IIIIIIIIIIIIIIIIIIIII I I r; l .'.q-r - ' . . £u.v‘....»--......; gal—.36 A; \ .A. _, O‘_A- - Tag“; 3 T".""""‘ rw L. 'v;~ .~ >. ~11,5’.&* h I '1 —~ , A- __A 5"! A. ’ ’ --.k WES!) ‘ ~.-._, an E: ‘? , '2“ 'J . 'i 'v OI-‘ifi. . ’ . :7 \Jné "0.35“. \ J w This is to certify that the dissertation entitled The Development of an Instrument to Study Sex Bias in Psychotherapy with Women Clients presented by Robin Sesan has been accepted towards fulfillment of the requirements for Ph.D, degree in Counsel ind Psychology [/Miw Major professor Date June '10? 1983 MSU i: an Affirmative Action/Equal Opportunity Institution 0- 12771 )V‘ESI.) RETURNING MATERIALS: Place in book drop to LIBRARJES remove this checkout from your record. flfl§§ will be charged if book is returned after the date stamped below. mamafi ‘ 10W . E00 02" - ;; THE DEVELOPMENT OF AN INSTRUMENT TO STUDY SEX BIAS IN PSYCHOTHERAPY WITH WOMEN CLIENTS By Robin Sesan A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY College of Education Department of Counseling, Educational Psychology, and Special Education 1983 Copyright by ROBIN SESAN 1983 ii ABSTRACT THE DEVELOPMENT OF AN INSTRUMENT TO STUDY SEX BIAS IN PSYCHOTHERAPY WITH WOMEN CLIENTS by Robin Sesan The primary purpose of the study was to develop an instrument to measure sex bias in psychotherapy. The Sex Bias in Psychotherapy Questionnaire (SBPQ) was developed and initial reliability and validity data obtained. The secondary purpose of the study was to investigate the phenomena of sex bias in psychotherapy with women clients. A descriptive research design was used to address the dual purposes of the research study. The SBPQ is a lOO-item, self-report instrument developed around the following themes of sex bias: foster- ing traditional sex roles, bias in expectations and devaluation of women, sexist use of psychoanalytic concepts, and responding to women as sex objects. Four sub-scales, reflective of the above themes, comprise the SBPQ total scale. The SBPQ was found to be a reliable instrument. In addition, content validity for the instrument was established and some evidence of construct validity was obtained. Robin Sesan Following the development of the SBPQ, a sample of 192 terminated women therapy clients completed the instru- ment and provided life status information. Therapist variables including sex, experience level, and agency affiliation were also identified for each client. The results indicated that women clients in general did not experience statistically significant amounts of sex bias in therapy. However, the following groups of women were found to experience the most sex bias in psychotherapy: women clients with less than a high school education, women clients with children, and women clients who had one to three therapy sessions. Additionally, the following sub-groups of women were found to experience sex bias in psychotherapy relative to their particular treatment needs: rape victims, incest survivors, and battered women. Acts of omission were the predominant form of sex bias experienced by women clients. The therapist's sex, experience level, and agency affiliation were found to have no effects on women clients' experience of sex bias in psychotherapy. Implications of the research are presented, along with a discussion of the results. ACKNOWLEDGMENTS To Bill Hinds, for his support, guidance, and care, especially during my early years of graduate study. His help during this transitional time allowed me to develop a firm foundation upon which to grow, and his continued support allowed me to risk. To my guidance committee, Linda Forrest, Doug Miller, Jim Snoddy, and Judith Taylor, for their support and trust in my competence. To Teresa Bernardez, for exposing me to new ways of understanding women's emotional difficulties, helping me to clarify my own sex-role biases, and for encouraging me to question the usefulness of current psychotherapy practices for women. To Mary Ann Stehr, a mentor and role model, a teacher and friend, for her wisdom and insight and for her trust in my judgment and convictions. To my expert judges, Martha Aldenbrand, Imogen Bowers, Jim.Heavenrich, John Powell, Mary Ann Stehr, and Terry Stein, for their diligent participation in establishing a defini— tion of sex bias in psychotherapy. To the therapists who participated in the study and left themselves vulnerable, and to the clients who allowed iii some intrusion into a very private aspect of their lives, I am grateful. To my parents, for their never-ending faith in me and my abilities, particularly in the face of my self-doubts. To Paul, for his patience and endurance through a difficult year. His optimism balanced my pessimism” enabling me to feel hopeful and excited during difficult momen t S . iv TABLE OF CONTENTS Page LIST OF TABLES . . . . . . . . . . . . . . . . . . . . vii CHAPTER I. STATEMENT OF THE PROBLEM . 1 Introduction . . . 1 Need for the Study . . 3 Importance of the Study. 6 Theory . . . 8 Sex- Fair Theories of Personality Development. . . . . . . . . . 9 Sex-Fair Counseling Theories . . . . . . . 15 Definition of Terms. . . . . . . . . . . . . l9 Purpose of the Study . . . . . . . . . . . . 19 Research Questions . . . . . . . . . . . . . 20 Overview . . . . . . . . . . . . . . . . . . 21 II. REVIEW OF THE LITERATURE . . . . . . . . . . . 22 Clinical Theory on Sex-Role Related Therapy Issues . . . . 22 Therapists' Sex- Role Stereotypes and Attitudes Towards WOmen. . . . . . . 28 Therapist Judgments and Evaluation . . . . . 38 Therapy Process. . . . . . . . . . . . . . . 51 Therapy Outcome. . . . . . . . . . . . . . . 72 Summary. . . . . . . . . . . . . . . . . . . 76 III. DESIGN OF THE STUDY. . . . . . . . . . . . . . 90 Selection and Description of the Sample. . . 90 Measures Used in the Study . . . 95 Sex Bias in Psychotherapy Questionnaire. . 95 The Stereotype Scale . . . . . . . . . 96 Procedures for Data Collection . . . . . . 100 Operational Definitions for Therapist Variables. . . . . . . . . . . . . . . . . 103 Design . . . . . . . . . . . . . . 104 Research Hypotheses. . . . . . 106 Procedures for Quantitative and Supple- mentary Descriptive Data Analysis. . . . . 108 Summary. . . . . . . . . . . 119 IV. VI. RESULTS: PART I - DEVELOPMENT OF AN INSTRU- MENT TO MEASURE SEX BIAS IN PSYCHOTHERAPY . Instrument Development. Fostering Traditional Sex Roles Bias in Expectations and Devaluation of Women. . . . . . . . . . . . . . Sexist Use of Psychoanalytic Concepts Responding to Women as Sex Objects. Tests of the Research Hypotheses. Summary . . . . . . . . . RESULTS: PART II . Tests of the Research Hypotheses. Supplementary Descriptive Analysis. Summary . . . . . . . . . SUMMARY, DISCUSSION, AND IMPLICATIONS . Summary . Discussion. . Instrument Development. . Phenomena of Sex Bias in Psychotherapy. Limitations of the Study. . . Implications of the Research. Directions for Future Research. Instrument Revision . . . Sex Bias in Psychotherapy . Conclusion. APPENDICES . A. WINDOW LETTER TO EXPERT JUDGES . CLIENT-SUBJECT QUESTIONNAIRE PACKET . CLIENT-SUBJECT FOLLOW-UP LETTER . THERAPIST-SUBJECT RESEARCH PACKET . THERAPIST-SUBJECT FOLLOW-UP LETTER. GUIDELINES FOR NON-SEXIST THERAPY . REFERENCES . vi Page 121 121 124 128 136 142 148 160 163 163 ' 184 192 196 196 I 201 201 I 211 224 . 226 . 228 . 228 . 230 231 233 . 233 . 235 244 245 255 . 256 . 258 TABLE 2. 3. 1 1 LIST OF TABLES Summary of the Research on Sex Bias in Psychotherapy . Description of Client-Subject Sample by Life Status Categories and Therapy Experience Variables for the Community Mental Health Center, Counseling Center, and Combined Populations. . . Description of Therapist- Subject Sample (N=49). . Procedures for Quantitative Data Analysis Objectives Used in the Development of the Sex Bias in Psychotherapy Questionnaire . A Summary of the Four Sub-Scales of the Sex Bias in Psychotherapy Questionnaire . Internal Consistency of the Sex Bias in Psychotherapy Questionnaire for the Total Scale and the Four Sub~Sca1es: Cronbach's Coefficient Alpha (N - 140) Correlation Matrix of the Total Scale and Four Sub-Scales of the Sex Bias in Psychotherapy Questionnaire . Two-Way Analysis of Variance of the Stereo- type Scale for Therapist Sex X Experience Level . . . . . . . . . . . . . . . Two-Way Analysis of Variance of the Stereo- type Scale for Therapist Experience Level X Agency Affiliation. Two-Way Analysis of Variance of the Stereo- type Scale for Therapist Sex X Agency Affiliation . . . . . . vii Page 82 94 96 118 146 149 154 156 158 158 159 Table 4.8 Correlation Matrix of the Stereotype Scale and the Sex Bias in Psychotherapy Questionnaire (N = 38) . . One Sample t- test Results for the SBPQ Total and Four Sub- Scales. . . . . . . . Item Means and Response Frequencies for the SBPQ General Items Prior to Instrument Revision . One-Way Analysis of Variance of the SBPQ for Client Age . . . . One-Way Analysis of Variance of the SBPQ for Client Education . . . . . One-way Analysis of Variance of the SBPQ for Client Relationship Status One- -Way Analysis of Variance of the SBPQ for Client Child Status. . . Two-Way Analysis of Variance of the SBPQ for Client Child Status X Therapist Agency Affiliation. . . . One-Way Analysis of Variance of the SBPQ for the Number of Therapy Sessions One-Way Analysis of Variance of the SBPQ for the Length of Time Since Termination of Therapy . . . . . . . . . . Group Means for the SBPQ for Hypothesis VI (a and b). . . . . . Two-Way Analysis of Variance on SBPQ Scores for Therapist Sex X Experience Level . Two-Way Analysis of Variance on SBPQ Scores for Therapist Experience Level X Agency Affiliation. . . . . . . . . Two-Way Analysis of Variance on SBPQ Scores for Therapist Sex X Agency Affiliation . Group Means for the Aggregate Mean SBPQ Scores for Hypotheses VII, VIII, and IX. viii Page 161 167 168 175 175 176 176 177 177 178 179 182 183 183 184 CHAPTER I STATEMENT OF THE PROBLEM Introduction During the last decade, the influence of the Women's Liberation Movement has impacted most every social institu- tion. The institution of psychotherapy is no exception. In the early 1970's feminists raised concerns about the patriarchical and Oppressive aspects of psychotherapy for women clients (Chesler, 1971; 1972). Voicing of these concerns coincided with a landmark investigation of clinicians' sex-role stereotypic concepts of mental health, which demonstrated that a double standard of mental health existed for men and women (Broverman, Broverman, Clarkson, Rosenkrantz and Vogel, 1970). Broverman et a1.'s investi- gation, reviewed in the following chapter, stimulated much research in the area of sex-role stereotyping as it relates to clinical judgment, treatment, and outcome in psycho- therapy. Yet, more than 10 years later, there is no con- clusive information about the treatment of women in therapy and the existence of sex-role stereotyping being translated into therapeutic actions. The concern with the treatment of women in psycho- therapy emerges from an awareness that clinicians hold many of the same beliefs and sex-role stereotypic attitudes as our culture at large and from a serious and critical examination of our psychological theories as they relate to women's development and experience (Rice & Rice, 1973; Howell, 1981; Collier, 1982; Gilligan, 1982; Greenspan, 1983). Additionally, it is suggested that cultural stereo- types of appropriate male and female behavior have directly influenced personality theory. Recently, issue has been taken with generalizing to women those personality theories deve10ped for men (Gilligan, 1979, 1982). Many psychologists no longer find this generalization acceptable and are disturbed by the negative view of women inherent in traditional psychoanalytic and developmental theories (Carlson, 1972; Rice & Rice, 1973; Barrett, Berg, Easton & Pomeroy, 1974; Voss & Gannon, 1978). Is it not possible that the biases inherent in these theories influence therapists' attitudes towards women, thus their treatment of women clients during the therapy hour? The problem in the research thus far has been in demonstration of the above relationship. Prominent figures in the field believe this to be the reality and, as stated by Carmen, Russo, & Miller (1981): It would be surprising if the pervasive cultural biases that denigrate women were not found in the training of mental health pro- fessional. and the delivery of mental health services for women . . . clinical theories of personality specify women's innate nature as passive, dependent, masochistic and childlike, and psychological treatment has often aimed at reducing her complaints about the quality of her life and promoting adjustment to the existing order. (p. 1324) The question which was first asked over 10 years ago remains unanswered. Do women clients experience sex bias in psychotherapy? The literature which attempts to answer this question abounds with methodological difficulties. Research in this area has been approached broadly and haphazardly, without a sound theoretical base from.which to choose dependent variables. A shotgun approach has been used to find possible dependent variables to correlate with therapist sex, as a means of demonstrating sex bias (Zeldow, 1978). In addition to these problems, the question has been studied indirectly through the use of analogue studies or indirect measures. Clearly, there has not been enough descriptive research from which to generate hypotheses. It may be that in the area of sex bias in psychotherapy, the approach to research needs to change. Various authori- ties in the field of women's research have suggested that movement to more qualitative modes of investigation is necessary, when studying women's issues (Carlson, 1972; Davidson & Abramowitz, 1980; Abramowitz, 1982a, 1982b). It is in this direction which the present research study moves. moves . Need for the Study In 1974, at the request of the American Psychological Association's Committee on Women in Psychology. a task force was formed to: (a) examine the extent and manner of sex bias and sex-role stereotyping in psychotherapeutic practice as they directly affect women as students, practi- tioners and consumers, (b) recommend actions to reduce sex bias and sex-role stereotyping in psychotherapy, and (c) develop materials and methods of dissemination. (APA, 1975, p. 1169) An open-ended questionnaire, designed to elicit instances of sex bias and sex-role stereotyping in psychotherapy, was sent to 2,000 women members in APA. From 320 responses, the task force was able to identify four general areas indicative of sex bias and sex-role stereotyping in psycho- therapy. These areas are (l) fostering traditional sex roles, (2) bias in expectations and devaluation of women, (3) sexist use of psychoanalytic concepts, and (4) respond- ing to women as sex objects. The results of the task force survey, by providing verbatim instances of sex bias in therapy, demonstrated a variety of instances of sexual discrimination and sex-role stereotyping in psychotherapy. This led to broad recommenda- tions aimed at alleviating discriminatory treatment of female clients. Included in the recommendations were needs for consciousness raising, development of guidelines for non-sexist therapeutic practice, criteria and procedures to evaluate training in the psychology of women and inclusion in the ethical standards statements regarding sexism.with the most important need being a "greater awareness of the problems of sex-bias and sex-role stereotyping in psycho- therapeutic practice" (APA, 1975, p. 1074). Although some of these recommendations have been implemented, others have not. The recommendations, because they are so broad, are difficult to implement without more specific and conclusive evidence of actual sexist practices in psychotherapy. The initial survey, with retrospective reporting by women psychologists, nonetheless points the way for researching women clients' experience of sex bias in their personal therapy. There is a need for the direct assessment of this experience. The need for ”documenting, in a systematic fashion, the kinds of biases and sex-stereotyping" (Waskow, 1976b, p. 96) which exist in therapy has been stressed by the Society of Psychotherapy Research and is seen as an "area demanding