make one attuned to social (i.e., external areas (Widiger et a1, 1980). The global style is conceptualized as an ability, or rather a lack of ability (cognitive ineffectuality), and a resulting need to be attuned to others for their help and support. A prediction task in which a judge is asked to put 12 himself in another's frame of reference is in one sense a cognitive task, but it is essentially a task that involves asking the judge to abandon his own internal frame of refer- ence and adOpt an external one (the subject's). Conceptual- ization, on the other hand, appears primarily to invoke the judge's internal frame of reference. It could be hypothe— sized that On person perception tasks primarily involving prediction, field-dependent persons would likely do well; whereas on conceptualization tasks, field-independent persons may do better. The weaker performance by psychologists on certain kinds of prediction tasks, eSpecially those involv- ing stereotypic accuracy may be due to psychologists' de- velOping an internal frame of reference on psychological and social dimensions (as, for instance, from theory and know- ledge) and they may be less likely or capable to adOpt a subject's frame of reference under the conditions imposed by a research analogue studies (e.g., brief presentation of subject or subject stimuli). Clinical Studies While the previous studies were primarily concerned with making comparisons between clinicians and other groups on person perception tasks, this section reviews studies which made within group comparisons among mental health pro- fessionals. First reviewed are the more discouraging findings which, in general, were earlier studies and then the larger group, mostly later Studies, of more encouraging findings. 13 The emergence and influence of client-centered and humanistic psychotherapy deemphasized the importance of the therapist to be a good judge of a client. This was the case at least, as far as the therapist having to be concerned or to judge the client's personality, symptomatology, or psychodynamic functioning. Humanistic psychotherapy fo- cused more on such qualities of the therapist as empathy and genuineness. It questioned any need for the therapist to compartmentalize the client either on nosological or personality dimensions. The influence of this viewpoint can be detected in Meehl's (1960) questionnaire survey. He found that only 17% of the 168 polled psychotherapists of various persuasions, believed that it greatly Speeded ther- apy if the therapist has prior knowledge of the client's dynamics and content from such devices as the Rorschach and TAT, and that 43% believed that warmth and real sympathy are much more important than an accurate casual understand- ing of the client's difficulty. Meehl in summarizing the reaponse style to other questions stated that "in the aggre- gate suggest minimization of the importance of the therapist forming a 'correct' picture of the client's psyche" (p. 20). While illustrative of some tendencies in psychotherapy thought in that era and perhaps also currently, Meehl's survey did not answer the question of, would it help or facilitate therapy the more the therapist was capable of accurately perceiving certain aspects of client functioning? Holt and Luborsky's (1958) study is directly related to this question and found essentially nonpositive results. These 14 authors were studying the personality characteristics of psychiatric residents. One variable investigated called "perceptual sensitivity" was defined as the "ability to be observant, to pick up subtle, yet psychiatrically important aSpects of the perceptual field" (p. 145). Holt and Luborsky found that while this skill characterized a few excellent residents, fair to poor perceptual sensitivity was so well diatributed throughout the range of talent that it did not distinguish acceptable from unacceptable candidates. The authors concluded that "the kinds of sen- sitivity that are most relevant to psychiatric diagnosis and treatment (often spoken of as observational ability, social sensitivity, or even intuition) are probably taken care of adequately by assessment of psychological-mindedness and empathy" (p. 323). Holt and Luborsky's (1958) conclusion certainly places a heavy sentence on the importance of psychiatric judging ability. This conclusion can be criticized because it was not based on a process and outcome study but only on the broad category of acceptable versus unacceptable psychiatric residency capacity. Furthermore, and this criticism applies to studies with positive findings as well, when one uses supervisor or peer ratings of therapist capacity, until this is cleared up by research into this area, one has to assume that such ratings are based on rating both field-dependent and field-independent persons well. In other words, a therapist could be rated well, not because of any particular therapeutic excellence but perhaps because of being liked 15 by the rater because he or she is a field-dependent person and may be very pleasing to the rater. A study should compare a therapist's analytic ability for its utility to the outcome (and process) of psychotherapy itself. Then whether analytic or conceptual ability augments the thera- peutic process or inhibits it because of itself or its relation to other qualities in the therapist can be better assessed. Abeles' (1967) study is interesting in this regard. He found a negative relationship between therapists' ratings of liking of clients presented in typescripted interviews and their scores on a form appropriateness dimension on the Holtzman Inkblot Test. This suggested that those thera- pists who are less able to discriminate correct form possess a higher degree of the capacity to like clients. Abeles prOposed that, "the pOpular notion that being accurate in judging others makes one effective in interpersonal relations needs to be reexamined" (p. 21). While Abeles' (1967) study appears to make discrimina- tory accuracy questionable for psychotherapists since there appears to be a correSponding deficit in therapists' capa- city to like others, this study is also readily incorporated into the preceding theoretical discussion. From.field dependence-independence theory, one would predict that therapists who are less able to discriminate correct form would Show a greater need or capacity to like others, pre- cisely the finding in Abeles' study. While this is admit- tingly an ad hoc explanation for the reported findings, it 16 is intriguing that the study's finding fit field dependence- independence theory so well. A follow-up research question could involve the direct assessment of therapists' discrimr inatory accuracy on the process and outcome of psychotherapy. The results of Abeles' study and related works to be re- viewed suggest that therapists' judging ability may impact on psychotherapy process. There are a number of clinical studies which have examined the ability among mental health professionals themselves to judge peOple and which have found encouraging findings. Gottheil, Exline and'Winkelmayer's (1979) study seems to have made an advance in this area with the distinc- tion between the ability to judge normal peOple and the ability to judge psychiatric patients. These authors questioned the early person perception findings which indi- cated that better judges of human behavior were more social- 1y detached and less psychologically oriented or trained. Gottheil and his associates believed that view neither reflected common sense nor did it relate with the experience and day-to-day observations of administrators and clinicians in mental health facilities. These authors reasoned that a possible explanation for the difference between the research findings and practical experience may be that the judging of the emotional messages of psychiatric patients is a different type of task, requiring different characteristics of the judges, than the more explicit behaviors of non- psychiatrics subjects. The authors also invoked the dis- tinction between stereotypic accuracy and differential 17 accuracy to suggest that judging the behavior of normals ‘may involve the former ability whereas differentiating and responding to the emotional messages of psychiatric patients such as schizophrenic subjects, may call for differential accuracy or sensitivity to individual differences. The authors found that student nurses' ability to relate to their patients, as judged by their nursing supervisors, was associated with the ability to accurately identify the non- verbal emotional communications of the schizOphrenic subjects but not of normal subjects as were presented on silent films. These authors' original hypothesis was also support- ed: accuracy in judging the emotional content of fihms of normal subjects was associated with social involvement and nursing interests. In other studies, Reid and Snyder (1947) found a .70 Correlation between a professor's ratings of students in his nondirective counseling course and the students' accuracy in recognizing client feelings. Regan et a1 (1967) found that a therapist's ability to perceive and identify the affective states in others has been shown to relate to ratings of sensitivity and/or counselor effectiveness as judged by supervisor and peer ratings (Rurtz and Grummon, 1972). Genthner and Saccuzzo (1977) found that high func- tioning psychologists compared to low functioning psycholo- gists as measured by the Carkhuff level of facilitation scale (LOP) were more accurate in their perceptions of the self-reported feelings and concerns of clients and therapist whom they observed in psychotherapy. Similarly, Cannon and 18 Carkhuff (1969) found that both level of functioning and experience in helping increased accuracy of discriminations of interpersonal processes such as empathy, respect, gen- uineness, and confrontations, and that level of functioning was more important than experience. In a less related study, Watley (1967) found that those school counselors rated most effective, as compared to those rated least or moderately effective predicted significantly better whether college students would graduate and would keep their major program selected at the time of admission. Cartwright and Lerner (1963) used an accuracy of per- ception task to compare the therapists’ empathic under- standing of their patients to therapists' ratings of outcome (improved vs. unimproved). The patients, after filling out the Kelly Role Construct Repertory test, were asked to describe themselves on ten discrete constructs that they had supplied. The therapists were asked to complete the same measure "as the patient sees himself" for their re- Spective patients after the second interview and final interview. The comparison of these measures, called "ther- apists empathic understanding," did not show a significant difference between improved and unimproved cases at the earlier interview. However, at the close of therapy, the therapists understood the self-image of the improved patients significantly better than those of the unimproved. A post hoc analysis discounted a possible alternative explanation that the patients had merely adopted their therapists' views over the course of treatment. These 19 findings suggest a process of improving accuracy of thera— pist perception in successful cases and progressive mis- perception in poorer outcomes over the course of treatment (Orlinsky & Howard, 1978). Psychoanalytic theory stresses the importance of the therapist to be an accurate perceiver or judge of his or her patient. Analytic candidates undergo their own analysis because this process of self-awareness is supposed to increase the candidate's perceptual and interpersonal sensitivity and objectivity (Cutler et a1, 1958). Schrier (1953), within a psychoanalytic framework, conducted a study on the relationship between psychothera- peutic process and outcome and the agreement or "congruence of perception" between the therapist and client. This focused on comparing the client's self-ratings and their therapist's ratings of them. The major difference between this and the variable central to the preceding study is that in Schrier's (1953) study, the therapist's rating of the client was based upon the therapist's own point of view. The congruence measure was comprised of twenty-two "person- ality" variables (e.g., achievement, aggression, abasement, blame avoidance, etc.). Schrier named this "syntonic identification" referring to the "syntony" (agreement) be- tween two peOple. Schrier considered this type of identification related to the Sharing of attitudes, distinguished from classical identification entailing the sharing of personality traits and characteristics. Schrier Operationalized classical 20 identification by comparing the self-ratings of the therapist and client. Schrier ranked the patients according to the magnitude of the correlation between the clients self-ratings and therapists' ratings of them (syntonic identification) at the end of therapy and found that this variable correlat- ed positively and significantly with ratings of therapeutic success and positive rapport as rated by the therapist and external judges as well as correlating significantly with this classical identification measure. However, from.the knowledge of Cronbach's (1955) methodology critique of person perception research, the point may be made that Schrier's classical identification index could be considered a measure of client-therapist similarity, and this alone could have been responsible for the findings. The reviewed literature suggests that variables such as "accuracy of perception" or "congruence of perception" may be worthwhile variables to investigate in psychotherapy research. Admittedly, all of the studies reported, if taken at face value, do not fit neatly together. Complex relation- ships appear to hold among such variables as this kind of task, nature of the judgment, length of relationship, subject characteristics, and Observer abilities. These variables and others would be worth investigating in a large social psychological study, perhaps using a systems approach. However, at this point there seems sufficient evidence to support further exploration of these variables in clinical settings. 