, ._,. v, .4 LIBRARY Michigan State University .‘ . ‘ ovsaous FINES: 'N-i“ ‘ ' 25¢ per do per item ,1 {WW}, f, RETURNING LIBRARY MATERIALS: 1 3;" "54' ‘1 Place in book return to remove ‘, “w i . charge from circulation records THE EFFECT OF LIKING OF THE THERAPIST ON SUCCESSFUL OUTCOME IN MARRIAGE THERAPY By Victor Kenneth De Jonge A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY College of Social Science 1980 in Nu m. me\\ W. ABSTRACT THE EFFECT OF LIKING OF THE THERAPIST ON SUCCESSFUL OUTCOME IN MARRIAGE THERAPY By Victor Kenneth De Jonge It has been theorized and demonstrated that there are a number of variables that contribute to successful or nonsuc— cessful psychotherapy and marriage counseling. One variable that appears to contribute to successful psychotherapy is that of the client's liking of the therapist. In this research, the therapist's likability by the client and its relationship to success was researched in a marriage counseling setting. A total of forty couples (eighty clients) participated in the research. Four social service agencies, serving mainly middle—class persons, in Grand Rapids, Michigan cooperated in the research. A total of seven therapists from these agen— cies helped in obtaining input from their clients in the filling out of the necessary questionnaires. Clients were asked to complete three questionnaires. At intake an Indi- vidual Problem Rating Questionnaire (IPR) was completed by clients. At termination or after six months of marriage counseling, a second Individual Problem Rating Questionnaire was completed by clients. In addition, clients completed Victor Kenneth De Jonge a Marriage Counseling Questionnaire (MCQ) at the time they completed the second IPR. A measure of client's liking of the therapist was included in the MCQ and a measure of the total number of counseling sessions was included in the MCQ. At termination or after six months of marriage counseling, the therapists also completed a questionnaire regarding the clients. This questionnaire asked the therapists their per- ception of success in the marriage counseling with regard to each person in the marital dyad that they counseled. The first hypothesis of the study attempted to see whether there was any significant correlation between the client's liking of the therapist and marriage counseling success. Results were inconclusive. Comparison of IPR scores, which were one measure of success, with liking scores showed no significant correlation between the two. Comparison of success responses from the MCQ and the Therapist Question- naire with liking showed a significant mean correlation of .28 (p <:.O5). The correlation of .28 does not reflect a strong linear, positive correlation between liking scores and success in marriage counseling. The data do suggest a positive correlation between success and liking of the thera— pist, but clearly additional research will have to be completed in order to confirm or reject this hypothesis conclusively. The second hypothesis stated that clients who had more counseling sessions would show significantly higher liking scores for their therapist than those who had fewer counseling sessions. This hypothesis was not confirmed. Kendall's tau Victor Kenneth De Jonge was .05 at the .283 level of significance. This showed that the number of counseling sessions had no effect on liking scores. Clients who had a low number of counseling sessions liked their therapist to the same degree as those clients who had a larger number of counseling sessions (up to twenty- nine counseling sessions). Dedicated to my wife, Ruth, and to my children--Carie. Todd, and David—-for their patience and encouragement. ii ACKNOWLEDGMENTS My deepest appreciation is extended to Dr. Victor L. Whiteman. He has patiently and diligently aided me as my thesis chairperson. I want to give special thanks to my guidance committee members who offered continued encouragement and support, Dr. Norman Abeles, Dr. David F. Gillespie, and Dr. Jay W. Artis. Also, I wish to give thanks to the marriage counselors who contributed to the research, Mr. Gordon L. Ritsema, Rev. Sidney Draayer, Dr. Donald L. Van Oostenberg, Mr. Glenn L. De Mots, and Rev. Terry A. Hager. Finally, I wish to give thanks to the clients, who took time to complete the questionnaires that made this research possible. iii TABLE OF CONTENTS Page LIST OF TABLES I I I I I I I I I I I I I I I I I I I I Vi LIST OF FIGURES . . . . . . . . . . . . . . . . . . . vii Chapter I. INTRODUCTION . . . . . . . . . . . . . . . . . 1 II. RELEVANT LITERATURE . . . . . . . . . . . . . 7 Client Likability . . . . . . . . . . . . . 7 Liking of the Therapist—-Transference . . . 11 Liking of the Therapist—-"Realistic Attitudes" . . . . . . . . . . . . . . 20 III. CONTRIBUTIONS FROM THE FIELD OF MARITAL THERAPY . . . . . . . . . . . . . . 32 IV. METHOD . . . . . . . . . . . . . . . . . . . . 37 Hypotheses . . . . . . . . . . . 37 Operational Definitions . . . . . 38 Independent Variables . . . . 5O Dependent Variables . . . . . . . . . . . . 51 Design and Procedure . . . . . . . . . . . . 51 V I RESULTS I I I I I I I I I I I I I I I I I I I 57 Hypothesis 1--Success of Marriage Counseling and Liking . . 57 Hypothesis 2--Length of Marriage Counseling and Liking of the Therapist . . . . . . . 73 VI. DISCUSSION . . . . . . . . . . . . . . . . . . 76 Explanation of the Results . . . . . . . . . 76 Alternative Explanations . . . . . . . . . . 77 Problems With the Research . . . . . . . . . 80 Implications of the Findings . . . . . . . . 83 Suggestions for Future Research . . . . . . 85 VII. CONCLUSION . . . . . . . . . . . . . . . . . . 90 iv Page APPENDICES . . . . . . . . . . . . . . . . . . . . 93 A. EXPLANATION OF QUESTIONNAIRES GIVEN TO RESPONDENTS . . . . . . . . . . 93 B. CONSENT FORMS . . . . . . . . . . . . . . . 94 C. QUESTIONNAIRES . . . . . . . . . . . . . . 98 BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . 110 Table 1. LIST OF TABLES Multiple regression analysis of pendency, and transference with Multiple regression analysis of pendency, and transference with Multiple regression analysis of pendency, and transference with Multiple regression analysis of pendency, and transference with, Multiple regression analysis of pendency, and transference with Multiple regression analysis of pendency, and transference with liking, de— Question 10. Question 11. liking, de— question 52. liking, de- Question 53. liking, de- Question 54. Question 55. Summed responses of Questions 10, 11, 52, 53. 54, and 55 compared to client liking of the therapist when dependency and transference are held constant. . Multiple regression analysis of liking, dependency, and transference with success as reported by the IPR. Multiple regression analysis of liking, dependency, and transference with suc- cess on the Therapist Questionnaire. vi Page 60 62 63 65 66 68 69 7O 72 LIST OF FIGURES Figure Page 1. How much have you benefited from your marriage counseling? . . . . . . . . . 6O 2. Everything considered, how satisfied are you with the results of your marriage-counseling experience? . . . . . . 61 3. How much do you feel you have changed as a result of marriage counseling? . . . . 63 4. On the whole, how well do you feel you are getting along now? . . . . . . . . 64 5. How adequately do you feel you are dealing with present problems? . . . . . . 66 6. To what extent have your complaints or symptoms that brought you to marriage counseling changed as a result of treatment? . . . . . . . . . 67 7. IPR scores compared with liking scores. . . 7O 8. Therapist Questionnaire-—Overall success of your marriage counseling? . . . . . . . . 71 9. Range of counseling sessions, number of clients in each range, and mean liking scores. . . . . . . . . . . . . 75 vii CHAPTER I INTRODUCTION Research in the field of psychotherapy regarding the variables that contribute to successful psychotherapy has in- creased substantially in the past years. The aim of much of this research was to test out many variables and contingencies that led to positive therapeutic results for clients. Much of the impetus for this type of research seemed to stem from the results that Eysenck obtained in his studies regarding the effectiveness or noneffectiveness of psycho— therapy (Eysenck, 1961). He stated that "we may conclude with some confidence that about two—thirds of severe psycho- neurotics show recovery or considerable improvement without the benefit of systematic psychotherapy. after a lapse of two years from the time that their disorder is notified, or they are hospitalized" (p. 711). Among other researchers, some agree with these findings (Malan, 1973, and Frank, 1973), and others do not (Rogers, 1954, Truax, 1963, Carkhuff, 1969, and Mills, 1969). The considerable confusion in the field leaves re- searchers attempting to define a satisfactory paradigm that can lend credence to variables that contribute either to the success or to the lack of success of psychotherapeutic 1 2 intervention. Kiesler (1966) stated: One of the unfortunate effects of the prolific and disorganized psychotherapy research literature is that a clear-cut, methodologically sophisticated, and sufficiently general paradigm which could guide investigations in the area has not emerged. Perhaps this is an unavoidable state of affairs in a new area of research. Yet, a perusal of this literature in- dicates that most of the basic considerations necessary for a general paradigm have appeared, albeit in many cases parenthetically, at some place or another. But to date no one has attempted to integrate empirical findings and methodological concerns in a way that might lead to a useful research paradigm (p. 110). This study will attempt to deal with one variable that is believed to aid in bringing about therapeutic change. Although a number of researchers have attempted to identify many variables that lead to the change of individual clients in therapy, few researchers have attempted to explore, identify, or isolate the variables that bring about change in married couples pursuing marital therapy. This present study is concerned with married couples' reaction to the therapist as a person, who can be liked or not, and the con- sequences of this liking for success. Client liking of the therapist as a necessary ingre- dient for successful psychotherapy has some linkage with both psychoanalytic and client-centered theories (Mills, 1969, p. 1). For example in the psychoanalytic framework "the doctor may be thought of as providing love . . . and trans- ference improvement may be achieved" (Fenichel, 1945, p. 559). In client—centered theory, client liking of the therapist is seen as necessary before positive therapeutic change can take place (Rogers, 1954, p. #25). 3 A number of problems are evident in a study of this undertaking. One problem concerns the number of clients available for sampling. Outpatient settings, such as the ones being utilized for this study, have a limited number of clients available for research purposes. Use of additional settings is a feasible alternative but is hampered by financial con- siderations. A second problem, as mentioned by other researchers (Collins, 1974, p. 3), is that clients come to an agency to solve their problems, not to aid the researcher. The moti- vation of clients to participate in this type of research could be inhibited by their lack of motivation to engage as subjects for research purposes. In fact, clients with a low degree of liking for their therapist may be extremely reluctant to participate in various aspects of the proposed research plan. Collins, in addition, states that "with clinical re- search one cannot keep outside factors from influencing out- come" (p. 3). In clinical research there is difficulty in controlling other variables that may contribute significantly to the outcome. There may be difficulty not only in controlling these variables but also in identifying them. The panel study utilized for this research may not alleviate these difficulties that a random assignment would alleviate. Hence, the con- tribution of these variables to the outcome may go undetected by the researcher. Garfield (1971), for example, pr0poses L, that many variables such as social class and intelligence are related to success and nonsuccess. In addition, client education, the therapist's preference for treatment modalities, and environmental contingencies are but a few of the variables that may contribute to success or nonsuccess. Another problem concerns the definition of success. The definition of success may be misleading and perhaps mis— understood by those involved in the research. For example, is success in marital therapy achieved when the couple simply remains married? Or is success measured by the therapist, by statements of the clients, or by persons in the clients' peer or reference group? This study will attempt to be operationally specific in the measurement of success as defined by the clients' observations and experiences. Also, the therapist's observations will be included as an additional measurement of success. By the process of triangulation, a variety of measures will be utilized to enhance the operational exact- ness of the dependent measure of success. Webb and Campbell (1966) state that by "triangulation of data procured from different measurement classes, the in- vestigator can most effectively strip of plausibility rival explanations for his comparison" (p. 174). Therefore, as many dependent variables as are practically feasible will be ex- amined to measure the degree of success in marital therapy as experienced by the client and by the therapist. Methods of measurement are limited, for, as these authors state, "When studying social change, the most practical method is to rely on available records, supplemented by verbal recall. If one wanted more control over the data, it would be possible to conduct a series of field experiments extending over a long period of years. But the difficulty of such an approach is evidenced by the scarcity of such longitudinal, original data studies in social science. . . . Cost factors make massive surveys prohibitive" (pp. 179-80). As with defining success, similar problems exist in defining liking of the therapist and defining marital therapy. The exact meaning of these terms perhaps varies among prac- titioners and clients. This study will define these terms more precisely in order to facilitate communication and under- standing for those involved in the research and for those whom this research will benefit. Yet, despite the difficulties of doing research in the "field" or in a clinical, outpatient setting, research of this type is needed. Success of psychotherapy needs to be examined where it takes place, be it in an inpatient setting or in an outpatient setting. For as Webb and Campbell (1966) state, "A multitude of operations, a combined collection of methods is needed to gather data to avoid sharing the same weaknesses" (p. 2) that data collection may incur if only one source were used (i.e., either laboratory or field). Although controlled laboratory research is valuable, research outside of the laboratory contributes other types of supplementary data that is useful for the researcher and practitioner. Thomas (1963) declares that the value of field ex- periments is in practical or experimental objectives. More specifically, he states: Practical objectives may involve the evaluation of the effectiveness of direct services to clients, of administrative means for improving services or community programs, or of the outcome of efforts "to engineer" knowledge about how services may be modified. The theoretical objectives may be to test selected hypotheses from social science of sig- nificance to social work or to test principles about diagnosis and treatment in social work. . . . The laboratory experiment contains manipu- lation of an independent variable and at least an implicit hypothesis about the relationship between the variables studied, but differs from the field experiment in studying individuals who have been removed from their customary sur- roundings and have been placed temporarily in a laboratory (p. 276). Research in a treatment setting can provide valuable information to clinicians eager to learn of the variables that help clients change. If a therapist is truly eager to be of service to his client, he will want to be knowledge- able of as much data as possible that can contribute to his efficacy as an advocate. Also, if administrative personell are aware of the staff variables that are associated with positive growth in clients, these administrators can be more effective in assigning or selecting staff members that are "suited" for the clients served by their agency. CHAPTER II RELEVANT IITERATURE Client Iikability An inevitable occurrence within the client and psycho- therapist's relationship is the phenomenon of the influence of the personality of the therapist on treatment outcome. In recent years, a number of studies have dealt with various aspects of the therapist's personality and its relationship to the prognosis of psychotherapy. Paul (1973) observed: The majority of psychotherapists, no matter what their theoretical persuasion, apparently agree that the distinguishing feature of psychotherapy lies in the "relationship" between a therapist and his patient. However, this global concept of the relationship too easily becomes an inarticulate and meaningless piece of jargon. The concept of what is a therapeutic re- lationship is a vague concept and needs refinement. We need to look at specific aspects within the thera- peutic relationship that makes the relationship posi- tive or negative. Then we need ask how do these specific characteristics of the therapist's personality help the client (p. 140). One of these specific aspects within the therapeutic relationship is the quality of client likability. Mullen (1969) defines liking as "definite feelings of caring for or 'loving' of an individual client by an individual therapist specifically within that dyadic relationship. . . specific feelings of liking for a particular client by a particular therapist and its expression within the context of their therapeutic relationship' (p. 12). 