researchers' attention” by Maracek and Johnson (1980). A handful of studies have attempted to document bias by obtaining therapy process and outcome data directly from.women clients and then correlating this data with different counselor-client gender pairings (Howard, Orlinsky & Hill, 1969; Orlinsky & Howard, 1976; Jones & Zoppel, 1982). To this date there has been only one study, in addition to that of the APA Task Force, which has directly assessed women clients' perceptions of sex—role issues in their therapy (Fabikant, 1974). There is a growing body of anecdotal literature, but nothing which looks at the problem in a systematic way. Yet in 1978 the APA Task Force recommended the "development of procedures for obtaining information from consumers about sexist practice in psychotherapy” (APA, 1978, p. 1175). It seems warranted that an instrument be developed and procedure established for exploring this phenomena and that women be questioned directly about sex bias in their psychotherapy experience. Importance of the Study There are several reasons why it is important to study sex bias in psychotherapy. Probably the most outstanding reason is the fact that the majority of clients seen in therapy are women while the majority of clinicians are men. Although this is changing, with 30% of APA membership being female, the disproportionate representation of male clients and female clinicians remains. There has been much research and speculation as to why there are so many more women in treatment than men. One hypothesis is that it is sex-role appropriate for women to seek help, making it acceptable for them.to be in therapy. Another reason which often appears in the literature is that women are responding to the stress associated with their devalued position in society and their prescribed sex-role with increased incidence of emotional disturbance and symptomatology of learned helplessness (Gove and Tudor, 1973; Cave, 1980; Carmen, Russo & Miller, 1981; Franks & Rothblum, 1983). If, in fact, this is the case, then perpetuation of sex roles within the therapy relationship will not alleviate the client's difficulty and encouragement to adapt to sex-role prescriptions can be viewed harmful to women. Sex-role stereotyping, defined by Broverman, et a1. (1970), as highly consensual norms and beliefs about the differing characteristics of men and women, has been shown to be detrimental to an individual's growth and develop- ment (Costrich, Feinstein, Kidder, Maracek, and Pascale, 1975). It is well known among educators that a teacher's attitudes and expectations for her/his student's achieve- ment and competence have an effect on their performance (Rosenthal and Jacobson, 1968). By extension, this under- standing is particularly appropriate to the therapeutic process. It can be assumed that the expectations and biases which the therapist brings to the therapy hour may in fact be transmitted to his/her client. If, for example, a therapist holds a stereotyped notion that women are supposed to be passive, submissive, etc., then by the self- fulfilling prophecy the client may adopt these character- istics. If, on the other hand, the therapist does not hold stereotyped notions for male or female behavior, the client has a wider range of behavior open to him/her. The therapist's expectations may direct the course of change or contribute to the client's stagnation. Some argue that the therapist does not have this kind of power in the therapy relationship. While some concession is given to this point, the weight of evidence falls on the fact that there is no value-free psychotherapy (Bart, 1971; Waskow, 1976). In fact, it is shown that during the course of therapy, a client moves closer to the values of her/his therapist (Rawlings & Carter, 1977). This being the case, exploration of the existence of sex-role stereotyping as it appears in therapy is essential. Once some reliable evidence is obtained as to the existence or nonexistence of sex-role stereotyping in therapy, the development of training materials regarding gender issues in therapy can be more adequately addressed. At the same time, this awareness will open up exploration of therapist's and trainee's biases regarding both men and women's roles which will lead to more sex-fair counseling and psycho- therapy. It is to this end that this research seems particu— 1arly important and useful. Theory A number of theories form the base for this research. The model of mental health proposed by the sex—fair theories of personality, including the theories of androgyny (Bem, 1974; Kaplan, 1976, 1979a; and Gilbert, 1981), sex-role . transcendence (Hefner, Rebecca and Oleshansky, 1975, and Rebecca, Hefner & Oleshanky, 1976) and self-actualization (Maslow, 1968) has formed a foundation for idea conception, instrument development, and research questions. Likewise, sex-fair theories of counseling, including the theories of nonsexist and feminist therapy (Rawlings & Carter, 1977; Gilbert, 1980; Collier, 1982) provided a theoretical base from which to explore the phenomena of sex bias in psycho- therapy. Additional theory relied upon has been critiques of Freudian and traditional psychoanalytic theory as relevant to women's personality development. This critique is presented in Chapter IV, in support of instrument development. Sex-fair theories of personality, specifically the models of mental health proposed by these theories and sex-fair theories of counseling, are presented in this section. Sex-Fair Theories of Personality Development An awareness of the implicit sex bias in more tradi— tional theories of personality lead theorists to search for and to develop sex-fair theories of personality develop- ment. Three theories of personality, those of androgyny (Bem, 1974; Kaplan, 1976, 1979a; and Gilbert, 1981), sex- role transcendence (Hefner, Rebecca and Oleshansky, 1975, and Rebecca, Hefner and Oleshansky, 1976) and self- actualization (Maslow, 1968) are observed to be sex-fair and propose similar models of mental health (Collier, 1982). Early theories of androgyny as a model for human development maintained that all individuals have a combina- tion of traditionally feminine or expressive and masculine or instrumental characteristics which may be relatively suppressed or developed. This theory posits that instru- mental and expressive qualities are present in any one individual along a continuum from masculine to feminine. Feminine characteristics refer to the communion of human existence and include qualities of nurturance, 10 interdependency and expression. Masculine qualities are ”agentic,' concerned with goal-orientation, competence, and achievement (Bakan, 1966). According to this early theory of androgyny, both men and women develop a person- ality style reflective of differing amounts of instrumentality and expressiveness (Jung, 1971 and Bem, 1974). Early theories defined androgyny as a balance between instrumental and expressive qualities, whatever their absolute strengths (Bem, 1974). More recent theorizing suggests a definition of androgyny somewhat different than that presented above. Androgyny is currently defined as the "possession of high degrees of masculine, instrumental attributes and feminine, expressive attributes" coexisting with one another (Gilbert, 1981). Healthy personality development is theorized to involve an integration of both high degrees of instrumentality and expressiveness. Kaplan (1976, 1979a) too proposes androgyny as a model of mental health for women, but takes issue with the polarity between masculine and feminine in the early theorflas of androgyny. Like Gilbert, she proposes as a goal for mental health "integration of masculine and feminine dichotomies, so as to coexist, to be tempered one by the other, to unite in the formation of truly integrated characteristics (p. 229). She theorizes that the integrated individual is one who is truly flexible, able to balance anger with love and dependency with assertiveness. This 31' "n 11 person would not have to choose between two qualities, but would incorporate both into an integrated whole. It is the integration and coexistence of instrumental and expressive qualities that is central to Kaplan's theory. She points out that an individual may appear to be androgynous, demonstrating assertiveness, competence, inter- dependency and so on, but not be able to integrate these opposing characteristics. Little flexibility is attained if this is so. Therefore, androgyny, as measured by an androgynous score on certain instruments, does not necessarily equal health (Kaplan, 1979a). Kaplan maintains that the temperance of masculine with feminine character- istics and flexibility of use depending on the situation are the essential components to using androgyny as a model of mental health. Sex-role transcendence (Hefner, Rebecca and Oleshansky, 1975, and Rebecca, Hefner and Oleshansky, 1976) is another theory in the evolution of the theory of androgyny which came forth in response to concerns about the sex-role polarities evident in early theories of androgyny. This model asserts that sex-role development proceeds through three stages: an undifferentiated conception of sex-roles (Stage I), a polarized or oppositional view of sex-roles (Stage II), and a flexible, ever-changing transcendence of sex-roles (Stage III). Stages I and II parallel the cognitive-developmental model of sex-role learning presented by Kohlberg (1966). Sex-role 12 transcendence theory takes Kohlberg's model one step further. According to sex-role transcendence theory, during the undifferentiated stage the child's thinking is global and s/he is not aware of sex differences or culturally defined sex roles. Towards the end of this stage, the child has an awareness of male and female sex differences and has begun to learn that there are different culturally valued behaviors for boys and girls. In Stage II, that of polarized sex-roles, the child actively accepts her/his assigned sex-role, in an attempt to gain admittance into the adult world. Conformity to sex— role prescriptions is encouraged and rewarded. While it is important for children to use an organizing technique of polarities to understand their world, rather than viewing this stage as temporary in one's development, our society views this stage as an end-product, thus reinforcing strongly sex-typed behaviors in men and women. Hefner, et a1. (1975), and Rebecca et al. (1976), maintain that healthy development proceeds to Stage III, that of sex-role transcendence. In this stage the individual can move freely from.one encounter to another with flexibility in behavioral and emotional expression, not characterized by sex-role restrictions. "There is a transcending of the stereotypes and a reorganization of the possibilities learned in Stage II into a more personally relevant framework" (Rebecca, et al., 1976, p. 95). This third stage is not seen as an end-product, but as a 13 dynamic process, allowing change and flexibility throughout life. Similar then to Kaplan's model, sex-role transcendance is a dynamic theory of flexible integration of instrumental and expressive qualities unrestricted by gender. Research supports the notions of androgyny and sex- role transcendence as models of mental health. In a study by Block (1973) greater maturity as measured by Kohlberg's Moral Judgment Test paralleled more androgynous, less sex- typed definitions of self. Likewise, women in Block's sample who demonstrated a balance with respect to agentic and communal concerns scored at higher levels of ego maturity on Lovinger's Sentence Completions. Maccoby (1966) found that optimal cognitive functioning in children of both sexes was dependent on an optimal balance between feminine and masculine orientations, and Hammer (1964) and Helson (1966) demonstrated that greater creativity depends on a balance of masculine and feminine traits. Block. on the basis of her research,points out the value of androgyny: If our social aim can become . . . the integration of agency and communion, the behavioral and experi- ential options of men and women alike will be broadened and enriched and we all can become more truly whole, more truly hunan. (Block, 1973, p. 526) The model of the mentally healthy person which emerges from the theories of androgyny and sex-role transcendence is similar to the model of self-actualization presented by Maslow (1968). Maslow (1942) was one of the first to observe that women high in self-esteem were also tolerant <>:f? (others, decisive, willing to take risks and initiative, \A Ii ‘1‘ Q.‘ 14 assertive, independent, ambitious, and self-reliant. His theory, in considering innate individual differences and the strong impact of culture and environment on personality development,is one that is useful for both men and women. According to Maslow (1968) the self-actualized person: - is autonomous and independent - has a realistic orientation - shows democratic values - tends to be spontaneous, open, and relatively free of unhealthy defenses - exhibits an unstereotyped appreciation for people and things - is problemrcentered rather than self-centered - resists conforming to cultural expectations, often through creativity - has intimate, fulfilling interpersonal relationships - enjoys privacy and solitude - does not exploit others - experiences a wide range of emotions All of these characteristics can be incorporated into the model of the mentally healthy woman and reflect an inte- gration of instrumental and expressive qualities used with flexibility. The three theories presented--androgyny, sex-role transcendence, and self-actualization--all are relatively free of sex bias. They represent sex-fair theories of 33ersonality development in recognizing the role of sex- Jrole socialization on development and in supporting the development in every individual of instrumental and expressive qualities. All three have as a goal of therapy an individual who is flexible, spontaneous, and open to choosing behaviors which are situationally appropriate without restrictions of superimposed sex-roles. 15 Sex-Fair Counseling Theory In response to documented sex bias in more tradi- tional personality theories and paralleling the emergence of more sex-fair theories of personality development, are theories of sex-fair counseling. Two approaches to treat- ment appear to be sex-fair, nonsexist .mmi feminist therapy. While these theories of therapy are in the early stages of development, they provide direction for therapeutic experiences that are relatively free of sex bias. A number of similarities are observed between the two approaches of nonsexist and feminist therapy. Neither approach relies upon a particular set of therapeutic techniques, theorizing that almost any existing therapeutic approach, with the exception of Freudian analysis, can be accommodated to nonsexist or feminist therapy. Both approaches reject biological bases for sex differences in favor of socio-cultural explanations and oppose restrictions placed on men and women due to gender. Neither theory prescribes to an adjustment model of mental health, nor maintains a double standard of mental health for men or women (Rawlings and Carter, 1977). Nonsexist and feminist therapists seek equality for men and women, use androgyny, sex-role transcendence, or self—actualization as models for Inental health, and believe that to gain equality a woman Inust "learn to know, to respect, and to use her own self- Lnnderstanding" (Collier, 1982, p. 36). ‘ds a": n; 16 The major distinction between nonsexist and feminist therapy theory is that feminist therapy incorporates the values and philosophy of feminism into its therapeutic values and techniques, while nonsexist therapy does not. ”A basic assumption underlying feminist therapy is that ideology, social structure and behavior are intricably woven” (Gilbert, 1980, p. 347). Feminist therapy not only incorporates an awarensss of the cultural Status of women and its effects on behavior, but advocates for the develop- ment of autonomous and self-actualized women and the eventual goal of establishment of a social structure more representative of the feminist ideology of egalitarianism (Lerman, 1974; Maracek and Kravetz, 1977). An under- standing of feminist philosophy and goals underlies the two main principles of feminist therapy. These principles are: (1) the person is political, and (2) the therapist-client relationship is egalitarian. In an attempt to further describe and outline the similarities and differences between nonsexist and feminist therapy the basic assumptions of the two approaches are presented. The basic assumptions of nonsexist therapy have been derived from a number of sources (Williams, 1976; Rawlings and Carter, 1977; Gilbert, 1980; Collier, 1982) and are as follows: 1. The therapist is aware of her/his values especially as they relate to sex-role issues. S/he works to heighten her/his awareness and to 7. 