THERAPEUTIC ALLIANCE AS A MEDIATING FACTOR BETWEEN COUPLE EXPECTANCY AND THERAPEUTIC OUTCOME Except where reference is made to the work of others, the work described in this thesis is my own or was done in collaboration with my advisory committee. This thesis does not include proprietary or classified information. _____________________________ Mikael Alicia Gray Certificate of Approval: _____________________________ __________________________ Thomas A. Smith Scott A. Ketring, Chair Associate Professor Associate Professor Human Development and Family Human Development and Family Studies Studies _____________________________ __________________________ Anthony J. Guarino Stephen L. McFarland Associate Professor Acting Dean Educational Foundations, Graduate School Leadership and Technology THERAPEUTIC ALLIANCE AS A MEDIATING FACTOR BETWEEN COUPLE EXPECTANCY AND THERAPEUTIC OUTCOME Mikael Alicia Gray A Thesis Submitted to the Graduate Faculty of Auburn University in Partial Fulfillment of the Requirements for the Degree of Master of Science Auburn, Alabama August 7, 2006 iii THERAPEUTIC ALLIANCE AS A MEDIATING FACTOR BETWEEN COUPLE EXPECTANCY AND THERAPEUTIC OUTCOME Mikael Alicia Gray Permission is to Auburn University to make copies of this thesis at its discretion, upon request of individuals or institutions and at their expense. The author reserves all publication rights. _____________________ Signature of Author _____________________ Date of Graduation iv VITA Mikael Alicia Gray, daughter of Michael and Rosalyn Gray was born May 6, 1982 in Atlanta, Georgia. She graduated from Fayette County High School in 2000. Upon completion of High School, Mikael attended Auburn University and graduated with a Bachelor of Arts degree in Psychology in May 2004. Immediately following graduation she entered graduate school at Auburn University, in Marriage and Family Therapy in August 2004. v THERAPEUTIC ALLIANCE AS A MEDIATING FACTOR BETWEEN COUPLE EXPECTANCY AND THERAPY OUTCOME Mikael Gray Master of Science, August 7 th 2006 (B.A., Auburn University, 2004) 69 Typed Pages Directed by Scott A. Ketring The purpose of this thesis was to examine how the therapeutic alliance mediated the relationship between couple expectancy for change and relationship satisfaction. The sample was comprised of 59 cases of males and females in committed relationships attending therapy at a marriage and family therapy training clinic at a southeastern university. The mediating effects of the therapeutic alliance were not significant with the current sample. However, statistical analyses revealed that female therapeutic alliance was a significant predictor of male and female change in relationship satisfaction. Results of this study suggest that the strength of the female therapeutic alliance could predict change in male and female satisfaction as early as the fourth session of therapy. The findings of this study propose the need for future exploration of the impact of the therapeutic alliance as a mediator and a moderator. Methodological issues, clinical implications and consideration for future research are addressed. vi ACKNOWLEDGMENTS I would like to thank my Lord and Savior Jesus Christ and my family for making it possible for me to attend school at Auburn University. The support and love of my family and friends is greatly appreciated throughout the completion of my Masters degree. I would also like to express my sincerest appreciation to Scott Ketring, my committee chair, for his guidance and perseverance in the course of this study. I also extend my sincerest gratitude towards Anthony Guarino for his patience, guidance and statistical expertise. Finally, I would like to thank Tom Smith for his guidance and flexibility during the course of this investigation. vii Style manual used: Publication Manual of the American Psychological Association, Fifth Edition. Computer software used: Microsoft Office Word, SPSS viii TABLE OF CONTENTS LIST OF TABLES ix LIST OF FIGURES x INTRODUCTION 1 REVIEW OF LITTERATURE 5 METHODS 23 RESULTS 29 DISCUSSION 38 REFERENCES 48 APPENDICES 55 APPENDIX A: Couple Therapy Alliance Scale 56 APPENDIX B: Revised Dyadic Adjustment Scale 58 ix LIST OF TABLES Table 1: Distribution of Variables for Males 26 Table 2: Distribution of Variables for Females 27 Table 3: Demographics of Participants 32 Table 4: Comparing Significance Tests of Drops Outs and Fourth Session Completers 34 Table 5. Correlations among Study Variables 35 x LIST OF FIGURES Figure 1: Hypothesized Model 4 1 INTRODUCTION Horvath refers to the therapeutic alliance as ?the quality and strength of the collaborative relationship between client and therapist? (Horvath, 2001). The therapeutic alliance has also been recognized as involving a sense of partnership in therapy between therapist and client, in which each participant is actively committed to their specific and appropriate responsibilities in therapy, and believes the other is likewise engaged in the process. Hatcher (1999) reports that the formation of the therapeutic alliance involves not only a sympathetic, accepting disposition toward clients but also the ability to develop a sense of shared commitment to the goals of treatment and the skill of facilitating their active and enthusiastic engagement in the actual work of therapy. It is commonly believed that a positive alliance provides the essential context within which unique therapeutic traditions build bridges that aid clients in negotiating the journey between emotional pain and mental health (Horvath, 2001). Extensive research has been conducted on the therapeutic alliance and its affect on therapy outcomes. There is a broad consensus among clinicians and researchers reporting that the alliance is an essential ingredient in the establishment of the therapeutic process. Horvath and Symonds (1991) concluded that the quality of the alliance, not withstanding some variation in the precise definitions of the concept, is a strong predictor of therapy outcome. Reviews of the research have consistently reported a positive relationship between the therapeutic alliance and outcomes across studies (Horvath and 2 Symonds, 1991; Lambert and Barley, 2001). The magnitude of this relationship appears to be independent of the form of therapy and whether the outcome is assessed from the perspective of the therapist or the client (Horvath, 2001). In looking at the therapeutic alliance and the features that impact its effectiveness, it is important to examine the effect of expectancy on the therapeutic outcome. The first step in examining this effect is to define client expectancy as it pertains to the therapeutic process. Garfield (1994) specifically uses the term client expectancy to describe general expectations clients may have regarding psychotherapy procedures, length of therapy and treatment effectiveness. The general concept of client expectancy is conceptualized by the client?s expectancy of treatment effectiveness or positive or negative outcomes in therapy. For the purposes of this study, client expectancy will be defined as the couple?s expectancy of change in therapy (relationship satisfaction). Goldstein (1962) and Shapiro (1981) have argued that therapeutic gain from therapy is contingent on the client?s expectancy of benefit from therapy. Austin and Vancouver (1996) have also studied how client expectancies can affect positive or negative outcome. According to Austin and Vancouver (1996), clients strive toward a goal as long as they expect that the goal can eventually be attained. Thus, positive expectancies predict persistent effort, whereas negative expectancies lead to disengagement or abandonment of the goal. Following this logic, clients? engagement in therapy depends on their expectations of treatment effectiveness. If they expect that treatment will lead to the desired outcome, clients will engage constructively in therapy, which will in turn have a positive affect on the therapeutic alliance. 3 Limited literature exists relating both therapeutic alliance and client expectancy to therapeutic outcome in therapy. Lambert (1992) in a review of literature identified expectancy as the third most influential of four classes of common factors, after patient factors and factors associated with a positive therapeutic relationship. In a study examining client expectancy, therapeutic alliance and therapy outcome; Joyce and Piper (1998) reported that expectancies predicted the quality of the therapeutic alliance as well as the treatment response. Client expectancy appears to have a significant influence on therapeutic outcome; however, evidence for a direct effect of expectancy on outcome is mixed and inconclusive within the literature (Joyce and Piper, 1998, Meyer et al., 2002). Although these relationships have been studied within individual psychotherapy, there has not been much research that examines the client?s expectancy for change, therapeutic alliance and marital adjustment in couple therapy. While theoretically it would be assumed that the couple?s expectancy would be related to the therapeutic relationship and therapy outcome, the dynamics of two antagonistic individuals working to join divergent views complicates this scenario. The current study investigates the therapeutic alliance as a mediating factor between couple expectancy for change and couple relationship adjustment. It is expected that couple expectancy will be related to change in relationship satisfaction, but that this relationship will be partially or fully mediated by the therapeutic alliance. Although, it should be expected that couple?s expectancy will impact therapy outcomes, it would be equally expected that the therapeutic alliance created in therapy would interact in a way that would mediate the relationship. 