7 Research in this area demonstrates that client likability is a significant variable affecting successful outcome. Therapists who like clients in the manner described by Mullen are significantly more successful than therapists who do not like clients. Stoler (1963), for example, reports results of a study where clients were rated for likableness by ten raters. Raters listened to and rated recorded segments from actual therapy interviews. In the study, the clients had been previously grouped into more-successful and less— successful groups. Results tended to show that client lik- ability could be reliably related to success in therapy/ Stoler defined success by the mean rating that the therapist received by independent raters. Ten raters were given 20, 2-minute, taped segments of therapist and client interactions drawn from 10 recorded therapy cases. Prior to this study, five of the cases were classified as more successful and five as less successful, on the criteria of therapist ratings of outcome, patient ratings of outcome, and a self-concept Q sort. Two segments were used from each case, one from an early interview and one from a late interview. The 20 segments were presented to each judge in a random order: this was made possible by having each 2-minute segment on a separate tape spool (p. 175). By utilizing an Analysis of Variance he foundaa3pOSi- tive correlation (p:<:.05) between client likability and success in therapy. "The more successful clients were liked significantly greater than the less successful clients, beyond the .05 level" (p. 175). Other researchers, such as Goz, agree that the thera- pist's personality inevitably colors the nature of treatment (Goz, 1975). In client—centered therapy, unconditional positive regard for the client is a vital ingredient in successful therapy (Rogers, 1959). Such regard includes liking, nonpossessive warmth, respect, empathy, and acceptance. Mullen (1969), in studying the liking of clients by thirty—six therapists at Michigan State University's Counseling Center, "concluded that the therapeutic conditions of high levels of liking and empathy are necessary to establish potential for change as a result of therapy. Low-level con- ditions do not permit the development of trust in a thera— peutic relationship that will be necessary for exploration of intensive conflict" (p. 80). Bent, Putnam, Kiesler, and Nowicki (1976) studied ninety—three outpatients in a Georgia outpatient facility over a period of one year. These researchers found that positive perception of the therapist by the patient was related to successful therapy outcome. Patients who were highly satis- fied with therapy, compared with those who were not, described their therapists as warmer, more likable, more active, and more involved and said they felt that therapy had a more noticeable effect on their behavior. More specifically, the authors stated: Subjects completed a questionnaire based on an instru- ment devised by Strupp et al. (1969). The items related to the clients' perceptions of both therapy and the therapist. The main analysis compared subjects' state- ments that they were satisfied (n = 24, top 25%) with those who stated that they were very dissatisfied with psychotherapy (n = 16, bottom 17%). Out of a possible forty-seven comparisons, thirty- one were found to differentiate between the groups (p-<:,O5). In essence, these differences revealed that those who were very satisfied with therapy 10 described their therapists as warm, active, likable, and involved in contrast to the description that those who were dissatisfied gave of their therapist (p. 149). In addition, other researchers and practitioners either directly point to the importance of the therapist's liking of the client or imply that liking, by being a part of nonpossessive warmth, respect, empathy, positive regard, and unconditional positive regard, is necessary for effective psychotherapy (Truax, 1967, Carkhuff, 1969, and Kell and Mueller, 1966). Although the variable of liking of the client con- sistently appears to be a necessary attribute of the therapist in psychotherapy, its importance and demonstration are not without potential negative consequences and problems. Mills and Abeles (1965) reason that "counselors may well have to bridle the expression of their nurturant and affiliative needs in order to meet the demands of the counseling situ— ation and of the clients appropriately" (p. 357). They also say: The area of counselin attracts peOple with initially high nurturant needs hence the high relationship between need for nurturance and liking-—the beginning therapist needs to nurture and to like people). How- ever, with increasing experience, the counselor is faced with a paradox—-he chose to do counseling in part presumably because of this need to take care of and to like people and yet he is forced "to limit the expression of his own needs and to derive his satisfactions essentially from meeting the needs of others rather than expressing his own except for nurturance. . . but his professional role limits how freely he may express his nurturant need lest he establish an unhealthy relationship" (Graten et al., p. 10). So with increasing experience in doing therapy, the counselor is forced to forego the satisfaction of his own nurturant need in part in order to enhance the well being and protection of his client (p. 355). 11 Too much nurturance and expression of liking is not good. Expressions of liking could be ill—timed and invoke client defensive reaction (Kell and Mueller, 1966). Although liking and expressions of liking can be helpful and lead to emotional growth and change, client likability must be per- ceived by the client to be a useful force contributing to a positive therapeutic relationship and growth. The expression and demonstration of liking and related therapist's variables that enhance the relationship must be conveyed with sound pro— fessional judgment. After all, a therapist's main purpose is to aid the client and any displays of liking (even though the client may be liked consistently) should be done with the benefits for the client serving as the main criteria. "Too much nurturance exposes the client to 'transference cures' (clients looking better only because of their relationship to the counselor) or, for the Rogerian or existentialist counselor, to the stifling of too much dependency upon the therapist" (Mills and Abeles, 1965, p. 356). Liking of the Therapist--Transference Client likability, reciprocal liking, and therapist likability are inextricably woven into the psychotherapeutic process. It is difficult to research, isolate, and study one of these variables without including their various combinations. It is this author's belief that if the therapist's liking of the client has been shown to be beneficial, then the client's liking of the therapist is also beneficial. Though there may be drawbacks and disadvantages in the area of liking, this 12 variable nevertheless seems to be a potent factor that can be useful for those practicing psychotherapy. The liking of the therapist by the client is an added dimension in the therapeutic relationship that can be helpful to promote positive client growth and change. This study will define liking of the therapist as "true" liking for the therapist as a person not only for his pro- fessional qualities but also for the qualities that he or she demonstrates as a person in the therapy relationship. Liking will include qualities of warmth for the therapist, respect, acceptance, and affiliation. The reasons for liking the therapist may stem from a variety of sources. Primarily two reasons appear to be promi— nent in the literature. The first reason entails the aspect of transference and the second entails the aspect of "reality" issues in the relationship. In describing transference, Freud (1920) stated: Instead of the patient remembering certain of the feelings and states of mind of his previous life, he reproduces them, lives through again such of them as, by means of what is called the "transference" may be made effective in opposition against the physician-and the treatment. . . these resistances of the transference are not an unforeseen danger threatening our analytic influence. No, we know that these resistances are bound to appear: we are dissatisfied only if we cannot rouse them definitely enough and make the patient perceive them as such (p. 301). Singer (1965), in referring to Freud's works and Freud's definition of the transference, said that Freud be- lieved "matter has a basic tendency to return to previous 13 states of organization and that man exhibits this tendency to return to past states in both physical and behavioral terms, that man tends to repeat previous acts over and over again" (D. 250)- Consequently, the patient's tendencies to view the therapist in the same light as an earlier figure, with shades of liking and disliking, are natural and something to be dealt with appropriately in the therapeutic process. Freud cited numerous examples of this phenomenon in his literature and frequently referred to the importance of dealing satis- factorily with these "resistances" (Freud, 1893, 1901, 1905, 1920). Since Freud's original writings, a variety of meanings have been attributed to the term transference. In a broad sense, according to Brammer and Shostrom (1960), the term refers to "any feelings expressed or felt by the client toward the therapist, whether a rational reaction to the personality of the therapist or the unconscious projection of earlier attitudes and stereotypes" (p. 209). In actuality, this definition is broader than a definition of transference attri- butable to most authors. As we shall see, "rational reactions" are generally not considered part of the transference as de- fined by a majority of researchers, writers, and practitioners. Other authors besides Freud expand, redefine, and restructure much of the material regarding transference. Garrett (1949) in discussing and defining transference stated: 14 The need to ask for help recreates to some extent in anyone a dependency situation analogous to one's infancy and thus tends to reactivate the character- istic way of handling problems which was developed at the time. . . . It is impossible for a person to place himself for long in such a dependency situation without a transference to this situation of his infan- tile attitudes (p. 225). Blanck and Blanck (1968) define transference as "the repetition in the present of unconscious feelings toward im- portant figures from childhood and is not limited to the therapeutic situation. . . . Transference feelings develop when the individual finds himself or herself in a position of need which reactivates the helpless, dependent feelings of childhood" (p. 167). Greenson (1967), summarizes the schools of Melanie Kleine and Franz Alexander in this way: The followers of the Kleinian school consider the interpretation of the unconscious meaning of trans— ference phenomena to be the crux of the therapeutic process. However, they believe that the patient's relationship to his analyst is almost entirely one of unconscious fantasy (Issacs, 1948, p. 79). Transference phenomena are regarded essentially as projections and introjections of the most infantile good and bad objects. Although these early introjections arise in a preverbial phase, the Kleinians expect their patients to com- prehend the meaning of these primitive goings-on from the beginning of the analysis (Klein, 1961: Segal, 1964). They do not analyze resistances as such, but instead make interpretations about the complex, hostile and idealized projections and intro- jections of the patient in regard to the analyst. It seems as though they expect to influence the internal good and bad objects in the patient's ego by inter- preting what they sense is going on. They do not communicate with a cohesive, integrated ego: they do not attempt to establish a work alliance, but seem instead to establish direct contact with the various introjects (Heimann, 1956) (pp. 169—70). Hinsie and Campbell (1970) see transference as "the phenomenon of projection of feelings, thoughts, and wishes 15 onto the analyst, who has come to represent the patient's past. . . . These feelings, although once appropriate, are inappropriate and anachronistic when applied to objects in the present" (pp. 780-81). Finally, Hollis (1964) also repudiates Brammer and Shostrom (1960) by expounding on the phenomenon of trans- ference as being something that originates from past feelings. She states: When we speak of transference reactions, we usually mean that the client displaces onto the worker feelings or attitudes that he originally experienced in early childhood toward a member of his family- most often but not necessarily his father or mother— and responds to the worker as if he were this person. A similar phenomenon can occur with displacement from later important associates. These are clear and specific transference reactions. Less specific is the client's bringing into treatment any distorted way of relating to people that has become a part of his personality, whether or not he identifies the worker in a direct way with early family figures (p. 154). Consequently, with but few exceptions, the transference relationship is seen to be a phenomenon that stems from the patient's past. However, in spite of the origins of the transference feelings, these feelings need not be viewed as something bad or negative. These reactions can be viewed as "positive or negative (in the sense of warm or hostile), and they may represent id, ego, or superego aspects of the personality" (Hollis, 1964, p. 154). When the patient ex- periences feelings of liking, warmth, positive regard, or other positive feelings that stem from transference phe- nomenom, these feelings are viewed as positive transference feelings. Angry, hostile, rejecting, and other negative 16 feelings resulting from the transference are referred to as negative transference. Therapists need to understand both types in order to deal more effectively with the patient and with the resistances that the patient brings into treatment. Some authors--Freud (1892), Greenson (1967), Stone (1961), and Singer (1965)--see all transference feelings, be- cause of their inappropriateness, as negative. The origin of the feelings themselves reflect the negative aspect of the feelings, even though, as Goldstein (1974) believes, "the feelings could be positive as well as negative" (p. 26). In the sense that transference is positive, with the inclusion of feelings of liking, the transference situation can serve to build the relationship. Also, negative trans- ference feelings can serve to "build" the relationship. Whether negative or positive, the transference feelings, expressed in an atmosphere of "rapport," allow the client, as Brammer and Shostrom (1960) point out, "to express distorted feelings without the usual counterdefensive responses" (p. 215). Other functions of the transference are to promote the client's confidence in the counselor through his wise handling of the transference feelings and to enable the client to receive insight into the origin and significance of these feelings (Brammer and Shostrom, 1960). Liking of the therapist, stemming from transference of feelings, is dealt with in the same way as other feelings stemming from the transference are dealt with, at least according to analytic and neo—analytic thinking. That is, 17 when appropriate, the therapist "interprets" to the client the meaning of these feelings and thus aids the client to grow and eventually become an autonomous individual (Paul, 1973). Liking of the therapist, stemming from transference, provides valuable knowledge in that these feelings provide important clues to the therapist concerning the patient's past. These feelings aid in building and sustaining the relation— ship, although "inappropriate," and provide insight into the patient's ways of relating. Nevertheless, these feelings present certain problems for the therapeutic relationship. Liking of the therapist may become congruent with excessive dependency on the therapist. Even though, as Rogers and others have said, "The two items which are rated at the end of therapy indicated that a high degree of mutual liking and respect between counselor and client is judged to be characteristic of the end phase of therapy" (Seeman, in Rogers and Dymond, eds., 1954, p. 104). Hollis (1964) confirms that liking or loving in the transference could create an unrealistic parent-child re- lationship. She states: When the client has strong transference feelings toward the worker, sustaining procedures will usually promote the positive side of the transference and will take on added significance to the client, for he will feel as if he is receiving reassurance or love from someone who was importnat to him in early life. Consistent emphasis on sustaining procedures tends to create a dependent type of parent-child relationship in both its realistic and its transference components (p. 157). Mills (1964) speculated that therapists with high- liking attitudes toward their patients will tend to "approach 18 dependency statements by the patient significantly more often than will low-liking therapists" (p. 19). This "approach" will serve as a reinforcer, thus tending to encourage dependency by the client. Mills' hypothesis was not confirmed, and he found that high nurturant therapists (with the exception of beginning therapists) tended to "train the patients for in- dependence" (Snyder, 1963) especially after the fifth inter— view (Mills, 1964, p. 73). Although Mills' research does not seem directly applicable to speculation on client dependency, this author would speculate that a high degree of liking of the therapist by patients would lead to dependency and a longer experience in psychotherapy. Fierman (1965), in discussing the thoughts of Hellmuth Kaiser, introduced two related concepts in reflecting on the problems of dependency. The first concept was the Delusion of Fusion. Here the patient "wants either to incorporate himself into the therapist and lose his own personality, or to incorporate the therapist and destroy the therapist's person- ality" (Fierman, 1965, p. xvii). The second concept was the Universal Symptom, which stems from the patient's efforts to achieve the delusion of fusion. However, Kaiser states that "man is essentially separate. No matter how close he may get to someone else, he cannot fuse with him" (Fierman, 1965, p. xviii). Kaiser and Paul (1973) declare that it is im- portant for the patient and the therapist to recognize the patient's possible attempt to fuse with the therapist. Both, then, must deal with this appropriately in therapy. 19 Thus, overdependency stemming from "positive" trans— ference can contribute to difficulties in the relationship before psychotherapy is completed. Other difficulties, besides overdependency stemming from "too" much liking, result when the client demonstrates liking for the therapist. Goldstein (1962) reflects on the beliefs that it is important that clients be attracted to therapists but warns that clients will change many of their expectations, actions, and performances to win the approval of the therapist (p. 70). Therefore, it is important that client and therapist have and demonstrate openness by frankly discussing their relationship and patient responsibility (Gold- stein, 1962, p. 81). Each must be aware of the value of liking and its limitations in the therapeutic process. Other authors (Carkhuff and Berenson, 1967: Heine, 1971: Kell and Mueller, 1966: et al.) reflect on the value and limits to the likability of both the therapist and the client. The value of likability seems demonstrable in numer- ous studies, yet "mature" therapists must be cognizant of the various manifestations of liking, positive and negative. If too much liking of the therapist exists, there could be an overidentification with the therapist and an inability to develop in an autonomous manner. Too much liking would be difficult to separate from too much de— pendency, and too much dependency could be fearful for the client. If liking, in the client's past was construed with Oppression by the persons he liked, then the 20 interrelationship of client and therapist could be construed by the client to be in opposition to liberation from what he sees as oppressive forces (Elrod, 1973). Fischer's find- ings indicate that claims by adherents of one or another of the various psychotherapeutic schools of therapy that their superior effectiveness is due to superior relationship abilities is probably baseless (Fischer et al., 1975). That is, the belief by any group of practitioners that they have a superior way of relating seems to be groundless. The ability to form positive relationships, as well as harmful relationships, appears to transcend various psychotherapeutic ideologies. Liking of the Therapist--"Realistic Attitudes" "Real" therapy comes out of the complex interaction between therapist and client in their relationship (Kell and Mueller, 1966). By itself, the therapist's liking for the client will probably be of little benefit to the client. Consequently, what is "meaningful in therapy has to do more with emotions of both client and therapist and their inter- action and effects on each other, and less with cognitive con- structs used by a 'scientist therapist,‘ with emphasis limited to insight and intellectual understanding" (Mullen, 1969, p. 2). "True" liking for the client along with re— ciprocal liking for the therapist aid in enhancing the thera- peutic relationship and set the stage for productive growth. One-sided liking may be helpful in forming a re- lationship, but perhaps the relationship will be short-lived 21 if there is not reciprocity. If the relationship is short- lived, the probability that the relationship will sustain con- tinued therapeutic intervention is considerably weakened. Reciprocity in liking will probably occur in the psychotherapeutic relationship. If there is a frequency of interaction between therapist and client, there will be a degree of liking that is dependent on this frequency of con- tact. Newcomb (1956) has emphasized that "the evidence is altogether overwhelming that, ignoring other variables, the prOposition is correct that as frequency of interaction in— creases, frequency of attraction will increase" (p. 576). This primarily is the definition of propinquity, which