21 The reviewed literature suggests that clinicians' _ ability to detect and understand certain states or disposi- tions, especially affective states in persons whom they have endeavored to treat psychotherapeutically, may well be an important clinical skill. Such a skill could be con- ceived to be of benefit to the client in at least two ways. A capacity to understand certain aspects of client function- ing may be a kind of interpersonal sensitivity or relatedness that influences or enhances the quality of the therapeutic relationship. And such a capacity to understand the client insofar as being able to provide relevant conceptualizations or to identify aSpects of the client's functioning or psychological states may in itself be therapeutic. The notion that it is important in a therapist-client relationship for the therapist to have an interpersonal sensitivity to the client in a way that reflects a "congru- ence" of thought receives support from a research study on premature termination of therapy by clients. Proctor and Duehn (cited in Garfield, 1978) compared clients who ter- - minated after the first interview with clients who continued for additional interviews. Stimulus-response congruence (whether the clinician's verbal reSponse acknowledged the content of the patient's preceding communication) and content-congruence (the clinician's verbal statements being consistent with the patient's expectations concerning what ‘was to be discussed) were the variables investigated. It was reported that the therapists were significantly more incongruent on both measures with the group of early 22 terminators. These findings are intriguing although the Proctor and Duehn's (1978) study is, for the most part, tangential to the line of thought being develOped for the present investi- gation. In it and in other studies of therapist-client similarity, such as on SES indices, it is certainly diffi- cult to separate values, expectations, therapist-client compatibility and the therapist's interpersonal sensitivity to the client. But one wonders whether differences in these variables between the therapist and client are re- flected or made apparent in a common way regarding the interpersonal relationship between the two. In other words, do substantial differences of these kinds between profes- sional and client result in a relationship in which the therapist is deficiently attuned to the client's individual characteristics or differences in a way that contributes to inferior results? Proctor and Duehn's findings, although related to premature termination and not the quality of the therapeutic relationship, suggest that, at least on a verbal interactive level, certain kinds of therapist sensi- tivity or attunement to the client is important. Methodological Considerations At this point, it seems useful to offer a preliminary and partial sketch of the present study's design. This study focuses on the agreement or congruence of ratings of client symptomatology between either client and intake worker or client and therapist. By focusing on the client's 23 symptoms which are an integral part of why the client seeks help, the conduct of therapy, and the termination of therapy, based on the literature reviewed thus far, it can be asked that while it seems important for the therapist to have an accurate understanding of client affective states, is is important for the therapist to perceive and understand symptomatology? In other words, this study considers variations in the nature of what is to be understood and perceived, an important dimension to vary in person percep- tion tasks (Oskamp, 1962). This study also uses for comparison, client self- ratings with therapist's ratings of them. This type of comparison was used in Schrier's (1955) study in which it was called "syntonic identification" and in the Derogatis et a1 (1976) study to be reported. While this focus seems a legitimate one in a psychotherapy outcome study, there are additional reasons for contending that such comparison is a preferred way to proceed in doing a clinical person perception study. Using the therapist's rating from his or her own point of view establishes the therapist to be free to express him/herself as compared to asking the therapist to reapond as he or she thinks the client will reapond. The former task appears to primarily involve conceptualiza- tion whereas the latter task appears to primarily involve prediction. While studies of predictive ability in a clinical setting seem worthwhile, this study focuses on the seemingly more important variable: the therapist as conceptualizer of client symptomatology. 24 This study uses the client's self-ratings in contrast to external ratings of clients or possibly using scores derived from projective or objective tests. In and of it- self, it seems reasonable to compare therapist's perceptions with client's perceptions, but it also seems reasonable that this may be the best way to proceed. Fer one, research has established that a person's self-report is often the most accurate and best predictor of how a person will behave (Gottman &'Markman, 1978; Mischel, 1972). Giedt (1958) in interviewing research, found that "ratings of intelligence, anxiety, dependency, and, to some extent, direction of feel- ings were usually accurate when based directly on what the patient said as to his feelings, preferences, or past be- havior." When the observers tried to use more devious interpretations or to infer from direct behavior, they tended to "err" (p. 405-06). Kurtz and Grummon (1972) found that client-perceived empathy linked more strongly to outcome than did any other kind of empathy measure. It seems clear that client self-perceptions about process and outcome should be an integral, perhaps primary part of psychotherapy research, especially when an accuracy of perception construct is being studied. It may be the case that clients with more severe psychOpathology such as psycho- tocism.do not have a realistic graSp of their own symptoms which makes self-reports somewhat unreliable. However, the present investigation is designed to monitor and investigate this hypothesis by the use of self-ratings of clients with various levels of psychOpathology together with the 25 therapists' ratings of them.and the congruence of perception measure . Physician-Patient Congruence Part of this study is designed to parallel Derogatis, Abeloff and McBeth's (1976) study which investigated physi- cian-patient agreement in the perception of psychological symptoms. These authors mainly focused on the doctor's perceptions of different symptoms to overall ratings by doctors of their patient's distress level for the cancer patients whom they were treating. Unfortunately, these authors did not compare the congruence of perception measure to any other kinds of patient ratings such as adjustment or degree of satisfaction with medical services or attending physician. The present study is interested in outcome measures . Derogatis et a1 (1976) reported that physicians' ratings of patients' global distress minimally correlated with ratings of depression by patients. As global distress ratings rose, discrepancies between patients' and doctors' perceptions of patients' symptoms increased. The more highly distressed patients characteristically reported substantially greater levels of depression than did their treating physician. The present investigation intends to report on the characteristic response tendencies of those psychologically trained and their clientele in the percep- tion of client symptomatology. HYPOTHESES The hypotheses in this study concern: 1) the relation- ship of congruence of perception to outcome; 2) whether con- gruence of perception is a significant variable in early termination; 3) whether there is increasing congruence of perception over time; 4) whether intake workers can be dis- criminated on the capacity to perceive accurately; 5) whether accuracy of perception is a therapist variable; 6) whether clients can be discriminated on their capacity to be per- ceived accurately; and 7) whether there is a difference in congruence of perception across the symptom dimensions and a global distress index. In addition to these hypotheses, results of exploratory hypotheses will be reported. Hypothesis 1: At termination, the therapist-client congruence of perception measures will be related to questions evaluating the therapy on posttherapy client and therapist questionnaires. Specifically, it is predicted that more congruent ratings between therapist and client on the client's symptoma- tic status will be correlated positively, at a significant level, with higher ratings related to therapeutic success. The posttherapy client questions are: l) How'much have you benefitted from your therapy? 2) Everything considered, how satisfied are you with the results of your psychotherapy experience? 3) What impression did you have of your therapist's level of eXperience? , 26 27 4) How much do you feel you have changed as a result of therapy? 5) 0n the whole, how well do you feel you are getting along now? 6) To what extent have your complaints or symptoms that brought you to therapy changed as a result of treatment? The posttherapy therapist questions are: 1) Compare the client with other clients whom you see in psychotherapy. a. defensiveness b. capacity for insight c. degree of symptomatic improvement d. overall success of therapy 2) How would you characterize your working relationship with this patient? 3) How satisfied do you think the patient was with the results of his therapy? Two congruence of perception measures are used. One is an average of the sum of the absolute value of the scaled T-score points discrepancies between client and therapist on the 10 dimensions divided by the number of dimensions. The second measure is the correlation of ther- apist and client ratings across the ten scales. If it be- comes apparent that the two measures are nearly synonymous or that one of the measures is clearly superior, then only one of the measures will be used. Hypothesis II: At intake, there will be significant differences on the congruence of perception measures between a com- bined group of clients defined as early terminators and a group of clients who had three or more therapy sessions after intake. 28 Hypothesis III: For those clients who had the same psychologist as intake worker and later as therapist, there will be more congruence at posttherapy than at intake. Hypothesis IV: At intake, there will be significant differences in intake workers' capacity to perceive clients' symptoms in con- gruence with clients' self-report. Hypothesis V: At termination, there will be signifi- cant differences in therapists' capacity to perceive clients' symptoms in con- gruence with clients' self-report. Hypothesis VI: Clients' congruence scores at intake and at posttherapy will be positively correlated. Hypothesis VII: There will be no significant differences in clinician-client agreement on client symptoms across the nine symptom.dimen- sions and a global distress index. This last hypothesis replicates part of Derogatis et al's (1976) study using therapists and clients instead of physi- cians and their patients. Additionally, most of his study will be replicated using therapists and clients and the find- ings will be reported. 