4 Hypotheses: Hypothesis 1: The therapeutic alliance measured at the fourth session will have a positive impact on the change in couple relationship satisfaction for males and females. Hypothesis 2: Measures of couple expectancy and the therapeutic alliance will both have significant direct relationships. Hypothesis 3: Therapeutic alliance will mediate the relationship between couple expectancy and male and female relationship satisfaction. Figure 1 Hypothesized Model His Expectancy Therapeutic Alliance Change Her Expectancy 5 REVIEW OF LITERATURE Literature within the field that explores the general therapeutic alliance is robust. However, there are few studies that address the role and efficacy of expectancy, therapy alliance and outcome collectively in the context of couple therapy. The purpose of this review of literature is to examine the current research concerning the relationship between the therapeutic alliance, couple expectancy and therapeutic outcome. In this literature review, the researcher will cover existing literature on the relationship between the therapeutic alliance, client expectancy for change and therapeutic outcome. First, the researcher will summarize literature on the therapeutic alliance and therapeutic outcome. Second, the researcher will speak to client expectancy and its effect on therapeutic alliance and outcome. Literature describing the relationship of the therapeutic alliance in a possible mediating role and the effect of couple expectancy on therapeutic outcome is also discussed. Therapeutic Alliance and Therapy Outcome In this section of the review of literature, the research findings of individual therapy and couple therapy have been summarized. The findings from empirical studies and review articles have been discussed from the individual therapy literature. Barber, Connolly, Christoph, Gladis and Siqueland (2000) examined the relationship between the therapeutic alliance and therapy outcome. In- treatment 6 symptomatic improvement was measured in a group of 86 psychotherapy patients with generalized anxiety disorders, chronic depression and obsessive- compulsive personality disorder. All patients met the primary Diagnostic and Statistical Manual of Mental disorders (3rd ed., rev.; DSM- III-R; American Psychiatric Association, 1987) diagnostic criteria for chronic depression; generalized anxiety disorders, or obsessive compulsive personality disorder and had completed the measures of alliance and depression at sessions 2 or 5. Therapists delivered treatments in one study protocol and had received group supervision by experts in the delivery of the therapy. The California Psychotherapy Alliance Scale (CALPAS; Gaston, 1991); a 21 item self-report, measured the strength of the patient- therapist therapeutic alliance. Patients completed the CALPAS at the end of sessions 2, 5 and 10. The patients also filled out the Beck Depression Inventory (BDI; Beck et al., 1988) at each assessment interview to measure change in therapy. The results indicated that the therapeutic alliance was strongly associated with prior change in symptom improvement at sessions 5 and 10, even when depression was ruled out. Three meta-analytic reviews in individual psychotherapy literature have implicated a consistent relationship between therapeutic alliance and client outcomes. Lambert and Barley (2001) summarized previous research findings and existing literature on the therapeutic relationship and psychotherapy outcome. The pan-theoretical theory on alliance was used as a basis for defining the determinants of the therapeutic alliance in this study. This theory suggests that the alliance is comprised of three major components; tasks, bonds and goals. Although this article does not specify a specific sample or 7 population, it does place emphasis on the general findings of the research studies. The percentages were derived by using a subset of more than 100 studies that provided statistical analyses of the predictors of outcome and averaging the size of the contribution each predictor made to the treatment outcome. Estimates from this article were derived from research findings that spanned extremes in research designs, and are representative of studies that allow the greatest divergence in the variables that determine outcome. The constructs studied in the review were factors that influenced therapeutic alliance and client outcome. Factors that yielded the strongest influence on the client outcome in therapy included social support, expectancy effects, specific therapy techniques and common factors such as empathy, warmth and the therapeutic alliance. Horvath and Symonds (1991) found similar results in their Meta analysis of 24 studies conducted between 1981 and 1990. The sample sized ranged between 8 and 144 different studies reviewed. The quality of the alliance was found to be a robust predictor of therapy outcome. Findings reported suggested that 26% of the variance in the rate of therapeutic success was due to the quality of the therapeutic alliance. Horvath replicated his previous review on the relationship between alliance and individual psychotherapy outcome in 2001. This study included a resulting data set of 90 independent clinical investigations. Using this type of design, the relationship between alliance and outcome was studied across different investigations that used different operational definitions of alliance, alliance and outcome ratings based on different 8 sources (self-report, therapist and observer). Therapy alliance assessments attained at different phases of therapy. Instrumental measures used to measure alliance were as follows: Helping Alliance (HA) ES=.27, Vanderbilt Instrument Scale ES=.24, Working Alliance Inventory (WAI) ES=.24, Therapeutic Alliance Rating Scale (TARS) ES= .17 and the California Psychotherapy Alliance Scale (CALPAS) ES=.17. These five instruments were used in 70% of the studies included in the data set that was studied. Although there was not specific mention of theoretical framework or sampling within this study; the focus was positioned on summarizing the existing literature regarding the relationship between client and therapist factors, alliance and therapy outcome. Horvath (2001) concluded that the therapeutic alliance was a strong variable linking change in therapy to therapeutic outcome with an effect size of .21. It was found that the relationship between the alliance and eventual therapeutic outcome is quite apparent as early as the third session of therapy. This relationship between alliance and the treatment results also seems to hold convincingly constant across various treatments, clinical diagnoses and client populations (Horvath, 2000). Similar findings were found in Martin, Garske and Davis?s (2000) meta- analytic review of the relation of the therapeutic alliance with outcome and other variables. The authors reviewed 79 studies that had been published between 1978 and 1996. Samples used in these studies were mostly outpatient therapy population; however, some patients had more severe symptomology. Most studies included both male and female patients, but the studies failed to break down the alliance ?outcome correlation by gender. Hence, 9 the Meta ?analytic techniques used were unable to differentiate whether patient gender affects the relation of alliance and outcome. A broad collection of instruments were used to measure the therapeutic alliance within these studies. Measures employed included the California Alliance Scale (Marmar, 1989), Working Alliance Inventory (WAI; Bordin 1979), Therapeutic Bond Scales (Saunders et al, 1989) and the Toronto Scales (Marziali, 1984). Martin et al. (2000) reported the average internal consistency of these scales to be .79. All alliance measures were rated by therapists, patients or observers. Therapy outcome was conceptualized in categories of outcome measurements: mood scales, symptom scales, global scales, specific outcome scales and termination status. The overall weighted alliance ?outcome correlation was .22 (n= 68, SD = .12) which demonstrates a significant relationship between the two constructs. The alliance ratings of patients, therapists and observers all had adequate reliability (.79). Interestingly, the patient?s ratings were found to be more consistent and significant than therapist?s or observers? ratings of alliance. Couple Therapy. Despite the fact that many therapists have underlined the importance of the relationship between the client and the therapist to the outcome of treatment, advances in understanding the role of the alliance in the therapeutic process have essentially been confined to individual psychotherapy. However, some studies exist that have found similar results within couple treatment. Pinsof and Catherall (1986) were the forerunners in developing clinical theory delineating the role of the alliance on marital therapy. They suggested that measures of 10 the therapy alliance that had been used in individual psychotherapy were not suitable for use in couple therapy research due to the fact that the therapist must work with two clients at a time. In this case, couples? therapists have the task of developing and maintaining the therapeutic alliance with both the husband and the wife individually as well as the marital dyad. According to Pinsoff and Catherall (1986), the Couples Therapeutic Alliance Scale constitutes the only existing measure designed to empirically asses the alliance in the context of marital therapy. The Couples Therapeutic Alliance Scale (Pinsof & Catherall, 1986) was used to measure the client?s view of the therapeutic relationship and contains two theoretical dimensions: content and interpersonal system. The results were scored as index of global alliance or broken down into six subscales: goals, engagement in tasks, agreement to therapy goals, and human systems of alliance (self, other, group). Brown and O?Leary (2000) investigated the therapeutic relationship as a predictor of success in group therapy for domestic violence in 70 marital couples. The spousal abuse within these couples was defined as the husbands abusing the wives. The conjoint treatment groups were led by a male ?female co-therapist team and comprised no more than 8 couples. The Dyadic Adjustment Scale (DAS) was used to measure global marital adjustment of couples. The Modified CTS (MCTS; Pan, Neidig and O?Leary, 1994) and Psychological Maltreatment of Women Scales (PMWS; Tolman, 1989) were used to measure aggression and psychological maltreatment of women. The Working Alliance Inventory- Observer (WAI-O) was used to measure the therapeutic alliance using observer code alliance rather than self- report. 11 Hierarchical regression analyses revealed that the therapeutic alliance was found to be significantly related to treatment outcome. Levels of psychological and physical aggression decreased as measured by the MCTS and PMWS. Interestingly, this study found that husbands? alliance was a better predictor of outcome than wives? alliance. Bourgeois, Sabourin and Wright (1990) examined the predictive validity of therapeutic alliance in group marital therapy. The objectives of their study were to 1) determine whether couple distress represented a stable predictor of alliance formation and 2) assess whether the quality of the alliance was a predictor of outcome in a group marital training program. This study included 63 couples with a mean age of 38.5. All couples were either legally married or cohabitating couples that had been living together for an average of 13 years. Six senior therapists and seven co-therapists who participated in the study were all licensed psychologists. Couples requested treatment in response to publicity in various media offering communication training which attracted distressed and non-distressed couples. Before the interview, couples completed the Dyadic Adjustment Scale (DAS; Spanier, 1976), the Potential Problem Checklist (PPCL; Patterson, 1976); the Marital Happiness Scale (MHS: Azrin, Naster and Jones 1973); and the Problem Solving Inventory (PSI; Heppner & Petersen, 1982). After the third therapy session, the Couples Therapy Alliance Scales was given to each spouse, while the therapist completed the Therapist Alliance Scale in another room. Each treatment group was led by one therapist and one co therapist and consisted of 4 couples: 2 distressed couples and 2 non-distressed couples. 