es— sentially states that "other things being equal, people are most likely to be attracted toward those in closest contact with them" (Newcomb, 1956, p. 575). Or as Homans (1950) ventured, "If the frequency of interaction between two or more persons increases, the degree of their liking for one another will increase" (Homans in Interpersonal Attraction by Newcomb, 1956, p. 576). This hypothesis appears to be correct in a wide range of situations, including the psycho- therapeutic relationship, where one would venture to state that there is close interpersonal contact. Newcomb (1956), in referring to behavior theory, sug- gested that opportunities for reciprocal reward occur with propinquity and with the motivations of potentially rewarding persons. He says that "the likelihood of being continually rewarded by a given person varies with the frequency with 22 which that person is in turn rewarded, and thus we have a proposition of reciprocal reward: the likelihood of being rewarded from a given person, over time, varies with the frequency of rewarding him" (p. 576). It would seem likely that clients would view supportive therapeutic procedures as rewarding and hence increase their liking of the therapist within their close interpersonal relationship. And if the relationship continues, it may very well continue because it is rewarding to persons in the relationship. Shaw and Costanzo (1970) state that the "dyadic interaction of any social relationship will most probably be continued and posi- tively evaluated if the participants view the interaction as being more rewarding than it is costly" (p. 69). Positive reinforcement or positive liking will tend to bring about positive liking by the other. Negative reinforcement of negative liking will tend to bring about negative liking by the other. Liking tends to be further increased, according to Newcomb, if actors within the relationship have perceived similarities regarding important and relevant objects. Thus, if therapist and client are agreed regarding the purpose and object of the therapy, liking between them should increase. Or, if a "contract" is mutually agreed upon, then liking between therapist and client should be accelerated. For as Newcomb states, "Proximity, alone, cannot account for attraction, but only to the degree that it facilitates the development of perceived similarity of attitude does it 23 . contribute to attraction" (Newcomb, 1954, p. 580). Adequate interpersonal communication regarding the status and priorities of the goals is essential, according to Newcomb, before liking can be enhanced. Similiarities in goals, adequate communication regarding goals, mutual reinforce- ment, and propinquity contribute to the client's liking of the therapist. To this, Newcomb adds that if a person thinks he is liked by another and if he perceives that this other per— son would describe him favorably, then the first person would like the other. Newcomb puts it more succintly when he states that the subjects in his study "tended to like those by whom they were liked, or by those who, they thought, would describe them in most favorable terms" (Newcomb, 1965, p. 586). If a client is perceived to be liked by the thera- pist, then there should be a tendency for the client to like the therapist. The validity of this concept is lent additional support by various researchers who discuss the positive impact of reciprocal feelings. Whitaker (1965) stresses that the depth of the patient's involvement is related to the depth of the involvement of the therapist as a person. The reciprocal impact of the therapeutic relationship has also received support from many other researchers who stress either directly or indirectly the importance of one member of a dyad liking the other and the relevance of exchange in a two-party interaction (Sequin, 1962: Whitaker et al., 1961: Nacht, 1962: Kell and Mueller, 1966: Paul, 1973; 24 Hollis, 1972: Rogers, 1954: Homans, 1950). Rogers (1954), for example, states that "the process of client-centered therapy, as caught in the factual evidence of these various studies, appears to be based on a warm re- lationship of mutual liking and respect" (Rogers in Rogers and Dymond, eds., 1954, p. 425). To be most effective, liking should be an ongoing process of reciprocity between client and therapist. One sided liking, does not appear to be use- ful as a therapeutic aid as is liking when reciprocity is present between the therapist and the client. Hollis (1964) refers to liking of the therapist that does not stem from the transference as "realistic attitudes and responses" (p. 149). Her insights deserve repetition be- cause of their valuable contribution regarding the etiology of the patient's feelings for the therapist that do not stem from the transference: Realistic attitudes, appropriate to the situation will differ among clients in accordance with variations in the significance the treatment situation has for them. Almost universally, when a person comes for help with interpersonal problems, he experiences some anxiety. . . . This is a realistic cause for anxiety. Characteristically he also experiences discomfort about entering into a dependent relationship. To come for help signifies weakness, despite the fact that the recognition of the need for help and the decision to come for it requires strength. . . . Such feelings are almost universal, but the intensity with which they are experienced will vary. . . . Many other types of client reactions can be realistic responses to varying circumstances. If the client has been overtly or subtly pressed to apply for treatment, he may feel anger as well as anxiety. . What the client knows about the agency or casework will also affect his initial attitudes. If he has heard favorable reports about the agency or an in— dividual worker, he will anticipate a sympathetic, 25 skillful reception. A previous bad experience or negative reports of others will lead him to expect the worker to be critical or hostile or dominating. Com- munity, ethnic, and class attitudes toward casework may affect his feelings as to whether coming for help is respectable or a degrading thing to do, and will condition his expectation of what treatment itself will be like. The less well—educated client who has not pre— viously experienced casework help will usually tend more to expect advice and a somewhat authoritative approach than will the college graduate who because he is apt to have at least a speaking acquaintance with modern dy- namic psychology will anticipate a more sympathetic and thought- provoking approach. . . As soon as client meets worker, the latter' 8 physi- cal appearance and manner set new reactions in motion. While class differences, including general education, may increase the confidence of the blue—collar worker in the professional ability of the caseworker, they may also make him fearful of being misunderstood and mis- judged and may increase feelings of anxiety and re- sentment concerning a situation in which he feels in- ferior and dependent. Sex differences also arouse different realistic reactions. The appearance of the worker's office and experiences in the waiting room will add their bit to the client's reactions. And all these responses take place independently of what the worker actually does! What the worker says and the way he acts when he says it are obviously the next set of reality factors affecting the client's realistic responses. Before jumping to the conclusion that a client is displaying either transference reactions or subjectively con- ditioned resistance, it is very important for the worker to make certain that he is not actually saying or doing something that is giving the client a real- istic basis for certain responses. Workers are some— times hostile, or at least critical or disinterested. Some workers, out of their own needs, act in a superior, overly impersonal way. Some enjoy a very subtle type of domination which puts the clients in an unneces- sarily dependent or inferior position. . . . Even the best of caseworkers will exhibit occasional "unthera- peutic" reactions. Caseworkers too are affected by mood changes, health, events in work or private life quite outside the particular treatment situation, and other factors (pp. 149-52).* *Used by permission of the publisher. 26 In addition, other authors reflect on the origin of liking and other feelings and perceptions that do not stem from transference phenonmena. Jourard (1971) says that "the amount of personal information that one person is willing to disclose to another appears to be an index of the 'closeness' of the relationship, and of the affection, love or trust that prevails between the two peOple. In more general terms, self— disclosure and liking for the other person may be correlated" (p. 13). Counseling relationships are not sterile situations unaffected by the "laws" of interpersonal intimacy. Con- sequently it would seem logical to this author that higher client self—disclosure would lead to higher therapist's likability or, conversely, that high likability would lead to more self-disclosure. Jourard continues by stating that "psychotherapists have long noted that when a patient likes his therapist he discloses himself more freely" (p. 13). Jourard was unable to show whether liking precedes disclosure, whether disclosure precedes, or whether both of these factors are determined by other variables (p. 15). In one interesting development of this research, Jourard found an exception to high self-disclosure leading to high liking. In one of his therapy groups, one subject (Dawn), was the highest disclosure person in the entire group. In spite of this, Dawn fell at the "least—liked" end of the average liking ranks. The therapist observed that she re— vealed herself "highly" in a consistent fashion without regard to social context. Jourard speculated that perhaps there was a .27 failure of this subject to vary self-disclosure with liking and this "betokens contrasting forms of interpersonal (and personal) maladjustment" (Jourard, 1971, p. 17). In a later study, Jourard (1971) found that a "subject's liking for each‘of his fellows was not a strong determiner of their disclosure to him" (p. 23). In other words, disclosure did not necessarily stem from liking. Other variables can contribute to liking, and when liking does take place, it does not necessarily mean that disclosure will follow. However, Jourard seemed to find that disclosure does seem to contribute to liking of the person to whom one discloses. Although the exact sequence of events, liking leading to disclosure or disclosure leading to liking, has been somewhat difficult to ascertain, Jourard does state that persons "who wish to become known and understood must disclose themselves" (Jourard, 1971, p. 23). With regard to sex differentials, Jourard found a "relatively minor role played by liking in men's disclosure that contrasted sharply with the major role of disclosure patterns in female subjects" (p. 25). Men seem to distrust their feelings and base transactions on cognitive factors, wheras women base many of their transactions on emotions and not on cognitive factors (p. 25). Even though the importance of self-disclosure is documented by many sources (Marlow, 1954: Jourard, 1971: Cozby, 1973, and others), self—disclosure must be ”appropriate" if it is to be a key factor in promoting mental health. 28 The context of the roles and situation "dictate" the ap- propriateness of self-disclosure. For example, Chaikin and Derlega (1974) state that disclosure from a low—status person to a high-status person is more appropriate than disclosure from a high-status person to a low-status person (p. 8). Also, when intimate disclosure "from a professional person takes place, to a client, there was a reduction in the client's intimacy" (Chaikin and Derlega, 1974, p. 8). Intimate self-disclosure, then, on the part of the therapist appears to be out of place. However, liking of the therapist will develop if the therapist self-discloses in the sense that he is genuine. Chaikin and Derlega (1974) in dis- cussing Truax and Carkhuff's works make the following state- ment: The genuine person says what he feels: he relates to peOple as honestly as he can, without defensive- ness and without hiding behind a facade or retreating into the security of a role-defined stereotype (e.g., the "objective" therapist). It is not essential for the therapist to disclose intimate information to the patient, but the responses of the therapist should be totally genuine. If the therapist's self-disclosure of some past experience is relevant to the patient's self-disclosure, and may indeed help the patient to understand his own experience better, such therapist's self—disclosure would serve as a model for the patient to imitate. In a later work, Carkhuff (1969) states more ex- plicitly that the therapist's self—disclosure is an aid to the therapeutic process, although he does not recommend it for early stages of therapy. The final level of facilitative self-disclosure, for use in mid to late stages of therapy, calls for the therapist to volunteer very intimate and often detailed material about his own personality. . . information that might be extremely embarassing under different circumstances p. 11 . According to these researchers, self—disclosure by 29 a therapist must be utilized with therapeutic wisdom, be appropriate, be well-timed, and contribute to the liking process. Chaikin and Derlega (1974) conclude their paper by sug- gesting that the "average" person will disclose intimacies to persons he likes. To reveal intimate feelings and thoughts to mere acquaintances would tend to make a person uncom- fortable and anxious (p. 27). Therefore, it appears from the literature reviewed that a person must like someone be— fore he will disclose himself. Once information is disclosed, liking increases and brings forth more disclosure, leading to an ever-broadening sphere of intimacy. Another variable that seems to contribute to the therapist's likability centers on a psycho-social dynamic called "ingratiation." Jones and Wortman (1973) define in— gratiation as "a class of strategic behaviors illicitly de- signed to influence a particular other person concerning the attractiveness of one's personal qualities" (p. 2). These authors believe that certain perSons will perform in such a manner as to make select target persons more at- tracted to them. For instance, in the therapeutic relation- ship, therapists by their manner, appearance, discussion, poise, compliments, and other acts will enhance their image and likability with clients. The rules of ingratiation seemingly apply to most dyadic relationships and are aimed at increasing attractiveness and likability. The authors say: 30 If we want a target person to respond to us with in- creased attractiveness, there are a number of things that we should keep in mind. We should take great care to avoid compliments under circumstances that will lead him to conclude that we merely "want some- thing" from him. We should also avoid complimenting others in front of the target person in such a way that he concludes that we are the kind of person who is positive to everyone. . . . We should compliment the target person so that he will conclude that we feel high positive regard for him--either because we are trying to "be nice" and not "hurt his feelings" or, preferably, because we genuinely believe the pos- itive things we are saying about him and are not uni— formly as positive to others (Jones and Wortman, 1973, p- 5)- Although the authors cite examples where compliments and other ingratiating acts might have negative consequences, they generally conclude that most ingratiating acts increase the target's liking of the ingratiator. As long as the in- gratiating acts are genuine, without ulterior motive, and seemingly follow after relatively negative or neutral com- ments, the target person is more likely to think positively of the ingratiator (Jones and Wortman, 1973, pp. 8-11). One part of ingratiating behavior, according to Jones and Wortman, entails self-disclosure. Interestingly, their conclusions seem to contradict findings by Jourard, Chaikin, and Derlega. Jones and Wortman (1973) declare, "The results that reveal that those who have made more intimate disclosures were better liked than those who had made less intimate dis- closures, should be interpreted with some caution. Later studies (post 1972) have not provided support for the hy- pothesis that high-disclosing others are liked more than low- disclosing others" (p. 14). These researchers believe, con- sistent with other researchers, that the setting in which 31 self-disclosures are made contribute significantly to the effectiveness of self-disclosures that contribute to liking. The norms and mores of the setting will dictate whether self— disclosure is called for. If these mores and norms call for self-disclosure, there is a high liklihood that self-disclosure will lead to liking. In our society, the setting of psycho- therapy sets the stage for self-disclosure and psychological intimacy. In this setting of high-potential dependence the "would-be ingratiator appears to have extra attraction credits for moderate disagreement or autonomy" (Jones and Wortman, 1973, p. 52). The probability of the therapist being liked, even if he remains "neutral," is high because of the dependence and because the setting "demands" self- disclosure. Even though negative aspects are possible (e.g., high dependence may elicit less attraction), the therapeutic setting potentially has many attributes that could lead to strong feelings of liking for the therapist. CHAPTER III CONTRIBUTIONS FROM THE FIELD OF MARITAL THERAPY Unfortunately, the amount of literature available in the field of marital therapy or marriage counseling re- garding the relationship of the therapist's likability to therapy outcome is considerably less than that of other areas. (The terms "marital therapy," "marriage counseling," and Pmarriage therapy" will be utilized synonymously for this report.) Because of this factor, data from research already cited will be assumed to be applicable to marital therapy. Even though generalization from a dyadic interpersonal re- lationship entails risks and potential inaccuracies, it will be assumed that generalizations and extrapolations can be made, for the phenomena of transference and other attitudes that a client may have toward the therapist are influenced by many variables. Primarily, transference, as Weiss and English (1957) believe, "is the attitude that the patient transfers toward the therapist from feelings that he has had in his childhood from important figures in his environment, particularly his parents" (p. 532). The fact that a couple is seen jointly may also affect the attitudes of the client toward the thera- pist, but the transference phenomena will also likely be present if a relationship is established. 