29 Before concluding this section, it seems desirable to give a more Specific sense of how each of the hypotheses is to be investigated or Operationalized. In hypothesis one, a comparison of the congruence of perception measures (explained in the next section) is made separately with each of the posttherapy therapist and client questions. The reSponses to these questions typically range from an "extremely dissatisfied" reaponse to an "extremely satisfied" response. The procedure for testing this hypo- thesis is to obtain correlations between the congruence of perception measures and the reSponses to each of the ques- tions for all clients who have had at least one therapy session and who have completed their posttherapy forms. ' Fer hypothesis two, early terminators and continuers (later defined) are compared on the intake congruence of perception measures. The comparison used is a t-test between the mean congruence scores for each of the groups. For hypothesis three, a comparison is made on the con- gruence of perception measures for a group of clients at intake and at posttherapy. Testing this hypothesis requires clients who have had the same person as both intake worker and therapist. A t-test is used comparing the mean congru- ence measures at intake with those at termination. Fer hypothesis four, the concern is to test for a trait-like capacity for intake workers to perceive symptoma- tology accurately (in congruence). Hypothesis five involves a similar test for therapists using data at termination. For hypothesis six, a trait-like capacity, to be perceived 30 accurately (congruently) is tested for among clients. Hypo- thesis five is considered a therapist variable in psycho- therapy and hypothesis six is considered a client variable. These variables are considered in their own right but these hypotheses have been designed to be used in relation to the first hypothesis. For instance, if congruence of perception emerges as a significant factor to the outcome of psycho- therapy, then it could be asked if it is primarily a thera- pist variable (hypothesis V), a client variable (hypothesis VI), or an interpersonal variable (hypothesis I). Hypothesis four is tested by an analysis of variance among intake workers' mean congruence scores. The average of each intake worker's congruence scores for three or more clients is used. Hypothesis five involves the same analysis but this time using therapists' maan congruence scores. The average of each therapist's congruence scores at termination for three or more clients is used. Hypothesis six is tested by obtaining a correlation between a client's congruence of perception score at intake and his or her score at termination. Finally, hypothesis seven is concerned with comparing congruence of perception among the ten scales. The congru- ence of perception measures used to study the previous hypo- theses were summed over all ten scales. Here, each scale is taken separately and viewed in relation to the other scales. A.mean congruence of perception measure is taken for each scale including both intake and termination data. This will provide data for an analysis of variance among the scales. This may permit identifying scales for which there is either a significantly high or low congruence of perception. METHOD 2555 This study uses data collected over the past two years (September, 1978 through June, 1980) for research purposes at the Michigan State Psychological Clinic, an outpatient clinic serving non-student members of the Michigan State University community and surroundings. The clinic is a training and research agency of the Department of Psychology and serves as a low cost clinic to adults, children, and families. The original data collection was based on the premise of obtaining relatively nonintrusive information on clients coming to the clinic. The purpose of the data col- lection was to provide later information on the process and outcome of therapy. This study and its hypotheses were designed after collection of the data was complete. This accounts for the variability of the number of subjects available for the testing of the different hypotheses. While the clients and therapists were aware that the forms they completed would be used for research purposes, it seems most unlikely that they. completed the forms with a knowledge of the possible way these forms were utilized in this study. The data for this study were taken at intake and at posttherapy. There were 114 intakes and 50 terminations from which a congruence measure was derived. For some hypotheses, these data were pooled together (§_- 164). In addition, there were 19 posttherapy therapist ratings on the SCL-9OR for which no posttherapy client ratings were available. This 31 32 provided the §,of 183 for one of the data analyses. The congruence measure required both therapist and client ratings on the SCL-90R. As suggested above, out of 69 terminations, only 50 cases were usable since 19 clients, for unknown reasons, did not fill out posttherapy forms. The possible effects on the present study of this condition are unknown. Clients A total of 115 adult clients (38 males and 77 females) who received an intake interview at the clinic, agreed to partake in the research and completed the forms correctly. Their mean age was 27.5 (range - 14-59), with a standard deviation of 8.4. Thirty-eight clients were defined as terminators " two "no shows" including twenty-two listed as "no therapy, for first therapy session, one cancellation, and thirteen who had two or less postintake therapy sessions. The "no therapy" group was comprised of persons who declined the offer of therapy or were not offered therapy, perhaps for a variety of reasons which are not listed in the files. This group was included with terminators, and a test will be conducted to ascertain if their data differed signifi- cantly from the other terminators. Sixty-eight clients, grouped as "continuers,' continued for three or more ses- sions after intake. 33 Therapists The therapist group was composed of all clinicians at the Clinic who consented to take part in the Clinic's re- search. They included the full range of MSU Psychological Clinic professional staff, from beginning practicum students to highly experienced PhD clinical psychologists. For hypo- thesis V, which compared the differential capacity of ther- apists to perceive client symptomatology, the group was composed of eight clinicians who had more than two termina- tions on file (range 3-5). Otherwise, data from termination was taken from clinicians who had at least one termination on file. Intake workers are mainly advanced clinical psychology PhD candidates serving an internship at the clinic. For hypothesis four, which compared the differential capacity of intake workers to perceive client symptomatology, the group ‘was composed of nine clinicians who had completed more than two intakes on file (range 4-23). Otherwise, data from intake was taken from clinicians who had at least one intake on file. Instruments A. Symptom Checklists (SCL-90R, Derogatis, 1977). (Appendices A and B). 1. Client form--(SCL-90R). This measure consists of 90 statements of problems. The problems comprise and load 9 symptom dimensions and a global severity index. The symptom dimensions are somatization, obsessive-compulsiveness, 34 interpersonal sensitivity, depression, anxiety, hostility, phobic anxiety, paranoid ideation, and psychoticism. Clients were instructed to check those statements that were current problems for them and to rate the degree of distress (0-4) associated with each problem. 2. Therapist form--This form (SOL-90A) consists of 9 symptom dimensions and one global pathology index. Clinicians were asked to rate the degree of symptomatology present (0-6) on each of the 10 scales (computerized version of the SOL-90A used by the clinic). B. Outcome Measures (Appendices C and D). 1. Client form--A 56-item.client form (Strupp, Lessler & Fox, 1969, shortened version) was given to clients at the termination of therapy. This form tapped the client's subjective beliefs about the effectiveness of their therapy. Six questions which appeared the best representative of relevant process and outcome variables were selected for use in this study. 2. Therapist form--A 33-item.therapist form in- cludes ten questions on the SCL-90A and 23 questions relating to the therapist's subjective belief about the effectiveness of therapy (Strupp et a1, 1969, shortened version). Six questions which appeared the best representative of relevant process and outcome variables were selected for use in this study. C. Congruence of Perception Measures. Clients' responses to the SCL-90R items were converted 35 to the degree of present symptomatology for nine symptom dimensions and a global severity index (Derogatis 1977). These scores were then converted to a T-score distribution based on norms from Derogatis (1977). Clinicians rated the dimensions directly and their scores were converted to a T-score distribution based on norms derived from the data in this study (Appendix E). Two congruence measures were derived from.these ratings by clients and clinicians. The first one was an average of the absolute value T-score points discrepancies between client and therapist across the 10 scales. The second was a correlation of client and clinician ratings across these ten scales. Both measures would be used in the study if they had good statistical prOperties. However, if one or both of the measures appeared deficient, then the better of the two measures would be used. Procedure Adults who requested therapy at the clinic received an intake interview. At this time, the clients who agreed to partake in the clinic's research, completed the SOL-90R among other questionnaires. Intake workers independently completed the SCI-90A after the intake. Intake workers were also required to write a case description and the case then became available for distribution. Clients were then selec- ted by therapists usually on an availability basis, and in consultation with supervisors. Some clients were assigned for therapy to their intake workers, but most were assigned 36 to other therapists. At termination, clients and therapists completed posttherapy questionnaires which included their versions of the SCL-90R. These research forms were coded and filed, and neither clients nor therapists had access to these completed forms. RESULTS* Preliminary Comments Before presenting the results, several methodological notes seem useful: A) Change from.two congruence measures to one: The original intent in this study was to use two congruence of. perception measures. It was hoped that the two measures would compliment each other if they measured different kinds of accuracy or congruence. These measures were designed in the planning stages of this study. It was not known whether these measures would be practical or reliable. If one or both of these measures had poor statistical or practical prOperties, the author's intent was to use the better of the two measures. The first congruence measure consisted of the average of the absolute value T-score points discrepancies between client and therapist across the 10 scales. This measure appeared to have good statistical prOperties with a mean for males of 9.20 (Sgg; - 3.87) versus a mean for females of 9.60 (Sggg - 4.37). The second congruence measure was a correlation between client and clinician ratings across the ten scales. This measure appeared to have poOr statistical properties with a mean correlation for males of .25 (§;2;,- .34) versus a mean correlation for females of .14 (S.D. - .36). FOr this second measure, the standard deviation was * Statistics were considered significant if they reached the p 5 .05 level by the two-tailed test. 37 38 considerably larger than the mean score which seems indica- tive of considerable instability of the measure. In other words, for example, for a female client this score could have a one standard deviation range of ISO to :22 which was too large a range to consider these scores as measuring any- thing meaningful or practical. Therefore, it was decided to use the one congruence measure. B) Interpretation of congruence measure: The congru- ence measure, while basically mathematical in origin, seems easily understood. A score of 9.20, for instance, represents an average of 9.20 T-score points discrepancy between client and clinician ratings of clients' degree of symptomatology across the ten symptom dimensions. Since