12 Hierarchical multiple regression analyses were conducted to measure the contribution of the therapeutic alliance, as viewed by the clients and the therapists. The results suggested that the couple?s perspective of the therapeutic alliance was a precursor of treatment outcome. The overall results of this study are consistent with growing evidence in individual psychotherapy research that early development and maintenance of a productive therapeutic alliance is predictive of positive outcome (Bourgeois et. al. 1990). Quinn, Dotson and Jordan (1997), also examined the positive relationship between therapeutic alliance and therapy outcome in couples therapy. The sample consisted of 17 couples who attended a university Marriage and Family therapy clinic. Measures utilized in this study were distributed dependent on the type treatment: CTAS for couple therapy or FTAS, for those seeking family therapy after the third session. Outcome measurement questionnaire statements asked the clients to (a) indicate the extent to which the goals of therapy had been met and (b) rate the extent to which they believed the changes in therapy were made in therapy would continue over the next 6 months. Treatment outcome was measured upon termination of therapy and consisted of a 5 ?point Likert scale questionnaire with ratings ranging from (1) ?not at all? to (5) ?completely?. A positive correlation was found between therapeutic alliance and therapy outcome in couple and family therapy. Interestingly, the results demonstrated a stronger association between therapeutic alliance and outcomes for wives (r=.74) than husbands(r=.56). However, therapy outcomes were found to be higher when the wife?s 13 alliance score was higher than the husbands, as compared to the husband?s outcome score being higher than the wife?s. The relationship between therapeutic alliance and therapy outcome is well -versed in the literature. Results from the previous studies have confirmed that the development and maintenance of the therapeutic relationship is known to be a primary curative component of therapy and provides the context in which specific therapeutic techniques exert their influence (Lambert and Barley, 2001). Taken together, it is safe to conclude that that therapeutic alliance has been found to significantly predict the outcomes in treatment. However, few studies have conducted the necessary procedures to establish a causal link between the alliance and change at the conclusion of therapy. In order for a study to conclude causation with a process variable such as alliance, it must exhibit three conditions. The study must rule out that a third variable is responsible for the changes in outcomes, something that can never be done using correlational data. As such, co variation between alliance and outcomes must be examined. Noting that the alliance accounts partially in predicting outcome, this leaves the possibility of other factors strengthening its relationship with therapeutic outcome. Client Treatment Expectancy, Therapy Alliance and Therapeutic Outcome In seeking to understand the causal link between therapeutic alliance and therapy outcomes, the relevant literature available studying the therapeutic alliance and client expectancy effects in individual and couple therapy is summarized. First, a review of 14 studies on client expectancy and outcome will be presented, followed by specific studies on individual and couple expectancy. Client?s expectations of treatment effectiveness have been found to be powerful predictors of outcome in psychotherapy (Meyer et al., 2002). Glass, Arnkoff and Shapiro (2001) conducted a review of the empirical evidence based on 76 studies of the relation between client expectations and therapy outcome. For the purposes of their review, the authors defined client preferences as similar to that of client expectations, with the exception of the behavior of the therapist or therapy being desired, as opposed to expected. The authors identified three types of client preferences: role preferences, preferences for types of psychotherapy, and preferences for demographic features of the therapist. Preferences were generally measured through factor ?based questionnaires, pre-therapy preference ratings, rank- ordering based on descriptions or videotapes of various therapy orientations, or ratings of treatment after therapy sessions (Glass et al., 2001). Results from these studies demonstrate that client outcome expectancies were positively related to the results of psychotherapy in most of the 24 studies reviewed. The majority of studies showed more positive or mixed findings than negative findings. A significant relationship between expectations and outcome was found in 12 studies ( 9 using self- report measures of outcome ,1 behavior change , 2 with independent clinician ratings, 1 with therapeutic alliance and 1 with a composite score of four types of outcome measures.) The subtotal was greater due to the fact that the several studies used more than one type of outcome measure. Glass et al. (2001) found the prediction of 15 the therapeutic alliance especially interesting, in that client expectancies were found to be the best predictor of the alliance after the first session, above therapist variables, client adjustment and symptoms. A significant relationship between expectations and outcome was found in 19 studies. Fifteen of these studies used continuation in psychotherapy as the measures of outcome, 5 used client self-report measures, 3 used therapy alliance report and 2 used independent clinician ratings. Again, the subtotals were greater due to the fact that the several studies used more than one type of outcome measure. Twelve of the 19 studies could be judged as using poor methodology and measurement such as unreliable reports or interviews with no quantification or consensus analysis of the qualitative data. The logic that clients need a sense of hope that the therapy or therapist can help in order to decide to seek therapy and remain in therapy is supported. Findings suggest that outcome expectations are an important topic for therapists to address with their clients. The authors agree with Garfield (1994), that the therapist?s ability to convey empathy and competence can elevate the client?s hope for change and improve the therapeutic alliance. Individual Therapy. Meyer et al. (2002) utilized prior analyses from the National Institute of Mental Health Treatment of Depression Collaborative Research Program in their examination of the relationship between treatment expectancies, patient alliance and therapy outcome. The authors hypothesized that patient?s expectations of treatment effectiveness would predict active engagement in therapy, which would then account for symptomatic improvement. 16 Participants were outpatients between the ages of 21 and 60 who met diagnostic criteria for major depressive disorder. 151 patients were selected for the study, which had completed 15 weeks of treatment. The BDI was used to assess outcome and the Vanderbilt Therapeutic Alliance Scale (VTAS; Hartley & Strupp, 1983) was used to evaluate the strength of the therapeutic alliance. Patient expectancy was measured with an item administered at the intake session: ?Which of the following best describes your expectations about what is likely to happen as a result of your treatment?? Responses ranged from (1) ?I expect to feel completely better; to (5) ?I don?t expect to feel any different? Patient expectancy was also measured using an expectancy- related item from the TDCRP (Krupnick, Sotsky, Simmons, Moyer, Elkins, Watkins & Pilonis, 1996). The first item asked about the patients? global expectancies: ?How do you think things will most likely be a year from now?? This item was also administered at intake, prior to the first session. Responses ranged from (1) ?absolute bottom- could not have been worse? to (11) ?absolutely tops- could not be better?) on an 11-point scale. Results revealed that patient?s treatment expectancies correlated moderately with their own and with the therapists? global expectancies. Patient?s treatment expectancies were correlated with the alliance measure and with therapy outcome (r = .50). Expectancy, therapeutic alliance and treatment outcome were also examined in short-term individual therapy by Joyce and Piper in 1998. Patient and therapist expectancies regarding the ?typical session? were measured during a controlled trial of short-term, time-limited individual psychotherapy. Patients were matched in pairs on age, 17 and gender, and then assigned to immediate or delayed therapy and to one of eight therapists. Sixty four out of 86 patients were chosen to participate in this study. Patients completed expectancy ratings as a part of the initial outcome assessment and therapists completed expectancy ratings after the second therapy session. The Stiles Session Evaluation Questionnaire (SEQ, 1980) was used to measure patient and therapist expectancy. To represent expectancies at pre-therapy and early therapy, respectively, the patient and therapist rated the SEQ items in response to the sentence stem, ?the typical therapy session will be?? Patients and therapists completed the SEQ form after each session. The two session evaluation scores were totaled across all sessions for each participant. The therapeutic alliance was measured by independently rated six 7 point items. Luborsky?s Concept of the Helping Alliance (1984) was used as the alliance measurement tool. The therapists rated four ?immediate? items after each therapy session, and two ?reflective? items were rated after each one-third of the therapy (at sessions 7, 14 and 20). Each of the six item ratings was totaled across sessions and subjected to a principal component analysis. One patient- rated alliance factor and two therapist ?rated alliance factors (immediate, reflective) were calculated. Patient, therapist and assessor ratings of target distress were included in the outcome battery. Three hierarchical regression analyses were conducted in examining expectancy and alliance as joint predictors of outcome. For the first outcome factor (general symptoms and dysfunction), the predictors were Quality of Object Relations (QOR), each of the alliance variables, and patient- expected comfort. QOR accounted for 7% of 18 outcome variance, however, when alliance and expectancy were in the equation, the direct effect of QOR was no longer significant. Alliance accounted for 7% to 13% of outcome variance; as each alliance variable provided for significant prediction in the regression. The patient expectancy rating, when included last, accounted for an additional 6 % to 14 % of outcome variance and was also a significant predictor in each analysis. Multivariate analyses studying the patient and therapist expectancy ratings as potential predictors of the therapeutic alliance and treatment outcome demonstrated that expectancies regarding the experience of therapy sessions were strongly and directly related to the quality of the therapeutic alliance. Relationships between expectancy and outcome were less strong, yet substantial. Significant relationships between expectancy and alliance were found. Expectancies of session usefulness for both clients and therapists were directly associated with the strength of the respective alliance ratings. Expectancy ?outcome relationships were notably smaller in absolute value than expectancy- alliance relationships. This discrepancy suggests that expectancies may have more direct effects on the establishment of the therapeutic alliance than on the actual outcome of treatment. Taken together, the results of these additional analyses suggest two conclusions. First, patient expectancies may be strong predictors of therapy outcome; however, therapist expectancies may not. Second, the patient?s capacity for a good relationship, expectancy that the therapy sessions will be valuable, and the actual experience of a strong therapeutic alliance all represent consistently strong determinants of therapy benefit (Joyce and Piper, 1998). 