32 33 The generalizations might appear valid for at least two other reasons. First of all, much marriage counseling is done in a one-to-one setting. Although some may question the validity of a one-to-one setting being called marriage counseling, the motivation for treatment and the agreed goals may qualify the setting as marriage counseling. For example, a therapist may decide that it is best to see the married couples on an individual basis and not jointly. This situ- ation would make the previously reported observations about liking of the therapist applicable in this setting. Second, even if the married couple is seen jointly, the patients are still relating to the therapist on an individual basis. In one sense, therapists cannot do mar— riage counseling but only counseling of the persons in the marriage. Therefore, regardless of whether or not the couple is seen jointly or separately, they nevertheless would not lose their individualistic way of relating to the therapist. In addition, marriage counselors utilize many of the same techniques that a therapist would utilize with a patient on a one-to—one basis. Haley (1963) states that "a marriage therapist, by dealing fairly with each spouse, deals differ— ently with them than others have. He does not let himself be provoked into condemning either marital partner" (p. 197). The marriage counselor practices acceptance, nonpossessive warmth, empathy, liking, and other qualities utilized by therapists in a one-to-one setting. For, as Herbert and 34 Jarvis (1959) maintained, "Unless a relationship in marriage counseling has been established-—unless, that is, the client has felt interest, warmth, and the attempt at understanding from a counselor--a series of interviews is unlikely to be sustained" (p. 55). Indeed, most practicing psychotherapists see in- dividuals on a one-to-one basis for individual psychotherapy or marriage counseling. These same therapists will regu- larly see couples in joint sessions for marriage counseling. Some therapists might argue that there are marked differences between psychotherapy and conjoint marriage counseling (Satir, 1964, et al.), but goals and techniques are somewhat similar. Ard (1964) puts it this way: Through procedures and techniques used in marriage counseling, which establishes a good therapeutic atmosphere, the client is encouraged to communicate (i.e., verbalize) his feelings and experiences, and will thereby gradually bring more and more of his significant feelings and experiences into the realm of awareness. In this regard, it may be said that there is no sharp, clear dividing line between marriage counseling and psychotherapy. Marriage counseling shades into depth therapy in imperceptible degrees. Marriage counseling may include certain aspects or methods which are different from those seen in various kinds of psychotherapy (e.g., psychoanalysis and client- centered therapy), but marriage counseling also in- cludes techniques and procedures which both these and other "schools" consider psychotherapy (p. 12). Ellis (1956) also reflects on this when he says: The best kind of marriage counseling that is now being done usually involves relatively short-term psycho— therapy in a face—to—face situation. . . . For if many or most of us are actually doing psychotherapy, it would certainly seem desirable that we fully recognize that this is what we are doing and that we have no illusions to the contrary. There may be good reasons 35 for continuing to say that we do "marriage counseling" instead of "psychotherapy.” But it is still most im- portant that, whatever the public may prefer to call us, we ourselves realize that we often, and perhaps almost invariably, do psychotherapy. . . . Unless we face this fact squarely, our marriage counseling effectiveness is likely to be minimal: and the herit- age that we pass on to counselors in training is likely to be of dubious value to them, to their po- tential clients, and to the entire marriage counseling profession (Ellis in Ard and Ard, p. 29). It is not the purpose of this research to belabor or expound on the differences of marriage counseling versus in- dividual psychotherapy. The researcher recognizes that there may be differences of opinion among practitioners and other researchers. It does appear unmistakably clear that many factors conducive to success in individual therapy are con- ducive to success also in marital therapy. One of these factors is that the therapist's likability is a key determi- nant to establishing a "working" relationship and eventual therapeutic success. In spite of the limitations of the literature re- garding the specifics of the therapist's likability to a marital setting, some authors do discuss its relevance. Jourard (1971) suggests that there may be a difference in the amount of self-disclosure that married couples make to third parties. Although no differences were found between married and unmarried subjects in total amount of self— disclosure, there was more disclosure to spouse than to any other target-person. Marriage thus appears to have the effect, not of increasing or decreasing the total extent to which subjects disclose themselves, but of producing a redistribution of self-disclosure. The married subjects "concentrated” self-disclosure upon the spouse, and became more reticent toward other persons (p. 9). 36 If this were true with married couples seeking marriage counseling, then it would appear as if the amount of liking of the therapist that is influenced by patient self-disclosure would be reduced. On the other hand, it seems to be a rela— tively safe assumption that with many marital couples seek- ing counseling, one of the primary problems concerns the rela— tive lack of self-disclosure toward each other. That is, although self—disclosure between mates within marriage appears to be the "choice" of married couples, this preference is not necessarily true for married couples who seek professional counseling. Belson (1975) also reflects on the importance of liking the client not only in a dyadic setting but in marital therapy with marriage pairs as well. Although the initial interview with a married couple comprises data collection, Belson believes that it is vitally important that in sub— sequent interviews attitudes of gentleness and "espirit-de- corps" should be fostered. CHAPTER IV METHOD Hypotheses The hypotheses for the study are as follows: Hypothesis 1: Clients who received marital therapy will have a significant, positive correlation between success and liking of the therapist. Liking is measured by ratings on the MCQ (Marriage Counseling Questionnaire). Success is measured by ratings on the MCQ and by changes in the IPR (Individual Problem Rating) from intake to termination (or at the end of six months). Success is also measured by responses on a Therapist Questionnaire completed by the therapist after the client had six months of marriage counseling or at the termination of counseling. I Hypothesis 2: Clients who received marital therapy will have a significant, positive correlation between liking of the therapist and number of counseling sessions. The concept of pr0pinquity essentially states that liking increases with the frequency of contact between persons. However, frequency of contact could very well increase as liking increases. (More discussion of this relationship occurs on p. 21.) Therefore, the variables in Hypothesis 2 are considered res ciprocal in that each continuously affects the other. Those who have more therapy sessions would have more time to develop a 37 38 liking for their therapist and more therapy sessions could lead to an increase in liking. Also, it seems feasible that too much contact may produce discomfort, and a reverse effect may follow. That is, the more often one person has contact with another, the more he will dislike the other. After a certain number of counseling sessions, the client has more time to dislike the therapist. A curvilinear relationship could very well occur with regard to liking and the number of counseling sessions. However, in this research, no curvilinear relationship was found. Persons were categorized according to the number of counseling sessions they attended which occurred weekly or bi-weekly. Only occasionally did appointments occur less frequently than this. Operational Definitions In any research treatise undertaken, it is vitally important to be precise in defining the terms used in the study. Underwood (1957) states that there is a threefold purpose in operationalizing research definitions: 1. Operational thinking removes the fuzz from the empirical concepts of the design. 2. Restrictions are placed on the design. New concepts are not added unless they are clearly differentiated from other concepts used in the research. 3. Operational thinking facilitates communication among scientists because the meaning of concepts so defined is not easily subject to misinterpre— tation (pp. 6—8). 39 At this time, four elements within the hypotheses will be defined: liking, therapist, success, and marriage therapy. Liking can be a total personal reaction on the part of the client to the therapist. For this study, liking of the therapist by the client is defined as definite feelings of caring for or "loving" an individual therapist by indi- viduals within the marital configuaration. These feelings in- clude acceptance, warmth, respect, trustworthiness, integrity, caring, interest, and concern. The client's perception of the therapist as not being cold, stiff, formal, or distant is also regarded as an indicator of liking. More specifically, a series of statements was elicited from clients, asking them to rate their therapist along a continuum in order that the degree of liking as well as the direction of liking could be determined (Appendix C, Questionnaire 2). Therapists are defined as those professionals in the outpatient clinics providing therapy. These include con- sulting psychologists (Ph.D.), psychologists (Ph.D. and M.A.), pastoral counselors (M.A.). and psychiatric social workers (M.S.W.). Seven therapists cooperated in obtaining help from their clients. These therapistsare graduates of "typical" schools that train professionals for the counseling professions. The therapists generally described themselves as eclectic, that is, they use traditional analytic, behavior- istic, Rogerian, and other types of "similar" models in com- binations to bring about change in clients. None of the 40 therapists described themselves as "purists." No therapist said he followed "just Freudian theory" or "just Rogerian theory" or any other specialized modality. Marriage therapy, for this study, is defined as the process of professional, psychotherapeutic intervention with marital pairs for the purpose of aiding the marital pair re— solve satisfactorily the difficulties that pertain to the marital adjustment. Both husband and wife must be involved in the marriage therapy for not less than one session. Joint sessions (husband and wife seen together) or single sessions are considered marriage therapy. Success is defined as that measurement which the clients consider positive outcome (Appendix C, Questionnaires 1 and 2). The clients' individual perception of positive out- come as indicated by submitted questionnaires determining de— grees of positive results were the determinants of success. In addition, therapists were asked to complete a very brief questionnaire as a third measure, in order that their observa- tions could be compared with the observations of the clients (Appendix C, Questionnaire 3). It is recognized that the definition of success, or the degree of success, is very arbitrary, subject to dis- agreement by client and practitioner. For example, Fenichel (1945) believes that a decrease in established neurosis occurs when: 1. The changed situation represents a decrease in the force of the repressed drive. This may be an absolute decrease, as in postclimacteric improvements, or a relat1ve one. 41 2. An increase in the warded-off forces results in a temporary victory of the ego if it succeeds in creating a more energetic and successful suppres- s1on. 3. Many spontaneous improvements occur when there is a decrease in the repressing forces sufficient to lower the entire defensive struggle to a level where it is not disturbing any more. 4. An intensification of the repressed drives creates an improvement if these drives are intense enough to break down the countercathexis altogether. (pp- 549-51). Bandura (1969), a behavorist, declares that success occurs when "behaviors are considered appropriate to the de- sired outcomes. In addition, objectives must often be further delineated by specifying the conditions under which one may expect the behavior to occur" (p. 73). He defines success primarily in terms of modification of behaviors. If behaviors change in the desired direction, success is obtained. Even though this brief description perhaps portrays an oversimpli— fication of the behavioristic position, the behaviorists never- theless view success more in terms of external, observable factors than in terms of internal states. Another definition portrays success in terms of better social functioning. Hollis (1964) states that "the aim of social casework is not only better social functioning, but also im- provement in inner psychological adjustment" (p. 11). She seems to incorporate a behavioristic concern (social behaviors) with an analytic approach (improvement of "internal" adjustment). Consequently, the arbitrary definitions of success are often subjectively determined by "observers" of the client. Any single definition of success is definitely limited unless, somehow, all significant variables are dealt with in measuring 42 outcome. It is possible that if one were to use multiple criteria of success, quite different relationships might be posited. This paper used three measures to reflect success. The first of these is an Individual Problem Rating Scale developed by David F. Gillespie and James R. Seaberg (Appendix C, Questionnaire 1). Gillespie (George Warren Brown School of Social Work) and Seaberg (University of Washington) developed a simple but sophisticated procedure for measuring individual client problems and the change of these problems over a designated time span. Even though this questionnaire has not been used in previous studies, the authors believe it is the type of questionnaire that lends itself to community mental health agencies (Gillespie and Seaberg, 1977, p. 25). In completing the questionnaire, clients list their problems and rate the degree of importance and degree of severity of each problem. The questionnaire was administered twice--at intake and at termination or after clients had received six months of marriage therapy. On page one of this three—page questionnaire, clients list their problems. Carbon paper duplicates this list on the next page. On page 2 the clients rate the Seriousness . of the problem on a scale from O to 100 and the degeee of Importance on a percentile basis, with the total percentile equalling 100%. In discussing the questionnaire, Gillespie and Seaberg write: 43 The Individual Problem Rating Questionnaire is based on two primary assumptions: first, that individuals referred for social services are able to describe the problems that led to referral (self or otherwise) and second, that individuals can distinguish among their problems in terms of the relative importance of each problem to them. IPR requires that individuals do the following: (1) in the course of an intake interview, specify the problems for which they would like help: (2) for each problem listed, rate the severity of the problem with appropriate instructions to treat the distance between each interval as equivalent on a scale of numeric values from 1 for "not at all severe" to 100 for "extremely severe:" and (3) among the set of problems listed assign weights of importance with appropriate instructions to treat the distance between each interval as equivalent as portions of 100 percent which total 100 percent (p. 23). These authors reflect that the IPR has an advantage over other types of questionnaires in that it is based on two reasonably clear criteria: severity of problem and im- portance of problem. Also, the questionnaire is relatively easy to administer, to complete, and to manage mathematically and statistically (pp. 24-5). A basic IPR score is calculated by the following formula: PS PI + PS PI + . . . . + PS PI _ 1 1 2 2 1 1 IPR score - N 100 where PS = severity of problem PI = importance of problem Np = number of problems listed Gillespie and Seaberg believe that Issues of validity and reliability must be approached empirically as well. The validity and reliability of the Individual Problem Ratings are, then, a major con- cern for potential considerations in adopting this measurement procedure. It will be necessary initially to investigate the concurrent validity of the changes observed over time by comparison with an accepted measure of interpersonal and intrapersonal adjustment. 44 Reliability can be determined by a derivative of coefficient alpha appropriate to linear combinations (Nunnally, 1967: 232-235) (p. 27). Consequently, because the IPR has not been used previously, its use is clearly exploratory and results obtained from its use are also exploratory. In addition to completing the IPR, clients were asked to complete the MCQ. which measured the client's perceptions of his or her liking of the therapist and length of time in marriage therapy (Appendix C, Questionnaire 2). The MCQ in- cluded a second measure of marriage-counseling success, be- sides the IPR. In this way the Individual Problem Rating Questionnaire was cross-checked with the Marriage Counseling Questionnaire and the therapist's rating (Appendix C, Question- naire 3) for consistency with regard to marriage-counseling success. The Marriage Counseling Questionnaire is a question-- naire developed by Strupp, Fox, and Lessler (1969). The questionnaire was modified for this research to make it more applicable to a marriage-counseling setting. Primarily, the words "marriage counselor" were substituted for the words "psychotherapist" or "therapist." The first part of the questionnaire measures success: the second the degree of the therapist's liking: and the third, the length of time in therapy. Questions 10, 11, 51, 52, 53, 54, and 55 measure success. Questions 18, 19, 22, 25, 29, 30, 31, 32, 37, 38, and 50 measure the client's liking for the therapist. Questions 2 through 7 measure the length 45 of time in marriage counseling and frequency of sessions. The questions measuring liking of the therapist primarily utilize a five-part scaling procedure whereby the client rates various aspects of the counseling, from a +2 (strongly agree) to a -2 (strongly disagree). In tabulating the data +5 was substituted for +2, +4 for +1, etc. With regard to statistical treatment of the MCQ, Strupp, Fox, and Lessler (1969) state: Ten item clusters were defined and formed the basis of cluster scores. In designing the Questionnaire, we wished to obtain, among other things, a clear picture of the manner in which the respondent viewed his psychotherapy experience. We, therefore, in- cluded a sizable number of items dealing with the patient's attitudes toward the therapist and the de— gree to which he felt respected by the therapist based upon the hypothesis that the "atmosphere" or "therapeutic climate" created by the therapist may play an important part in determining the course and outcome of psychotherapy. We also intended to measure a given attitude from slightly different vantage points. An important ob- jective was to determine empirically whether several items could be grouped under a single heading so that these composites could be used to define and measure a given attitude more reliably. At the same time, we anticipated that certain items--of course, we did not know which ones--woumddbe relatively independent statistically and thus should be retained as single items. . . . The statistical analysis, then was in- tended to refine the original questionnaire and to develop a more precise and incisive instrument that might prove useful in future investigations. The steps followed in the earlier investigations and in the present one included: (1) study of response frequencies for each item: (2) inter-correlations (Pearson's rs) among all structured items: (3) system- atic study of the statistical interrelationships: (4) isolation of item clusters: (5) computation of cluster scores based on items included in each com- posite: and (6) correlations among cluster scores, individual items, and other measures (apart from the Questionnaire) (pp. 193-99). 