this number repre- sents the absolute value discrepancies, it is a sum of clinicians overratings and underratings of client in relation to client self-report. C) Pooling of male and female data: The overall means for male and female clients on the congruence measure were not significant statistically from each other. T-tests were also conducted comparing the mean scores for males and fe- males on each of the ten scales and again no differences proved significant statistically. Therefore, it was decided to pool these data which had been intended originally. Hypothesis I At termination, the therapist-client congruence of per- ception measure on symptom dimensions will be related to questions evaluating the therapy from.posttherapy client and 39 therapist questionnaires. The results are shown ianable 1. Ten of the twelve questions correlated significantly with the congruence measure, all in the expected direction. The results indicate a moderate relationship between the congruence of perception measure at termination with evaluations of the therapy by therapists and clients. Thus, the greater their agreement about the clients' symptoma, the higher the ratings of ther- apeutic success. Specifically, congruence was related positively (p <.05 or above) in order of magnitude to: l) therapists' ratings of clients' degree of symptomatic improvement; 2) clients' ratings of satisfaction with the therapy; 3) therapists' ratings of overall success of the therapy; 4) therapists' ratings of expectation of client satisfaction; 5) clients' ratings of improvement of complaints or symptoms from.therapy; 6) therapists' ratings of the working relation- ship with the client; 7) clients' ratings of benefit from therapy; 8) clients' ratings of change from.therapy; 9) therapists' ratings of clients' capacity for insight; and 10) clients' ratings of therapists' level of experience. The congruence measure did not correlate at the .05 level of statistical significance to therapists' ratings of clients' defensiveness or to clients' ratings of how well they are presently getting along. 40 So. vars. Jo. v m3 “3G .9. nH.n «an. «eon. sewe. arson serum cc. «mu. Son. seam. «see. kkkan. Amrmmvaoauoaouuou an ac cm a: an a: ac on we on me on 2 neoco>waoomoa usages“ wow huuooeoo «uaoauom away and: Oununouuoaou mauxuoz HOOA onawouoouoao so» capo: so: smashesu mo «nausea osu new: nos ucoauom use sedan :Oh on ooamewuoa so: Leonora mo euoooso HHouo>o uaoao>oueaa ouuoaoOeaho mo seamen ”avocado nonuo ou OOuwuomaoo aw uaoqao «any oauuH oeuom uoanouoaa ~30: MGOHO waquuow one no» Hoom pom on HHo3_so: .oaoaa one no poocoauoexo mo Ho>o~ o.ue«aouonu Room mo o>s£ no» owe coaueoueau uses pzmsuoau mo uaaeou o no Camcoso o>os OOh doom no» on 5055 so: sumowoau scum vouuumo:oo sou o>s£ 50:5.302 eusoauoowu mo unseen o no OOwONMO anemone on 90% unwooun usnu oEOOe a no oquOHmaoo noon o>os uaouxo yoga OH peacowwoexo anouonu050hen woo» sown sou one poamouuoe 3o: .oowoowucoo «nuhuo>w 283 e33. 238 .owcquou anouonuueoe .euoweouonu one .Ouaouao one mouooe coaueoouoe mo oocoauwcoo costume ecouuououuoo "H sages 41 Hypothesis 11 At intake, there will be significant differences on the congruence of perception measure between a combined group of clients defined as early terminators and a group of cli- ents who had three or more therapy sessions after intake. The hypothesis was tested using intake data because ear- ly terminators provided sparce posttherapy data. Early ter- minators were a diverse set, including: five clients who had less than two therapy sessions, eight who had one session, two "no shows,‘ one cancellation, and twenty-two persons listed as "no therapy.’ In a preliminary analysis on the congruence measure, the twenty-two person "no therapy" group had a mean of 9.49 (_S_D_ . 4.91), and all others had a mean of 8.82 (S2.- 4.10). This difference did not approach statis- tical significance difference (g(36) - 0.44). The main find- ings, shown in Table 2, indicate no general difference on the congruence measure at intake between early terminators and those clients who continued for three or more therapy sessions. Table 2: Comparison of early terminators and continuers' congruence of perception scores at intake. Group 91 Mean §_l_)_ g 51; Early Terminators 38 9.21 4.54 -.22* 101 Continuers . 65 9.39 3.90 * p (.85 42 Hypothesis III For those clients who had the same psychologist as intake worker and later as therapist, there will be more congruence at posttherapy than at intake. A t-test for dependent means was used to test this hypothesis. The results for this hypothesis, shown in Table 3, indicate that there was significantly more congruence of perception at posttherapy than there was at intake for the group of clients who had the same person as intake worker and therapist. Table 3: Comparison of congruence of perception scores at intake and at posttherapy for clients who had the same person for intake worker and therapist. E Learn a 2:. 9.: Intake 14 11.59 4.91 2.33* 26 Posttherapy 14 8.43 4.62 * Ef(.05 Hypothesis IV At intake, there will be significant differences in intake workers' capacity to perceive client symptoms in con- gruence with clients' self-report. The data are shown in Table 4. As Table 5 indicates, there were no significant differences in intake workers capacity to perceive client symptoms in congruence with clients' self-report. The results did indicate, however, 43 that there was a marked tendency for intake workers to show differences on this variable. This trend would have been significant if a l-tailed test of significance had been utilized. Table 4: Intake workers' mean congruence of perception scores at intake.* Intake WOrker Number ogIntakes Mean SE B** . 4 14.39 7.11 I 5 12.10 7.39 G 9 8.07 2.16 A 10 10.17 5.14 F 11 8.28 2.45 H 11 9.56 4.21 D 13 8.48 3.28 C 21 7.96 2.79 E 23 8.98 3.61 *Limited to all intake workers with more than two intakes on file. **Clinicians research code numbers have been given a letter code. Table 5: Analysis of variance of intake workers' mean congruence of perception scores. £22529. as. .DE 14.8. It Between 220.39 8 27.55 1.85 * Within 1,458.20 98 14.88 Total 1,678.60 106 *pg<.10, (not significant with 2-tailed test.) Perusal of the data in Table 4 suggests that the number of intakes done by the psychologists may affect their mean congruence score. By rank-ordering the intake workers 44 according to the number of intakes and by correlating the order with this mean congruence scores, a rank-order correla- tion of .58 was obtained (pg(.06). This result of better congruence scores with increasing number of intakes suggests that practice in intake interviews, or possibly, amount of experience with the ratings may favorably affect congruence scores . Hypothesis V At termination, there will be significant differences in therapists' capacity to perceive clients' symptoms in congru- ence with clients' self-report. These data are listed in Table 6. The results shown in Table 7 indicate that while there is a slight trend for ther- apists, at posttherapy, to show differences on the congruence of perception measure, no statistically significant findings emerged. Table 6: Therapists' mean congruence of perception at posttherapy.* Therapist Number of Terminations Mean Score SQ C 3 5.09 1.82 D 3 9.63 8.69 H 3 6.23 0.53 K 3 14.75 4.08 A 4 11.62 5.05 J 4 5.43 0.68 B 5 11.08 5.77 E 5 9.22 3.79 *Limited to all therapists with three or more posttherapy congruence scores. 45 Table 7: Analysis of variance of therapists' mean congruence of perception scores at postt erapy. Source SS 2:, M§, 2 Between 286.21 7 38.32 1.83* Within 460.05 22 20.91 Total 728.26 29 #21:.20 Hypothesis VI Clients' congruence scores at intake and at post- therapy will be positively correlated. This hypothesis was designed to determine whether clients' congruence scores across two psychologists, intake worker and therapist are correlated. Results indicate that there is virtually no correlation between the two, 5(32) - .01, ns. Hypothesis VII* There will be no significant differences in clinician- client agreement on client symptoms across the nine symptom dimensions and global distress index. As mentioned, male and female data were pooled. Table 8 shows the mean discrepancy scores for each of the dimensions. A F-test indicates no significant differences in clinician- client agreement across the ten dimensions, {(9, 1590) - .13, 1‘18. *In proceeding with the data analysis for this hypothesis, methodological and statistical problems became apparent which question the value of the results. See discussion section. 46 Table 8: Mean T-Score points discrepancies between clinician and client for each of the ten dimensions* Dimension H53, _§_ S.D. Somatization 162 2.69 12.28 Obsessive-Compulsive 160 - .06 13.75 Interpersonal Sensitivity 162 -1.26 10.43 Depression 164 .80 10.40 Anxiety 162 -2.01 14.61 Hostility 159 .04 12.13 Phobic Anxiety 159 1.49 12.28 Paranoid Ideation 159 .37 12.49 Psychoticism. 159 .59 11.49 Global Severity Index 163 .70 12.00 *For this analysis, the directional value was taken on the discrepancy scores. This enabled determination of whether therapists, on the average, overrated (+) or underrated (-) the clients' symptoms in relation to clients' self-report. **Intake and posttherapy data pooled together. Additional Data Analyses Additional data analyses were conducted for two reasons. One was to undertake a more comprehensive replication of Derogatis et al's (1976) study from which hypothesis seven was derived. The other reason was to provide some additional information about the congruence measure. Table 9 and Table 10 contain data analyses designed from Derogatis et al's (1976) study. The Table 9 data consists of correlations between clinicians' global distress ratings of clients' symptomatology and discrepancy scores between clini- cian and client ratings on each of the symptom dimensions. The results indicate that as clinicians' ratings of clients' global distress rose, there tended to be larger discrepancies between clinicians' and clients' ratings on the symptom dimensions of hostility (p ('05): paranoid ideation (p4<.01) 47 and psychoticism.(pg<.05). As global distress ratings rose, clinicians tended to rate clients higher on these dimensions than clients rated themselves. Table 9: Correlations between clinicians' global distress ratings of clients' symptomatology and discrepancy scores for individual symptom dimensions. Smtom Dimensions _N__ Correlationa Somatization 161 .09 Obsessive-Compulsive 160 .11 Interpersonal Sensitivity 161 -.00 Depression 161 .13 Anxiety 161 .15 Hostility 158 .16* Phobic Anxiety 158 -.O6 Paranoid Ideation 158 .29** Psychoticism 159 .20* *p_<.05, **p_<.Ol aWhen comparing Derogatis et al's results (1976), reverse the positive and negative signs. The data in Table 10 is composed of correlations between clinicians' global distress ratings of clients' symptmmatology and their ratings of clients on the separate symptom scales. The findings indicate high and significant correlations be- tween the clinicians' global distress ratings and eight of the nine symptom scales. Depression, anxiety, and psychoti- cism ratings by clinicians correlated the highest with the global ratings suggesting that these factors weigh somewhat more heavily in the global ratings. Interpersonal sensitiv- ity, obsessive-compulsive, hostility and somatization ratings were also related to global distress ratings. Only phobic anxiety ratings appear to be a non-significant factor in global distress ratings. 48 Table 10: Correlations between clinicians' global distress ratings and their ratings of clients on the separate scales. Symptom Dimension _§_ Correlation Somatization 183 .34* Obsessive-Compulsive 181 .37* Interpersonal Sensitivity 182 .47* Depression 183 .57* Anxiety 182 .52* Hostility 179 .37* Phobic Anxiety 180 .15 Paranoid Ideation 180 .34* Psychoticism. 180 .52* *p_< .001 Finally, in other data investigations, congruence of perception at posttherapy was found uncorrelated with the total number of psychotherapy sessions (5(46) - .09,ns). This finding indicates that sheer amount of psychotherapy contact between therapist and client does not seem to produce better congruence of perception. In another data analysis, correla- tions were obtained between clinicians' global distress ratings and congruence of perception scores (5(163) - .24, pg<.01), and between clients' global severity index and congruence scores (5(164) - .29, p.