19 Couple Therapy Al-Darmaki and Kivlighan (1993) studied the relationship between congruence in client- counselor expectations for their relationship and the working alliance. In this study the focus was on relationship expectations defined as the client expecting to (or the counselor expecting that the client will) spontaneously self-disclose in the context of a comfortable relationship with the counselor. The hypotheses studied included (a) higher levels of client or counselor expectations for relationship would be related to higher working alliance ratings; and (b) congruence in client and counselor relationship expectations would predict ratings of the working alliance after the effects of client and counselor relationship expectations had been controlled. This study consisted of 25 counselors ?client dyads (19 women and 6 men) in the counseling center of a large midwestern university. Of the clients, 18 were women and 7 were men. Their ages ranged from 18 to 30 years old with the M=21.76. Twenty four were Caucasian and 1 was a minority (unspecified). The measures used in this study are client and counselor expectations of behavior in counseling and the working alliance. The PEI-R, Revised Psychotherapy Expectancy Inventory (Berzins, 1971) was used to measure the client-counselor expectancy. The PEI- R was considered reliable; with a coefficient alpha of .87.Test-retest coefficients obtained from clients and counselors were .68 within a 1-week interval and .76 within a 4-week interval. The Working Alliance Inventory (WAI: Horvath and Greenberg, 1989) was used to measure the therapy alliance. Counselors volunteered to participate in this study. After agreeing to participate, they were asked to recruit one of their clients to participate as 20 well. Participants were identified by a code number and instructed to complete the PEI-R (Berzins, 1971) and the WAI (Horvath & Greenberg, 1989) after their third counseling session. The research design in this study incorporated correlational analyses and six multiple regression analyses. Counselor-rated agreement on goals was significantly related to client-rated agreement on tasks and goals. Clients? expectations for relationship were significantly related to all aspects of client-rated alliance, but not to any aspect of counselor-rated alliance. Counselors? expectations for relationship were significantly related to all aspects of counselor-rated alliance, but not to any aspect of client rated alliance. This study also found that congruence in client and counselor expectations that the client will self-disclose in the context of a comfortable relationship accounted for a significant amount of variance in counselor and client ratings of working alliance after the effects of counselor expectations for relationship were controlled. As stated earlier, Abouguendia, Joyce, Piper and Ogrodiniczuk (2004) indicated that the therapeutic alliance served as a mediator of expectancy effects in group therapy. In this study the authors use the mediation model as the primary theoretical base. This mediation model held that the effect of patient outcome expectancies is expressed through the therapeutic alliance and has an indirect influence on outcome in couple therapy. After meeting inclusion criteria for complicated grief, 107 patients were randomly assigned to therapists in the study. 77% were women and 23% were men. The therapist group consisted of one psychologist, one female social worker and one female 21 occupational therapist. The psychologist conducted four therapy groups, and the other two therapists conducted six groups each. The constructs used in this study were expectancy and therapeutic alliance. A pre- therapy rating of expected improvement averaged across objectives and served as a measure of patient outcome expectancy. An 11 point Likert ?Scale was used to measure the patient rated expected improvement variable. The Cronbach?s alpha coefficient was .75 indicating moderate reliability. After each of the 12 group sessions, the patient and therapist each rated four items on a 7 point Likert- type scale ranging from very little (1) to very much (7), which measured the therapeutic alliance. High internal consistency of the alliance rating items was demonstrated by Cronbach?s alpha rating of .91 for the patient-rated alliance and .92 for the therapist?rated alliance. Assessment of outcomes included 14 measures (questionnaire and interview) that covered 15 variables in the areas of grief symptoms, interpersonal distress, social role functioning, psychiatric symptoms, self esteem, life satisfaction, and physical functioning. The data from this study was analyzed using hierarchical linear modeling approach. The analysis addressed the relationships among expectancy, alliance and outcome at the group level that accounted for variation in the relationships at the level of individual patients within groups. At both the individual and group levels, patient outcome expectancy ratings were found to be directly associated with improvement on two of the three outcome factors (general symptoms, target objectives and life satisfaction). In looking at the therapeutic alliance, patient ratings of the quality of the collaborative relationship with the therapist were found to be associated with benefit on 22 all three outcome factors at the individual and group levels. However, the therapist- rated alliance was found to be associated with only a single outcome factor at the individual level, and this relationship was less obvious. Therapist perceptions of the alliance were not significantly related to patient outcome expectancy. Conversely, the patient-rated therapeutic alliance played a mediating role accounting for substantial amounts of the direct effect of patient expectancy on the outcome factors. Results also revealed that the therapeutic alliance served as a curative factor-both directly and as a mediator in group therapy. These findings also showed that the alliance represented a central mechanism of therapeutic change, which operated in a similar way across different types of treatments and orientations. The findings regarding the role of the therapeutic alliance as a mediator of expectancy effects in group therapy served as a cross-validation of the recent findings for individual therapy. (Abouguendia et al., 2004) 23 METHODS This study attempts to better understand how the role of the therapeutic alliance interacts with couple expectancy of therapeutic change and therapy outcome as measured by relationship satisfaction in couple treatment. The methods of this study will be presented in the following order. First, the data collection procedures for this study will be presented. Second, each measure including internal consistencies will be discussed. Third, the distributions of variables for males and females in committed relationships will be presented. Fourth, this section will conclude with the researcher?s plan of analysis. Procedure Data from the Auburn University Marriage and Family therapy Center was used (MFT Center files) of marital couples who received therapy services and whose file been closed between the years of (March) 2004 and (May) 2006. The self-report questionnaires used in this study were compiled by members of the MFT Center faculty for clinical, administrative, and research purposes. Participants completed intake forms at the beginning of therapy. Follow- up paperwork was completed at the 4th session. These forms and reports assessed the client?s rate of expectancy for therapy outcome (Item # 21 on the Process of Change Scale of the AUMFT Adult in Committed Relationship Intake Paperwork), the therapeutic alliance 24 (CTAS), and the change in relationship satisfaction (RDAS). All clients were informed of the purposes of survey completion at the beginning of therapy and signed agreements to release information for clinic sponsored research. The current data utilized for this research project came from confidential data which was transformed into an anonymous data set. Measures Couple Therapeutic Alliance (See Appendix A). The Couple Therapeutic Alliance Scale is a 40 item scale, used to assess and measure the therapeutic alliance. This 7 point Likert- scale is a self report measure designed to assess the client?s perception of their relationship with the therapist (CTAS; Pinsof & Catherall, 1986). The CTAS is comprised of three sub scales: bonds, tasks, and goals. Statements such as ?The therapist cares about me as a person? and ?I trust the therapist? can be found in the bonds subscale. The tasks subscale includes statements such as ?The therapist is helping my partner and me with our relationship? and ?The therapist is not helping me?. Items in the goals subscale include: ?The therapist understands my goals for therapy? and ?The therapist is in agreement with the goals that my partner and I have for ourselves as a couple in this therapy?. Responses range from (1) ?completely disagree? to (7) ?completely agree? with (4) being a ?neutral? response. The CTAS requires reverse scoring on half of the items on each subscale, and then a sum is taken of all scores for a final total score. The test-retest reliability for this measure is reported to be r = .84 (Pinsof & Catheral, 1986). Heatherton and Friedlander (1990) examined the internal consistency of 25 the scale and report an alpha of .93 for the total score. Alpha levels for the bonds, tasks, and goals sub-scales are .85, .88, and .70. Content validity is the only form of validity that has been established for this scale as reported by Pinsoff and Catherall (1986). The internal consistency for the sample of this study was ? = .96 for males and ? = .95 for females. Couple Expectancy Male and female expectancy of the clinic?s ability to treat their problems will be measured by the Auburn University Family Therapy Center Adult in Committed Relationship Intake Paperwork (See Appendix B). This form includes a 32 item Process of Change scale which asks questions such as ?I think I might be ready for some self- improvement?? and ?It might be worthwhile to work on my problem?? Expectancy that the clinic is able to treat the couple?s problems includes items labeled as: I?m hoping that this place will help me to better understand myself?? and ?I hope that someone here will have some good advice for me.? For the purposes of this study, client expectancy will be measured by item number 21; ?Maybe this place will be able to help me.? Scores for the AUMFT center follow up form range from (1) ?strongly disagree? to (5) ?strongly agree?. Therapy Outcome: Relationship Satisfaction (See Appendix B). The Revised Dyadic Adjustment Scale (RDAS; Busby, Crane, Larson, & Christiansen, 1995) is used to measure the dyadic adjustment in distressed and non- 26 distressed couples. The RDAS consists of three subscales: the Dyadic Consensus Subscale, the Dyadic Satisfaction Subscale, and the Dyadic Cohesion Subscale. This scale is brief, only consisting of 14 items; 18 items less than the original Dyadic Adjustment Scale (DAS; Spanier, 1976). The scores