46 Ten cluster items were used in the original study by Strupp, Fox, and Lessler. Three will be discussed be— cause they pertain particularly to this research. Cluster 1: Therapist's Warmth (five items). This cluster appears to represent a broad estimate of the therapist's attitude during therapy, as the patient recalled it. The major dimensions appeared to be warmth vs. coldness, and closeness (informality) vs. distance (formality). Cluster 2: Therapist's Interest, Integrity, and Respect (six items). This cluster measures the patient's perceptions of the therapist's interest, respect, and acceptance, as well as his trust in the therapist's integrity. This cluster is considered of crucial importance because the attitudes expressed here seem to permeate all other responses throughout the questionnaire. Cluster 3: Amount of Change and Present Adjust- ment (seven items) (pp. 193-94). Clusters 1 and 2 represent client's liking of the therapist and Cluster 3 represents success on the MCQ. Strupp, Fox, and Lessler believe that the MCQ, when used in their research, was reliable when they compared re— sults of the MCQ with independent rating systems. "To study reliability of the MCQ, 20% of the sample's charts, randomly selected, were reviewed by independent raters" (p. 181). These raters compared success as measured by the MCQ with file charts. "Over all success, one of the key variables, was found to be highly correlated. A correlation of .80 (r = .80) was found regarding over all success of treatment" (1% 194)- In the study by this researcher a Split-Half tech- nique was used to measure and document reliability of the MCQ. Questions pertaining to the client's liking of the therapist were randomly divided in a way that was practical 47 while at the same time assuring the researcher of random- ization. Using Pearson's r, a correlation of .86, at the .01 level, was found, indicating a very high correlation of liking scores. Questions pertaining to success on the MCQ were also submitted to a Split-Half technique. Here a Pearson's r corre— lation of .67, at the .01 level, was found again indicating a positive correlation but not to the extent that liking scores were correlated. Validity was also established by the method of com— paring the results of Question 55 of the MCQ, "To what extent have your complaints or symptoms that brought you to marriage counseling changed as a result of treatment?" with the IPR scores. Using Pearson's r, a correlation of .30 (.01 level of sig.) was found, indicating a slightly positive correlation between IPR scores and question 55 of the MCQ. Since there was no way of knowing the clients' true position, validity, in terms of accuracy of the clients' responses with regard to their liking attitudes toward the therapist, is indeed more difficult to document. Only the MCQ measured these responses, and other measures, such as direct interviewing, were not used because of the difficulty in establishing this type of technique. It was not feasible to have anyone question the clients regarding their attitudes toward the therapist because of the confi— dentiality of the counselings Consequently, a major assumption in this research is that the clients' responses were correctly reported. The Split—Half technique established that responses between equally rated items were indeed comprable. 48 A third measure used to ascertain the success of marriage counseling was obtained by a Therapist Questionnaire. Seeman (1954) believes that "not only is the therapist's judgement about outcome a readily available method, but it is a sensible one to use" (p. 99). Therapists were asked to evaluate the results of the marriage counseling in order to compare their responses with client responses. In this way validity of success was substantiated. In addition to the therapist's rating of success, therapists were asked to rate the clients with regard to dependency and transference. This researcher is aware of the ambiguity and overlap in the phenomena of liking, transference, and client dependency. Therefore, a measure was introduced in the Therapist Question- naire in an attempt to separate liking of the therapist from transference and dependency. Therapists were asked to rate clients with regard to dependency at the time therapy began and after therapy was concluded or after six months. Thera- pists were also asked to rate the degree of transference that they believed the clients experienced during the course of the marriage counseling. The scales that were used were comparable to the scales Seeman (1954) introduced to measure similar reactions of clients toward the therapist. The results of the transference and dependency measure were compared with clients' perceptions of their therapist. Multiple regression was used to indicate the separate contributions of transference, dependency, and liking on success. By holding transference and dependency "constant," 49 an accurate measure of the contribution of liking to success could be obtained. It was impossible, within the limits of this study, to identify all of the variables that could con— ceivably contribute to success. However, it was felt that de- pendency on the therapist and transference should be explored because of the ease in which these two variables could be con- fused with liking (Appendix C, Questionnaire 3). It would be difficult to accept the hypothesis that liking was causally associated with success unless liking were found to be associ— ated independently of dependency or transference. Also, since this research is more representative of a survey design than of an experimental design, there are more opportunities for extraneous variables to enter in and influence outcome. It is beyond the scope of this paper to attempt to control or measure all of these possible extraneous variables. However, it is recognized that these extraneous variables could very well effect the results. In outline form, then, the following items were sub- mitted to clients: I. At the Intake Interview A. Brief Letter of Explanation (Appendix A) B. Consent Form (Appendix B) 0. Individual Problem Rating (IPR) (Appendix C, Questionnaire 1) II. At Termination or at Six Months A. Individual Problem Rating B. Marriage Counseling Questionnaire (MCQ) 50 C. Therapist Questionnaire (Appendix C, Question— naire 3) Independent Variables One independent variable in the study is the degree to which the client liked the therapist. This was measured by using the Marriage Counseling Questionnaire (MCQ), which is a modified version of the Strupp, Fox, and Lessler (1969) liking form (Appendix C, Questionnaire 2). A second independent variable was the degree of trans- ference experienced by the client toward the therapist. This was measured by having the marriage counselor rank the degree of transference he believed the clients were experiencing toward him during the course of marriage counseling. In order to obtain some consistency with regard to the counselors' definition of transference, the counselors were asked to use the definition of transference given by Hinsie and Campbell (1970) as a guide when completing that aspect of the question- naire (Appendix C, Questionnaire 3). A third independent variable was the degree of de- pendency experienced by the client toward the therapist. De- pendency on the therapist was rated in two ways. First, the therapist rated what he thought was the degree of dependency when the client began marriage counseling. Second, the thera- pist rated what he thought was the degree of dependency when the client concluded marriage counseling or after six months of treatment. As with transference, an attempt was made to 51 have the therapist use the same criteria of dependency when rating this aspect of the marriage counseling. Therefore, a definition of dependency given by Brammer and Shostrom (1960) was incorporated into the questionnaire as a guide to the therapists when rating dependency. Dependent Variable The dependent variable is the degree of success ex- perienced by the client as a result of marital therapy. This was measured by scores resulting from answers that clients gave on the Individual Problem Rating Questionnaire, the Marriage Counseling Questionnaire, and on the questionnaire completed by the therapist. Reciprocal Variables Regarding Hypothesis 2, the length of time that clients remained in marital therapy, as measured by the number of counseling sessions, is considered a reciprocal variable along with the variable of client liking of the therapist. These two reciprocal variables are considered symmetrical variables. That is, one cannot say which variable is the cause and which is the effect and each variable is considered to be continuously affecting the other. Design and Procedure A panel design was used for the study. A panel de- sign is defined by Kerlinger (1973) as a technique by which "a sample of respondents are selected and interviewed, and then reinterviewed and studied at a later date" (p. 413). 52 The advantage, according to Kerlinger, is that a "great deal of information can be obtained from a large population" (p. 422). In addition, cost can be reduced, and information is generally accurate. (Disadvantages are that these designs do not usually get at "depth" material, high cost could possibly result and not low costs, they are demanding of time, and there could be a sampling error similar to sampling errors that occur with other types of research.) Moser and Kalton (1972) point out that the advantages of the panel design are that there is a relatively minor problem with precision: and it can measure "turnover" changes, that is, change over time can be documented (pp. 138-39). This latter aspect particularly makes the panel technique a suitable strategy for this study where change is measured from the beginning counseling sessions to the final counseling sessions. Post hoc measures were used to measure liking. Be- cause of the way the structure of the research was pr0posed, liking could be measured only "after the fact." Also, post hoc measures were used to ascertain the therapist's evaluation of the client's dependency and of the client's degree of trans— ference, the therapist's rating of success, and success as measured by the MCQ. In this particular research, most of the clients were middle-class persons. Exploration of the epidemiological characteristics of clients being served by the counseling agencies reflects a predominance of white, middle-class Americans. Also, all of the clients who completed the 53 questionnaires listed themselves as Caucasian. Although this in itself is not an indicator of middle-class socio-economic status, the subjective observations by the subjects' therapists confirmed that clients were middle—class persons. Clients requesting counseling at the Christian Counsel— ing Center, Bethany Christian Services, Community Counseling and Personal Growth Ministry, and Human Resources Associates were asked to participate in the study. All of these agencies are located in Grand Rapids, Michigan. They were selected because of their close proximity to one another. Also, these agencies are engaged in marriage counseling, and their adminis- trative personnel indicated a willingness to aid in the re- search. However, the questionnaire responses cannot be con— sidered representative of persons in Grand Rapids who requested marriage counseling for a number of reasons. First, although fifteen therapists from these agencies said they would cooperate in obtaining help from their clients, only seven therapists actually aided in the research. The other eight therapists did not follow through because of "lack of clients" or for other reasons. It was not possible to elicit all of the reasons for lack of follow-through, but apparently a lack of com- mitment by the therapists was one of the reasons. Second, not all clients who were asked to participate in the research did so. Approximately ten couples would not or could not follow through on the project. One male patient stated that it was impossible to put his feelings into words 54 and consequently he did not complete the IPR during the time of the intake interview. Other persons also had difficulty conceptualizing their problems and thus had trouble completing the questionnaires. Third, some couples dropped out of therapy after one or two counseling sessions. Attempts to have them complete questionnaires were futile. They were not interested in co- operating. This researcher suspects that these couples were disappointed in counseling results. Their lack of cooperation in completing questionnaires is regretted, as it is felt that they would have made a significant contribution to the re— search. Because of the lack of cooperation by some therapists and clients, generalizing the results to any parameter cannot be done. That is, it cannot be assumed that these subjects are representative of clients seeking marriage counseling in Grand Rapids. Even though the clients who participated in the research were believed by their therapists to be representative of clients who sought help at the agencies over the past years, the therapists' observations are questionable because of the client drop-out rate. Furthermore, the precision to which this sample accurately reflects the average clientele of these agencies is difficult to document because of the professional confidentiality regarding clients maintained by the therapists. The total number of clients who were asked to participate but did not and of the couples not asked because of unknown cir- cumstances (such as the therapist "forgetting" to ask certain 55 couples) is not known. Even though the total number of couples who did not participate is felt, by this researcher, to be small, it could possibly be considerable--perhaps as many as fifty couples. Client drop-out rate, the failure to ask clients, and the refusal of some clients to be involved in the research could have significantly effected results if they were included in the sample. The "sample" used is that of clients over a period of approximately one year. Clients were included if they initiated counseling some time between October 1978 and Sep- tember 1979. Clients were asked by their therapist during the intake interview if they would be willing to participate in the research project. The study was explained to them orally and in writing. There was no coercion. It was made clear to clients that they did not have to participate and also that they could cease their participation in the research at any time if they desired. Clients signed a consent form (Appendix B) agreeing to participate in the study and agreeing to allow the researcher to use information for research pur- poses. Only married couples in which both husband and wife received marital therapy were included in the study. Married couples were included if they were being seen separately by one therapist or in a combination of joint sessions and sepa- rate sessions. Couples who participated in the study saw only one therapist. Individuals seen were relatively free of severe 56 interpersonal conflicts. According to the separate therapists who were involved in the study, their clients, from a medical- model standpoint, would probably not be labeled psychotic. If they were to be diagnosed medically, the diagnosis would probably indicate psychoneurosis, personality disorder, or perhaps transient situational disturbance. CHAPTER V RESULTS The results of the study are organized into two main sections. The first section discusses Hypothesis 1, the statistical tests used in computing scores, and verification or nonverification of the hypothesis. The second section discusses Hypothesis 2, the sta- tistical tests used in computing scores, and verification or nonverification of the hypothesis. Hypothesis 1-—Success of Marriage Counseling and Liking The first hypothesis stated that clients who received marital therapy will have a significant, positive correlation between success and liking of the therapist. An arbitrary distinction was made regarding liking categories. Liking scores were measured on a continuum. The highest liking score possible was 55. Theoretically, the low— est liking score obtainable was 11. However, the lowest liking score that was given by a client was 41. The highest liking score, given by a number of subjects, was 55. The mean liking score of the eighty subjects was 51.23. Generally, it 57 58 is conceded that clients genuinely seemed to like their marriage counselor. No clients had a "low" liking score in an absolute sense. That is, no client had the lowest liking score possible (11). In fact, no client came near it. However, there were gradations of liking among clients. For this re- search, the actual lowest score of 41 is considered the low score. More specifically, scores between 41 and 45 were con— sidered indicative of low liking. Scores between 42 and 48 were considered indicative of moderate liking, and scores between 49 and 55 indicated high liking. These categories are assigned arbitrarily only for purposes of clarity. Scores ranged from 41 to 55 and were computed as separate integers when compared with success scores. Liking scores were compared to responses reflecting success on the MCQ, the IPR, and Therapist Questionnaire (TQ). Liking scores were considered to be ordinal data. Liking of the therapist was considered from least liking to greatest liking. Also, questions reflecting success On the MCQ, IPR, and TQ were considered to be ordinal data. Kendall's tau was used when comparing liking scores with measures of success on the MCQ, IPR, and TQ. Because this researcher believed that transference and dependency of the client onrthe therapist could be confused with the client's liking of the therapist, transference and dependency scores were measured along with liking in order to obtain a liking-success correlation when dependency and transference were Hhéld constant." Pearson's r was utilized to see whether dependency and 59 transference were correlated with liking. The first de- pendency score, indicating dependency of the client when counseling began, was compared with liking. No significant correlation was found (r = .02, sig. = .421), indicating no correlation between liking and dependency. The second dependency score, indicating dependency of the client at termination or after six months of counseling, using Pearson's r was .04 (sig. = .358), indicating no correlation with de- pendency and liking after treatment was completed. Pearson's r, correlating transference and liking, was .07 (sig. = .249), indicating no significant correlation between transference and liking scores. Each individual question on the MCQ that reflected success, all the success scores combined on the MCQ, success conclusions from the IPR, and success answers on the TQ were analyzed and compared with liking, using Kendall's tau and multiple regression. Question 10 on the MCQ asked the client, "How much have you benefited from your marriage counseling?" The re- sponse that the client could make to this question was: 1. A great deal, 2. A fair amount, 3. To some extent, 4. Very little, or 5. Not at all. Statistical analysis of this question with liking scores found a positive correlation between liking and success as illustrated by Figure 1. Using Kendall's tau, a .29 correlation was found at the .001 level of significance. 6O 55 .p U) or! a. was 0:4 mfi350 $3 E4 .,-1 ::m ...: ..c: has 45. I Not at Very To Some A Fair A Great All Little Extent Amount Deal n = 3 9 13 25 30 N = 80 Mean Liking Scores 53.0 47.3 50.0 51.0 52.7 Figure 1. How much have you benefited from ‘“ i”‘ your marriage counseling? The multiple regression analysis indicates that liking is significantly correlated with success when dependency scores and transference scores are held constant. The standardized regression (Beta) score for liking with Question 10 was .28 (F = 6.84). Table 1 summarizes the analysis among liking, de- pendency, and transference as determined by answers to Question 10. Table 1. Multiple regression analysis of liking, de— pendency, and transference with Question 10. Independent Variables Beta F Liking .28 6.84* 1st Dependency Score .26 2.60* 2nd Dependency Score -.29 3.93* Transference .19 1.50 Constant = 6.23 * Significant at .05 level 61 Question 11 asked the client, "Everything considered, how satisfied are you with the results of your marriage—counsel- ing experience?" The reSponse the client could make to this question was: 1. Extremely satisfied, 2. Highly satisfied, 3. Moderately satisfied, 4. Fairly satisfied, 5. Fairly dis- satisfied, 6. Moderately dissatisfied, or 7. Extremely dis- satisfied. Statistical analysis of Question 11 with liking scores again found a positive correlation between success and liking as illuatrated by Figure 2. Kendall's tau was .21 at the .009 level of significance. Question 11 indicates a lesser positive correlation between liking and success than does Question 10. However, Question 11 also demonstrates that as liking of the therapist increases, success increases. 55 .p m or! 9. $1M ° z: 3:50 l If-I ‘— #:m .H;: 919 451 . . . Ex. Mod. Fa1rly Falrly Mod H1ghly Ex. Dis. Dis. Dissat. Sat. Sat. Sat. Sat. n = 2 3 10 14 17 24 10 N = 80 Mean Liking Scores 55 48.8 49.3 49 52.6 52 52.6 Figure 2. Everything considered, how satisfied are you with the results of your marriage-counseling experience? 