<.01). These results suggest that as both clients' and clinicians' global distress ratings increased, there was an increasing discrepancy be- tween clinicians' and clients' ratings of client symptoms (Table 9 listed the specific scale discrepancy increases for the clinicians' global distress ratings). DISCUSSION The major focus of this study involved the generation of an Operational definition for a hypothetical construct "congruence of perception," to assess broadly the construct's relationship to factors in psychotherapy. Furthermore, the particular sector selected for appraising congruence was the clients' symptomatic status as Operationalized and measured by the SCL-9OR'S ten dimensions. Overall, favorable relation- ships were found between the congruence measure and selected posttherapy client and therapist evaluations of the outcome of psychotherapy. Integration and interpretation of the findings on this variable is presented in the first part of this section. The second part of this section is a report on another aspect of this study which was the collection and analysis of data on the SOL-90R as used by clinicians and psycho- therapy clients. This part of the study paralleled selected analyses done by Derogatis et a1 (1976) on the SCL-90R, which had been done in a setting using cancer patients and their treating physicians. Some comparisons between these studies are made, but the discussion focuses primarily on the present findings. Egggruence of Perception Finding; A. Thgggpy Outcome. I The congruence of perception measure reflected the degree of agreement between the client's SOL-90R self-report, (scaled on nine symptom.dimensions and a global distress 49 50 index) with a clinician's direct rating of the client on the same scales. This measure correlated positively with post- therapy appraisals of the therapy by both client and therapist. Positive correlations were found between the congruence measure and appraisals of the therapy such as clients and therapists' ratings of the degree of clients' symptomatic improvement, benefit and satisfaction of the therapy, overall change, as well as the therapists' ratings of the working relationship between the two. Therapists' ratings of clients' capacity for insight was also positively correlated with the measure. The correlational nature of this research limits inter- pretation of these findings. The magnitude of the correla- tions between congruence and outcome factors suggests that there may be a direct, or at least closely linked, relation- ship between the two variables. But the possibility that there is another more immediate variable between better congruence and better outcome should not be ruled out. There are a number of reasonable interpretations which may be proposed to account for the congruence-outcome correla- tion. One is that the therapist's sensitivity to a particular client results in better outceme along with better congruence. In this kind of interpretation, another variable is posited in this way such as the interpersonal relationship, therapist- client complimentarity, clients' capacity for insight, clients' capacity to identify accurately and communicate his or her problems, etc. Also, the consideration that better congruence is just a by-product of better outcome is a similar type of interpretation. 51 . Other interpretations could be prOposed as leading to better congruence and better outcome where the latter two variables are conceived to be connected in a more relevant and important way. All of the possibilities mentioned above could be considered. For instance, a good interpersonal relationship leads to a better understanding or congruence on client symptoms, and this may, in turn, lead to better outcome. Of the two sets of interpretations above, the latter set seem more likely. The moderate correlations suggest that congruence of perception on client symptomatology and better outcome appear more directly related than the first line of interpretations suggest. The pattern of correlations between congruence and outcome measures indicated that better congru- ence was correlated more highly with symptom improvement and overall success ratings than it was with other variables such as ratings of the working relationship and clients' capacity for insight. There is another set of interpretations which may be important to consider. In this set, the congruence of per- ception measure is considered to indicate the primary and direct influence on outcome. Fer instance, one may propose that the implicit identification and accurate judgment of clients' symptomatic states by therapists as also perceived by the client, facilitates gain in psychotherapy and relief from these troubling states. In other words, something about the congruence of perception itself may lead directly either to variables such as better outcome via better therapy 52 or increased quality of relationship, or in itself be thera- peutic for relieving clients symptomatic status. The pos- sibility exists that in the course of clients' expressions of problems, therapists who are more attuned to or under- standing of clients' level of symptomatic distress (as ex- perienced by the client), enable better and fuller relief for the client. The writer prefers an interpretation between better congruence and better outcome based on a combination of some of the hypotheses suggested. First, it could be hypothesized that the therapist is capable of being more attuned to a particular client or that there is more compatibility, or cmmplimentarity to begin with or that develops in the course of the therapeutic relationship. From this base, the there- pist is able to gain a better understanding of or has a better capacity to be sensitive to and detect the level and degree of clients' symptomatic status. This may be due to the client being more Open and expressive with this particular therapist. This resulting better congruence may "snowball" in the therapy and contribute to better therapy and outcome; or this better congruence becomes the condition upon which therapeutic change and symptomatic relief are enabled. These results are not free from questions of internal validity. Interpretations could be considered which question the findings and one cannot expect to dispel all possibilities of artifactual influences. For instance, it could be hypothe- sized that given decreasing congruence on symptoms at higher levels of distress particularly on psychoticism, paranoid 53 ideation, and hostility ratings, better or worse congruence at posttherapy just reflects that clients' posttherapy sympto- matic status. This argument can be rebutted by considering an inter- esting nonsignificant finding in this study. The congruence measure did not correlate with how well the clients "thought they are presently getting along" or with therapists' ratings of clients' defensiveness. In fact, the nature of relation- ships between the congruence measure and the posttherapy questions suggest that clients' ratings of "how well they are presently getting along" was unrelated to ratings evaluating improvement and evaluating the psychotherapy itself. Fur the above argument one would have to assume that in relation to other clients, more distressed clients reported less psycho- pathology than expected by therapists' ratings, concurrently with reporting more dissatisfaction with the therapy and less improvement but functioning currently just as well as other clients. In other words, one has to assume that these clients misrepresented or misunderstood all questions relating to their symptomatology but honestly reporting no improvement in symptomatology from the therapy. At the same time, one has to assume that therapists rated them as less improved but not more defensive. While such a situation is not too unlikely, it seems to require more singular reasoning than is warranted. B. Congruence of Perceptions - Person factors. One of the more important questions to ask in psychotherapy research when a variable is found related to outcome, is whether the variable is a therapist or client 54 factor. If it is neither, then perhaps the variable is an interpersonal or situational one. Two hypotheses using different data bases (intake and posttheraPY) were developed to determine whether clinicians differed in their capacity to perceive client symptomatology in congruence with client self-report. At intake, it was found that there was a marked tendency for clinicians to show differences on the congruence variable but no statis- tically significant differences emerged using the two-tailed test (§(8.98) - 1.85, py<.10). Similarly, at posttherapy, findings showed a slight tendency for therapists to show differences on the congruence measure but the findings did not reach statistical significance (E(7.22) - 1.83, 23¢.20). These results, both approaching statistical signifi- cance, are suggestive of a therapist factor in congruence of perception. The possibility exists that the therapist var- iable may have reached statistical significance under differ- ent conditions such as a larger number of scores per clini- cian. To test this hypothesis as a post hoc analysis, clinicians' intake and posttherapy congruence scores were pooled together (Tables 4 and 6 combined). An analysis of variance on this data indicates that the therapist variable emerges as a statistically significant finding (£(9,128) - 2.16, p¢<.05). These findings are suggestive that the capa- city to perceive client symptomatology in congruence with client self-report varies with clinicians. An incidental finding related to intake workers was obtained. It was found that the number of intakes the 55 intake workers had conducted correlated positively with better congruence at near statistical significance (p_< .06) . This suggests that there may be a practice or experience variable related to clinicians' congruence scores. In sum, these findings suggest that a therapist variable may be important in the congruence-outcome findings. Results suggest that there is a marked tendency for clinicians to show significant differences on the congruence variable. There is also some indication that practice and/or experience may be involved in the clinicians' capacity to be congruent on client symptoms. Future research on the congruence vari- able, eSpecially as it relates to outcome, should monitor and further examine congruence as a therapist variable. Another hypothesis was develOped to test for a client variable in congruence of perception. A correlation was obtained between clients' intake and posttherapy congruence scores for those clients who did not have the same psycho- logist as intake worker and therapist. The correlation obtained was near zero. This finding suggests that in the relationship of congruence of perception to outcome, a client variable is not a significant factor. C. -Congruence of Perception and Early_Termination. One hypothesis was designed to assess whether there was less congruence of perception on client symptoms for early terminators. For this hypothesis, a comparison was made at intake between a group of clients defined as termin- ators (5:2 therapy sessions after intake) and a group of clients defined as continuers (2:3 therapy sessions). 