17 explore conscious and unconscious beliefs about "male" and "female." The therapist has no preconceived notions about appropriate sex-role behaviors. S/he encourages the client to behave in ways which are effective and to make decisions based on individual goals rather than on what is expected of them because of their sex. Role-reversals in clients are accepted and not labeled pathological. Diagnoses are not based on the client's success or failure in achieving culturally prescribed sex-roles. Both instrumental and expressive characteristics in male and female clients are valued. Men and women are seen as individuals with equal capa- bility of being autonomous, assertive, inter- dependent, tender, and expressive. The dominance of biology in determining sex differences is rejected in favor of a socio- cultural explanation for sex differences. The therapist does not use his/her power to subtly reinforce sex-role appropriate or inappro- priate behaviors. Sex—biased testing instruments are not used. The basic assumptions of feminist therapy incorporate those of nonsexist therapy. Other assumptions are added to those of nonsexist therapy reflective of the underlying philosophy of feminism. The assumptions of feminist therapy have been derived from a number of sources (Williams, 1976; Rawlings and Carter, 1977; Gilbert, 1980; Collier, 1981) and are as follows: 1. The inferior status of women is due to the power differential between men and women. The primary source of women's emotional problems is social, not personal. The client is helped to understand the relationship between sociological factors and psychological factors in contributing to her distress. 18 3. Although the main source of a woman's problems is seen as external to her, this does not relieve the individual of responsibility for her choices. The individual can change herself and her external environment. 4. Relationships with women are as valued as relation- ships with men. Both men and women are victims of sex-role socialization. Other women are seen as sources of social support in attempts to change, rather than adjust to, the status quo. Through relationships with women, validation of the female experience is achieved and self-nurturing is encouraged. 5. Psychological and economic autonomy are important goals for women. The client is encouraged to experience a sense of personal power through greater self-confidence, expression of anger, self- direction and autonomy. While both theories presented incorporate humanism, it is clear that nonsexist therapy has its base in humanistic psychology. while feminist therapy has as a base socio- logical or political ideologies. Neither approach is judged as better or worse, but it has been shown that nonsexist therapy may be more useful than feminist therapy for women who are more traditional and less attuned with the philosophy of feminism. Feminist therapy has been found to be useful for female clients who express dissatisfaction with role constrictions, have some commitment to feminism and are actively seeking alternatives (Rawlings and Carter, 1977; Maracek, Kravetz and Finn, 1979). Any form of effective therapy begins where the client is and moves with the client to wherever s/he wants or chooses to go. Nonsexist and feminist therapies are no (tifferent from.other therapeutic approaches in this respect. 19 Definition of Terms Sex Bias in Psychotherapy For the purpose of this research the APA Task Force's (1975) definition of sex bias in psychotherapy will be adopted. Sex bias in psychotherapy is defined as fostering of traditional sex-roles, demonstrating bias in expecta- tions and devaluation In? women clients, sexist use of psychoanalytic concepts and treatment of women clients as sex objects. The development of the Sex Bias in Psycho- therapy Questionnaire (SBPQ) is based on this definition and provides further elaboration for each of the four areas of perceived bias. Sex-Role Stereotype Broverman, et a1. (1970, 1972), define sex-role stereotypes as "highly consensual norms and beliefs about the differing characteristics of men and women" (p. 1), unsupported by research on sex differences. Purpose of the Study The primary purpose of this study is to determine whether or not women clients perceive and report sex bias in their psychotherapy. Therapist characteristics and client life status categories associated with perceptions and report of sex bias will be explored. In order to accomplish this purpose, an instrument is needed which ciirectly assesses women clients' experience of sex bias in psychotherapy. The second purpose of the study is to a a 20 develop the Sex Bias in Psychotherapy Questionnaire (SBPQ) and to provide some initial reliability and validity data for the SBPQ. Research Questions Two sets of research questions guide this study. The first set of questions focus on the reliability and initial validity of the Sex Bias in Psychotherapy Questionnaire. The second set of questions center around the actual phenomena of sex bias in psychotherapy. Ten main questions are researched. These questions are stated generally in this section and in statistical form in Chapters IV and V. Instrumentation Questions 1. Is the SBPQ a reliable measure of sex bias in psychotherapy? 2. Is the SBPQ measuring a unidimensional construct? 3. Does the SBPQ have content validity? 4. Is there evidence of construct validity on the SBPQ? Phenomena of Sex Bias in Psychotherapy Questions 5. Do women clients experience sex bias in psycho- therapy? 6- Do women in different life status categories and with different therapy experiences re ort different amounts of sex bias in psyc otherapy? 21 7. Do women clients experience more sex bias in psychotherapy with male therapists than with female therapists? 8. Do women clients experience more sex bias in psychotherapy with inexperienced therapists than with experienced therapists? 9. Do women clients experience more sex bias in psychotherapy with Community Mental Health Center therapists or with Counseling Center therapists? Overview In Chapter I, the problem to be investigated was presented along with the need for, importance of, and theory framing the research. In Chapter II, a compre- hensive review of the literature is presented. Chapter III presents the overall design and methodology for the study. Chapters IV and V contain the results of the research. In Chapter IV instrument development of the Sex Bias in Psych- therapy Questionnaire is discussed, along with an analysis of the reliability and validity results. Chapter V contains an analysis of the results pertaining to the phenomena of sex bias in therapy. In Chapter VI a summary is presented and conclusions drawn from the results. Impli- cations of the research are discussed. CHAPTER II REVIEW OF THE LITERATURE The purpose of this literature review is to present theory and research relevant to the area of sex bias in psychotherapy. The five broad areas covered are: clinical theory on sex-role related therapy issues, therapists' sex-role stereotypes and attitudes towards women, therapists' judgments and evaluations, therapy process, and therapy outcome. A summary is provided with conclusions drawn from the research reviewed. Clinical Theory on Sex-Role Related Therapy Issues As early as 1938, Thompson in a clinical paper recognized the importance of the sex of the analyst. Of the many factors significant in the choice of an analyst, including age, cultural background and personality makeup, sex of the analyst was viewed as very siginficant in this choice. For some clients, the sex of the therapist has little to do with his/her choice; for others the choice is made primarily around the sex of the therapist. According ‘to Thompson, there are any number of reasons for a client Choosing a therapist of one sex over the other. Some Clgients seek out the person with whom they feel most 22 23 comfortable and are most capable of relating to intimately. Others choose someone who is not likely to fulfill unpleasant life patterns. There are also clients who, in reaction to issues of authority and power, choose an analyst with the assumption that the analyst's sex-role stereotype will meet their needs. Clearly, the choice of an analyst because of gender sometimes stems from neurotic conflict and can serve a defensive function. Yet, the client's preference with regard to sex of the therapist is most often honored. In light of this, Thompson stresses the importance of the resolution of the analyst's own sex-role conflicts. If these sex-role conflicts are not resolved, Thompson sug- gests that countertransference issues may prevent the reso- lution of the client's conflicts. The impact of sex-role related countertransference on women clients has been discussed by Bernardez (1975). Her theory is that unconscious beliefs about women affect their psychotherapy. While many therapists consciously deny stereotypic notions about female behavior, Bernardez notes it is often in their actions that unconscious biases about ‘women become apparent. From her own work with clients and in the work of therapists she has supervised, Bernardez identifies areas "betraying prejudicial views of women." One area concerns the reaction to openly hostile behavior or a domineering attitude in women clients. Bernardez observes that therapists experience strong feelings in response to this behavior but often restrain these feelings during the hour. It is in supervision that the stereotypic labels "castrating," "destructive," and "competitive" are often heard. The therapist is covertly disapproving of the client, while acting with approval and understanding. In the session, "interpretations" are made which accentuate the client's guilt, by pointing to the destructiveness of her feelings. No understanding of the woman's defensive or protective use of this behavior is offered. Bernardez finds this particularly apparent in family therapy, where strong reactions occur towards women deviat- ing from sex-role stereotypes. The domineering wife does not receive help in learning assertive ways to meet her needs, while the submissive husband often receives train- ing in the very behavior disapproved of for the wife. Bernardez theorizes that the revulsion experienced in reaction to women's hostile or domineering behavior origi- nates in unconscious fears of women's power and destructiveness, as discussed by Lerner (1974). As further support for this theory, Bernardez cites examples of therapists' behavior with compliant, submissive, and self-effacing women clients. Here the problem is one of (unission. Therapists too often do not confront or question this type of behavior in their women clients, which, although pathological, is in accordance with sex-role stLereotypic cultural norms. It is theorized that the 25 therapist implicitly reinforces this behavior out of fear of its opposite, an independent, assertive stance. As stated by Thompson, Bernardez too feels it essential for therapists to be aware of their own sex-role related conflicts. This means that therapists need to closely examine their unconscious beliefs about women. Kaplan (1979b) has analyzed sex-role related issues in the therapeutic relationship, by focusing on sex-role socialization theory and concepts of androgyny. She explores countertransference issues related to male and female therapists' preparation to deal with the aspects of authority and empathy inherent in their role. Her theory maintains that due to sex—role socialization, male and female clinicians are differentially trained to deal with issues of authority and experience empathy in the therapeutic relationship. Likewise, male and female clients hold different expectations for their male and female therapists in regard to these issues. Kaplan's theory is particularly relevant for female therapists in the role of authority. She observes that women therapists, who are rarely trained for roles of authority, struggle with discomfort concerning their authority and sometimes belittle or undermine their own power. Kaplan perceives this in women clinicians as idndecisiveness, a tendency to smile too much or giggle and a. willingness to relax the therapeutic boundaries and relate more as a peer to clients. At the same time, women 26 therapists often have to deal with clients' subtle challenges to their competence and authority. These challenges include patterns of evasiveness and tentative resistance to the female therapist's authority. Different countertransference issues arise for women therapists around clients' expectations and acknowledgment of the female therapist's empathic abilities. Kaplan theorizes that a complex countertransference situation can develop, which stems from the female therapist's unconscious wish to receive. This desire to be cared for is translated into unconscious wishes for loyalty, acceptance, and personal validation from the client, in return for her empathic understanding. The effect of this on the thera- peutic relationship is seen in the client's ambivalence about closeness. The wish;finrcloseness and nurturance, combined with the fear that this closeness and acceptance of the therapist's empathy may inhibit anger, creativity and independence, is observed in the client's subtle distantation from the therapist,varied with requests for more closeness. Kaplan does not address the issues of authority and empathy in as much depth for male clinicians as for female clinicians. Nonetheless, she maintains that sex-role training differentially prepares male clinicians as well as female clinicians for the use of authority and empathy in psychotherapy. Kaplan observes that male clinicians, unlike female clinicians, feel comfortable with their authority. 