62 The multiple regression analysis indicates that liking is significantly correlated with success, as indicated by Question 11, when dependency scores and transference scores are held constant. The standardized regression (Beta) score for liking with Question 11 was .23 (F = 4.70). Table 2 summarizes the analysis among liking, dependency, and trans- ference as determined by answers to Question 11. Table 2. Multiple regression analysis of liking, de- pendency, and transference with Question 11. Independent Variables Beta F Liking .23 4.70* lst Dependency Score .38 5.70* 2nd Dependency Score -.26 3.00* Transference .10 .43 Constant ‘ 7.98 * Significant at .05 level Question 52 of the MCQ also asked the clients about their perception of success. The question was, "How much do you feel you have changed as a result of marriage counseling?" The response that the client could make to this question was: 1. A great deal, 2. A fair amount, 3. Somewhat, or 4. Very little. Once again a positive correlation was found between liking and success as illustrated by Figure 3. Kendall's tau was .23 at the .006 level of significane. 63 55 +3 m ot-l 9. $1M °t w£50 GB or! ::m min rJP 45 Very Somewhat A Fair A Great Little Amount Deal n = 9 16 35 20 N = 80 Mean Liking Scores 48.6 50.6 51.3 52.7 Figure 3. How much do you feel you have changed as a result of marriage counseling? Once again, the multiple regression analysis indicated that liking is significantly correlated with success when de- pendency scores and transference scores are held constant. The Beta score for liking with Question 52 was .27 (F = 6.61). Table 3 summariazes the analysis among liking, dependency, and transference as determined by answers to Question 52. Table 3. Multiple regression analysis of liking, de- pendency, and transference with Question 52. Independent Variables Beta F Liking .27 6.61* 1st Dependency Score .19 1.74 Transference .20 1.87 Constant = 5.71 * Significant at .05 level 64 Question 53 asked the client, "On the whole, how well do you feel you are getting along now?" Seven responses were possible: 1. Extremely well, 2. Very well, 3. Fairly well, 4. Neither well nor poorly, 5. Fairly poorly, 6. Very poorly, and 7. Extremely poorly. Statistical computation of this question does not show a highly significant, positive corre— lation between liking and success. Using Kendall's tau, a correlation of .12 was found at the .088 level of significance. Figure 4 presents a graphic presentation of the relationship of Question 53 with client liking of the therapist. It must be concluded that Question 53 does not show a positive corre- lation between liking and success. 55 .p U) u-l a m m 0 a 2.95 50 u-i M o .r-i ,C: g-p 45 Ex. Very Fairly Neither Fairly Very Ex. Poor Poor Poorly Nor Well Well Well n = 9 6 2 7 28 25 3 N = 80 Mean Liking Scores = 50.5 49.8 50 54 49.9 51.9 55 Figure 4. On the whole, how well do you feel you are getting along now? 65 Multiple regression analysis of Question 53 with liking indicates that liking is not significantly correlated with success when dependency and transference are held con- stant. The Beta score for liking with Question 53 is .10 (F = .92). As with Kendall's tau analysis, multiple regression analysis indicates that there is no positive, significant corre— lation with success, as indicated by Question 53, and the client's liking of the therapist. Table 4 summarizes the multiple regression data. Table 4. Multiple regression analysis of liking, de- pendency, and transference with Question 53. Independent Variables Beta F Liking .10 .92 1st Dependency Score -.21 1.45 2nd Dependency Score .04 .06 Transference .22 i 1.70 Constant = 5.77 No significant F scores at .05 level Question 54 asked, ”How adequately do you feel you are dealing with present problems?" With this question, four responses were possible: 1. Very adequately, 2. Fairly ade- quately, 3. Neither adequately nor inadequately, and 4. Very inadequately. Kendall's tau correlation was .21 at the .013 level of significance as illustrated by Figure 5. As with Question 10, 11, and 52, a significant, positive correlation was found between liking of the therapist and marriage counsel- ing success. 66 55 g - I, .ll .350 m:m 0:. (D am: $264 -.-1 $3245 H49 40 ery Neither Fairly Very Inadequately Nor Adequately Adequately n = 3 9 51 17 N = 80 Mean Liking Scores 43-5 51 50.8 51.2 Figure 5. How adequately do you feel you are dealing with present problems? The multiple regression analysis also indicates that liking is significantly correlated with success, as indicated by Question 54, when dependency scores and transference scores are held constant. The Beta score for liking is .30 (F = 7.66). Table 5 summarizes the analysis among liking, dependency, and transference as determined by answers to Question 54. Table 5. Multiple regression analysis of liking, de- pendency, and transference with Question 54. Independent Variables Beta F I I * L1k1ng .30 7.66 lst Dependency Score -.12 .62 Transference Constant = 4.28 .03 .04 * Significant at .05 level 67 Question 55 asked the client, "To what extent have your complaints or symptoms that brought you to marriage counseling changed as a result of treatment?" In this instance, six responses were possible, but only five re- sponses were used by clients. The six possible responses were: 1. Completely disappeared, 2. Very greatly improved, 3. Considerably improved, 4. Somewhat improved, 5. Not at all improved, and 6. Became worse. None of the respondents stated that their complaints or symptoms completely dis— appeared. Kendall's tau was .20 at the .011 level of sig- nificance as illustrated by Figure 6. 55 .p U) ---| Q: / mam -—— O :3: i 5:: B u-l :10 was R49 45 Became Not at Somewhat ConSid. Very Great Worse All Imp. Improved Improved Improvement n = 12 18 29 17 N = 80 Mean Liking Scores = 49.5 49.5 51.3 50.8 53.2 Figure 6. To what extent have your complaints or symptoms that brought you to marriage counseling changed as a result of treatment? 68 The standardized regression correlation (Beta) was .23 (F = 4.15), indicating that liking was slightly but posi- tively correlated with success on Question 55 when dependency and transference were held constant. Table 6 summarizes the analysis among liking, dependency, and transference as de- termined by answers to Question 55. Table 6. Multiple regression analysis of liking, de— pendency, and transference with Question 55. Independent Variables Beta F Liking .23 4.15* 1st Dependency Score -.06 .14 2nd Dependency Score -.01 .O2 Transference .21 1.80 Constant = 6.76 * Significant at .05 level When the answers to Questions 10, 11, 52, 53, 54, and 55 were summed and compared to liking scores, a correlation of .22 was found at the .004 level of significance using Kendall's tau. The standardized regression correlation (Beta) of these summed responses was .28 (F = 6.81), in- dicating a slight, positive correlation of client's liking of the therapist with questions reflecting success on the MCQ when dependency and transference were held constant. Table 7 summarizes the analysis of liking, dependency, and transference as determined by responses indicating success on the Marriage Counseling Questionnaire. 69 Table 7. Summed responses of Questions 10, 11, 52, 53, 54, and 55 compared to client liking of the therapist when dependency and transference are held constant. Independent Variables Beta F Liking .28 6. 81* 1st Dependency Score .10 .37 2nd Dependency Score -.12 .69 Transference .21 1.85 Constant = 36.74 * Significant at .05 level As a further indicator of success, an IPR Questionnaire was also used (Appendix E, Questionnaire 1). Clients were asked to write down and rate their problems at the time of the intake interview. The clients rated their problem with regard to the Seriousness of the problem and the Importance of re- solving each problem. At termination or after six months of counseling, clients were asked to complete an IPR Questionnaire once again. Page one of the original IPEIWaScreturned to them for reference. This page listed the original problems but not the ratings of the problems. With regard to the listed problems, the percentage of change that took place from the first question- naire to the second was calculated. The percentage of change was then compared to the liking scores. Kendall's tau was found to be .02 at the .374 level of significance, indicating no significant correlation between client liking of the therapist and success as reflected by the IPR. Figure 7 illustrates this relationship graphically. 7O 55 P __. U) oc-I 81 a’///’7”/rt can 0 a: 355° u-l 3:0 ”4: FJP 45 Minus % N0 1 - 33% 34 - 66%II67 - 100% Improve, Change Improve. Improve. Improve. n = 10 17 19 23 11 Mean Liking Scores 51.2 53.6 53.8 50.2 53.2 Figure 7. IPR scores compared with liking scores. The Beta correlation coefficient of IPR scores with liking, when dependency and transference were held constant, was .07 (F = .45). This also indicates that there is no sig— nificant correlation of client liking of the therapist with success as shown by the IPR. Table 8 summarizes the analysis among liking, dependency, and transference as shown by IPR responses. Table 8. Multiple regression analysis of liking, dependency, and transference with success as reported by the IPR. Independent Variables Beta F Liking .07 .45 2nd Dependency Score .27 3.70* Transference -.23 2.62* Constant = —4.57 * Significant at .05 level 71 Finally, liking scores were compared with responses reflecting success as indicated by each marriage counselor. The counselors were asked to evaluate the success of marriage counseling, judged by its results with regard to the husband and wife whom they counseled (Appendix E, Questionnaire 3). The counselors were asked, "How would you evaluate the overall success of your marriage counseling with the husband and the wife whom you counseled?" The five responses possible for the marriage counselor answering the question were: 1. Very great, 2. Fairly great, 3. Moderate, 4. Some, and 5. Very little. Kendall's tau indicates a slight, positive correlation of .18 at the .018 level of significance, between liking and success. Figure 8 depicts this correlation graphically. 55 +3 m -H 9.. sis 054 tw£i50 $2 {—1 or! #10 was prp 45 Very Some Moderate Fairly Very Little Great Great n = 11 19 20 21 9 N = 80 Mean Liking Scores = 48 50.9 52.4 51 53.1 Figure 8. Therapist Questionnaire--Overall success of your marriage counseling? 72 The Beta correlation coefficient of marriage counse- lors' success responses with liking, when dependency and transference were held constant, was .29 (F = 7.15). This indicates that there is a slight, linear, positive corre- lation with client liking of the therapist and success. Table 9 summarizes the analysis among liking, dependency, and transference in regard to success as indicated by the therapist. Table 9. Multiple regression analysis of liking, dependency, and transference with suc- cess on the Therapist Questionnaire. Independent Variables Beta F Liking .29 7.15* 1st Dependency Score .27 2.78* 2nd Dependency Score —.02 .03 Transference -.31 3.86* Constant = 7.40 * Significant at .05 level Hypothesis 1, which stated that significantly more success was attributable to clients who expressed higher liking scores for their therapist than to those who ex- pressed lower liking scores, appears to be confirmed, cautiously. Six of the eight responses that dealt with success (Question 10, 11, 52, 54, and 55 of the MCQ and success as reported on the Therapist Questionnaire), when compared with liking, were significant at the .05 level of significance or lower using Kendall's correlation 73 coefficient. Summed responses from the MCQ that reflected success, when compared with client's liking of the therapist, showed a correlation of .22 (Kendall's tau) at the .004 level of significance. A Beta correlation of .28 (F = 6.80) was obtained from these summed responses, when dependency scores and transference scores were held constant. However, even though a significant .28 (Beta) correlation reflects a positive, linear correlation between liking and success, .28 cannot be considered high. The correlation coefficient of .28 would reflect only a slight, positive linear relationship between liking and success. In addition, the IPR shows a standardized regression correlation (Beta) of .07, which would indicate that the positive correlation of the other responses to the questionnaires also has to be accepted cautiously. Therefore, regarding Hypothesis 1, it is con- cluded that the null hypothesis (there is no difference be— tween success levels and liking levels) is rejected only for the MCQ and TQ. That is, there is a significant difference between liking scores and degree of success as represented by the sample utilized in this research. Hypothesis 2——Length of Marriage Counseling and Liking of the Therapist The second hypothesis stated that the clients who received marital therapy would have a significant, positive correlation between liking of the therapist and number of counseling sessions. As with Hypothesis 1, low-moderate- or high—liking categories are arbitrarily determined. 7L: Liking scores were measured on a continuum with the highest liking score being 55 and the lowest liking score again being 41. For statistical purposes, liking scores and number of interviews were treated as ordinal data. A total of 650 interviews were conducted with 80 clients. The mean number of interviews was 8.12 with a standard deviation of 6.61. Most of the couples were seen separately in noncon— joint therapy sessions. Eighty-three percent of the counseling sessions were separate sessions, with only the husband or wife present. The longest any one person was seen was for 29 sessions (one client) and the least frequent number of sessions was 1 (six clients). Kendall's tau was the statistical measure used to correlate number of sessions with liking scores. Kendall's tau was .05 at the .283 level of significance, showing no correlation between number of sessions and liking of the therapist. Therefore, it is concluded that the null hypo- thesis is not rejected. That is, there is no significant difference in liking of the therapist by clients when num- ber of counseling sessions is compared with liking scores. Figure 9 presents a graphic depiction of the results. 75 55 .p U) or! o a I a: 50 ---l 31m was sup 45 Range of 1-5 6410 11-15 16-20 21-25 26-30 Interviews Number of 36 23 10 6 3 2 Clients Mean Liking 52.1 51.4 51.6 49.6 52.8 54 Scores Figure 9. Range of counseling sessions, number of clients in each range, and mean lik1ng scores. CHAPTER VI DISCUSSION This chapter is divided into five major sections. The first section deals with the explanation of the results, the second discusses possible alternative explanations for the results, the third discusses problems with the research, the fourth discusses implications of the findings, and the fifth discusses suggestions for future research. Explanation of the Results In testing Hypothesis 1, an attempt was made to see if, in a marriage counseling context, there was a significant correlation between the client's liking of the therapist and success as reported by the client and by the therapist. Al- though significant associations were generally found, these associations were low, the mean significant association (Beta) being 28.5 (MCQ and TQ). The fact that there were positive associations seems to support beliefs by other researchers (Carkhuff and Berenson, 1967: Rogers, 1957: Kell and Mueller, 1966: et al.) that liking is an important ingredient in coun— seling. However, apparently none of these researchers completed their research in a primary, marriage-counseling context. Con- sequently, in a strict sense, their research cannot be compared to this research. 76 77 The findings in this research could probably best be described as "suggesting" that liking of the therapist by the client is correlated with success in marriage counseling. As no strong correlations were found, the most that could be said is that liking appears to be of some value with regard to suc- cess in marriage counseling. If success in marriage counseling is to be obtained, a number of variables will probably have to be used in a positive manner to bring about success. Regarding Hypothesis 2, which stated that liking of the therapist would be increased with an increase in the num- ber of counseling sessions, it is apparent that it is not confirmed. The findings indicate that essentially no differ- ence is found in liking of the therapist whether the client has a low number of sessions (one to ten), a moderate number of sessions (eleven to twenty), or a high number of sessions (twenty-one to twenty—nine). The total number of sessions, according to this research in a marriage counseling setting, makes no difference in the degree of the therapist being liked by the client. Alternative Explanations Liking of the therapist by the client could to a de- gree account for marriage-counseling success. However, suc- cess could be explained by other variables. Experience of the counselor, types of therapies, environmental contingencies, and seriousness of the "illness" are but a few of the factors that researchers say contribute to positive or negative outcomes in psychotherapy (Strupp et al., 1977: Ricks, 1974: 78 Malan et al., 1968: Waskow and Parloff, 1975: et al.). Even though Truax (1963) obtained support for his hypothesis that low levels of therapist-offered empathy, warmth, and genuine- ness were predictive of negative outcome, other researchers found these conditions either equivocal or clearly not sig- nificant (Truax and Carkhuff, 1967: Garfield and Bergin, 1971). Consequently, other variables contribute to success besides liking or contribute to success when liking of the therapist is not necessarily "high." In addition to the fact that the therapist's likability may not necessarily be coordinated with success, the high— liking scores that were given by the clients in this research could have been caused by a number of additional factors that may not imply liking. For one, the clients may have been "nice" and attributed higher liking scores to their therapist than they actually felt. Even though clients were not told the primary purpose of the research, they nevertheless may have "guessed" that this is what their therapist was looking for and consequently scored the questionnaires accordingly. Second, clients may not have been answering a question regarding liking, but the questions that implied liking may have in actuality implied other attributes of the therapist, such as the therapist's natural command of respect. Respect and liking are not necessarily the same, for a client may re— spect his or her therapist without liking the therapist. Again, regarding Hypothesis 2, other factors may have 79 accounted for the lack of positive correlation between the number of counseling sessions and liking. The theory of propinquity, as described by Homans (1950), essentially states that other things being equal, interpersonal attraction increases as frequency of contacts increases. Perhaps in a marriage—counseling relationship "other things" are not equal. Other variables may contribute to what appears to be a rather consistently high level of liking, regardless of the number of interviews. Maybe, as Jourard (1971) discussed, self— disclosure from couples to a third party does not take place to the same degree and intensity as it does from one party to another without another person in the relationship. There may be less "intense" self-disclosure by individuals in a marriage dyad to a marriage counselor than to a counselor in psycho- therapy when marriage problems are not necessarily the primary focus of treatment. Consequently, assuming that self-disclosure leads to liking, higher liking scores would not take place in marriage counseling after successive interviews because the degree of self-disclosure may not intensify. Or, perhaps the theory of propinquity needs further examination. Perhaps length and number of interpersonal con- tacts does not lead to liking. A high number of contacts could just as well lead to nonliking or work against higher- liking relationships. A type of "flooding" may take place with an increase in interpersonal contacts that could very well work against liking or, at the very least, not contribute to an intensity of liking. Stagner (1967) stated that indeed, 80 "if an individual is in close proximity to another, with an intensity of separate goals, conflict will take place" (p. 145). A high number of interpersonal contacts will lead to a greater propensity for conflict. Stagner elaborated by stating that conflicts manifested in unmet expectations, polarization of ideas, regulation of social roles, and discrepancy in power could very well be enhanced with increased frequency of dyadic involvement (pp. 144-59). Also, liking may not have actually been measured in the research. A faulty measurement may have described person- ality manifiestations other than liking. If a more precise measuring instrument were to be used, it may well indicate a higher correlation between number of counseling sessions and liking scores. Problems With the Research Occurring with the research were a number of problems that may very well have altered or modified the results. First of all, although eighty subjects (forty couples) completed the questionnaires, an unknown number of persons refused to participate in the research or did not complete questionnaires after they stated that they were willing to do so. Often these clients came for only one or two interviews and they may very well have made a meaningful contribution to the research if they had been willing to participate. Second, a high number of clients reported success or reflected moderate success with the results of treatment. 