for the RDAS range from 0 to 69 with higher scores representing better relationship adjustment. Construct validity levels are found to be acceptable with this scale (Chi square = 78.73 (56, p=.024) and the GFI was .95. Other advantages of the RDAS include adequate internal consistency (Cronbach?s alpha, .90) and excellent split-half reliability (Busby et al., .95). The reality coefficients for Dyadic Consensus, Satisfaction and Cohesion subscales are reported to be .89, .88 and .80 respectively (Busby et al., 1995). In addition to this, the RDAS demonstrates multidimensionality, strong correlation to other reliable and commonly used marital adjustment scales, and the precision in discriminating between distressed and non distressed couples. Internal consistency for this sample was ? = .87 for males and ? = .86 for females. Distributions and Transformations of Data The distributions of all variables for females and males were examined to verify that each one exhibited normal distribution. All measures in this study were normally distributed with minimal skewness. (See Tables 1 and 2). Table 1. Distributions of Variables for Males Expectancy RDAS 1 RDAS 4 Therapeutic Alliance Mean 3.9 36.3 42.5 220.84 SD 0.71 9.3 9.3 37.5 Skewness -0.16 0.3 -0.29 -0.3 Kurtosis 0.457 0.412 2.78 -0.46 27 Table 2. Distributions of Variables for Females Expectancy RDAS 1 RDAS 4 Therapeutic Alliance Mean 4.1 32.8 41.1 222.71 SD 0.61 10.1 10.2 33.9 Skewness -0.315 -0.033 -0.609 -0.106 Kurtosis -0.596 -0.335 0.103 -0.771 Approach to Data Analysis The hypothetical model of this study proposes that the therapeutic alliance acts as a mediating factor between couple expectancy and relationship satisfaction. The independent variable couple expectancy is measured at time one and the possible mediating variable, therapy alliance is measured at the fourth session. The dependent variable is designated as change in relationship satisfaction. To create this variable, the researchers used the fourth session relationship satisfaction scores after controlling for the level of satisfaction at intake. By taking into account the change in the couple relationship then the relatedness with independent variables takes on added significance. Baron and Kenny (1986) define a mediator as an intervening variable which accounts for the relationship between the predictor and the outcome. For this study, in order for therapeutic alliance to be considered a mediator, the variable must account for the relationship between male and female expectancy and couple relationship satisfaction. In testing the potential mediator hypothesis, the recommendations of Baron and Kenny (1986) will be followed. Potential mediation is determined by a series of three analyses: a) male and female expectancy (IV) must be correlated with change in 28 relationship satisfaction (DV) b) male/female expectancy (IV) must be correlated with male/female therapy alliance (M) and c) male/female therapy alliance (M) must be correlated with change in relationship satisfaction (DV). If the criteria of all three analyses are met, it will be possible to test whether the effect of the male/ females expectancy is significantly reduced by controlling for the therapeutic alliance. In conclusion, the possible model to be proposed is that the therapeutic alliance will act as a mediating variable explaining the impact of couple expectancy at the beginning of therapy on change in relationship satisfaction for males and females. The objective with this analysis is to determine if the hypothesized relationship is consistent with the data we have and determine the strength of the association. 29 RESULTS This study investigated the possible mediating effect of the therapeutic alliance between couple expectancy and therapeutic outcome in couple therapy. The therapeutic alliance was measured using the Couple Therapy Alliance Scale (CTAS; Pinsof & Catherall, 1986). Change in scores on the Revised Dyadic Adjustment Scale (RDAS; Busby Crane, Christensen, Larson, 1995) from session one to session four was used to measure the outcome in couple therapy. Couple expectancy was measured by item number twenty- one on the Process of Change Scale within the Auburn University Family Therapy Center Adult in Committed Relationship Intake form. The results of this study will be presented in the following order. First, the participants? demographic background will be presented. Second, Analysis of attrition will be discussed. Third, the analysis of the variable correlations will be presented. Fourth, the results of the individual hypotheses will be presented. Finally, concluding remarks will be made regarding findings. Participants The sample for the present study included clients and therapists from the Auburn University Marriage and Family Therapy Center. The Auburn University Marriage and Family Therapy center was accredited by the American Association for Marriage and Family Therapy?s Commission on Accreditation for Marriage and Family Therapy 30 Education. Therapy conducted at the MFT center is regularly live supervised by the MFT faculty, and observation and/or team consultation by other therapists in training. One hundred and thirty two individual men and women in committed relationships participated in therapy at the Auburn University MFT training center from 2004-2006. For the current study, a dropout would be considered someone who completed intake paperwork but failed to complete fourth session paperwork. Out of the 132 cases, 16 terminated therapy services before the fourth session. Of the remaining cases, 31 males and 26 females failed to complete the required paperwork. Of these remaining cases, 47% of males and 56% of females completed fourth session paperwork. 52% of males and 44 % of females did not complete fourth session paperwork, which yields a 48 % retention rate among males and a 56% retention rate among females in committed relationships. Therefore, this study included 59 cases of males and females in committed relationships. Participant Demographics Of the cases in this study, the age of participants ranged from 19 to 59 years. Forty?one European American females (83.7%), 40 European American males (75.5%), 7 African American females (14.3%), 9 African American males (17.0 %), 1 Hispanic /non- White females (2.0%), 1 Hispanic/ non-white males (1.9%), participated in this study. The remaining 5.7 % of males identified in the ?other? category. Client annual income ranged from less than $ 5,000 to $40,000+ annually. The under $10,000 annual income category contained 22.7 % of the females, and 14.3 % of 31 the males. Approximately 11.4 % of the females and 19.6 % of the males reported in the $10,000 to $20,000 range. The $20,000 to $30,000 category contained 11.3 % of the females and 21.4 % of the males. The $30,000 to $40,000 range 17.9 % of the females and 17.9 % of the males. The remaining 36.4% of the females and 26.8 % of the males reported in the $40,000+ category. The education of the clients ranged from the completion of grade school to the completion of advanced degrees. Of those who had completed information concerning education, 30% of the females and 26.8 % of males graduated from high school or received a GED, 20% of females and 19.6 % of males had completed a technical or an associates degree, 28 % of females and 32.1 % of males completed a bachelors degree, 12% of females and 8.9 % of males completed a masters degree, and 10 % of females and 12.5 % of males indicated an education of ?other?. See Table 3. 32 Table 3.Demographics of Individual Participants Males Females Racial/Ethnic Group NPercent N Percent White/Non Hispanic 40 75.5 41 83.7 African American 9 17.0 7 11.9 Hispanic/Non White 1 1.9 5 2.0 Other 3 5.7 0 0 Education Level Completed NPercent N Percent Junior High 0 0 0 0 High School/GED 15 26.8 15 30 Tech/Assoc Degree 11 19.6 10 20 Bachelors Degree 18 32.1 14 28.0 Masters Degree 5 8.9 6 12.0 Other 7 12.6 5 10.0 Income NPercent N Percent Under $10,000 8 14.3 8 22.7 $10,000 -$20,000 11 19.6 5 11.4 $20,001-$30,000 12 21.4 5 11.3 $30,001-$40,000 10 17.9 10 22.7 Over $40,000 15 26.8 16 36.4 Age NPercent N Percent 19-29 22 38.6 46 51 30-39 20 35.1 18 35.3 40-49 12 21 4 7.8 50+ 3 5.3 3 5.9 33 Attrition of Participants Attrition within research studies is important to examine due to the fact that there could be a difference among those couples who completed the study and those who did not. Measuring and understanding this difference could explain and prevent potential threats to the validity of this study. For this study drop out was defined as someone who completed initial paperwork and four sessions of therapy, but failed to complete the fourth session paperwork. Using this definition 52% of males and 44% of females dropped out of the study. The main reasons for dropout are: participants who failed to complete fourth session paperwork, participants who moved to another area and therapists? non- compliance with procedures. A Chi- square analysis was used to test for differences between study completers and non-study completers compared across the demographic variables of race, education and income. Non ?study completers were characterized by comparing those who did complete four sessions and those who did not complete four sessions. The categories within race, education and income were collapsed in order to meet the requirements of the Chi-square test. The non-dropouts were not statistically different from the drop outs of the study on the variables of race, education and income. Results for the Male demographic variables include: education ? 2 (3) = 4.77, p = .18, race ? 2 (1) = .041, p = .84, income ? 2 (4) = 1.86, p = .76. Chi ?square results for females include: education ? 2 (3) = 1.60, p = .66, race ? 2 (1) = 1.58, p = .21, income ? 2 (4) = 4.69, p = .32. None of the analyses yielded significant results. 34 T-tests were used to test for a difference between participants remaining in the study and those who dropped out on the expectancy pre-therapy scores and marital adjustment pre therapy scores before the fourth session. No significant difference was found between couples who remained in therapy longer than four sessions as compared to those who dropped out before completing four sessions. T-scores have been reported in Table 4 below. No attrition bias was found in this study. Table 4. Comparing Significance Tests of Male and Female Drop-outs and Fourth Session Completers. Males Females Sig. Sig. t-score 2-tailed t-score 2-tailed Age 0.11 0.92 0.38 0.70 Intake Expectancy -0.76 0.45 0.09 0.93 Intake Relationship Satisfaction -0.13 0.21 1.60 -1.10 * p< .05 level; ** p, .01; *** p< .001 Correlation Analysis Pearson?s correlation coefficients were calculated to examine the relationships between male/female expectancy for change, therapy alliance and change in relationship satisfaction for males and females. Table 5 presents the correlations of the variables available for the mediation analysis. Correlation Results for Males and Females Therapy Alliance None of the male correlations measured in this model were significantly related with any of the variables of interest. As expected, the female therapy alliance was related (r = .46, p < .01) to male therapy alliance. In addition, female?s expectancy scores were positively related to both male expectancy scores (r =. 