27 This comfort has implications for the treatment of sub- missive as well as assertive women clients. Male thera- pists, because of their comfort with authority, often exercise this authority in a hierarchical relationship which reinforces feminine sex-role stereotypic behaviors of passivity, submission, and dependency. Kaplan further theorizes that male therapists feel uncomfortable with female anger and assertiveness because of its implied challenge to this hierarchical relationship. She observes that male therapists often react to anger in women in a distant manner which becomes less so when the anger is turned to tears and self-blame. Minimal theorizing about male clinicians' issues with empathy is offered by Kaplan. Her premise regarding male clinicians' empathy is that due to sex-role socialization, they are less empathic and have a more difficult time accurately assessing their clients' emotional experience. Kaplan makes a strong point for the incorporation of both masculine and feminine characteristics in the therapist, as a way of dealing with and eliminating these counter- transference dynamics. She notes, though, that for women the integration of masculine characteristics feels like growth, whereas for men the incorporation of feminine characteristics and the letting go of some masculine czharacteristics feels like a compromise and can be threaten- izng. Nevertheless, Kaplan advocates for more adrogynous tIIerapists. She concludes her paper with a statement "that 28 a knowledge of basic patterns of sex—role socialization can attune the clinician to pertinent clinical dynamics that might otherwise go unrecognized or misperceived. It is suggested that sucha sex-role analysis be added to the framework with which we attempt to understand the nature of the therapeutic relationship" (p. 118-119). In summary, the theories presented cover theory focusing on client dynamics in the choice of a therapist, theory regarding therapists' unconscious beliefs about women which affect psychotherapy,and socialization theory which provides an understanding of the differential prepara- tion of male and female clinicians for their role as therapist. All theorists stressed the importance of the therapist's awareness of his/her sex-role conflicts, whether they be unconscious or socially determined. This awareness as well as resolution of their own sex-role conflicts seems essential if therapists are to aid in the resolution of their clients' conflicts. The theories, while originating in different schools, overlap and provide a basis from which to explore the possible biased treatment of women clients and sex-role stereotyping within the psychotherapeutic relationship. Therapists' Sex-Role Stereotypes and AtEitudes TOwards WOmen Research has been conducted on therapists' sex- rxale stereotypic attitudes towards men and women. The land- muark investigation in this area, and one often cited in the 29 literature is a study by Broverman, Broverman, Clarkson, Rosenkrantz, and Vogel (1970). The purpose of their study was to determine if sex-role stereotypes influenced clinical judgments of mental health. They developed the Stereotype Questionnaire, an instrument with 122 bipolar adjectives describing behavior traits, to measure stereotyping among counselors. Of the 122 items, 37 were considered stereo- typic of men or women and are the items scored. The researchers administered the Stereotype Question- naire to 33 female and 46 male psychologists, psychiatrists and social workers. One of three sets of instructions was given to the clinicians. One group was asked to fill out the Questionnaire for the healthy male. They were to determine where on each item a mentally healthy male would fall. The second group was to do the same for the mentally healthy female, the third for the healthy adult. Broverman, et al., found that clinicians held different standards of mental health for men and women. Relative to men, a healthy adult woman was seen as more sub- missive, less independent, more easily influenced, less aggressive, less competitive, more excitable in minor crises, having their feelings hurt more easily, more emo- tional, more conceited about their appearance and less objective. The healthy adult was seen as similar to the liealthy male, but as significantly different from the healthy female. The ratings between male and female clinicians did not differ. Both groups were found to 3O maintain a double standard of mental health :flm: men and women. The authors suggest that this double standard of health stems from an "adjustment" notion of health. For a woman to be considered healthy, She must conform to sex- role norms even if this is less socially desirable and viewed as less healthy for the competent adult. Maslin and Davis (1975) replicated Broverman, et a1.'s study with counselors-in-training. They administered a shortened version of the stereotype questionnaire to 45 male and 45 female counselors-in-training. Their results partially replicated those of Broverman, et a1. (1970). Maslin and Davis found no difference between men's and women's perceptions of the healthy adult or the healthy male, but found a.sex difference between the male and female counselors-in-training perceptions of the healthy female. Female counselors-in-training held the same set of expectations for all healthy persons. Male counselors-in- training held somewhat more stereotypic expectations for healthy females as compared with standards of health for males and adults. The difference in these findings as compared to the Broverman, et al., findings was attributed to historical change and the differential impact of the women's movement on male and female clinicians. In a recent replication of the Broverman study, Bflaxfield (1982) analyzed results on the Stereotype Chaestionnaire from 147 male and 107 female clinicians. Half of her sample was trained before 1970 (pre-Broverman), 31 the other half after 1970. She found that healthy males and females were still seen in sex-role stereotypic ways by both male and female clinicians, a confirmation of Broverman et a1.'s original finding. Contrary to Broverman et a1.'s finding, characteristics of the healthy adult were not limited to masculine qualities. Rather, the healthy adult was seen as androgynous, possessing both masculine and feminine characteristics. No difference was found between those clinicians trained before 1970 and those trained after 1970. Maxfield concludes that although standards of mental health remain sex-role stereotypic for males and females, the androgynous view of the healthy adult represents progress in eliminating sex-role stereotyping within the psychology profession. Delk and Ryan (1975) studied the relationship between A-B therapist status and attitudes regarding sex roles for the mentally healthy male and female. Subjects were asked to rate a healthy male and healthy female on the Stereo- type Questionnaire. Both male and female ratings were on the same form, which is different from the procedure of having subjects rate males and females on separate forms used by Broverman, et a1. (1970). In using this parallel rating it was assumed that clinicians would make more conscious decisions about stereotyping. "The results indicated that A therapists (those who work better with schizophrenic clients) stereotyped to a greater degree than 28 therapists (those who work better with neurotic clients). 32 In an attempt to broaden the application of Broverman, et a1.'s Stereotype Questionnaire, Aslin (1977) asked therapists to fill out the Stereotype Questionnaire for wife, mother, healthy female, and healthy adult. The purpose of the study was to see if the roles of wife, mother, healthy female,and healthy adult were perceived as compatible. Her sample consisted of 75 female and 55 male community mental health clinicians and 82 feminist therapists. One of four sets of instructions was randomly given to the therapists. They were either asked to fill out the Question- naire for the healthy adult, woman, wife, or mother. Aslin found that female community mental health and feminist therapists maintained one standard of mental health for females, wives, mothers, and adults. Male community mental health therapists showed significant differences in their perceptions of the healthy adult and wife when cempared to women and mothers. They perceived the healthy adult and wife in more male-valued terms than they did mother and female. Although differences were found between male and female clinicians' perceptions, none of the groups responded to the roles with feminine sex-role stereotypic terms. The male therapists responded to the adult roles with significantly more masculine sex-role stereotypic items than any other role. Contrary to Broverman, et al., but similar to Maslin and Davis, Aslin found that sex of the judge effected ‘therapists' perceptions of normal mental health for a variety 33 of roles. She also found female therapists to be more similar than different to feminist therapists. Even though male clinicians in this study maintained different standards of mental health for the roles of wife, mother, female, and adult, no clinician saw women in stereotypic terms, but rather as incorporating both masculine and feminine characteristics. This finding as Maxfield's find- ing suggests that clinicians are more able to recognize the value of androgyny for healthy functioning. Kahn (1976) also compared male and female clinicians with feminist therapists on clinical judgments and sex- role stereotypic notions of mental health. Her sample consisted of 45 male clinicians, 44 female clinicians, and 40 feminist therapists. Therapists were mailed an instru- ment which incorporated many items from the Stereotype Questionnaire, items pertaining to attitudes towards women and clinical judgments about written vignettes. Her find- ings indicate that clinicians not only expect the same characteristics in a mentally healthy woman as were previously expected for a healthy adult man or adult, but that they hold relatively non-traditional views of women. The male therapists were the most likely to reveal stereo- typic biases, whereas the feminist therapists were the most egalitarian. It was noted that in areas where men's needs or privileges might be adversely affected, male and female 'therapists were less interested in the woman client's needs t:han were feminist therapists. Kahn concludes that feminist 34 awareness is a more important influence on clinical judg- ment than sex of therapist. Hayes and Wolleat (1978), in a simulated counseling interview, attempted to determine if sex-role stereotypic judgments would be made about male and female clients by therapists. Twenty male and 20 female graduate students in counseling psychology listened to a five-minute tape of a simulated therapy session with either a male or female client. The sessions were identical except that one session was with a male client, the other with a female client. After listening to the tape, the therapists were asked to fill out the Stereotype Questionnaire for the client. While no effect was found for the sex of the counselor, sex of the client elicited differences. There was a decided trend for female clients to be seen as leaning more towards the masculine end of the continuum than the male client. Stereotypes were consistently reversed for the clients presented. These findings are contrary to those of Cowan (1976). In her study with therapists in practice, she found that the problems of women clients were perceived along sex-role stereotypic dimensions. Women clients were perceived to have too few masculine characteristics and too many stereo- typical feminine characteristics. The problems of male clients were not seen in sex-role stereotypic dimensions. A recent naturalistic investigation by Jones and Zoppel (1982) requested male and female therapists to rate 35 clients on an adjective checklist. This study differs from most in that rather than requesting the clinician to make a judgment about the generic male or female, the clinician was being asked to describe a real person, a client with whom s/he had worked. This research study was much less sensitive to social desirability, and the results reflect this fact. The data reported by Jones and Zoppel suggest that male therapists, in contrast to female therapists, endorse less positive adjectives in describing clients, particularly women clients. Women therapists' characterization of their female clients were more sympathetic and reflected an aware- ness of the woman's competencies and strengths. Male therapists were consistently more severe and stringent in their descriptions of clients than female therapists. These results suggest that women therapists may be more accepting and tolerant of clients, thus able to display more uncondi- tional positive regard. There have been a few studies, which, rather than look- ing at stereotyping, have looked at therapist attitudes towards women. In 1975, Brown and Hellinger found that therapists' attitudes towards women were normally dis- tributed among their sample of psychiatrists, psychologists, social workers, and psychiatric nurses. Female therapists, though, were found to have more contemporary attitudes towards women than male therapists. 36 Davenport and Reims (1978) used a revised version of Brown and Hellinger's Attitudes Towards Women Scale to explore the possible associations between theoretical orientation, sex of clinician, and traditional or contempo- rary attitudes towards women's roles. Their sample was composed of two groups of clinicians-in-training. One group was being trained in an intrapsychic/dynamic orienta- tion, the other in a more sociological/systems framework. It was hypothesized that the clinicians with an intra- psychic orientation would hold more traditional attitudes towards women than those from a more sociological orienta- tion, and that male clinicians would hold more traditional attitudes towards women than female clinicians. Theoretical orientation produced no difference. However, sex of the therapist was found to be signfiicant in the hypothesized direction. Using a more reliable and valid scale, Helwig (1976) found differences in male and female employment counselors' attitudes towards women. Helwig administered Spence and Helmreich's (1972) Attitudes Towards Women Scale to 60 male and 20 female employment counselors. Although employment counselors as a whole were found to have more liberal attitudes towards women than the general population, male counselors held significantly less liberal attitudes towards women than did female counselors. There was also a correlation found between age and attitudes towards women. 37 Older counselors held more traditional attitudes towards women than younger counselors. Gilbert (1981b) studied the influence of client gender and student-counselor attitudes towards women on counselors' perceptions of sex-role behaviors. Twenty-one female and 16 male students in graduate-level counseling classes were used as subjects in this study. They were asked to read a 400-word written vignette describing either a male or female client's reaction to his/her first term at college. Follow- ing this, they filled out an evaluation form which had embedded items regarding role behavior. Two weeks earlier the student-counselors had filled out Spence and Helmreich's Attitudes Towards WOmen Scale. The results showed that counselors viewed the male client as more likely to succeed in education and work roles and less likely to perform well in spouse and parent roles than the female client. The client's sex did not influence counselor judgments about the likelihood of obtaining a degree. Both clients were rated as equally likely to obtain their college degree. In considering counselor attitudes towards women, it was found that counselors with liberal attitudes held more egalitarian beliefs about male and female clients' abilities to perform well in work and educa- tion roles than did moderate counselors. Attitudinal differences did not influence counselor judgments regarding family roles. Both liberal and conservative counselors Inade similar predictions about success in family roles, (I) I ll) 1 IN AH 38 suggesting that these roles may be more firmly established and more resistant to change. While the results of this study are limited due to the analogue, small sample, and neglect of the counselor sex variable, it is one study which examines the relationship between attitudes towards women and sex-role stereotyping in a counseling Situation. It is a well designed and executed study which provides some basis for further exploration into the relationship between therapist attitudes and values and the treatment of female clients. In summary, results of studies on therapists' sex-role stereotypic attitudes towards men and women indicate that a double standard of mental health exists for men and women. Clearly, progress has been made in this area, for therapists are now able to recognize the value of androgyny for normal, healthy adult functioning. Due to the inconsistencies in research findings in this area, further conclusions would be premature. Therapist Judgments and Evaluations It is assumed that therapist judgments and evaluations of clients are affected by the client's sex and therapist's gender. The results of research in this area have not supported this notion. Overall, findings Show few main effects, but some noteworthy interaction effects. It is important to note that the majority of research in this area is analogue research. Consequently, there are limitations pm 5‘ ‘ E8? '1' r) ab PC '1.) n: 'C’ [1 39 to these findings and results must be interpreted with prudence. One of the earlier studies in this area was done by Abramowitz, Abramowitz, Jackson and Gomes (1973). Their purpose was to clarify the role of political bias in clini- cal evaluations of women clients. Seventy-one professionals in mental health or education were asked to read a case history and then to rate the client's psychological adjust- ment. The case histories were identical with the exception of the client's sex and political orientation. These variables were systematically varied to yield four different case histories. Therapists were also asked to complete questions measuring their relative liberalism on foreign affairs. Neither the politics of the examiner nor the sex of the client alone affected the ratings of the client's adjust- ment. An interaction between the politics of the examiner and the sex of the client was observed. Clinical inferences about men were less strongly related to the evaluator's political philosophy than for women. The left-oriented woman client was judged more harshly than the male client by the less, not the more liberal examiners. The researchers conclude that the "bias against the left-oriented woman raises the spector of covert discrimination against the 'liberated' woman, unintentional though it may be, on the part of certain workers holding unsympathetic socio- political views" (p. 88-89). 