81 This researcher has no reason to doubt the success of treat- ment: however, a low number of "failures" or cases of "non— success" may have made it difficult to see the effects of failure or nonsuccess on liking scores. Third, as Strupp (1977) points out, "there may be an inappropriate selection of therapists, patients or treatment modalities which compromises the generalizability of results" (p. 27). The fact that the clients were middle-class persons makes generalization to other class groups inapprOpriate. "The hypothesis that the treatment experience, particularly the relationship with the therapist, contributed to an ex— acerbation of the patient's condition cannot be adequately tested in studies in which the therapeutic contact was very limited--mean number of sessions from one to ten sessions" (p. 31). Since the mean number of sessions in this research was eight, it would seem that a setting where the mean was considerably higher would be necessary before the hypotheses in this research could be tested adequately. Fourth, a control group, although difficult to imple- ment, would have been helpful to see whether success was caused by treatment or some other type of variable. "It is essential that the experimenter provide some comparison group if he wishes to assert with any degree of certainty that either adaptive or maladaptive changes were attributable to the psychotherapy experience" (Strupp, 1977, p. 37). Fifth, the length of the questionnaires may have made clients somewhat reluctant to participate in the research. 82 Also, clients may have had some difficulty in completing parts of the questionnaires. For example, some clients had trouble conceptualizing their problems with regard to completing the IPR. The authors of the IPR, GilleSpie and Seaberg, mentioned that some persons might vary in their levels of abstraction and have difficulty articulating problems (GilleSpie and Seaberg, 1977, p. 26). This seems validated in this research. A number of clients, perhaps 10 percent, had difficulty in completing the IPR and needed aid from the therapist or receptionist. This "aid" could very well have distorted the results. Lastly, the fact that the questions regarding success on the MCQ elicited quite different responses perhaps reflects difficulties of the MCQ to measure success accurately. For example, Question 10 asked the clients, "How much have you benefited from your marriage counseling?" This question re- ceived a standardized, significant Beta score of .28 (F = 6.84), whereas Question 53, "On the whole, how well do you feel you are getting along now?" received a nonsignificant, standard- ized Beta score of .11 (F = .925). The discrepancy perhaps can be explained by the fact that these questions do not deal with success in the same way. Question 53 may not take into account that although clients have one set of expectations when they enter counseling regarding success, they nevertheless leave counseling with another set of expectations and criteria for success. For example, couples may enter counseling with 83 "saving the marriage" as their primary desire and goal. How— ever, during the course of counseling their views regarding the saving of the marriage may change and they may no longer see that as an appropriate goal. The questionnaires used in the research may not have adequately addressed the issue of change of goals and the relationship of this change to an accurate measure of success. Only the IPR seemed to address this issue adequately because of its pretherapy test and posttherapy test. Yet the IPR did not correlate (.07 Beta, F = .450) with liking to the degree that Questions 10, 11, 52, 54, and 55 of the MCQ and the TQ correlated with liking. However, the IPR did corre- late slightly with these questions, with the exception of Question 54 of the MCQ. Using Kendall's tau, the IPR had a mean correlation of .26 Cp<< .05) with Questions 10, 11, 52, 53, and 55 of the MCQ and the question regarding success on the Therapist Questionnaire. Regarding Hypothesis 2, one main problem seemed to be the face that not enough counseling interviews were completed to adequately address the issue of a correlation between client liking of the therapist and number of interviews. A higher mean number of counseling sessions than eight would have been preferable. However, this did not occur, and consequently the conclusions regarding the fact that there is no relationship between number of counseling sessions and liking must be accepted tentatively. When further research is completed, it may show a different result. 84 Implications of the Findings The first implication of these findings is that the therapist's likability by the client in marriage counseling plays a significant role with regard to marriage counseling outcome. However, while the correlation is significant, the findings should not be overinterpreted in view of the small correlation (mean = 28.5) between success and client liking of the therapist. An average mean correlation of .28.5 is not reflective of a strong, linear relationship, considering the fact that .00 reflects no relationship and a +1 reflects a perfect positive relationship. The second implication is that a generalization of these results, applying them to a larger group would not be appropriate. The data is descriptive of the subjects in the study and is not inferential. Only middle-class persons par- ticipated in the research and whether we can generalize and declare the data applicable to other middle-class persons is questionable. Since mostly middle—class persons seem to seek marriage counseling (Wallis and Booker, 1958, p. 132) it would seem that the results are applicable to middle-class persons who seek marriage counseling. On the other hand, perhaps the most that can be said is that the subjects and the findings represent middle-class experiences, but we do not know how accurately. Perhaps such generalization would have been more definitive if a more representative sample had been obtained. In addition, the sample was very much dependent on the subjectivity of the therapists with regard to the selection of 85 clients. No control over the selection of clients was possible once the agencies agreed to cooperate in the study. The in- ternal workings of the agencies dictated the sample, and conse— quently accurate generalization to include any larger number could only be speculative. Suggestions for Future Research The first recommendation is that in future research of this sort a more representative sample be obtained. Repli- cation with more social-service agencies and with participation of more therapists in these agencies could make the data infer- ential rather than descriptive. A second recommendation is that the criteria defining both liking of the therapist and marriage-counseling success be further refined. Perhaps a second measure, such as the therapist's perceptions, could be incorporated to measure liking besides the measure used on the MCQ. Also, specific questions could be geared to measure both success and changing attitudes with regard to success as reported by the client over the course of counseling. Perhaps the IPR should be further refined so that the client would be fully aware that changes in his or her goals can be clearly incorporated into the second completion of the IPR at termination of counseling or during successive completions of the IPR while the client remains in counseling. Specific measures would have to be introduced to measure success as well as deterioration effects occurring 86 during the course of counseling. Deterioration effects, or negative results, as described by Strupp, Hadley, and Gomes- Schwartz (p. 12), can occur during counseling and these nega- tive effects may have little or nothing to do with the thera- pist or the counseling relationship. Bergin and Garfield (1971) state that some patients "have already begun to deteriorate and they can't be helped. However, there is another group who have already attained a neurotic equilibrium that is upset by the therapist, resulting in the initiation of a new cycle of deeper deterioration" (p. 249). "A significant proportion of patients experience negative effects from psychotherapy" (Strupp et al., p. 10). Whether positive or negative effects take place in psycho- therapy, Strupp believes that "because of the great com- plexity of the subject matter and the youth of psychotherapy research, we are not permitted to say how these positive or negative effects are achieved" (p. 12). Fischer (1977) also discussed client deterioration and noted that client deterioration can very well take place during the course of professional social work intervention. "In the research of deterioration, MSW level social workers, their clients do worse on one or more measures than clients of nonprofessionals (non—MSWs) or people in no treatment con- trol groups" (p. 245). Dollard, Ellis, Friedman, Garfield, and other comment— ing in Psychotherapy, for Better or Worse (Strupp et al., 1977) conclude that client deterioration or negative effects can and 87 do take place in psychotherapy. The causes, according to these practitioners, include poor training of therapists, dis- illusionment with the therapist, incompetence, client ex- ploitation, and therapist's insensitivity. Future research would have to measure success, nega— tive effects, and their causes with more precision. Also attempts should be made to incorporate measures regarding the qualities or lack of qualities of the therapist and the effect of these qualities on therapeutic results. In conjunction with this, a correlation of liking scores with success, as they pertain to individual therapists who participate in the research project, might be helpful to see if there is a difference between summed responses and individual responses on the part of the clients. Confidentiality, wide variation in number of counselees per therapist, and the fact that this approach could be perceived as "threatening" to therapists may make it difficult to implement this type of research. A third recommendation is that techniques be incorpo— rated to include more accurate knowledge of the failures of those persons who were not interested in participating in the research. This of course would have to be accomplished within the ethical guidelines established by the professional groups of those therapists participating in the research. It is suspected by this researcher that if data concerning knowledge of the persons who did not complete the questionnaires or who did not begin the research, even though their aid was requested, 88 would have been incorporated into the findings, significant differences might have occurred in these findings. Perhaps also additional and prior discussion with agency staff members regarding the study and the approaches to understanding success and liking would have been helpful in further specifying the meaning of these variables. A fourth recommendation is that the design incorporate questions that could differentiate between liking and pro- fessional legitimation of the therapist by the client. The consistent high—liking scores possibly suggest that what is being measured is the professional legitimation of the therapist by the client rather than liking of the therapist as an indi- vidual person. That is, perhaps the therapists, because of their professional role, obtain consistently high-liking scores that in actuality reflect legitimation. Therapists may le- gitimately "deserve" a high—liking score because he or she is a professional person. In a sense, the role that the therapist is in may be liked more than the person in the role. If liking were measured at the beginning stages of marriage counseling, as well as at the end of counseling, it could very well aid the researcher in obtaining a more reliable measure of liking and possibly aid the researcher in distinguishing between legitimation and "real" liking. A fifth recommendation is that techniques be incorpo- rated into the design that will aid the researcher in ascer- taining if success could contribute to liking. Once again, an earlier measure of liking at the outset of counseling, and a measure of liking in different ways may aid the researcher in 89 establishing whether success "leads" to variations in liking scores. A last recommendations is that in future studies, success and liking be defined according to the sex of the client, age, race, nature of the marital problems, and se- verity of the marital problems. Also, when correlating suc— cess and liking scores, it would be helpful to make a dis- tinction between the sex of the therapist and the sex of the client. Differentiation of liking and success of persons according to these varied characteristics could very well show a significantly higher (or lower) correlation between liking and success than was obtained in this research. CHAPTER VII CONCLUSION It has been theorized and demonstrated that there are a number of variables that contribute to successful or nonsuc- cessful psychotherapy and marriage counseling. One variable that appeared to contribute to successful psychotherapy was that of the client's liking of the therapist. However, this variable of the therapist's likability and its relationship to success has never been researched primarily in a marriage-counseling setting. It is to this issue that this research was addressed. A total of forty couples (eighty clients) participated in the research. Four social service agencies, serving mainly middle-class persons, in Grand Rapids, Michigan cooperated in the research. A total of seven therapists from these agen- cies helped in obtaining input from their clients in the filling out of the necessary questionnaires. Clients were asked to complete three questionnaires. At intake an Indi- vidual Problem Rating Questionnaire (IPR) was completed by clients. At termination or after six months of marriage counsel- ing, a second Individual Problem Rating Questionnaire was com— pleted by clients. In addition, clients completed a Marriage Counseling Questionnaire (MCQ) at the time they completed 90 91 the second IPR. A measure of client's liking of the therapist was included in the MCQ and a measure of the total number of counseling sessions was included in the MCQ. At termination or after six months of marriage counseling, the therapists also completed a questionnaire regarding the clients. This questionnaire asked the therapists their perception of suc- cess in the marriage counseling with regard to each person in the marital dyad that they counseled. The first hypothesis of the study attempted to see whether there was any significant correlation between the client's liking of the therapist and marriage counseling success. Results were inconclusive. Comparison of IPR scores, which were one measure of success, with liking scores showed no significant correlation between the two. Comparison of success responses from the MCQ and the Therapist Question- naire with liking showed a significant mean correlation of .28.5 (p‘<:.05). The correlation of .28.5 does not reflect a strong linear, positive correlation between liking scores and success in marriage counseling. The data do suggest a posi- tive correlation between success and liking of the therapist, but clearly additional research will have to be completed in order to confirm or reject this hypothesis conclusively. The second hypothesis stated that clients who had more counseling sessions would show significantly higher liking scores for their therapist than those who had fewer counseling sessions. This hypothesis was not confirmed. Kendall's tau was .05 at the .283 level of significance. This showed that 92 the number of counseling sessions had no effect on liking scores. Clients who had a low number of counseling sessions liked their therapist to the same degree as those clients who had a larger number of counseling sessions (up to twenty- nine counseling sessions). From this research it is apparent that additional refinement of the design is needed and that further testing is needed before one can conclude that liking of the therapist is significantly related to success. Also, additional re- search will have to be completed if one were to conclude that number of counseling sessions was related to liking of the therapist by the client. APPENDICES APPENDIX A EXPLANATION OF QUESTIONNAIRES GIVEN TO RESPONDENTS APPENDIX A EXPLANATION OF QUESTIONNAIRES GIVEN TO RESPONDENTS EXPLANATION OF QUESTIONNAIRE In order that we may be more knowledgeable about the counseling experience and informations exhanged between the counselor and the clients, we would appreciate your help in completing the following questionnaire to the best of your ability. It is important that all the questions be answered as completely and honestly as possible. The questionnaire should only take a few moments and when completed please return it to the receptionist. Enclosed also is a release—of-information form that gives us consent to utilize information for research purposes. All answers will remain confidential. If at any time you wish to withdraw as a participant in this research, you are free to do so. Thank you for your cooperation. Sincerly, Victor K. De Jonge, M.S.W. 93 APPENDIX B CONSENT FORMS 95 CONSENT FORM I, , freely volun- teer to part1cipate 1n a research project designed to evaluate the effects of the counseling experience between the counselor and the client, to be conducted by Victor K. De Jonge, M.S.W., Licensed Marriage Counselor. I understand the research project which has been explained to me, and I have been advised that any questions I have pertaining to the research will be answered. I also give permission to Community Counseling and Personal Growth Ministry, and to Victor K. De Jonge, M.S.W., Licensed Marriage Counselor, to use information pertaining to the Individual Problem Rating Questionnaire, and the Marriage Counseling Questionnaire for research purposes in cooperation with the College of Social Science at Michigan State University. I understand that in no way will I or my family be identified in the results of the study, that my confidentiality will be strictly protected at all times, and that I am free to withdraw as a participant, without penalty, at any time. I understand that there are no beneficial effects of the research for myself, but that the results of the study will be made available to me, at my request, within the restrictions of the research at the conclusion of data collection. Signed Date Witness Date APPENDIX C QUESTIONNAIRES 98 .m: :pps pompsoo 930% wGHHSG mopsp mmH msp p50 HHHM op umxmm mp Haws so» .mHSUmoonm map p0 mopm comm pm cmcp>onm mam mQOHposnpmsp opppomnm .smaponm comm o>aom op 50% op mp pp pQMpsoQSH 30: m: damp .Emanong oso swap macs cmpmpa 50% pp .cnpnp ©s¢ .pp smcpmsoo Sch MMMNMW So: cmpmpa amanoug comm sop m: Hamp .csoomm .mspp psmmmsn map pm p:opm vmssmosoo can 50% mamanosm map pmpa zapmpsn .pmnpm «mmmpm cmnsp mam maize .mSpp mo pepsin m nm>o mmsmno has meoanonm mmmsp So: campwnmuss m5 mama osm mamaponm psommsg mach pmpa op can; insemoonm manepn n mp AmmHv ozHasm smqmomm q¢DQH>HmzH UZHBHQZH H mppmscoppmmso o xHszmm¢ / rvv/ V, I. I fr .1 I u . a. . o i 1. .. ._ . .. . I - \ u . . i . # I v v. . V ~U. t . k I I I I - ' ' - I~ J a .. I .7 . (.Ht...