27 p < .01) and female therapy alliance scores (r = .36 p< .01). Most importantly, female therapy alliance was found to 35 be significantly related to male and female change in relationship satisfaction (r = .37 p < .01) and (r = .31 p < .01). See Table 5. Table 5. Correlations among Study Variables Variables 1 2 3 4 5 6 1. Male Expectancy 1.000 2.Female Expectancy 0.274** 1.000 3.Male Therapeutic Alliance 0.211 -0.160 1.000 4.Female Therapy Alliance 0.255 0.358** 0.455** 1.000 5.Male Relationship Sat. -0.037 -0.084 0.244 0.365** 1.000 6.Female Relationship Sat. 0.030 0.135 0.079 0.306** 0.133 1.000 * p < .05 ** p < .01, Test of Potential Mediating Relationship The following variables needed to be correlated in order to qualify for possible mediation: male/female expectancy, therapy alliance and male and female change in relationship satisfaction. Male/female expectancy (independent variable) and change in relationship satisfaction (dependent variable) were not found to be significantly related. Male/female expectancy (independent variable) and therapy alliance (mediating variable) also yielded non-significant correlations. Finally, only 1 significant correlation was found between therapy alliance (independent variable) and change in relationship satisfaction (dependent variable). A significant relationship was found between female expectancy and female therapy alliance. A test of the mediation model was not possible with any of the variable measures in this study. However, additional relationships that 36 warrant further investigation are indicated by the relationship between female therapy alliance and male and female change in relationship satisfaction. Additional Relationship Testing of Male and Female Alliance Due to the significant finding regarding the relationship between male and female therapy alliance and female therapy alliance and change in male and female relationship satisfaction; a simple regression was conducted in order to test for direct effects and significant relationships. Two key findings were discovered when looking at male and female therapy alliance. First, male therapy alliance was not related to any other variables in the study. Secondly, the model reported that female therapy alliance was a significant predictor of male satisfaction (R 2 = .13) and female satisfaction ( R 2 = .09). This means that the female therapy alliance variable explains 13% of variation in male change in relationship satisfaction and 9% of the variation in change in relationship satisfaction for females. The objective of this data analysis was to examine the relationships between male/female expectancy, therapeutic alliance and change in relationship satisfaction. After conducting the correlation analyses, it was found that more of the female variables of interest were significantly correlated than the male variables of interest. Not only was female therapeutic alliance correlated with male therapy alliance; it was correlated with female expectancy as well. A significant relationship was also found between male and female expectancy. Perhaps, among the most valuable significant relationships discovered, female therapy alliance was also found as a significant predictor of male and female change in 37 relationship satisfaction. This finding demonstrates that the female?s relationship with the therapist could be especially influential in therapeutic process and outcome in couple therapy. However, another interesting finding was revealed in the data analyses. Male change in relationship satisfaction was not significantly correlated with female change in relationship satisfaction. This could possibly infer that different aspects of the therapeutic process contribute to male and female change in therapy. Further implications of these findings will be discussed in the next section. 38 DISCUSSION This purpose of this study was to explore the therapeutic alliance as a possible mediating variable between couple expectancy and change in relationship satisfaction. Specifically, the impact of the therapeutic alliance on change in relationship satisfaction for males and females was investigated. In order to effectively understand the findings of the study, a summary of the results will be provided. Following the summary, the implications of the findings in relation to the separate hypotheses presented at beginning of the study will be discussed. The final section will highlight limitations of the study, future implications for research, clinical implications and conclusions. Summary of Results The major finding in the current study was that couple expectancy was not related to the therapeutic alliance or change in relationship satisfaction. Male and Female therapy alliance was related. When looking at male and female variables separately, male variables were not found to be related to any other variables. However, the female expectancy and therapy alliance variables were found to be related to change in relationship satisfaction. In fact, the female therapeutic alliance was found to be the only consistent predictor of change in relationship satisfaction. 39 Hypothesis 1: Therapeutic Alliance and Therapy Outcome The first hypothesis suggested that the therapeutic alliance measured at the fourth session would have a positive impact on the change in couple relationship satisfaction in males and females. Previous literature concluded that the better the collaborative bond between the client and therapist the better the alliance in therapy, which accounted for greater therapeutic outcome (Pinsof & Catherall, 1986). The findings from this study supported this hypothesis for females only. Female therapy alliance was positively related to male and female change in relationship satisfaction (r = .365 p < .01, r = .306 p < .01). As a result, it is offered that within this sample, the female?s relationship with the therapist predicted how the couple therapy proceeded. This signifies that females could be the driving force in couple therapy. As expected, therapeutic alliance was a significant predictor of change in relationship satisfaction for both males and females (R 2 = .13, R 2 = .9). These results supported previous findings in the literature that stated that the therapeutic alliance is related to positive outcomes in therapy. Horvath (2001) found in his review of meta- analytic studies that the therapeutic alliance was a strong variable which linked change in therapy to therapeutic outcome. Interestingly, female therapy alliance also had a stronger impact on male change in relationship satisfaction than female change in relationship satisfaction. This implies that the stronger the female therapy alliance in couple therapy, the greater the change in relationship adjustment at the end of therapy. This is supported in the literature by Quinn et al. (1997); who reported that a positive correlation was found between therapeutic 40 alliance and outcome in couple therapy. In addition, Quinn et al.?s results demonstrated a stronger association for wives (r = .74) than husbands (r = .56). Contrastingly, Brown and O?Leary found that when taking into account the level of symptom distress at intake, that although therapeutic alliance was found to be related to treatment outcome, the husband?s alliance was a better predictor of outcome than wives? alliance (2000). A possible explanation for this difference could be that this study took into account the level of relationship satisfaction for both males and females at intake, whereas, Brown and O?Leary only measured symptom distress at intake. Perhaps, this suggests that change in relationship satisfaction is impacted by different therapeutic constructs for males and females. Hypothesis 2: Significant Relationships between Variables Hypothesis two speculated that measures of couple expectancy would be significantly related to measures of therapeutic alliance and would both have direct relationships. The findings of the current study partially supported hypothesis two. A significant relationship between female expectancy and female therapy alliance was discovered ( r = .358* p < .05). However, this study did not fully support the prediction of hypothesis two because no relationship was found between male expectancy and the male and female therapeutic alliance. The reader should be cautious in reading too much into the results due to the small sample size of this study (n = 59). In contrast, several of the previous studies indicate that a significant relationship exists between individual and couple expectancy and therapeutic alliance. Meyer et al. (2002) found that individual?s treatment expectancies correlated significantly with 41 therapy alliance. A strong relationship was also found between patient alliance and outcome (2002). In another study, Al- Darmaki and Kivlighan (1993) discovered that client?s expectations for relationship were significantly related to all aspects of the therapeutic alliance. Hypothesis 3: Therapeutic Alliance as a Mediator Hypothesis 3 speculated that the therapeutic alliance would mediate the relationship between male and female expectancy and male and female relationship satisfaction. Because significant relationships were not found between the variables of expectancy, therapeutic alliance and relationship satisfaction, this mediation hypothesis could not be tested. One reason for the lack of significance is due to small sample size of this study. Within the current sample size, a substantial amount of statistical power could be lost in looking for relationships within the variables of interest. Another reason that the therapy alliance did not mediate the relationship between expectancy and outcome is because of the measure of the expectancy variable. The expectancy variable was measured by one item on the Process of Change scale administered at intake stating: ?Maybe this place will be able to help me.? This measure essentially lacked enough strength to measure expectancy within this study. Although it was not possible to test for possible mediation effects of the therapeutic alliance in the present study between male /female expectancy and relationship satisfaction, the significant relationships between these variables within the literature lends promise to future research. There is definitely a need to replicate the findings with not only a larger sample size, but a stronger expectancy measure as well. 42 Upon comparing the simple regression findings for males and females, more of the variance was explained by the female variables than the males. This indicated that a female?s expectancy within the couple relationship could not only have a significant impact on her partner?s expectancy for therapy but also strengthen her alliance with the therapist. As noted earlier, female therapy alliance also had a stronger significant impact on male change in relationship satisfaction than female change in relationship satisfaction. It is important to note that one study found that male intake symptom distress scores were significantly related to male therapy alliance formation, but female scores were not (Stephens, 2006). This suggests that for males, the level of symptom distress was predictive of male alliance formation, and possibly impacts the change in relationship satisfaction. Whereas, for females, it appears that the