40 In an effort to extend the data base on sex bias in evaluations of clients, Abramowitz, Weitz, Schwartz, Amira, Games and Abramowitz (1975) looked at judgments of women with and without medical school aspirations. A small sample of counselors and graduate students in counseling, half of whom were women, were asked to read two case histories. The case histories were either a male or female aspiring to medical school or a male or female client with- out these goals. Each counselor received two case histories, one case being that of a client aspiring to medical school. The counselors, after reading the Cases, made a psychologi- cal adjustment rating and completed a measure of traditional moralism. Assessor sex was found to be an important variable in ratings of adjustment. Women assessors were more lenient in their ratings than male assessors. In addition, the researchers found that more traditional assessors tended to make harsher judgments of help-seeking behavior than did more liberal counselors. Not surprisingly, medical school aspiring women received more stern judgments from tradi- tional than liberal examiners. Games and Abramowitz (1976) continued to study sex- related bias in relation to therapists' traditionalism and clients' sex-role normative or counternormative behavior. Once again they prepared two case histories, one male and one female, which were identical except for the last paragraph. The last paragraph couched the client's 41 presenting problem in either sex-role stereotypic or counter sex-role stereotypic terms. This variation yielded four case histories: one male sex-role stereotypic, one male sex-role deviant, one female sex-role conforming, and one female sex-role deviant. The case histories, along with a Client Evaluation Form measuring the client's emotional maturity, social adjustment, mental disorder and prognosis, as well as an Attitudes Towards Traditional Sex-Roles scale, were responded to by 99 male and 83 female therapists. No consistent effects were found for any of the variables. However, some interaction effects were noted. The sex-role deviant female was perceived as more mature than the sex-role conforming client and both male clients. In addition, female clients were rated higher than males in social adjustment by the more traditional clinicians. Male judges in general gave a better prognosis to female clients than female judges. These results seem to contradict those found in Abramowitz, et a1. (1975). The absence of con- sistent effects was attributed to the possible transparency of the study as well as to enhanced professional sensitiza- tion to issues of sex-role stereotyping. Thomas and Stewart (1971) also looked at judgments of female clients with feminine (conforming) career goals or traditionally masculine (deviate) career goals. Sixty-two practicing counselors listened to three simulated inter- views of female clients on audio tape. The groups heard ' identical client interviews but to one group the client was 42 introduced as having a deviate career goal and to another group the same client was introduced as having a conforming career goal. After listening to the client, counselors were asked to fill out the Cough Adjective Checklist as a measure of acceptance of the client, to rate the appropriateness of the career choice and to determine the need for further counseling. The counselors as a group did not show any difference in their acceptance of clients with deviate or conforming career goals. However, experienced male therapists were more accepting of women clients with deviate career goals than were inexperienced male counselors. An experience effect was also noted for the counselor's rating of the appropriateness of the client's career goal. Inexperienced male and female counselors but not experienced counselors perceived a conforming career choice as more appropriate than a deviate career goal. Interestingly, it was only the experienced female counselors who perceived women clients with deviate career goals to have greater need for counseling than women with conforming goals. This may be an indication of sex bias, but may also reflect the experienced women clinicians' knowledge about the diffi- culty of pursuing a non-traditional career choice. The results of this study only minimally support the existence of sex bias in counseling. Nonetheless, they raise important issues concerning the interaction of sex bias with experience level of the counselor. St 'L) Y?- C t”: 43 Recognizing the importance of interactions in studying sex bias in clinical judgment, Schwartz and Abramowitz (1975) factorially varied client sex and race with psychiatrist traditionalism and experience in an analogue study. The purpose of their study was to clarify the role of political bias in forming psychiatric impressions. One hundred-and-two psychiatrists, predominantly male, made clinical judgments from an extensive case history which varied only on client sex and race. Therapists were asked to make judgments as to the severity of maladjustment, prognosis, desirability of hospitalization, ECT, chemo- therapy, and insight-oriented treatment. Psychiatrists also filled out the Traditional Beliefs Form as a measure of political orientation. Although some interaction effects were noted, the major finding of this study was the inability to detect psychiatric bias against clients identified as black or female. Male clients were judged to be more maladjusted than female clients by the more traditional psychiatrists. It was also generally observed that experienced psychiatrists gave more negative appraisals of women clients. Experience in this study did not act as an insulator against bias. More recently, Wright, Meadow, Abramowitz and Davidson (1980) found sex differences in psychiatric diagnosis to be a function of assessor sex. Moving from the lab to a natural setting, the researchers examined actual diagnoses made by 26 clinicians for 200 clients randomly assigned to 44 clinicians for intake interviews at a community mental health center. The clinicians included physicians, psycholo- gists, social workers, and nurses. Complex interactions were found between clinician discipline, clinician sex, and client sex. Women clinicians tended to make more lenient diagnoses for the clients than male clinicians. This was especially true when the client was female. Women clinicians were disinclined to regard their own women clients as psychotic. For male clinicians, discipline was a factor, in that male psychiatrists made the most psychotic diagnoses, male psychologists the next highest, and so on. This finding did not hold for social workers. In concluding, the researchers suggest that the female clinician's leniency in clinical judgments about client pathology may lie in their sex-role socialization which creates a conflict for women clinicians between nurturance and authority. Bearing in mind that therapists are not only asked to make initial judgments as to client pathology and level of functioning, researchers have attempted to study sex-role stereotyping as it relates to treatment goals and planning. Stearns, Penner and Kimmel (1980) provided practicing clinicians with data on client history, presenting problem, symptomatology, and gender. Their purpose was to determine if different standards of mental health were operating for male and female clients in regard to treatment recommenda- tions. One of eight videotaped case conferences was viewed 45 by 86 psychotherapists. The client presented was either male or female with symptomatology of aggression or depression. After viewing the tape, opinions regarding the client's presenting problem, current level of functioning, and prognosis were gathered along with treatment recommendations. The results showed no difference between clinicians in regard to treatment recommendations, initial perceptions of the client or prognoSis. Among this group of clinicians, though, men who sought therapy and acknowledged emotional problems were viewed as more distressed than women with the same symptoms. The authors conclude that when variables other than sex are independently manipulated, they take precedence over gender in decision-making. They speculate that it may be when symptomatology is vague that sex-role stereotypes are most operative. This assumption finds some support in a study by Miller (1974) designed to investigate the effects of a client's sex on clinical impressions and treatment planning. A case analogue was devised presenting an "ambiguous enough clinical picture to allow some variability in clinicians' judgments" (p. 92). All variables except the sex of the client were held constant. The most marked clinical feature of the case was the client's obvious passivity. A group of social workers, psychiatrists, and psychologists made judg- ments about the client's level of functioning, diagnostic classification, major difficulties, areas designated for the focus of treatment and type of treatment. 46 The results showed a consistent tendency for the female client to receive more favorable judgments and to be seen as healthier than the male client. In all cases, the client's major problem was seen as passivity. In view of this, the results regarding the focus of treatment are interesting. Treatment of the client's passivity was the overwhelming choice for the focus of treatment for the male client, while only half of the clinicians saw this as the necessary focus of treamment for the female client. No differenceszu1diagnostic classification or type of treat- ment recommended were observed nor were any clinician sex differences noted. While there are limitations to this study, in that only one case was presented, the findings regarding the client's passivity support a sex-role stereotypic notion of mental health for female clients. The finding that passivity in the female client would less often be a focus of treatment than for the male client, suggests that these clinicians adopted an adjustment notion of mental health. While passivity in a woman client may be growth inhibiting, it is sex-role appropriate and seen as an area less in need of change than for a male client. Miller's finding poses a question as to how clinicians would view a female client with a normal amount of adaptive aggression, if passivity is accepted as role appropriate, and raises concerns about the therapist's covert reinforcement of a client's sex-role conformity in psychotherapy. 47 A study by Bowman (1982) sheds further light on the differential treatment of ”activity" in men and women receiving counseling. Bowman designed an analog study in which 61 male and female psychotherapists read a fictional intake summary of a male or female client portrayed as "active" in work, sex, and interpersonal relationships. The presenting problem of the client was defined as a concern about a deteriorating relationship with her/his spouse since taking a new job. The client was struggling with career and family concerns. After reading the case summary, each therapist came to a formulation of the client's problem, suggested issues to be explored in treatment, and specified treatment goals. Additionally, therapists were asked to choose either individual dynamic, couples or "other" therapy as a treatment modality for the client. After the treatment plan was finished, the short form of the Attitudes Towards WOmen Scale was administered to the therapists. Significant differences were found between client sex on therapist assessment. The problem for the male client was more often defined as ”mutual anger" between husband and wife, with the problem for the female client more often seen as being a conflict about sexual identity and dominance in marriage. Congruent with problem formulation, developmental issues were identified for exploration with the female client much more often than for the male client with the goal for the female client being to achieve a1 better balance 48 of roles. Individual insight-oriented therapy was most often the treatment of choice for the female client; couples therapy most often the choice for the male client. No inter- action was observed between client assessment and therapist attitudes towards women. Bowman concludes that therapists do respond to the client's sex when making treatment plans. Further, she cites evidence of bias against activity in the woman client observed in the expectation that she compromise her career ambition to achieve a balance with her career and family roles. Bias was also observed in the understanding of the woman client's problem to be intrapsychic, while the identi- cal problem presented by a man was viewed from a system's perspective. One wonders what type of bias would be present for a woman with this problem who also had children. The results of this study point to the importance of therapists examining their conceptualization of appropriate roles for men and women and the consequences of their beliefs for clinical practice. In a similar study Billingsley (1976) studied the extent to which a pseudo-client's sex and presenting pathology influenced treatment goals. She hypothesized that treatment plans for a woman client would emphasize an increase in stereotypic "feminine" behavior and for a male client an increase in stereotypic "masculine" behavior. Two case histories, one of an explosive client and one of a restricted client, were developed. Therapists read either 49 two cases designated as male or two cases designated as female. Treatment interventions were measured by requesting that therapistschoose six initial therapy goals from a check- list of 18 items taken from the Stereotype Questionnaire. Of the 18 items, nine were masculine-valued characteristics, nine were feminine-valued items. Billingsley found that clinicians chose more feminine goals for the explosive client than for the restricted client. WOmen therapists, however, chose a greater number of mascu- line treatment goals for all clients and male therapists chose more feminine treatment goals. Client sex did not influence choice of treatment goals. The researcher concludes from these results that therapist sex as well as client pathology is closely related to treatment planning and goal setting. Male and female therapists in choosing different types of treatment goals may inadvertently be supporting sex—role stereotypic behaviors in their opposite- sex clients. Clinician sex effects were also noted by Oyster-Nelson and Cohen (1981). Using as a model a program of psychologi- cal peer review, these researchers had 119 psychologists review prepared case histories and make judgments regarding the perceived severity of the client's problem, the necessity of psychological treatment, the degree to which a client is a good candidate for psychological treatment, the number of sessions needed for successful resolution of the problem and the appropriateness of a number of 50 treatment procedures. Each psychologist received one of three case histories. The cases were identical except that sex and symptomatology were varied. One case was of a client having relationship difficulties, another achievement- related problems, and a third neutral concerns. Few differences were found on the dependent measures. Male and female clients were seen as equally good candidates for treatment, requiring a similar number of sessions