“ ... 4 u u . . . . .. . . . o r . . .. c . I . 1 a I O l: I .M. U ... . .. . U l . l. . f . a.” .u. w a t . a . r .. Q r I . . i .I p) I . a. . : I: p: .1 . . . - ' ._4 ’33 J .1- (l‘ )- r 99 .oppps soplmm ohms Gaoo mmosm omwopm ow mmoo Conme m mspxme on ppps mapppps kaow .soh nonpon pwmsop o: pmzp mEopnopm map pso mwono so mamppopm So: can his 30% .opopmp pmpp mpnp oopopmsoo o>ms 3oz pH .omho map mp pmgp pp .zoe op Mpmp pos mmoo oswnmsn\mpp3 ha: mppps p:®pa no» :msmpponm ommppnws: mapppps po ommpmsp .opnsmxm pom .opnpmmom mm opppommm mm on op has .ppop map pm mispp map so msmpnonm psoz pmpp ommmpm mZOHBoDmBmzH pnpg empponm UZHB¢m Empmomm QHQZH omMpppmz Mo on opmn namppi v L . :: c mzapqomw prod pmpp mmmmhp o i II II I,l I'll: . I r 1’ all \I w. 121.;L; aiunfillfp an. .qHUp m opm pampe so» mpne .non Ado» omop opsoo 50h omSMomn pauppog nap opos oonoopmsoo in pampe nnop apps map:MMm .:mmon app; woman: can :omSomm apps pampp. ompnpp mso pp .opmsmxm pom Keep on opsonm mopr snoopmm pmsop>posp one .&00p map Mo ppma o>pmomp Upcosm .ooGMpnomEp mpp so wspusmmmo .Emp rnopg comm .&00p op m: mspoom mm ompmpp msopponm ppm po moss ppnonep was no espnp .&00p mo puma m amp inopm comm mapsmpmmm an osoo on sac mpne .mapp pso among map pm mHoSpo op oopmmsoo sons mp ampponm zoom map> upommp paupnomep Son ppmp immopm .pQMpp map op mospp map so onppoSppmsH .30» Mom was zmcp msoppmm ho own 30: ms mapppmp .ponsss AsppEpm no pamnmpppo m o>mn UHSoo smpnonm comm .hpmpmpmmmm Eopnopm comm psonn gopnp .Amzoppom zpmeohpxmv 00p OpAmSOppmm ppmlwm pony 0 Soup mp amp unopm,£me mSopnom 30: m: ppmp ppmp mpmpoossp onp p< .ompmpp mempnopm map 90 hmoo m mp ppmp cap map pa mGoppoSppmsH .p muoppom mSoppom spmemnpxm\pp« on poz 00p 0 mmmsmSoppmm pnpq emppocm UZHB¢m Empmomm HHQZH —. 2.1.4 . Questionnaire 2 MARRIAGE COUNSELING QUESTIONNAIRE Initials Date Date of Marriage Sex: Male Female Race Please give the information about your most current marriage-counseling experience. If more than one counselor was involved, limit your answers to the counselor with whom you spent the most time. 1. Date of birth Beginning date of marriage counseling: Mo. Yr. Termination.date Total number of counseling hours Typical frequency of sessions Number of times seen separately Number of times seen together (EN) oxmpmm (a) If you had only one period of marriage counseling, did you ever feel a need for futher marriage counsel- ing? (Check one) Never Very rarely A humber Of times . Often Very often (b) If you felt a need for futher marriage counsel— ing but did not seek it, what were your reasons? 101 192 9. If you have terminated, what led to your termination of marriage counseling? My decision My therapist's decision Mutual agreement External factors (describe briefly) Other (describe briefly) 10. How much have you benefited from your marriage counseling? A great deal A fair amount To some extent Very little Not at all 11. Everything considered, how satisfied are you with the results of your marriage counseling experience? Extremely dissatisfied Moderately dissatisfied Fairly dissatisfied Fairly satisfied Moderately satisfied Highly satisfied Extremely satisfied 12. +2 +2 +2 +2 +2 +2 +2 +2 +2 +2 +2 +2 +2 ' 10.5 Please indicate to what extent each of the following statements describes your marriage—counseling experience. Disregard that at one point or another in marriage counseling you may have felt differently. Use the following code and circle your answer. +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 +1 Strongly agree Mildly agree Undecided Mildly disagree Strongly disagree 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. . My marriage counseling was an intensely emotional experience. My marriage counseling was often a rather painful experience. On the whole, I experienced very little feeling in the course of marriage counseling. There were times when I experienced intense anger toward the marriage counselor. I feel the marriage counselor was rather active most of the time. I am convinced that the marriage counselor respected me as a person. I feel the marriage counselor was genuinely interested in helping me. I often felt I was "just another patient." The marriage counselor was always kindly attentive to what I had to say. The marriage counselor tended to be rather stiff and formal. I felt that he often didn't understand my feelings. I feel he was extremely passive. His general attitude was rather cold and distant. + 2 Strongly agree + 1 Mildly agree 0 Undecided - 1 Mildly disagree - 2 Strongly disagree +2 +1 0 -1 -2 26. +2 +1 0 -1 -2 27. +2 +1 0 -1 -2 28. +2 +1 0 -1 -2 29. +2 +1 0 —1 —2 30. +2 +1 0 -1 -2 32. +2 +1 0 —1 -2 33. +2 +1 0 -1 —2 34. +2 +1 0 —1 -2 35. +2 +1 0 -1 -2 36. +2 +1 0 -1 -2 370 +2 +1 0 -1 -2 38. 104 I often had the feeling that he talked too much. Nothing the marriage counselor said or did ever decreased my self-respect. I would not want to be without marriage counseling experience for anything in the world. I was never sure whether the marriage counselor thought that I was worthwhile person. I had a feeling of absolute trust in the marriage counselor's integrity as a person. I felt there usually was a good deal of warmth in the way he talked. The tone of his statements tended to be rather cold. A major emphasis in treatment was upon my attitudes and feelings about the marriage counselor. A major emphasis in treatment was upon my relationship with people in my current life. I was almost never given any reassurances by the counselor. I had the feeling that the counselor sometimes criticized things I did or said. I usually felt I was fully accepted by the marriage counselor. I never had the slightest doubt about the marriage counselor's interest in helping me. + 2 Strongly agree + 1 Mildly agree 0 Undecided - 1 Mildly disagree - 2 Strongly disagree +2 +1 0 -1 -2 390 +2 +1 0 -1 —2 40. +2 +1 0 -1 -2 41. +2 +1 0 -1 -2 42. +2 +1 0 -1 -2 43. +2 +1 0 -1 -2 44. +2 +1 0 -1 -2 45. +2 +1 0 —1 —2 46. +2 +1 0 -1 —2 47. +2 +1 0 —1 -2 48. +2 +1 0 -1 -2 49. +2 +1 0 -1 -2 50. 105 I was often uncertain about the marriage counselor's real feelings toward me. I feel the emotional experience of marriage counseling was much more important in producing change than intellectual understanding of my problems. I remember very little about the details of the marriage counsel- ing experience. My marriage counselor almost never used technical terms. The marriage counselor often used very abstract language. He very rarely engaged in small talk. The marriage counselor's manner was quite natural and unstudied. A major emphasis in treatment was upon childhood experiences. A major emphasis in treatment was upon gestures, silences, and shifts in my tone of voice, and bodily movements. My marriage counselor showed very little interest in my dreams and fantasies. My marriage counselor stressed intellectual understanding as much as emotional experience. The therapist's manner of speaking seemed rather formal. 106 Please complete these last remaining questions — 51. How severly disturbed did you consider your marriage at the beginning of marriage counseling? Extremely Very much Moderately Somewhat Very slightly disturbed disturbed disturbed disturbed disturbed 52. How much do you feel you have changed as a result of marriage counseling? A great deal A fair amount Somewhat Very little 53. 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 54. How adequately do you feel you are dealing with present problems? Very adequately Fairly adequately Neither adequately nor inadequately Very inadequately cont. 107 55. To what extent have your complaints or symptoms, that brought you to marriage counseling, changed as a result of treatment? ___ Completely disappeared Very greatly improved ___ Considerably improved Somewhat improved Not at all improved Became worse 56. In general, how would you describe your attitude toward the marriage counselor? Finished finally. Thank you for taking time to complete this questionnaire. Your help is sincerly appreciated. Questionnaire 3 THERAPIST QUESTIONNAIRE It is respectfully requested that each therapist complete this very brief questionnaire pertaining to each couple they have seen, and who have participated in this research. Date 1. Your sex-—Male Female 2. How would you characterize the form of counseling you conducted with this couple? Analytic Behavioristic T.A. Gestalt Other (please specify) 3. Questions pertaining to the husband counseled: Husband's initials Very Some Mod- Fairly Very Overall success of your little erate great great marriage counseling. 4. Questions pertaining to the wife counseled: Wife's initials Very Some Mod- Fairly Very Overall success of your little erate great great marriage counseling. 108 B I BL IO GRAPHY 109 In order to obtain some idea of dependency, and transference occurring within the marriage-counseling re— lationship we would like the therapist to answer questions 5, 6, 7, and 8 after reading the definitions preceding questions 5, and 6, and after reading the definitions pre- ceding questions 7, and 8. Dependency, as defined by Brammer and Shostrom (1960), refers to the client who insists that the therapist "take over his decisions, and self-management. . . . The dependent client wishes to prolong therapy. . . looks for support with- out desiring insights. . . and resists taking responsibility for progress in therapy." Keeping this in mind, please answer questions 5, and 6 making a B for where the husband was (question 5) at the beginning of counseling and an E for where the husband was at the end of counseling. Do the same for the wife in question 6. 5. Dependency of husband: 1 2 3 4 5. 6 7 8 9 low 1 high 6. Dependency of wife: 1 2 4 4 5 6 7 8 9 low high Transference, as defined by Hinsie and Campbell (1970), is a "phenomenon of projection of feelings, thoughts, and wishes onto the therapist, who has come to represent the patient's past. . . . These feelings, although once ap- propriate, are inappropriate and anachronistic when ap- lied to objects in the present." Keeping this definition in mind, to what extent would you say the husband's relation— ship with you was indicative of this? Place an X in the apprOpriate box. Do the same for the wife in question 8. 7. Husband's degree of transference: 1 low high 8. Wife's degree of transference: 1 2 3 456 7 82 low | high BIBLIOGRAPHY Ard, B.H., and Ard, C.C., (eds). Handbook of Marriage Counseling. Palo Alto: Science and Behavior Books, 1969. Bandura, A. Principles of Behavior Modification. New York: Holt, Rinehart and Winston, 1969. Belson, R. The importance of the second interview in marriage counseling. Counselinngsychologist, 19759 V01- 5(3): Pp' 27'31- Bergin, A.E., and Garfield, S.L., (eds). Handbook of Psychotherapy and Behavior Change: An Empprical Analysis. New York: John Wiley and Sons, 1971. Bent, R.J., Putnam, D.G., Kiesler, D.J., and Nowicki, S. Correlation of successful and unsuccessful psycho— therapy. Journal of Consulting and Clinical Psychology, 1976, (Feb.), Vol. 44(1), p. 149. Blanck, R., and Blanck, G. Marriagg and Personal Develop- ment. New York: Columbia University Press, 1968. Brammer, L.C., and Shostrom, E.L. Therapeutic Psychology. Englewood Cliffs: Prentice-Hall, 1960. Carkhuff, R.R., and Berenson, G.G. Beyond Counseling and Therapy. New York: Holt, Rinehart and Winston, 1967. Carkhuff, R.R. Helpingyand Human Relations, Chicago: Holt, Rinehart and Winston, Vol. I, 1969. Chaikin, A.L., and Derlega, V.J. Self—Disclosure. University Programs Modular Studies. Morristown: Learning Press, 1974. Collins, M.A. A comparative study evaluating the diff- erential effectiveness of behavior mofification treatment groups and groups with an added component, rational emotive therapy. Unpublished doctoral dissertation, Michigan State University, 1974. Cozby, P.C. Self-Disclosure: A literature review. Psychological Bulletin, 1973, Vol. 79: pp. 73-91. 110 111 Elrod, N. Reflections on my psychotherapeutic practice. Human conteXt, 1973. V01. 5(3). pp. 601-6210 English, 8.0., and Finch, S.M. Introduction to Psyphiatpy. New York: W.W. Norton and Co., 1957. Ethical Principles in the Conduct of Research with Human Participants. Washington: American Psychological Association, 1973. Eysenck, H.J. The effects of psychotherapy. In H.J. Eysenck (ed). Handbook of Abnormal Psychology: An Experimental Approach. New York: Basic Books, 1961, pp. 697-725. Fenichel, O. The Psychoanalytic Theory of Neurosis. New York: W.W. Norton and Co., 1945. Fierman, L.B. (ed). Effective Psychotherapy--The Contribution of Hellmuth Kaiser. New York: The Free Press, 1965. Fischer, J. Isn't casework effective et? Social Work. 1977. (May), Vol- 24(3). pp- 2 5-247- Fischer, J., et al. The relationship between theoretical orientation and therapists' empathy, warmth, and genuineness. Journal of Counseling Psyphology,_ 1975. (Sept-) Vol. 22(5). PP- 399-403- Frank, J.D. Persuasion and Healing. Baltimore: Johns Hopkins University Press, 2nd edition, 1973. Freud, S. A General Introduction to Psychoanalysis. New York: Liverright Publishing, 1920. *Freud, S. Studies on hysteria. (1893-95). Standard Edition, 1955, 2, pp. 1—205. The dynamics of transference. Standard Edition 1912, 12: pp. 97-108, 1958. Fragment of an analysis of a case of hysteria. (1905—1901). Standard Edition, 7: pp. 3-122, 1953. Garrett, A. The worker-client relationship. American Journal of Orthopsychiatry, Vol. XIX, No. 2, Pp. 22u-2380 *The Standard Edition of the Complete Psychological Works of Sigmund Freud, 24 Volumes, Translated and edited by James Strachey. London: Hogart Press and the Institute for Psychoanalysis, 1953. 112 Gillespie, D.F., and Seaberg, J.R. Individual Problem Rating: A proposed scale. Administration in Mental Health, 1977, (Fall/Winter), Vol. 5(1), pp. 21-29. Goldstein, A. The client social worker relationship: transference behavior and interpretation. Unpub- lished doctoral dissertation, Michigan State University, 1974. Goldstein, A.P. Therapist Patient Expectancies in Psycho- therapy. New York: Macmillan Co., 1962. Goz, R. On knowing the therapist as a person. International Journal of Psychoanalytic Psychotherapy, 1975, Vol. 4, pp;'437-58- Grater, H.A., Kell, W.L., and Morse, J. The social service interest: roadblock and road to creativity. Journal of Counseling Psychology, 1961, Vol. 8, pp. 9-120 Greenson, R.R. The Technique and Practice of Psychoanalysis. Vol. 1, New York: International University Press, 1967. Haley, J. Marriage therapy. In Familnyherapy by Erickson, G., and Hogan, T. Belmont, Calif.: Wadsworth Publishing Co., 1970. Heine, R.W. Psychotherapy. Englewood Cliffs: Prentice Hall, 1971. Herbert, W.L., and Jarvis, F.V. A Modern Approach to Counseling. London: Methuen and Co., 1959. Hinsie, L.E., and Campbell, R.J. Psychiatric Dictionary, Fourth Edition. New York: Oxford University Press, 1970. Hollis, F. Casework, A Psychosocial Treatment. New York: Random House, 1964. Casework, A Psychosocial Therapy. New York: Random House, 1972, Second Edition. Homans, G.G. The Human Group. New York: Harcourt, Brace and Co., 1950. Jones, E.E., and Wortman, C. Ingratiation: an attributional approach. University Programs Modular Studies. Morristown: General Learning Press, 1973. Jourard, S.M. Self-Disclosure. New York: John Wiley and Sons, 1971. 113 Kell, W.L., and Mueller, W.J. Impact and Change: A Study of Counseling Relationships. New York: Appleton- Century— Crofts, 1966. Kerlinger, F.N. Foundations of Behavioral Research. Second Ediiion. Chicago: Holt, Rinehart and Winston, 1973. Kiesler, D.J. Some myths of psychotherapy research and the search for a paradigm. Psychological Bulletin, 1966, Vol. 65, No. 2, pp. 110-136. Kiresuk, T. Goal attainment scaling at a county mental health service. Evaluation, Special Monograph, NO- 1. (1973)- Malan, D.H. The outcome problem in psychotherapy research: A historical review. Archives of General Psychiatry, 1973. V01- 29: PP- 719-29- Malan, D.H., Bacal, H.A., Heath, E.S., and Balfour, E.G. A study of psychodynamic changes in untreated patients. British Journal of Psychiatry, 1968, Vol. 114, pp. 525-551. Maslow, A.H. Motivation and Personality. New York: Harper and Row, 1954. Mills, D.H. Liking as a therapist variable in the psycho- therapeutic interaction. Unpublished doctoral dissertation, Michigan State University, 1969. Mills, D.H., and Abeles, N. Counselor needs for affiliation and nurturance as related to liking for clients and counseling process. Journal of Counseling_Psychology, 1965, Vol. 12(4). Moser, C.A. Survey Methods in Social Investigation, New York: Basic Books, 1972. Mullen,.J.A. An investigation of the variable of liking in therapy: its relation to the variable of outcome, empathy, and therapist experience. Unpublished doctoral dissertation. Michigan State University, 1969. Nacht, S. The curative factors in psychoanalysis. Inter- national Journal of Psychoanalysis, 1962, Vol. 4 , No. 206, pp. 2061211. Newcomb, T.M. The prediction of interpersonal attraction. The American Psychologist, 1956, Nov., Vol. 11. 114 Paul, I.H. Letters to Simon. New York: International Press, Inc., 1973. Ricks, D.F. Supershrink: methods of a therapist judged successful on the basis of adult outcomes of adolescent patients. In Life History Research in Psychopathology, Vol. 3, ed. Ricks, D., Roff, M., and Thomas, A. Minneapolis: University of Minnesota Press, 1974, pp. 275—97. Rogers, C.R. A theory of therapy, personality and inter- personal relation, as developed in the client centered framework. In S. Koch (Ed.), Psychology: A Study of a Science, Vol. III, New York: Mc Graw- Hill, 1959, pp. 184-256. Satir, V. Conjoint Family Therapy. Palo Alto: Science and Behavior Books, Inc., 1964. Seaberg, J.R., and Gillespie, D.F. Goal attainment scaling: a critique. Social Work Research and Abstracts, 1977, (Summer), Vol. 13(2), pp. 4—9. Seeman, J. Counselor judgements of therapeutic process and outcome. In Psychotherapy and Personality Change, ed. Rogers, C.R., and Dymond, R. Chicago: University of Chicago Press, 1954, pp. 99-109. Sequin, C.A. Love and psychotherapy: the psychotherapy eros. Acta. Psychotherapy, Psychosomatic., 1962 Vol. 10(3), pp. 173-92. Shaw, M.E., and Costanzo, P.R. Theories of Social Psychology. New York: Mc Graw-Hill, 1970. Singer, E. Key Concepts in Psychotherapy. New York: Random House, 1965. Skinner, B.F. Verbal Behavior. New York: Appleton- Century-Crofts, 1957. Snyder, W.U. Dependency in Psychotherapy. New York: Macmillan, 1963. Stagner, Ross. The Dimensions of Human Conflict. Detroit: Wayne State University Press, 1967. Strupp. H.H., Fox., and Lessler, K. Patients View Their Psychotherapy. Baltimore: John Hopkins Press, 1969. 115 Strupp, H.H., Hadley, S.W., and Comes-Schwartz, B. Psychotherapy for Better or Worse. New York: Jason Aronson, 1977. Thomas, E.J. Field Experiments and Demonstration. In N.A. Polansky (Ed.), Social Work Research. Chicago: University of Chicago Press, 1963. Truax, C.B., and Carkhuff, W. Toward Effective Counseling and Psychotherapy: Training and Practice. Aldine Press: New York, 1967. Truzx, C.B. Effective ingredients in psychotheapy: an approach to unravelling the patient-therapist interaction. Journal of Counseling Psychology. 1963, Vol. 10(37. Underwood, B.J. Psychological Research. New York: Appleton-Century-Crofts, 1957. Wallis, T.H., and Booker, H.S. Marriage Counseling. London: Ruthledge and Kegan Paul, 1958. Waskou, E.I., and Parloff, M.B. (eds.). Psychotherapy Change Measures. Rockville, Maryland: National Institute of Mental Health, 1975. Webb, E.J., Campbell, D.T., Schwartz, D.R., and Sechrest, L. Unobtrusive Measures: Nonreactive Research in the Social Sciences. Chicago: Rand Mc Nally and Co., 1966. Whitaker, C.A., et a1. Rational and non-rational psycho- therapy: a repl . American Journal of Psychotherapy, 1961, V01. 15(4 , pp. 212-220. Whitaker, C.A. Doctor-patient relationship in therapy. American Journal of Psyphoanalysis, Vol. 15(3), 1965.