satisfaction scores at intake were not the indicators of therapy outcome, rather, therapeutic alliance scores in therapy are related to change in relationship satisfaction for females and also their spouses. This can be conceptualized in terms of the primary consumer in therapy. In the initial stages of marital therapy, generally, one client in the dyad is the primary consumer for change in therapy. In this case, if the female had a strong alliance with therapist, this meant that both the female and the therapist could work together to influence the couple agreement on tasks, bonds and goals in therapy. As the therapeutic alliance developed, this influenced the greater change in male relationship satisfaction. The trends toward which can be seen in the data. 43 Furthermore, it is strange that a significant relationship between male and female change in relationship satisfaction was not found. One would assume that the changes in satisfaction between males and females would be correlated. If one partner?s improves or worsens, then the other would respond in somewhat similar manner. In conjunction with this finding, it is odd that male therapy alliance is not related to relationship satisfaction; yet female therapy alliance is related. Again, this finding could be a result of small sample size of this study. Another reason could be due to misunderstanding the relationship between males and females in therapy. Perhaps, this suggests that change in relationship satisfaction is impacted by different therapeutic constructs for males and females. Following this logic, when looking at male change in relationships the central focus could be on the level of symptom distress at the beginning of therapy. For females in this clinic, expectancy and the therapeutic alliance could be the primary focus at the beginning of therapy in order to facilitate greater change in relationship satisfaction for not only herself, but her partner as well. Although casual inferences could not be made in this study, further future investigation of these findings is encouraged. Study Limitations Limitations of the present study can be noted. The small sample size is the first major limiting factor to the current study. The small sample size limited the power of the statistical analyses creating challenges in finding significant relationships between the variables. 44 The second limiting factor to the study is the lack of generalization of the sample. Because the sample was compiled from a southeastern university based clinic, the majority of the sample was Caucasian. This presents problems when attempting to generalize the results of the study to other racial and ethnic groups. Another shortcoming was the single measurement of expectancy before therapy. More substantial effects might have been identified using standardized, multidimensional measures of expectancy (Joyce et al., 2003). Assessment of client expectancy in this study occurred prior to the couple?s first meeting with the therapist. However, it is unlikely that a client?s expectancy remains static throughout therapy. One would assume that it is more probable that a client?s expectancy could be altered after the first session of therapy. Finally, all of data in this study was gathered by self-report measures. Therefore, the findings are comprised only of information that participants are willing to share. Moreover, the social desirability effect could also be a factor. Social desirability refers to the client?s responding by what they feel is socially acceptable to the therapist. Strengths of Study and Clinical Implications Although this study has a significant amount of limitations, it offered support to previous research findings. The current study also contributed to the limited research in couple therapy involving the client expectancy, couple therapeutic alliance and relationship satisfaction. This study has offered useful information regarding the impact of male and female alliance on couple therapy outcomes. As stated earlier, the results of this study 45 confirm that female therapy alliance has a stronger impact on expectancy and the change in couple satisfaction with their male counterparts. This information could be useful for therapists in working toward a strong alliance in the initial stages of therapy. Additionally, the study has clinical implications to offer to the field of Marriage and Family Therapy. The study of the alliance as a potential mediator suggests that a strong therapeutic alliance can explain the change in the relationship between husband and wife. In other words, the formation of a strong therapeutic alliance with a couple; particularly with the female, could possibly explain the relationship between what the couple expects from therapy (therapy clinic, therapist, therapy procedures) and the change in relationship satisfaction over the course of therapy. Hence, the therapist could have the tools and control within the vehicle of the alliance to provide an even stronger sense of motivation, collaboration and hope for change. Future Research Directions This research study attempted to establish association between couple expectancy, therapy alliance and relationship satisfaction. Even with the limitations, this study offered important findings which require future inquiry into alliance formation and change in therapy. Although therapy alliance was not found to be a mediator of expectancy effects; it was found to be a significant predictor of outcome, as reported in previous studies. Future research should include a larger more representative sample size. Not only will this allow for more generalizability to the population, but it will also allow for stronger, more statistically significant relationships and findings. 46 The findings also suggest that therapy alliance may represent a central mechanism of therapeutic change that operates in a similar fashion across different modalities and therapy orientations. It is important for future researchers explore the impact of the therapeutic alliance in couples; specifically differences between males and females in committed relationships. Future research should also continue to explore the possibility of mediating and moderating effects of the therapeutic alliance as it related to therapy outcome. More is to be understood in relation to the impact of client factors on the therapeutic alliance. In examining the relationship between male and female expectancy, therapy alliance and therapeutic outcome in couple therapy, it is critical to have standardized, multidimensional measures of expectancy. It would be also be beneficial to track changes in expectancy over the course of therapy, not solely before the initial session. Specifically, measuring client expectancy at specific intervals in the therapy process ;( i.e. following the second, fourth and sixth sessions). Furthermore, identifying associations between these changes and in?session events may be useful as well. This study provided a glimpse into how the interaction of different dynamics affects change in relationship satisfaction. However, due to methodological limitations, the current study could not focus on the mediating role of the therapeutic alliance between expectancy and relationship satisfaction. Additional research is needed in order to truly comprehend the distinctive differences between males and females within committed relationships. Overall, an improved understanding of the therapeutic alliance, expectancy and changes in relationship satisfaction among males and females within 47 committed relationships will add depth to our understanding of the important collaboration between clients and therapists in facilitating a positive therapeutic outcome. 48 REFERENCES Abouguendia, M., Joyce, A., Piper, W., & Ogrodiniczuk, J. (2004). Alliance as a mediator of expectancy effects in short-term group psychotherapy. Group Dynamics: Theory, Research, and Practice, 8, 3-12. Al-Darmaki, F., Kivlighan, D. (1993). Congruence in client-counselor expectations for relationship and the working alliance. Journal of Counseling Psychology, 40, 379-384. American Psychiatric Association. (1987). Diagnostic and statistical manual of mental disorders (3rd ed., rev.). Washington, D.C.: Author. Arbuckle, J.L. (1999). Amos 4.0 [Computer Software]. Chicago: Smallwaters. Austin, J. T., & Vancouver, J.B. (1996). Goal constructs in psychology: Structure, process and content. Psychological Bulletin, 120, 338-375. Azrin, N., Naster, B. & Jones, R. (1973). Reciprocity counseling. A rapid learning based procedure for marital counseling. Behavior Research and Therapy, 11, 365-382. Barber, J., Connolly, M., Cristoph, P., Gladis, L., & Siqueland, L. (2000). Alliance predicts patients? outcome beyond in-treatment change in symptoms. Journal of Consulting and Clinical Psychology, 68, 1027 -1032. Baron, R. M., & Kenny, D. A. (1986). The moderator- mediator variable distinction in social psychological research: Conceptual, strategic and statistical considerations. Journal of Personality and Social Psychology, 51, 1173-1182 49 Beck, A., Steer, R., & Garbin, M. (1988). Psychometric properties of the Beck Depression Inventory: Twenty-five years of evaluation, Clinical Psychology Review, 8, 77- 100. Beutler, L. E., Machado, P., & Allsetter N.: Therapist variables in Handbook of Psychotherapy and Behavior Change, 4th edition, edited by Garfield S. L., Bergin, A.E. New York, Wiley, 1994, 229-269. Berzins, J. I. (1971). Revision of Psychotherapy Expectancy Inventory. Unpublished manuscript, University of Kentucky, Lexington. Bonner, B. L., & Everett, F. L Influence of client preparation and problem severity on attitudes and expectations in child psychotherapy. Professional Psychology: Research and Practice 1986; 17: 223-229. Bordin, E. S. (1979). The generalizability of the psychoanalytic concept of the working alliance. Psychotherapy: Theory, Research, and Practice, 16, 252-260. Bourgeois, L., Sabourin, S., & Wright, J. (1990). Predictive validity of therapeutic alliance in group marital therapy. Journal of Consulting and Clinical Psychology, 58(5), 608-613. Brown, P.D., & O?Leary, K.D. (2000). Therapeutic alliance: Predicting continuance and success in group treatment for spouse abuse. Journal of Consulting and Clinical Psychology, 68, 340-345. Busby, D. M., Christensen, C., Crane, D. R., & Larson, J. H. (1995). A revision of the dyadic adjustment scale for use with distressed and nondistressed couples: construct hierarchy and multidimensional scales. Journal of Marital and Family Therapy, 21, 289-308. 50 Fedders, L., Pinsof, W., Mann, B. (2004). The formation of the therapeutic alliance in couple therapy. Family Process, 43, 425-442. Garfield, S.L. (1994). Research on client variables in psychotherapy. In A .E. Bergin & S.L. Garfield ( EDs.), Handbook of psychotherapy and behavior change ( 4 th ed., pp. 190-228). New York: Wiley. Garfield, R. ( 2004). The therapeutic alliance in couple? therapy: Clinical considerations. Family Process, 43, (4), 457-465. Gaston, L. (1991). Relaibility and criterion-related validity of the California Psychotherapy Alliance Scales. Psychological Assessment, 3 68-74. Gelso, C. J. & Carter, J. (1985). The relationship in counseling and therapy: Components, consequences, and theoretical antecedents. The Counseling Psychologist, 13 155- 243. Glass, C., Arnkoff, D., & Shapiro, S. ( 2001). Expectations and Preferences. Psychotherapy, 38, 455-461. Goldstein, A. P. (1962). Therapist- Patient expectancies in psychotherapy. New York: MacMillan. Hatcher, R. L. (1999). Therapists? view of treatment alliance and collaboration in therapy. Psychotherapy Research, 9, 405-423. Heaherington, L., & Freidlander, M.L. ( 1990). Couple and Family therapy alliance scale: Empirical considerations. Journal of Marital and Family Therapy, 16, 299- 306. 