and equally appropriate for a variety of therapeutic interven- tions. The male client was seen as more disturbed and more in need of treatment than the female client and the rela- tionship problem was seen as the most severe problem. A trend was observed for female psychologists to rate insight- oriented therapy as more appropriate for males, whereas male psychologists placed more value on systematic desensitiza- tion and cognitive-behavioral interventions. The researchers suggest that this trend may be related to the female cli- nician's viewing relationship problems as more serious than the other two problems presented. This study fails to confirm the existence of sex bias in treatment recommenda- tions but points to some clinician sex differences. Taken as a whole, the findings of these studies fail to clearly confirm the existence of sex bias in clinical judg- ments and evaluations. Many studies found that the male seeking treatment was seen as less healthy. At first glance this appears to be sex bias towards the male client and may in fact be just that. It also suggests that less 51 may be expected of women when they enter treatment; thus they are seen as more healthy because of the double standard of mental health which is applied to them. While this is speculation, it is not speculation that in this group of studies, complex interaction effects were observed and clinician sex effects were found to be an important variable in clinical decision-making. The next group of studies looks at these variables in the therapy process as they relate to sex bias and sex-role stereotyping. Therapy Process The studies to be reviewed in this section are more balanced between analogue and naturalistic research. The external validity of these studies is better than those reviewed thus far. Findings, though, of sex-role stereotyping and sex bias in actual or analogue studies of therapy have been inconsistent. Again, there are few main effects but many complex interactions. Abramowitz, Roback, Schwartz, Yasuma, Abramowitz and Gomes (1976) examined the issue of sex-related counter- transference and bias in therapy in an analogue study with group psychotherapists. A detailed clinical profile of a client presenting performance as well as hostile dependency conflicts was read by 65 male and 75 female group therapists. Only the sex of the client was varied. After reading the case, therapists completed measures which determined their clinical impressions and degree of liking of the client. 52 From a list of 19 therapist responses, adapted from the Wile Group Therapy Questionnaire, therapists also rank ordered the three responses they would most likely use with the bogus client and completed the Traditional Moralism Scale. Differences with regard to therapist sex were found. WOmen therapists were consistently higher than male therapists in their expressed empathy for the client, in their use of group dynamics questions, psychodynamic inter- pretation, subtle guidance and behavior change responses. Male therapists used more clarification and confrontation than female therapists. The impact of client sex was slight. No significant results were obtained but some trends were noted. WOmen clients received a slightly better prognosis and more empathy than male clients, especially with women therapists. WOmen therapists also tended to use behavior change interventions more with women than male clients. Role-playing was more often used in opposite-sex pairs. In their discussion, the researchers note a parallel between women therapists' use of interventions supported by theory (group process and psychodynamic interpretation) as well as covertly controlling responses (subtle guidance and behavior change) and conventional feminine sex-role prescriptions for assertion. They suggest that "professional training and status may not guarantee insulation of thera- peutic reactions from incorporated sex-role ideals" (p. 708). In r—i 53 Also discussed is women therapists' enhanced interpersonal sensitivity and ability for empathy. This, too, parallels traditional feminine sex-role behavior and may provide women therapists with an important quality for therapy. Both in this study and another (Abramowitz, Abramowitz and weitz, 1976), men were not able to be as empathetic as women with their clients. In an attempt to make the analogue study more similar to actual therapy, researchers have moved to using video- tape and audiotape analogues in place of the written analogues so widely used in the early studies of sex bias and sex-role stereotyping in therapy. Johnson (1978) used videotaped stimulus material to examine the influence of counselor gender on reactivity to clients exhibiting gender appropriate and cross gender affects. Equal numbers of male and female psychologists viewed two fourdminute videotape vignettes of a 45-year-old couple in conflict resolution paired with either a 30-year-old depressed male or female or a 30-year-old angry male or female client. The psychologists were asked to respond verbally to the client at pauses in the videotape. They were to describe their immediate subjective feelings towards the client, what they would say and then completed four self-rating scales measuring their liking for the client, client attractiveness, counselor empathy and counselor comfort. Their verbal responses were judged by independent raters for 54 sympathy for the client, identification with the client, defensiveness and/or anger towards the client. Only two significant findings were obtained. Female counselors rated themselves as more empathic than male counselors and female counselors were rated by the judges as more angry than male counselors. No findings supported client gender or client affect effects or interactions. Both the male and female client, whether depressed or angry was found to be responded to in a similar fashion. Reasons for the lack of significant findings focus on the limita- tions of the analogue method, short length of the vignettes, counselor self-report and the sensitization of counselors to issues of gender so that they avoid making sexist responses. A number of researchers have commented on clinician sensitization to sex-role issues. While the research sug- gests that therapists' attitudes and sex-role stereotypic notions about women are changing, it seems possible that this sensitization helps therapists to appear non-biased in analogue research where they are being observed, but may not be the case in actual therapy. It is in analogue studies that the issue of the social desirability of responses as well as the sensitivity of clinicians to the researcher's questions becomes apparent and may influence the results. In an attempt to address these issues, Buczek (1981) investigated sex-bias in counseling using incidental memory tasks as a non-reactive measure. She was concerned with discovering if counSelors were less attentive to 55 concerns of female or male clients and whether vocational issues for women were minimized while family concerns and social issues were overemphasized. A large group of internship-level clinical psychology trainees listened to an audiotape of either a male or female client with chief complaints of loss of energy, anxiety, and depression. Each tape contained 24 social, 24 voca- tional, and 72 general facts. After listening to the tape the 89 therapists were asked to write down every fact remembered (recall), were given a true/false task containing facts from the interview (recognition), developed addi- tional questions they wanted to ask the client, evaluated the importance of the facts, and generated hypotheses con- cerning the etiology of the client's difficulties. Although some minor differences between male and female counselors were observed on recognition, counselors of both sexes remembered a similar number of vocational facts for the male and female client. Buczek assumes from these findings that where vocational issues are the concern, counselors are equally responsive to the needs of their male and female clients. More specific counselor gender and client gender interactions were found for the social scores. Male counselors, as evidenced by their remembering more facts and the content of their additional questions, showed more interest in the social concerns of women clients than male clients. Conversely, women counselors 56 showed more interest in the social concerns of male clients than female clients. Overall, findings on the total scores revealed that women counselors recalled and recognized more facts than male counselors. Counselors of both sexes, though, remembered more facts presented by the male than the female client. The fact that counselors remembered less about the female client than the male client suggests that the concerns of women clients may not be taken as seriously as those of male clients. This may be evidence of women's devaluation in the therapy process. At the same time, the consistent finding of female therapists' better retention of clients' concerns implies that women may be more attentive to the needs of their clients and in turn have more information available for treatment planning than male counselors. These findings demonstrate that gender has an indirect, if not direct influence on women's experience in therapy. Using client-confederates Shapiro (1977) explored the psychotherapeutic treatment of sex-role typical and sex- role atypical women clients. The purpose of her study was to determine whether counselors contribute to sex-role sociali- zation and whether male counselors tend to be more biased than female counselors. For the study, eight female and eight male master's-level trainees interviewed one of two client-confederates. One client role-played a traditionally feminine sex-role, the other deviated from this role in a 57 masculine direction. Interviews were videotaped and then analyzed for verbal and non—verbal reinforcement and verbal and non—verbal extinction. Counselors' general impressions of the client were obtained through a questionnaire and client perceptions of the counselors were also elicited. Additionally, counselors filled out the Bem Sex Role Inventory (BSRI) for the mentally healthy adult male and female two to three weeks following the interviews. There were no significant differences between male and female counselors' behavioral bias when responding to the two client-confederates. Counselors as a whole were more biased with the sex-role stereotypic client than with the sex-role non-stereotypic client, a finding in the opposite direction expected. This finding was supported by the analysis of the BSRI results, which showed that the healthy woman was seen as possessing more masculine characteristics than the healthy man, although both fell within the androgynous range. Shapiro noted important trends in her results. She found that female counselors were more reinforcing, less punishing, and less behaviorally biased than male counselors with female clients. Female counselors also seemed more accepting of the non-traditional client than male counselors, and were rated more positively by clients in both roles. Many of Shapiro's findings were in the direction opposite that predicted. The finding that the atypical 58 client was more positively valued suggests that the sex- role stereotypic female client in therapy may actually be treated in a more biased manner than the more "liberated" female client. The findings of this study contradict find- ings of other studies presented. Shapiro concludes that the findings of this investigation are inconclusive due to the small sample and analogue method, and that further research needs to be done in an effort to determine whether or not women are treated in a biased manner in psychotherapy. Helms (1978) in an effort to extend the external validity of the analogue therapy research regarding counselor reactions to female clients, replicated a study by Hill, Tanney, Leonard and Reiss (1977) in a naturalistic setting. In Hill, et al., 88 master's, doctoral, and post-doctoral- level mental health professionals viewed four videotaped vignettes of a 20-year-old and a 35-year-old client, presenting either personal-social or career development concerns. The personal-social concerns were either feared rape or existential anxiety and the career development concerns focused around a decision to major in either social work or engineering. After each client statement on the videotape, therapists were asked to respond in writing to the client. At the end of each vignette, counselors completed an empathy measure and a reaction form which asked for an estimate of problem severity, ability of the client to profit from counseling, client attractiveness, and the number of sessions needed. 59 The researchers found that personal problems were consistently seen as more serious than vocational problems. For the feared rape problem, women counselors viewed treat- ment as more profitable and were more empathic than male counselors. The 35-year-old client with this problem.was perceived as needing more therapy than the younger client, yet the younger client was responded to with more empathy. For the client with existential anxiety, women counselors rated the younger client's problem as more serious than that of the older client. This finding was reversed when the client's presenting problem was a career choice of social work or engineering. For the client considering engineering, women counselors viewed treatment as more profitable than male counselors. No other differences were found on dependent measures. In a partial replication of this study in a natural setting, Helms (1978) randomly selected and reviewed 32 client files of women who had terminated counseling within a six-month period. Selected clients were put into two groups, 25 years and younger, and 25 years and older. Problem type was designated either personal-interpersonal, educational or vocational. The two dependent variables used in this study were problem importance as rated by the counselor and number of sessions the client was seen in therapy. Of interest to the researcher was the relationship between these dependent variables and counselor sex, client sex and client age. 60 The findings for problem importance were similar to those in the analogue study. Naturalistic counselors perceived personal issues as most important, followed by vocational and educational concerns. For the younger clients, though, this finding did not hold. Contrary to the analogue results, vocational problems were seen as more important for younger clients than personal-interpersonal issues. In addition, the naturalistic counselors did not perceive any of the problems to be as important as did the analogue counselors. A sex difference was found in Helms' study which was not found in Hill, et a1. Female counselors perceived women clients to have significantly more problems and saw women clients for more sessions than did male counselors. No other effects were found. Helms concludes that for the most part analogue and naturalistic results were similar. In terms of the number of sessions, male and female analogue counselors came close to predicting the average number of sessions used by naturalistic women counselors. Male counselors in the natural setting used only half these number of sessions. This finding supports the finding of Hiller (1958) and McNain, Lorr and Callahan (1963) that female therapists appeared more successful than male therapists in retaining clients, but contradicts the finding of Abramowitz, Abramowitz, Roback, Corney and McKee (1976). .In their study, also naturalistic, a review of client records indicated that male therapists saw female clients for longer periods 61 than male clients. The treatment duration for male and female clients with women therapists was equal. There is controversy in the research as to whether or not male therapists foster dependency in women clients by retaining them in treatment longer than male clients. Clearly this was not the case in Helms' study, but has been demonstrated by other researchers. Either result applies to Helms' sug- gestion that "there may be an ideal number of sessions desired by counselors, but that something occurring during the process prevents