56 No meaningful differences between these two groups at intake was revealed by the congruence measure. The generalizability and usefulness of this finding may be limited. For one, the group of terminators was a diverse group of clients among which a large number listed as "no therapy" are really indeterminable as to the reason for not beginning therapy. Some of these clients may have been referred to other agencies. In general, the division and definitions of terminators and continuers was somewhat arbitrary, being based on the physical characteristic of number of therapy sessions. Often, clients may terminate "early" after five, six or even ten sessions. A.much better assignment of subjects could be based on obtaining the neces- sary information from therapists and clients regarding the reasons for termination. Finally, this finding was based solely on intake data, and consequently no inferences can be made regarding congruence of perception and the termination of psychotherapy. It can be stated that there appeared no ‘meaningful differences at intake between the two groups. The availability of posttherapy data on early terminators through a follow-up system would be necessary for a better test of this hypothesis. Posttherapy data of clients who prematurely discontinued therapy was generally not available in the research files.' D. Qgggruence of Perception - Intake vs. Posttherapy. The last hypothesis on the congruence measure was designed to determine whether there was better congruence at posttherapy than there was at intake. More specifically, the 57 investigation was to determine whether there was an increase in congruence of perception at posttherapy when compared to intake. For this reason, the comparison was made for clients who had the same psychologist as intake worker and therapist: a sample of 14 out of 50 completed cases. The results indicated that there was significantly better congruence at posttherapy than at intake for this group of clients. This finding suggests that there is an increase in therapists' accuracy of clients' symptomatic status after a therapeutic relationship when compared to clinicians' accuracy at intake. Greater contact appears to have led to greater agreement about symptomatology although sheer amount of psychotherapy contact at posttherapy (i.e., number of sessions) had been found unrelated to the congruence measure. E. Igplication for Future Research. Accurate empathy to clients' current feelings in psychotherapy has been of major research interest (Kurtz and Grummon 1972; Truax and Carkhuff, 1967). Far instance, accurate perceptions both of clients' self-reported feelings and concerns have been related to ratings of therapists' level of facilitation (Grethner and Saccuzzo,l977). Besides clients' current feelings, therapists' accuracy in other areas has been found important. Therapists' rated level of functioning was related to increased accuracy of discriminations of interpersonal pro- cesses such as empathy, respect, genuineness. and confronta- tions (Cannon and Carkhuff, 1969). Schrier (1953) found that 58 the congruence of perception between therapist and client on twenty-two "personality" dimensions (e.g., achievement, aggres- sion, abasement, blame avoidance, etc.) was related to ratings of therapeutic success and positive rapport. Cartwright and Lerner (1963) found that therapists' capacity to see the client "as the client sees himself" on the Kelly Role Construct Reper- tory Test at termination, but not at the beginning of therapy, was significantly related to ratings of improved vs. unimproved outcome. Orlinsky and Howard (1978) have suggested that this latter finding may indicate improving accuracy of therapist perception in successful cases and progressive misperception in poorer outcomes over the course of treatment. Taken together, these studies suggest the possibility that accuracy or congruence of perception between therapist and client across other aSpects of client functioning besides current feelings, may be an important factor in psychotherapy.* In other words, a general accuracy or congruence may be more important then accuracy in any one particular area such as affect or personality. An interesting study could be designed comparing the relationships of accuracy or congruence of per- ception across various aSpects of client functioning and their relationships to the process and outcome of psychotherapy. Future research could also address the specific finding in this study of better congruence on symptmms and better *It should be highlighted that accuracy may be a matter only as far as the client perceives it (Kurtz and Grummon, 1972). The term "congruence of perception" suggests that a therapist's accuracy is related to a client's own perceptions. 59 outcome. Does congruence of perception specifically on client symptomatic status facilitate therapy? Fer instance, accuracy in identifying affect may facilitate the eXpression of clients' affect and lead to better outcome. IS there a process related to symptom identification by the therapist which enables re- lief? Answers to these future research questions appear impor- tant and intriguing to psychotherapy research in particular and to the study of psychology in general. The Use of the SCL-9OR.in a Psychotherapy Setting Three separate data analyses were conducted on the SOL-90R as used by clinicians and clients. These analyses were based on similar analyses by Derogatis et a1 (1976) in a study comparing cancer patients and their treating physicians. The analyses were: 1) a comparison of the mean discrepancy scores between clinicians and clients for each of the nine symptom dimensions and a global distress index; 2) correla- tions between clinicians' global distress ratings of clients and discrepancy scores for each of the nine symptom dimen- sions; and 3) correlations between clinicians' global distress ratings and their ratings of clients on each of the nine symptom scales. These analyses are shown, respectively, in Tables 8, 9, and 10 of the results section. A) Discrepancies between clinicians and clients on the SCL-9OR'S Ten Scales. For all three analyses, data was collected at intake and at posttherapy and pooled together. For the first analysis, which involved a comparison of discrepancy scores between clinician and client ratings of client symptoms on each of 60 the ten dimensions, an overall F-test showed no significant differences in clinician-client discrepancy across the ten dimensions. Derogatis et a1 (1976) also reported no significant findings in this analysis. Some comments need to be made regarding the data for this finding (hypothesis seven). These comments seriously question the value of the data and hence subsequent analyses. It became apparent in deriving the data, that in converting the raw scores of clients and clinicians on the SCL-90R, a problem arose. Specifically, each raw score for each symptom dimension gets a particular T-score. Let us say, for example, that on one symptom dimension client and therapist raw scores of "3" receive T-values of 50 and 60, whereas for another symptom dimension, the same raw scores receive T-values of 55 and 59. In other words, the different symptom dimension mean differ- ence scores are confounded with these systematic error load- ings. This error variance, conceptually, plays a small part in a congruence of perception measure taken across the ten scales. However, direct comparisons between the symptom scale difference scores (as in hypothesis seven) are confound- ed by this artifact. . Another problem with this particular analysis was the high standard deviations obtained for the discrepancy scores. The problem with this is that any significant differences ' between the scales would be nearly impossible to detect in a statistical test. Systematic differences are difficult to obtain with such within scale fluctuations. The reason for such large standard deviations is this. For one, as mentioned 61 before, each discrepancy score contains error from.scaling. But the major reason appears to be that it is very easy for therapists to either overrate or underrate a client on scaled T-score points. Disparate raw-score ratings between therapist and client (e.g., "3" vs. "6") can produce a thirty to forty T-scales points discrepancy. Consequently, it must be concluded that because of ‘methodological problems encountered in the investigation of the data for the hypothesis (seven) concerning whether there are larger discrepancies on some symptom dimensions between clinician and client, no determination can be made. While an F-test showed no statistically significant differences in discrepancy across the ten dimensions, the validity of this finding is seriously questioned. B) Correlations between clinicians'global distress gggiggs and separate scale discrepancies. The other two data analyses, also done by Derogatis et a1 (1976), appear much less problematic and seem to provide information which is both interesting and sensible. One analysis consisted of correlations between clinicians' global distress ratings of clients and the discrepancy scores for each of the nine symptom dimensions (Table 9 of Results). The reasoning or hypothesis behind this investigation would be to determine if there is a relationship between clinicians' and clients' discrepancy of particular symptomatology and the clients' level of distress as rated by the clinician. The results indicated that as clinicians' ratings of clients' global distress rose, there was a significantly larger 62 discrepancy between therapists' and clients' ratings on the symptom dimensions of paranoid ideation (pp<.01), psychoticism (p<.02), and hostility (p< .05). As would be expected, the characteristic pattern was for clinicians to rate the clients higher on these dimensions than the clients perceived or re- ported themselves. There appear to be at least two competing explanations or interpretations of these results. One is that clients do not either perceive or report higher levels of paranoid idea- tion, psychoticism, or hostility, while clinicians perceive higher levels of these symptoms when perceiving higher levels of client distress. In other words, at higher clinician per- ceived distress levels, clients do not perceive or report these symptoms as might be expected. Another possible interpreta- tion is that as clinicians perceive increasing levels of hostility, paranoid ideation, and psychoticism, they tend to be oversensitive to these dimensions as to overrate their presence in relation to client self-report. It appears that the first interpretation considering client misperception or misreport is more likely. It is a well-known and established finding in the field of psycho- pathology that more disturbed clients, particularly psychotic persons, are not as aware of their symptomatology as perceived by others. One explanation of this finding usually offered by theorists is that the person needs to dissociate or disengage from the overwhelming anxiety, distress, fear, and terror involved in such conditions. 63 The finding that at higher levels of distress, there is more discrepancy between clinician and client on the hostility dimension is an interesting finding. While the strength of this finding is somewhat less as on the psychotic and paranoid ideation dimension, one can suggest that this may be another dimension where client misperception or misreport may take place. It is likely that different defense mechanisms may be involved, such as denial or a more conscious self-deception. This finding along with the finding that there is some tendency for more discrepancy on other symptom.dimensions such as anxiety and to a lesser extent, depression suggest decreasing levels of defensive structure while also suggesting their presence. This finding is very sensible based on current theory of psychotic and neurotic defensive functioning and offers the possibility that a similar analysis using the SCL-9OR in studying such phenomena may be a fruitful investi- gation. C) Correlations betweengglobal distress ratings and separate scale ratingp. The other data analysis involved obtaining correla- tions between clinicians' global distress ratings and their ratings of clients on each of the nine symptom scales. The goal was to determine what symptmm dimensions on the SOL-90R influenced in clinicians' global distress ratings and eight of the nine symptom.scales. Depression (.57), anxiety (.52), and psychoticism (.52), ratings by clinicians correlated the highest with global distress ratings followed by interpersonal sensitivity (.47), obsessive-compulsive (.37), hostility (.37), somatization (.34), and paranoid ideation ratings (.34). 64 Only phobic anxiety ratings (.15) did not appear to be a significant consideration in global distress ratings. These findings indicate a good balance of relationships between separate scale ratings and global distress ratings by the clinicians in this study. These findings also suggest that clinicians show good use of the SOL-90R symptom scales and global distress index; and conversely, the SOL-90R appears to be a well-designed and well-suited instrument for clinical use. D) Summary and Comparison with Derogatis et al's (1976) Study. Taken together, the results of this investigation on the use of the SCL-90R in a clinical setting appears to in- dicate especially that the SCL-9OR is a very useful and well- developed instrument. The pattern of results in this investi- gation on this instrument seem quite sensible. Similarly, another indication of this investigation is that clinicians utilize the instrument very well. These findings are sub- stantiated further when comparing the results of this study with those of the other study by Derogatis et a1 (1976) which contain similar analyses. Besides the difference in the number of subjects between the present study and Derogatis' study (§_- 160 to 180 depend- ing on analysis for this study vs. a H.0f 23 for the other study) and the different pOpulations (clinicians and clients vs. cancer patients and treating physicians), the two studies appear Open to comparison. In obtaining correlations between physicians' global distress ratings and discrepancy scores, 65 Derogatis found that with increasing global distress, physicians tended to overrate interpersonal sensitivity (5,- .52, pg<.02) and anxiety (5.