51 Heppner, P. P. & Petersen, C. H. ( 1982). The development and implications of a personal problem solving inventory. Journal of Counseling Psychology, 29, 66- 75. Holmbeck, G. (1997). Toward terminological, conceptual, and statistical clarity in the study of mediators and moderators: Examples from the child-clinical and pediatric psychology literatures. Journal of Consulting and Clinical Psychology, 65, 599- 610. Horvath, A.O. (2001).The alliance. Psychotherapy: Theory, Research, Practice, Training, 38, 365-372. Horvath, A. O., & Greenberg, L. (1989). Development and validation of the Working Alliance Inventory. Journal of Counseling Psychology, 36, 223-232. Horvath, A.O., & Symonds, B. D. (1991). Relationship between working alliance and outcome in psychotherapy: A meta-analysis. Journal of Counseling Psychology, 38, 139-149. Joyce, A.S., & Piper, W. E. ( 1998) . Expectancy, the therapeutic alliance, and treatment outcome in short- term individual psychotherapy. Journal of Psychotherapy Practice and Research, 7, 236-248. Joyce, A. S., Ogrodniczuk, J. S., Piper, W. S., & McCallum, M. (2003). The alliance as a mediator of expectancy effects in short-term individual therapy. Journal of Consulting and Clinical Psychology, 4, 672-679. Knobloch- Fedders, L.M., Pinsof, W.M., & Mann, B.J. (2004). The formation of the therapeutic alliance in couple therapy. Family Process, 43, 425-442. 52 Krupnick,J. L., Sotsky, S.M., Simmens, S., Moyer, J., Elkin, I., Watkins, J., & Pilkonis, P. A. ( 1996). The role of the therapeutic alliance in psychotherapy and phramacotherapy outcome: Findings in the National Institute of Mental Health Treatment of Depression Collaborative Research Program. Journal of Consulting and Clinical Psychology, 64, 532-539. Lambert, M. J. (1992). Implications of Outcome research for psychotherapy integration. In J.C. Norcross & M. R. Goldstein (Eds.), Handbook of Psychotherapy integration (pp. 94-129). New York: Basic Books. Lambert,M., & Barley,D.( 2001). Research summary on the therapeutic relationship and psychotherapy outcome. Psychotherapy: Theory, Research, Practice, Training. 38, 357-361. Luborsky, L. Principles of Psychoanalytic Psychotherapy: A Manual for Supportive- Expressive Treatment. New York, Basic Books, 1984. Marmar, C.; Weiss, D.; and Gatson, L. ( 1989). Toward the validation of the California Therapeutic Alliance Rating System: Psychological assessment. Journal of Consulting and Clinical Psychology. Martin, D.J., Garske, J.P., & Davis, M.K. ( 2000). Relation of the therapeutic alliance with outcome and other variables: a meta-analytic review. Journal of Consulting and Clinical Psychology. Meyer, B., Krupnick, J., Simmens, S., Pilkonis, P., Egan, M., & Sotsky, S. ( 2002). Treatment Expectancies, patient alliance, and outcome: Further analyses from the National Institute of Mental Health Treatment of Depression Collaborative 53 Research Program. Journal of Consulting and Clinical Psychology, 70, 1051- 1055. Pan, H., Neidig, P.H. & O?Leary, K.D. (1994). Male- Female and aggressor-victim differences in the factor structure of the Modified Conflict Tactics scale. Journal of Interpersonal Violence, 9, 366-382. Patterson, G. R. (1976). Some procedures for assessing changes in marital interaction patterns. Oregon Research Institute Research Bulletin, 16 (7). Pinsof, W. M., & Catherall, D. R. (1986). The integrative psychotherapy alliance: family, couple, and individual therapy scales. Journal of Marital Therapy, 12, 137-151. Quinn, W. H., Dotson, D., & Jordan, K. ( 1997). Dimensions of therapeutic alliance and their association with outcome in family therapy. Psychotherapy Research, 7, 429-438. Saunders, S. M., Howard, K. I., & Orlinksy D. E. (1989). The therapeutic Bond Scales: Psychometric characteristics and relationship to treatment effectiveness. Psychological Assessment. 1, 323-330. Shapiro, D. A. (1981). Comparative credibility of treatment rationales: Three tests of expectancy theory. British Journal of Clinical Psychology, 21 111-122. Snyder, D.K. (1997). Manual for the Marital Satisfaction Inventory- Revised. Los Angeles: Western Psychological Services. Spanier, G. B. (1976). Measuring dyadic adjustment: New Scales for assessing the quality of marriage and similar dyads. Journal of Marriage and Family Therapy, 38, 15-28. 54 Stephens, M. (2006). Marital adjustment as a mediating factor between symptom distress and therapeutic alliance formation in couple therapy. Unpublished master?s thesis, Auburn University, Alabama. Stiles, W. B., Measurement of the impact of psychotherapy sessions, (1980) Journal of Consulting Clinical Psychology; 48,176-185. Toleman, R. (1989). The development of measure of psychological maltreatment of women by their male partners. Violence and Victims, 7, 159-177. Wilkins, W. (1973). Expectancy of therapeutic gain: An empirical and conceptual critique. Journal of Consulting and Clinical Psychology, 40, 69-77. 55 APPENDICES 56 APPENDIX A Couple Therapy Alliance Scale Instructions: The following statements refer to your feelings and thoughts about your therapist and your therapy right NOW. Please work quickly. We are interested in your FIRST impressions. Your ratings are CONFIDENTIAL. They will not be shown to your therapist or other family members and will only be used for research purposes. Although some of the statements appear to be similar or identical, each statement is unique. PLEASE BE SURE TO RATE EACH STATEMENT. Each statement is followed by a seven-point scale. Please rate the extent to which you agree or disagree with each statement AT THIS TIME. If you completely agree with the statement, circle number 7. If you completely disagree with the statement, circle number 1. Use the numbers in-between to describe variations between the extremes. Completely Agree Strongly Agree Agree Neutral Disagree Strongly Disagree Completely Disagree 7 6 5 4 3 2 1 1. The therapist cares about me as a person 7 6 5 4 3 2 1 2. The therapist and I are not in agreement about the goals for this therapy. 7 6 5 4 3 2 1 3. My partner and I help each other in this therapy. 7 6 5 4 3 2 1 4. My partner and I do not feel the same ways about what we want to get out of this therapy. 7 6 5 4 3 2 1 5. I trust the therapist. 7 6 5 4 3 2 1 6. The therapist lacks the skills and ability to help my partner and myself with our relationship. 7 6 5 4 3 2 1 7. My partner feels accepted by the therapist. 7 6 5 4 3 2 1 8. The therapist does not understand the relationship between my partner and myself. 7 6 5 4 3 2 1 9. The therapist understands my goals in therapy. 7 6 5 4 3 2 1 10. The therapist and my partner are not in agreement about the about the goals for this therapy. 7 6 5 4 3 2 1 11. My partner cares about the therapist as a person. 7 6 5 4 3 2 1 12. My partner and I do not feel safe with each other in this therapy. 7 6 5 4 3 2 1 13. My partner and I understand each other?s goals for this therapy. 7 6 5 4 3 2 1 14. The therapist does not understand the goals that my partner and I have for ourselves in this therapy. 7 6 5 4 3 2 1 57 15. My partner and the therapists are in agreement about the way the therapy is being conducted. 7 6 5 4 3 2 1 16. The therapist does not understand me. 7 6 5 4 3 2 1 17. The therapist is helping my partner and me with our relationship. 7 6 5 4 3 2 1 18. I am not satisfied with the therapy. 7 6 5 4 3 2 1 19. My partner and I understand what each of us is doing in this therapy. 7 6 5 4 3 2 1 20. My partner and I do not accept each other in this therapy. 7 6 5 4 3 2 1 21. The therapist understands my partner?s goals for this therapy. 7 6 5 4 3 2 1 22. I do not feel accepted by the therapist. 7 6 5 4 3 2 1 23. The therapist and I are in agreement about the way the therapy is being conducted. 7 6 5 4 3 2 1 24. The therapist is not helping me. 7 6 5 4 3 2 1 25. The therapist is in agreement with the goals that my partner and I have for ourselves as a couple in this therapy. 7 6 5 4 3 2 1 26. The therapist does not care about my partner as a person. 7 6 5 4 3 2 1 27. My partner and I are in agreement with each other about the goals of this therapy. 7 6 5 4 3 2 1 28. My partner and I are not in agreement about the things that each of us needs to do in this therapy. 7 6 5 4 3 2 1 29. The therapist has the skills and ability to help me. 7 6 5 4 3 2 1 30. The therapist is not helping my partner. 7 6 5 4 3 2 1 31. My partner is satisfied with the therapy. 7 6 5 4 3 2 1 32. I do not care about the therapist as a person. 7 6 5 4 3 2 1 33. The therapist has the skills and ability to help my partner. 7 6 5 4 3 2 1 34. My partner and I are not pleased with the things that each of us does in this therapy. 7 6 5 4 3 2 1 35. My partner and I trust each other in this therapy. 7 6 5 4 3 2 1 36. My partner and I distrust the therapist. 7 6 5 4 3 2 1 37. The therapist cares about the relationship between my partner and myself. 7 6 5 4 3 2 1 38. The therapist does not understand my partner. 7 6 5 4 3 2 1 39. My partner and I care about each other in this therapy. 7 6 5 4 3 2 1 40. The therapist does not appreciate how important my relationship between my partner and myself is to me. 7 6 5 4 3 2 1 58 APPENDIX B Revised Dyadic Adjustment Scale Most persons have disagreements in their relationships. Please indicate below the approximate extent of agreement or disagreement between you and your partner for each item on the following list. Always Agree Almost Always Agree Occasional Agreement Frequently Disagree Almost Always Disagree Always Disagree 1. Religious matters 5 4 3 2 1 0 2. Demonstrations of affection 5 4 3 2 1 0 3. Making major decisions 5 4 3 2 1 0 4. Sex relations 5 4 3 2 1 0 5. Conventionality (correct or proper behavior 5 4 3 2 1 0 6. Career decisions 5 4 3 2 1 0 All the time Most of the time More often than not Occasionally Rarely Never 7. How often do you discuss or have you considered divorce, separation, or terminating your relationship? 0 1 2 3 4 5 8. How often do you are your partner quarrel? 0 1 2 3 4 5 9. Do you ever regret that you married (or live together)? 0 1 2 3 4 5 10. How often do you and your mate ?get of each other?s nerves?? 0 1 2 3 4 5 59 How often would you say the following events occur between you and your mate? Never Less than once a month Once or twice a month Once or twice a week Once a day More often 12. Have a stimulating exchange of ideas 0 1 2 3 4 5 13. Work together on a project 0 1 2 3 4 5 14. Calmly discuss something 0 1 2 3 4 5 From: Busby, D.M., Crane, D.R., Larson, J.H., & Christensen C. (1995). A revision of the Dyadic Adjustment Scale for use with distressed and nondistressed couples: Construction hierarchy and multidimensional scales. Journal of Marital and Family Therapy, 21, 289-308 Every Day Almost Every Day Occasionally Rarely Never 11. Do you and your mate engage in outside interests together? 4 3 2 1 0