male counselor—female client dyads from reaching this ideal" (p. 198). Abramowitz, et al., point to this as sex-role-related countertransference. Several of the analogue and naturalistic studies reviewed thus far have examined the effect of a client's presenting problem on judgments and therapy process. Few have used diagnostic category as an independent variable. Stein, Del Gaudio and Ansley (1976) examined treatment differences for male and female neurotically depressed clients. Their sample consisted of 44 male and female community mental health center clients, diagnosed as neurotic depressive. These clients completed instruments assessing length of stay in treatment, symptoms, moods, interpersonal styles as well as a social desirability scale. The results of this study found no evidence that male and female clients entered treat- ment with different levels of reported distress, different interpersonal styles or needs. Despite this finding, male clients had fewer therapy sessions than female clients, and 62 a higher proportion of female than male clients received some form of antidepressant medication. Most of the medica- tion prescribed to male clients was in the major tranquilizer category. Neither finding is surprising for, as previously discussed, women often stay in therapy longer than men. The medication finding also finds support in the literature. Numerous surveys in this area demonstrate that women are much more likely than men to receive medication for treatment of emotional disturbance (Cooperstock, 1978, Hughes and Brewin, 1979 and Fidell, 1981). Del Gaudio, Carpenter and Marrow (1978) attempted to replicate Stein, et al., with a broader range of clients. Their purpose was to determine if male and female treatment differences could be generalized across various diagnostic categories. The same instruments as used in Stein, et al., were given to 122 community mental health center clients. The clients comprising the sample had a broad range of diagnoses, including neurosis, transient situational dis- order, psychosis, personality disorder, and addictive states. Again, it was found that there were no differences between clients when they entered therapy on any of the variables. Unlike Stein, et al., these researchers failed to find treatment differences between male and female clients in number of sessions or medication prescribed. The researchers conclude that when a client sample is broadened beyond one diagnostic category, no treatment differences emerge and suggest "that limits be placed on the generalization that 63 differential psychiatric treatment is afforded to men and women" (p. 1578). This conclusion seems to be premature and. considering the evidence already presented, unwarranted. Unfortunately, the researchers did not explore their results for possible interaction effects between diagnosis, sex of the client, and sex of the therapist. This seems to be an area in need of investigation. A handful of studies have attempted to investigate sex bias in therapy through observation of actual therapy and data collected from clients in psychotherapy. These are the most naturalistic studies in the field and as such are assumed to capture the more meaningful dimensions of psycho- therapy. The results from these studies are more easily generalized to actual clinical practice. The first study presented is the one study which examines clients' perceptions of therapist sex-role attitudes. The remaining studies presented examine sex bias more indirectly, by examining client report related to different client-counselor gender pairings. The results of these studies are complex and have implications for many client-therapist dyads. One of the earlier naturalistic studies in this area was a study by Fabrikant (1974). He used as subjects both therapists and clients, his purpose being to investigate both therapist sex-role attitudes and clients' perceptions of the therapist's sex-role attitudes. Additionally, demographic variables and length of time spent in therapy were obtained from therapists and clients alike. Although the sample for 64 this study was relatively small, a fairly equal number of male and female subjects, varied across life status cate- gories,were represented. Some support for the differential treatment of men and ‘women in therapy was found. Female clients' report indi- cated that they had been in therapy over twice as long as the male clients. A similar finding was noted for therapists themselves as clients. Fabrikant states that these " . results most strongly support the feminist vieWpoint that females in therapy are victimized by a social structure and therapeutic philosophy which keeps them dependent for as long as possible” (p. 96). This finding is an important one and adds further support to the notion of reinforced dependency in the therapeutic relationship for women. The strength of Fabrikant's research, though, is in his findings regarding client and therapist sex-role attitudes and client perceptions of the therapist's sex-role attitudes. The overall results of this study suggest that as a group therapists have become more egalitarian in their attitudes towards roles for men and women. Nonetheless, some biases were reported. It was found that while both male and female therapists feel marriage should be a partnership, female clients felt that the attitude conveyed was that the male should dominate the marital relationship. Both male therapists and clients agreed that the majority of women could be satisfied and fulfilled in the role of wife/mother. Female therapists and clients strongly disagreed with this. 65 Disagreement was also observed by the sexes in the matter of family income. Males felt that the male should be the provider, while women conveyed a less traditional attitude. A difference was also noted in the area of sexual standards. While both male and female therapists expressed an attitude of an equal standard for sexual experiences for men and women, male clients in particular saw therapists maintaining a double standard, reflecting an attitude that a wife should be more dependent on her husband for sexual satisfaction than he on her. In the area of gender characteristics of men and women, the results indicate that everyone, therapist and client alike, still hold many stereotypes. Male character- istics continue to be seen as positive, female as negative. Fabrikant's study in presenting very complex and often confusing results finds some evidence of bias in therapy. Another naturalistic investigation is a study of the effects of the sex of the client and counselor on various counselor and client behaviors in actual therapy sessions (Hill, 1975). Male and female counselors, half at the practicum level and half with at least two years of therapy experience, audiotaped the second therapy session with a male and female client. Immediately following the session, both counselor and client completed a questionnaire assess- ing their satisfaction with the therapy hour. The therapy tapes were then rated for counselor verbal behaviors, per- ceived counselor empathy, client self-exploration and activity levels. 66 The results of this study demonstrate a strong inter- action between experience level of the counselor, counselor gender and client gender. In general, inexperienced counselors, both male and female, were most empathic, active, and elicited more feelings with the same-sex clients than in opposite pairs. Experienced counselors paired with their own sex were more focused on feelings and more empathic than with opposite-sex clients. Counselors at all levels of experience had difficulty being as empathic with opposite- sex clients and experienced counselors were often the most active, directive, and least focused on feelings with these clients. Client self-report indicated that clients were most satisfied with women counselors. For counselor satisfaction, though, an interaction between sex of the counselor and experience level emerged. The most satisfied counselors were inexperienced males and experienced females, supporting an overall finding that these therapists were the most empathic, active, and satisfied therapists. Hill suggests that this finding may have something to do with women needing to gain experience in order to feel competent and satisfied with their work, whereas men tend to lose interest in counseling skills once acquired. While this study does not provide conclusions regarding sex bias, it does point to some counselor sex differences in the treatment of male and female clients which vary with the experience level of the therapist. 67 A series of studies by Howard, Orlinsky and Hill (1969) and Orlinsky and Howard (1976) provide valuable information concerning women's experiences in therapy. The studies reviewed here are based on data gathered for a much larger study on the process of therapy which focused on the subjective experiences of both therapist and client. The format of the study was for clients and therapists to fill out a Therapy Session Report (Orlinsky and Howard, 1966) following each therapy hour. This is a detailed report, tapping many aspects of therapy, including: dialogue, feeling process, therapeutic relationship, experience dur- ing the hour, and session development. The clients studied were all women who had been seen in therapy at a community mental health center for anywhere from.eight to 26 sessions. Therapists were equally divided between male and female. In analyzing the results to the question, "How did you feel during this session?" from the Therapy Session Report, Howard, Orlinsky and Hill (1969) related the therapist's feelings in the therapy session to the client's experience in therapy. Overall, male therapists were found to experi- ence more negative affects with their female clients and female therapists reported the most positive and gratifying feelings. These findings, however, varied as a function of client experience. Women therapists were more responsive than male therapists to female clients' ”positive and negative transference." Negative transference elicited feelings of 68 preoccupation, intensity, and resignation for female therapists, whereas for these same therapists feeling good, and calm accompanied clients' positive transference. Male therapists had a more difficult time than female therapists with clients' ”painful self-exploration," and felt little nurturant warmth and uneasy intimacy with their female clients who were high in painful self-exploration. Both male and female therapists felt uneasy when the client was "courting the rejecting therapist." Probably the two most significant findings were the male therapist's feelings to the female client's experience of "erotic transference resistance” and the female therapistfis feelings to the client's "seeking therapist cues" and "intrusive dependence." Although both male and female therapists were highly effectively responsive to erotic transference resistance, they responded with different feelings. Male therapists felt disturbing sexual arousal and did not feel good. Female therapists felt uneasy intimacy, good, resigned, nurturant warmth, and suffering. Women therapists felt more comfortable with their clients' erotic transference, but had a difficult time with their female clients' dependency demands, whether these were expressed passively or intrusively. WOmen therapists with these clients experienced a sense of failure, disturb- ing sexual arousal and suffering as well as little nurturant warmth. Male therapists appeared to be more comfortable with their female clients' dependency needs and demands. The 69 researchers understand this difference as reflective of women therapists' feeling more threatened by their clients' dependency needs than male therapists. It seems likely that female therapists feel uncomfortable with female clients' dependency because of its Sex-role stereotypic nature, the exact reason men feel more comfortable with this type of client experience. Carrying their analyses further, Orlinsky and Howard (1976) reported results regarding the effects of the sex of the therapist on the therapeutic experiences of women. Women clients paired with male therapists as compared with those paired with female therapists were found to talk more about the opposite sex and their involvement with the therapist. In addition, they experienced a greater desire to gain insight, were more concerned with identity issues, and felt more eroticized affection, anger, inhibition, and depression. The therapists of these women were experienced as more demanding, detached, and less expansive than the female therapists. In therapy with male therapists, these women clients experienced themselves as being less self- possessed, less open, and more self-critical than those clients with female therapists. Women paired with women therapists felt more supported and experienced more encourage- ment from their therapists than those paired with male therapists. The therapists continued to analyze these results by breaking clients down into life status categories. The 70 categories included: young single women, single women, independent women, young married mothers, young divorced mothers, and family women. From this breakdown, it was observed that the above results were largely due to the effects of the sex of the therapist on young women who were neither wives nor mothers. Young single women, between the ages of 18 to 22, reported most of the feelings discussed above for pairing with male therapists. Single women, those 22 to 28 years of age, reported much satisfaction and expansiveness of feeling with female therapists. After breaking the data down by life status categories, diagnostic categories were considered. It was found that clients with depressive reactions were most reactive to the therapist's gender. Depressed clients seemed to have a more positive therapy experience with female therapists. Clients in other diagnostic categories were not reactive to the gender of the therapist. These results taken in considera- tion with those of Stein, et a1. (1976), and Del Gaudio, et a1. (1978), stress the importance of the interaction of not only diagnostic category, but of the client's life status category with therapist gender. Carmen, Russo and Miller (1981) also advocate for an understanding of women in therapy as a heterogenous group of women with differing needs and problems. While there are a number of limitations to Howard, at a1.'s (1969) and Orlinsky and Howard's (1976) research, including use of an all-female client population, client and 71 therapist self-report and lack of control for the experi- ence level of the therapist, the results warrant serious consideration about the experiences of women in therapy with male and female therapists. Jones and Zoppel (1982) provide further insight into the differences of experience in psychotherapy as a function of gender pairing. In an interview study with 99 former therapy clients, they found men and women in therapy with women therapists more often established a positive thera- peutic alliance than those who were treated by male therapists. Women clients in treatment with women therapists also found therapy to be more of an emotionally intense experience than those women clients in treatment with a male therapist. Additionally, clients in sameogender pairings were more likely to experience their therapist as neutral and nondirective than those clients in crossesex pairs.