- .34, pg<.05) and to underrate depression sub- stantially (E - .78, p¢<.001) in relation to patients' self- report. Psychotherapists in the present study tended to over- rate paranoid ideation (£_- .24, p,<.01), psychoticism (5'- .20, p<.02) and hostility (E' .16, p<.05). These results are Open to comparison. For one, just considering the magnitude of the correlations suggests that clinicians are much more accurate than physicians on assessing the degree of symptomatology present in relation to self-report. When considering that clinicians are psychologically trained and physicians are not, the results make more sense. Physi- cians tend to go most astray on the more subtle psychological dimensions (depression, anxiety, and interpersonal sensitivity). Clinicians, on the other hand, probably working with a much more psychological disturbed pOpulation, and a much greater percentage of psychotic persons, show the greater discrepancies on paranoid ideation, psychoticism and hostility ratings of clients. The fact that, in comparing both studies, no sympton dimension occurs twice, suggests good design in the develOpment of the scales. For another comparison, Derogatis found that in correla- tions between physicians' global distress ratings and ratings on the separate scales; only anxiety, hostility and psycho- ticism appeared to relate to global distress ratings. He also found that somatization ratings correlated negatively with global distress ratings, but as he stated, physicians probably 66 tended to discount somatization in their ratings because of the clear presence of physical disease. As mentioned in this study, eight of the nine symptom dimension ratings correlated substantially with global distress ratings. The comparison suggests that clinicians have a better sense than physicians in weighing the degree of symptomatology present on the symptom dimensions with the global distress ratings although physicians appeared to do reasonably well. (Mean correlation .30 (physi- Cians) versus .41 (clinicians).) With the caveat that the specific hypothesis that clin- icians would be better judges than physicians on psychological symptoms was not investigated and considering also the post hoc nature of these comparisons, the results seem provocative. As Derogatis stated, the issue of the relation between psycho- logical symptoms and their effect on malignant diseases, is one of mounting importance. The "increased efficacy of medical treatment elevates the psychological components of cancer into an increasing significant role" (Derogatis et a1, 1976, p. 201). With this the case, and with the exploratory comparisons of physicians and clinicians use of the SCL-90R, an instrument for the assessment of psychological symptoms, one may conclude that the role of psychologically trained personnel in such situations needs to be examined further. SUMMARY This study examined congruence of perception (clini- cian's judgment and client self-report) on client symptoma- tology and its relationship to outcome ratings of psycho- therapy. The construct, congruence of perception, was defined as the degree of agreement between therapist and client through independent ratings based on the Symptom Distress Check List's (SCL-90R) ten symptom dimensions. The major hypothesis stated that: "congruence of perception on client symptoms at posttherapy is related to positive out- come as rated by therapist and client." This hypothesis was supported. Additional analyses on the congruence measure are reported, as well as analyses on clinicians' use of the SOL-90R. Findings suggest that therapists' accuracy or congruence to clients' symptomatic status may be an impor- tant factor in psychotherapy. Future research in this area is suggested. 67 LIST OF REFERENCES LIST OF REFERENCES Abeles, N. Liking for clients - Its relationship to thera- pist' a personality. unexpected findingL Ps chothera : Theopy, Research and Practice, 1967 (l), I9-21. Bronfenbrenner, U., Harding, J., & Gallwey, M. The measure- ment of skill in social perception. In D.C. McClelland (Ed. ) Talent and Society. New York. VanNostrand, 1958. Cannon, J.R., & Carkhuff, R.R. Effects of rates level of functioning and experience upon the discrimination of facilitative conditions. Journal of Consultin and Clinical Psychology, 1969, EE’ZZS, I89-I94. Cartwright, R.D., & Lerner, B. Empathy, need to change, and improvement with pay ychotherazy. Journal of Consulting Psychology, 1963, p27 138 14 Cline, V. B. , & Richards, J. M., Jr. Components of accuracy of interpersonal perception scores and the clinical and statistical prediction controversy. The Psychological Record, 1962, 12, 373-381. Cronbach, L. J. Processes affecting scores on "understanding of others" and' 'assumed similarity" Psychological Bulletin, 1955, 52, 177- 193. Cronbach, L.J. PrOposals leading to analytic treatment of social perception scores. In R. Tagiuri & L. Petrullo (Eds.) Person Perce tion and Inter ersonal Behavior. Stanfor : tan or vers ty ress, Crow,'W. J. The effect of training upon accuracy and vari- ability in interpersonal perception. Journal of Abnormal Ppychology, 1957, 55, 355- 359. Cutler, R..L. Bordin, E. S. , Williams, L & Rigler, D. Psychoanalysts as expert Observers of t e therapy pro- ceSL Journal of ConsultingPsychology, 1958 335- 340. Derogatis, L.R. The SOL-9O Manual 1: Scorin Administration and Procedures for the SOL-90. Baltimore: John HopEIns Universéty SOhool ofMedicine, Clinical Psychometrics Unit, 1 77 68 69 Derogatis, L.R., Abeloff, M. & McBeth, C. Cancer patients and their physicians in the perception of sychological symptoms. Psychosomatics, 1976, 11, 1 7-201. Estes, S.G. Judging personality from expressive behavior. Journal of Abnormal and Social Psychology, 1938, 33, 2173236. Fancher, R.E., Jr. Accuracy versus validity in person per- gg tégg. Journal of ConsultingPsychology, 1967, §l_(3), Garfield, S.L. Research on client variables in sychotherapy. In S.L. Garfield & A.E. Bergin (Eds.) Hand ook of s chothera and behavior chan e: An empiricallanalyses. New York: John Wiley 5 Sons, 1578. Genthner, RnW., & Saccuzzo, D.P. Accuracy of perception of psychotherapeutic content as a function of observers' level of facilitation, Journal of Clinical Psychology, 1977, 33 (2), 517-519. Giedt, F.H. Cues associated with accurate and inaccurate interview impressions. Psychiatry, 1958, 21, 405-409. Gottheil, E., Exline, R.V. & Winkelmayer, R. Judging emotions of normal and schzophrenic subjects, American Journal of Psychiatry. 1979. 136 (8), 1049-1051. Gottman, E., & Markman, H.J. Experimental designs in psycho- therapy research. In S.L. Garfield & A.F. Bergin (Eds.) Handbook of psychothera and behavior chan e: An empirical analysis. New ork: John Wiley é Sons, 1978. Grosz, H.J. & Grossman, K.G. Clinician's response style: A source of variation and bias in clinical gud ents. Journal of Abnormal Psychology, 1968, Z; (3 , 07-214. Holt, R.R., & Luborsky, L. PersonalityPatterns of Psychia- trists: A Study in Selection Techniques, Vol. 1, New York: Basic Books, 1958. Kagan, N., Krathwohl, D., Goldber , A.D., Campbell, R.J., Schauble, P.G., Greenberg, §.S., Danish, S.J., Resnikoff, A., Bowes, J., & Bandy, S.B. Studies in Human Interaction: Interpgrsonal Process Reca Simulatedub Videot e. East Lansing, Michigan State University Educational Publication Services, College of Education, 1967. Katz, MQM., Cole, J.O. & Lowery, H.A. Studies in the diagnos- tic process: The influence of symptom perception, past experience, and ethnic background on diagnostic de- gisiozs. American Journal of Psychiatry, 1969, 125 (7) 37-9 7. 70 Rurtz, R.R., & Grummon, D.L. Different approaches to the measurement of therapist empathy and their relationship to therapy outcome, Journal of Consulting and Clinical Psychology, 1972, 33 (15, 106-115. Meehl, P.E. The cognitive activity of the clinician. American Psychologist, 1960, 33, 19-27. Mischel, W. Direct verses indirect personality assessment: Evidence and im lications, Journal of Consulting and Clinical Psycho ogy, 1972, 433 Z35, 319-324. Orlinsky, D.E. & Howard, K.I. The relation of process to out- come in psychotherapy. In S.L. Garfield & A.E. Bergin (Eds.) Handbook of Psychgthera and Behavior Changg: An empirical analysis. New YorE: John Wiley E—Sons, 1978. Oskamp, S. The relationship of clinical experience and trainin methods to several criteria of clinical predict on. Psychological Monographs, 1962, 13 (28, Whole No. 547). Reid, D.K. & Snyder, W.U. Experiment on "recognition of feeling" in non-directive psychotherapy. Journal of Clinical Psychology, 1947, 3, 128-135. Schrier, H. The significance of identification in therapy. American Journal of Orthopsychiatry, 1953, 33, 585-604. Strupp, H.H., Lessler, R. & Fox, R. Patients View Their Psychotherapy. Baltimore: John op ns ress, Taft, R. The ability to judge people, Psychological Bulletin, 1955, 33, 1-23. Taft, R. Multiple methods of ersonality assessment, Psychological Bulletin, 959, 33, 333-352. Truax, C.B. & Carkhuff, R.R. Towards Effective Counseling and Psychotherapy. Chicago: Aldine, 1967. Wallins, P. The prediction of individual behavior from case studies, Social Science Research Council Bulletin, 1941, 532.. 181-2375. Watley, D.J. Counselor predictive skill and rated counselor effectiveness, Personnel and Guidance Journal, 1967, $2, 579-584. Weiss, J.H. Effects of professional training and amount and accuracy of information on behavioral prediction. Journal of Consulting_Psychology, 1963, 31 (3), 257-262. Widiger, T.A., Knudson, RJM. & Rorer, L.G. Convergent and discriminant validity of measures of cognitive style and abilities, Journal of Personalit and Social Ps cholo , 1980, 33 (15, 116-129. APPENDICES APPENDIX A SYMPTOM DISTRESS CHECKLIST CLIENT FORM (SOL-90R) Copyright © Leonard R. Derogatis, Ph.D., 1975 APPENDIX A SYMPTOM DISTRESS CHECKLIST CLIENT FORM (SCL-9OR) INSTRUCTIONS: Below is a list of problems and complaints that people sometimes have. Please read each one carefully. After you have done so. please circle one of the numbers to the right that best describes hoW'much that problem has bothered or distressed you during the past can le weeks including today. Circle only one num er for each problem.and do not skip any itemm. Please read the example before beginning. CATEGORIES: O - Not at all 1 - A little bit 2 - Mbderately 3 - Quite a bit 4 - Extremely EXAMPLE: Homeuch were you bothered by l. Backaches.... By circling #1. this person answered that he/she was a little bit bothered by backaches. l. Headaches 2. Nervousness or shakiness inside 0 l 2 3 4 3. Unwanted thoughts. words, or ideas that won't leave your mind 0 l 2 3 4 4. Faintness or dizziness 0 l 2 3 4 5. Loss of sexual interest or pleasure 0 l 2 3 4 6. Feeling critical of others 0 l 2 3 4 7. The idea that someone else can control your thoughts 0 l 2 3 4 71 72 APPENDIX A (cont'd.) CATEGORIES: O - Not at all 1 - A little bit 2 - Moderately 3 - Quite a bit 4 - Extremely 8. Feeling others are to blame for most of your troubles 9. Trouble remembering things 10. worried about sloppiness or carelessness 11. Feeling easily annoyed or irritated 12. Pains in heart or chest 13. Feeling afraid in open spaces or on the streets 14. Feeling low in energy or slowed down 15. Thoughts of ending your life 16. Hearing voices that other people do not hear l7. Trembling 18. Feeling that most people cannot be trusted 19. Poor appetite 20. Crying easily 21. Feeling shy or uneasy with the Opposite sex 22. Feeling of being trapped or caught 23. Suddenly scared for no reason 24. Temper outbursts that you could not control 0 73 APPENDIX,A.(cont'd.) CATEGORIES: O - Not at all 1 - A little bit 2 - Moderately 3 - Quite a bit 4 - Extremely 25. Feeling afraid to go out of your house alone 26. Blaming yourself for things 27. Pains in lower back 28. Feeling blocked in getting things done 29. Feeling lonely 30. Feeling blue 31. worrying too much 32. Feeling no interest in things 33. Feeling fearful 34. Your feelings being easily hurt 35. Other peOple being aware of your private thoughts 36. Feeling others do not under- stand you or are unsympathetic 37. Feeling that people are unfriendly or dislike you ' 38. Having to do things very slowly to insure correctness 39. Heart pounding or racing 40. Nausea or upset stomach 41. Feeling inferior to others 42. Soreness of your muscles 43. Feeling that you are watched or talked about by others 0 000000 00000 ...: HHHHHH HHHHH N NNNNNN NNNNN w wwuwuu uwuww b J-‘J-‘b-PJ-‘Jl‘ b-l-‘J-‘bk 74 APPENDIX A (cont'd.) CATEGORIES: O - Not at all 1 ~ A little bit 2 - Moderately 3 - Quite a bit 4 - Extremely 44. Trouble falling asleep 45. Having to check and double- check what you do 46. Difficulty making decisions 47. Feeling afraid to travel on buses. subways, or trains 48. Trouble getting your breath 49. Hot or cold spells 50. Having to avoid certain things, places, or activities because they frighten you 51. Your mind going blank 52. Numbness or tingling in parts of your body 53. A lump in your throat 54. Feeling hopeless about the future 55. Trouble concentrating 56. Feeling weak in parts of your body 57. Feeling tense or keyed up 58. Heavy feelings in your arms or legs 59. Thoughts of death or dying 60. Overeating 61. Feeling uneasy when people are watching or talking about you 75 APPENDIX A (cont'd.) CATEGORIES: 0 - Not at all 1 - A little bit 2 - Moderately 3 - Quite a bit 4 - Extremely 62. Having thoughts that are not your own 63. Having urges to beat, injure, or harm someone 64. Awakening in the early morning 65. Having ideas or beliefs that others do not share 66. Sleep that is restless or disturbed 67 Having urges to break or smash things 68. Having ideas or beliefs that others do not share 69. Feeling very self-conscious with others 70. Feeling uneasy in crowds such as shopping or at a movie 71. Feeling everything is an effort 72. Spells of terror or panic 73. Feeling uncomfortable about eating or drinking in public 74. Getting into frequent arguments 75. Feelin nervous when you are left a one 76. Others not giving you prOper credit for your achievements 76 APPENDIX A (cont'd.) CATEGORIES: 0 - Not at all 1 - A little bit 2 - Mbderately 3 - Quite a bit 4 - Extremely 77. Feeling lonely even when you are with people 78. Feeling so restless you couldn t sit still 79. Feelings of worthlessness 80. Feeling that familiar things are strange or unreal 81. Shouting or throwing things 82. Feeling afraid you will faint in public 83. Feeling that peOple will take advantage of you if you let them 84. Havin thoughts about sex that other you a lot 85. The idea that you should be ‘punished for your sins 86. Feeling pushed to get things done 87. The idea that something serious is wrong with your body 88. Never feeling close to another person 89. Feelings of guilt 90. The idea that something is wrong with your mind 0 APPENDIX B SYMPTOM DISTRESS CHECKLIST CLINICIAN FORM (SCL-9OA) Copyright © Leonard R. Derogatis, Ph.D. and ' Nick Mellisaratos, 1976 APPENDIX B SYMPTOM DISTRESS CHECKLIST CLINICIAN FORM (SCL-9OA) HOPKINS PSYCHIATRIC RATINGS l. Somatization O 1 2. Obsessive-Compulsive 0 l 3. Interpersonal Sensitivity 0 l 4. Depression 0 l 5. Anxiety 0 l 6. Hostility 0 l 7. Phobic Anxiety 0 l 8. Paranoid Ideation 0 l 9. Global Pathology Index 0 1 CATEGORIES: 0 - None 1 - Slight 2 - Mild 3 - Moderate 4 - Marked 5 - Severe 6 - Extreme 77 NNNNNNNNN wwwuwuwww .l-‘J-‘bbJ-‘b-L‘J-‘b MUIUiLnUIUUUIUIUI O‘O‘O‘O‘O‘O‘O‘GO‘ APPENDIX C POSTTHERAPY CLIENT ' QUESTIONNAIRE APPENDIX C POSTTHERAPY CLIENT QUESTIONNAIRE For each item choose the answer which you feel best describes your therapy experience. 1. *4. Homeuch in need of further therapy do you feel now? No need at all Slight need Could use more Considerable need ery great need What led to the termination of your therapy? My decision My therapist's decision utual agreement External factors How much have you benefitted from.your therapy? A great deal A fair amount To some extent Very little Not at all Everything considered, how satisfied are you with the results 0 your psychotherapy experience? Extremal dissatisfied derately dissatisfied Fairly dissatisfied Fairly satisfied derately satisfied Highly satisfied Extremely satisfied *Questions used in this study. 78 *5. 79 APPENDIX C (cont'd.) What impression did you have of your therapist's level of experience? Extremely inexperienced Rather inexperienced Somewhat experienced Fairly experienced Highly experienced Exceptionally experienced How well did you feel you were getting along before therapy? Very well Fairly‘well Neither well nor poorly Fairly poorly ery poorly Extremely poorly How long before entering therapy did you feel in need of professional help? Less than 1 year -2 years 3-4 years 5-10 years ll-lS years 16-20 years How severely disturbed did you consider yourself at the beginning of your therapy? Extremely disturbed Very much disturbed Moderately disturbed Somewhat disturbed Very slightly disturbed How much anxiety did you feel at the time you started therapy? A tremendous amount A great deal A fair amount Very little None at all *Questions used in this study. 10. *ll. 12. *13. 80 APPENDIX C (cont'd.)- How great was the internal "pressure" to do something about these problems when you entered psychotherapy? Extremely great ery great Fairly great Relatively small ery'smal Extremely small How much do you feel you have changed as a result of psychotherapy? A eat deal A air amount Somewhat Very little Not at all How much of this change do you feel has been apparent to others? (a) PeOple closest to you (husband, wife, etc.) A great deal A fair amount Somewhat Very little Not at all (b) Close friends. A great deal A fair amount Somewhat Very little Not at all (c) Co-workers, acquaintances, etc. A great deal A fair amount Somewhat Very little Not at all On the whole, how well do you feel you are getting along now? Extremely well Very well Fairly well Neither well nor poorly Fairly poorly Very poorly Extremely poorly *Questions used in this study. 14. *15. 16. 17. 81 APPENDIX C (cont'd.) How adequately do you feel you are dealing with any present problems? Very adeguately Fairly a equately Neither adequately nor inadequately Somewhat inadequately Very inadequately To what extent have your complaints or s toms that brought you to therapy changed as a resu t of treatment? Completely disappeared ery greatly improved Considerably improved Somewhat improved at at all improved Got worse How soon after entering therapy did you feel any marked change? weeks of therapy (approximately) How strongly would you recommend psychotherapy to a close friend with emotional problems? would strongly recommend it ould mildly recommend it 'Would recommend it but with some reservations ould not recommend it Would advise against it Please indicate to what extent each of the following statements describes your therapy eXperience. Disregard that at one point or another in therapy you may have felt differently. l - Strongly agree - Mildly a ree Undecide Mildly disagree Strongly disagree \ONUIUD I The following questions were rated on the above scale. 18. 19. 20. My therapy was an intensely emotional experience. My therapy was often a rather painful experience. I remember very little about the details of my psychotherapeutic work. *Questions used in this study. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 82 APPENDIX C (cont'd.) My therapist almost never used technical terms. On the whole I experienced very little feeling in the course of therapy. There were times when I experienced intense anger toward my therapist. I feel the therapist was rather active most of the tbme. I am convinced that the therapist respected me as a person. I feel the therapist was genuinely interested in helping me. I often felt I was "just another patient." The therapist was always keenly attentive to what I had to say. The therapist often used very abstract language. He very rarely engaged in.small talk. The therapist tended to be rather stiff and formal. The therapist's manner was quite natural and unstudied. I feel that he often didn't understand my feelings. I feel he was extremely passive. His general attitude was rather cold and distant. I often had the feeling that he talked too much. I was never sure whether the therapist thought I was a worthwhile person. I had a feeling of absolute trust in the therapist's integrity as a person. I felt there usually was a good deal of warmth in the way he talked to me. The tone of his statements tended to be rather cold. The tone of his statements tended to be rather neutral. I was never given any instructions or advice on how to conduct my life. 43. 44. 45. 46. 47. 48. 49. 50. 51. 52. 53. 54. 55. 83 APPENDIX C (cont'd.) The therapist often talked about psychoanalytic theory in my sessions. A.major emphasis in treatment was upon my attitudes and feelings about the therapist. A.major emphasis in treatment was upon my relationships with people in my current life. A major emphasis in treatment was upon childhood experiences. A.major emphasis in treatment was upon gestures. silences, shifts in my tone of voice and bodily movements. I was almost never given any reassurances by the therapist. My therapist showed very little interest in my dreams and fantasies. I usually felt I was fully accepted by the therapist. I never had the slightest doubt about the therapist's interest in helping me. I was often uncertain about the therapist's real feelings toward me. The therapist's manner of speaking seemed rather formal. I feel the emotional experience of therapy was much more important in producing change than intellectual understanding of my problems. My therapist stressed intellectual understanding as ‘much as emotional experiencing. APPENDIX D POSTTHERAPY THERAPIST QUESTIONNAIRE CATEGORIES: APPENDIX D POSTTHERAPY THERAPIST QUESTIONNAIRE None Sli ht M113 Moderate Marked Severe Extreme ChthuNU-‘O llllll. HOPKINS PSYCHIATRIC RATINGS 1. 'rd <3 \0 (D \l 0‘ U'l # U N o o o o o o o o Somatization Obsessive-Compulsive Interpersonal Sensitivity Depression Anxiety Hostility Phobic Anxiety Paranoid Ideation Psychoticism :3 <3 c> <3" c> <3 c: <3 c: <3 :a P‘ p. ,3 :4 F‘ rd ea F‘ p. h) be to no no u: no to h) to I» «a u: no :3 u: in «a u: to «b b- a-¢~ a~ 4> -b .9 a- a~ MUU'IMUIUIUIUIUIUI Global Pathology Index Please rate each of the following items, comparing the client with other clients whom you see in psychotherapy using the following scale: 84 0000000000 *11. 12. 13. 14. *15. 16. l7. l8. 19. 20. 21. *22. 23. 24. 25. 85 APPENDIX D (cont'd.) 1 - Very little 3 - Some 5 - Moderate 7 - Fairly great 9 - Very great Defensiveness Anxiety Ego Strength Degree of disturbance Capacity of insight Overall adjustment Personal liking for patient Motivation for therapy Improvement expected (Prognosis) Degree to which counter- transference was a problem in therapy Degree to which you usuall enjoy working with t is kind of patient in psychotherapy Degree of symptomatic improvement Degree of change in basic personality structure Degree to which you felt warmly toward the patient How much of an "emotional investment" did you have in this patient? rd rd rd rd rd rd ...: * Questions used in this study. uwuuuu U UIUIUIUIUIUI Ul \INNNNN \1 000000 \0 26. *27. *28. *29. 30. 86 APPENDIX D (cont'd.) Continue rating your client in comparison to other clients on the following scale: Very little Some Moderate Fairly great Very great U'IwaH lllll Degree to which you think the patient felt warmly toward you 1 3 5 7 9 Overall success of therapy 1 3 5 7 9 How would you characterize your working relationship with this patient? 1 3 5 7 9 1 - Extremely poor 3 - Fairly poor 5 - Neither good nor poor 7 - Fairly good 9 - Extremely good How satisfied do you think the atient was with the resu ts of his therapy? 1 3 5 7 9 Extremely dissatisfied Fairly dissatisfied Neither satisfied nor dissatisfied Fairly satisfied Extremely satisfied \ONUINH lllll How would you characterize the form of psychotherapy you conducted with this patient? 1 3 5 7 9 l 3 5 7 9 Largely supportive Intensive analytical * Questions used in this study. 31. 32. 33. 87 APPENDIX D (cont'd.) Do you recall any strikingly pleasant experiences that you had during the therapy sessions with this patient? If yes, please mark the number that best indicates the degree of pleasantness. Otherwise mark '0" for No. l 1 2 3 4 5 2 3 4 5 6 7 8 9 7 8 9 Mildly pleasant Do you recall any strikingly unpleasant experiences you had with this patient? If yes, lease mark the number that est indicates the de- gree of unpleasantness. Otherwise mark "0" for No. 1 Extremely pleasant 2 3 4 5 6 7 8 9 7 8 l 2 3 4 5 Mildly unpleasant ' Overall, how would you characterize your exper- Extremely pleasant iences with this patient? 1 2 3 4 5 6 7 8 9 l 2 3 4 5 6 7 8 9 Unpleasant Pleasant APPENDIX E STANDARD (T-SCORE) NORMS FOR MICHIGAN STATE PSYCHOLOGICAL CLINIC CLINICIAN RATINGS OF MALE AND FEMALE OUTPATIENTS ON THE SCL-9OAIS SYMPTOM DIMENSIONS AND GLOBAL PATHOLOGY INDEX. 88 ~.HA o.no «.0m Se o.~¢ ~.am m.o~ mum H.m0H H.mm H.nw H.nh A.mo H.nn H.m¢ NMM c.mOH m.¢¢ «.mm «.mh «.mo H.mm me new a.na n.5m A.ma ox m.Ho e.~n m.n¢ poem «.03 o.~w «.mm o.m@ «.5m N.m¢ Ho mom m.¢m a.mh «.hc w.mn m.om N.H¢ «.mm mmd m.om o.~s a.ao ~.am m.ae m.Ha H.¢n mmm N.¢o H.oe o.on o.Hn c.5c o.m¢ m.mm uaH H.0w ¢.Hm o.mn ¢.mo H.wm ¢.om o.~¢ sow OHme'flnxO ouoom 3mm u NHszmm< .Ameuzv «ca-aom war no xmozu roogo=r