‘ Regardless of therapist gender, women were more likely than men to experience deprecation in the therapy experience. As part of this study correlations between outcome measures obtained from these clients and their therapists were observed. A more detailed description of that study will be presented in the next section. In general, though, the results of the outcome study suggest that women clients do particularly well with women therapists. This most probably has to do with the fact that women clients paired with women therapists viewed their therapists as more accepting, attentive, and comprehensible, facilitating the 72 establishment of a positive therapeutic alliance, which in turn resulted in a more positive outcome. The overall results of studies on sex bias and sex- role stereotyping in the therapy process are complex, and suggest that the term sex bias may not do justice to the complexity of the impact of gender in psychotherapy. Few consistent main effects are noted, but many interactions are observed. There is evidence that women therapists are more empathic, attentive and accepting of their clients, and that women clients may fare better in therapy with women cli- nicians. These findings, though, are complicated by experience level of the therapist, client life status cate- gory, and diagnosis. Findings in this area need to be interpreted cautiously with regard to actual therapy process, for only a handful came close to actual therapy process studies. No studies were found which explored manifest content in the therapy hour, for possible evidence of sex bias and sex-role stereotyping. Further research in the area of therapy process including studies of content analy- sis are needed. Therapy Outcome There are very few studies to be found in the litera- ture which address the effects of the therapist's gender on the outcome of therapy. Persons, Persons and Newmark (1974) looked at helpful therapist characteristics and client improvement as related to therapist and client gender. At 73 termination, clients of trained undergraduate therapists were asked to make a list of the characteristics of their therapists that were either helpful or nor helpful and to describe the changes they made as a function of therapy. Although the original intention of the study was not to study sex-related variables, the data indicated that sex was an important factor in therapy outcome for these clients. Both male and female clients wrote a similar number of responses for helpful therapist characteristics. How- ever, when women clients were paired with women counselors, they gave three times as many responses as male clients paired with female counselors. In addition to this, male clients paired with male therapists felt the therapist to be more interested and concerned, self-disclosing and help- ful with sexual identity concerns than those paired with female therapists. Female clients in same-sex dyads saw their therapists as more perceptive and insightful, encouraging of risk-taking, warm, friendly, helpful with sexual identity concerns, self-disclosing, supportive, and able to offer more honest feedback than those female clients paired with male therapists. All clients felt they received more assistance in same-sex dyads than in opposite-sex pairs. From these results, the researchers conclude that at least with a college population it may be preferable to pair women clients with women therapists, male clients with male therapists. Considering the developmental issues with 74 which college students are struggling, this conclusion seems to be a valid one. A study by Howard, Orlinsky and Hill (1970) with community mental health center female clients extends the generalizability of the above finding. Taking into consideration client and therapist life status categories, the researchers found that women were generally more satisfied with their therapy when in treatment with a female therapist. Kirshner, Genack and Hauser (1979) also studied the effects of gender on the outcome of short-term therapy with a university population. An analysis of responses to questionnaires sent to clients after termination revealed significant gender effects on clients' self-rated improvement and satisfaction with therapy. When compared with male clients, female clients reported significantly greater improvement in their attitudes towards their career. A trend was also noted for greater improvement for women in academic motivation, academic achievement, and family rela- tions. By and large, clients of both sexes reported greater satisfaction with female than male therapists, even when experience level of the therapist was statistically controlled. The results of this study, while finding some client sex differences, did not find reasons to support Specific gender pairings in brief therapy. In an outcome study, Orlinsky and Howard (1980) reanalyzed data frem client charts collected in the 1960's to assess the effect of therapist gender on treatment outcome 75 for women clients. No differences were found in outcome when the women clients were seen as a homogeneous group. When clients were divided into diagnostic groupings, this finding held for depressed clients and those with personaligr disorders. Female clients with anxiety reactions or schizophrenic clients did considerably better in therapy with female therapists. As mentioned previously, life status categories were also found to be an important intervening variable. The only women who did better with a male thera- pist were single mothers. Those women who were especially helped by a female therapist were single women, providing support for Orlinsky and Howard's earlier results (1976). Also found to be an important variable was experience level. Highly experienced male and female therapists as well as moderately experienced female therapists did equally well with their clients. Moderately experienced male therapists did not. Clearly, then, when considering gender and outcome, it is important to take into account numerous variables and interactions. A recent study on the impact of client and therapist gender on psychotherapy outcome was undertaken by Jones and Zoppel (1982). The researchers had a large group of clinicians complete an adjective checklist and two therapy outcome indexes for a group of randomly selected terminated clients. A number of controls were used, including screening out of clients with serious psychopathology and number of sessions to reduce variance in the results. 76 Some significant differences were observed as a function of therapist gender, client gender, and gender pairings on the outcome measures. Male and female therapists rated their male and female clients as having equally benefited from treatment. On rating scales concerned with specific areas of improvement, though, women therapists consistently rated clients as more improved than male therapists. Same-gender pairings were rated to have higher degrees of improvement than cross-sex pairs. These results correlated with client perceptions of outcome. Greater agreement was attained between client and therapist perceptions of the overall benefits of therapy than on the specific areas of change or improvement. Research in the area of therapy outcome suggests a relationship between gender pairings and therapeutic out- come. Gender by itself, either that of the client or therapist, does not appear to be an adequate prediction of outcome but suggests that for certain groups of clients specific gender-pairs may be appropriate and helpful. This is an area of research warranting further study. Summary Research in the area of sex bias and sex-role stereo- typing in psychotherapy is broad. Many aspects of the therapeutic process have been studied, including: therapist attitudes and sex-role stereotypes, judgments and 77 evaluations, therapy process and outcome. In recent years, this vast area of research has been reviewed by a number of authors. Stricker (1977), in one of the earlier reviews, critically examines the research on sex bias in psycho- therapy. His review fails to find evidence of sex bias or negative evaluations of women clients in therapy. He criticizes the research as primarily analogue, with other approaches producing "unsystematic, anecdotal and polemic results" and advocates for the use of descriptive and field approaches in researching the treatment of women in therapy. Stricker does not rule out the possibility that sex bias in therapy exists and suggests that the conclusions regarding sexist practices are premature in light of the existing data. While Stricker's criticism of the research is valid and his caution regarding the often-cited conclusions of less than valid findings is well taken, he neglects evidence from the more naturalistic research which demonstrates treatment differences between men and women clients as well as clini- cian sex differences (Hare-Mustin, 1977, and Gilbert, 1977). Other reviews of the literature draw similar con- clusions to Stricker (Whitely, 1979 and Smith, 1980). Yet, in reviewing these reviews, it seems clear that the subtleties involved in assessing sex bias in therapy are not recognized and that important trends towards documentation of sex bias in therapy are overlooked in these broad over- view conclusions. A more specific review of these studies 78 reveals almost as many positive as negative findings. Granted, within this complex area of study few main effects are found. Nevertheless, intricate interactions between clinician gender, client gender, and other relevant therapy variables are consistently noted (Sherman, 1980 and Davidson and Abramowitz, 1980). As stated earlier, one of the serious problems with research in this area is the lack of a sound theoretical framework from which to examine issues of sex bias in therapy. In addition, research in the field has abounded with methodological problems, including: weak designs, small and uncontrolled samples, social desirability, reactive measures, sensitization of clinicians to the issues being studied, lack of consistent and appropriate inde- pendent and dependent variables, and poor external validity due to the predominance of analogue studies. In spite of all these difficulties, results have emerged which, though not dramatic, indicate the need for further investigation of sex-bias and sex-role stereotyping in psychotherapy. The conclusions which can be tentatively drawn from a review of the literature are as follows: 1. In the area of therapists' sex-role stereotypes and attitudes towards women, it is apparent that a double standard of mental health for men and women still exists. a. Clinicians still hold sex-role stereotypic notions regarding characteristics of the healthy man and the healthy woman. 79 The healthy adult, sex unspecified, previously seen as similar to the healthy male, is now seen as more androgynous. The healthy adult is perceived as possessing both masculine and feminine characteristics. There are inconclusive results regarding clinicians' attitudes towards women. Some studies show that clinicians with more conserva- tive attitudes towards women are more biased, other studies fail to confirm this finding. In the area of therapists' judgments and evalua- tions, clinician gender was more often than not found to be an important variable. Clinical judg- ments appear to be influenced by the client's gender, specifically in the area of treatment planning. a. Female therapists, as compared to male thera- pists, appear to be more lenient in their diagnoses and assessments of both male and female clients. Minimal support is found for the notion that the sex-role-inappropriate client is judged more harshly than the sex-role-conforming client. In fact, some findings in this area demonstrated that the sex-role-inappropriate client was seen as healthier than her sex-role-conforming counterpart. There is evidence of sex-bias in treatment planning and goal setting, suggestive of an adaptation notion of mental health for women clients. In the area of therapy process few main effects for client gender or therapist gender are noted, but many complex interactions are apparent. a. WOmen therapists, as compared with male thera- pists, are consistently found to be more empathic and accepting of their clients. Women therapists are also found to be more attentive to their clients' concerns than male therapists. Evidence from naturalistic studies suggests that gender-pairing and experience level of the therapist may be important variables in the therapy process. Inexperienced therapists seem to engage in a more productive therapy process with their same-sex clients than in opposite-sex pairs. A similar trend is noted for ‘ ‘ 80 experienced therapists. This finding may be further influenced by client's life status, diagnosis and client experience during the therapy hour. Male and female therapists tend to experience different affects as a function of female clients' experience during the therapy session. Male therapists experience more negative affects with their female clients' erotic transferences than female therapists, but female therapists are more uncomfortable with female clients' experience of dependency. Conservative therapists are more biased than liberal therapists in their treatment of women with regard to issues they explore and areas of focus for the therapy. There is inconclusive evidence concerning whether or not women clients stay in therapy longer than male clients, and if the duration of treatment is a function of the therapist's gender. Studies which find evidence of women staying in therapy longer than men, particularly with male therapists, suggest that this is indicative of reinforcement of the notion of female dependency. In the area of therapy outcome, a relationship between therapeutic outcome and gender pairings is noted. a. For a college population, same-sex therapy dyads seem to be more helpful, with a similar trend noted for female same-sex dyads for a community mental health center population. Diagnostic grouping interacts with the sex of the therapist in regard to therapy outcome. Female clients with anxiety reactions or schizophrenia do better in therapy with female therapists. This finding does not apply to depressed clients. Experienced male and female therapists and moderately experienced female therapists do best with their clients. Clients are generally more satisfied with their therapy when paired with a female thera- pist than a male therapist. 81 In spite of these conclusions, controversy concerning the issues of sex bias in therapy continues to dominate the literature. For some the overall results are far reaching; for others they seem radical and are viewed with skepticism. Clinical analogues continue to produce unimpressive results, but naturalistic investigations have turned up findings which are more consistent with sex bias formulations. As Abramowitz and Abramowitz (1977) question: In which group of studies are we to put the most credence, the analogues which vindicate clinicians from.charges of sex bias or the archival- correlation investigations, which provide some, althou h not univocal,support for such claims? (9. 89) Apparently, although this area of research has received much attention during the last decade, there are many unanswered questions. Despite the inconsistent, inconclusive and sometimes negative findings, feminists continue to find their personal experience of sex bias and sex-role stereotyping in therapy more compelling. It is essential that this area continue to receive investigation and that methodologies be employed which can more adequately address the issues at hand. Qualitative approaches have been suggested. The complexity of the area as well as the subtleties involved warrant research which is sensitive enough to detect sex bias in psychotherapy. A summary of the studies reviewed in this section is found in Table 2.1. 82 .cmEmufimoom umoe mums mumwemuofi mono—Emmi. 3853 .835 ummcefim “moan Owauoououm dome/on 3 3on que mumwdmnmfi ogmom :95: moms mummdmnofi 3m: .8803 we m5? #30333”. ”.3563 manage $2532 Amouuocm? 1:0: 39338 30: use .332... no one .3383 £533.53 mam 8mm $8 3338. m 5 mm SEC: 338: 333.8. m 5 35:3 mo :33? couuwnzv 83333320 98m 93 uuodxo Eco no: 83358 cam 01...: p832 mega—2 333 53.x .modzuoouonm Salaam eggs—ow :33 mode» 93 3 omecamon museum of mo mcoz 5253 com .359: mummdmnmfi Ext 953 omogglofime once 5 comm mos. out: new ”.3553 Snow 1333; ofi Swanson—u 3m: pom .mfisom com @5835 $9.39: .333 5853 saw 52mm; douse... uo pudendum .mumamnofi moo poemmucwm: mumwamfimfi D3583 mam 390m wowowuomum 095 3333332 :33 :33 .mumwemnmfi m coco mumwimuofi oniootonm oHoulxom 3 box: one. mum 339985 < m e < 8% 3339532 383 comm pom ion Tum um .mzogwouocm once we 8383a cflfimsoum mo mw 25mm 133mm: 05. .983 veguoououm mfioutxmm 33353 SUmoZemmv E 8388a :E... as SEE E... was. 3:3: messages 8% osmzauamz 38: Emma: .332... .8 $0de now omen 836mm gimme new mcoEmoooexo 0338.83 once noggm 3...; EB .28... Rs 32 2:8; 2... to est 3.. d 1&8qu .593 E muodmmcnoo 39:3 3 $3.5m oumz 5.5905 mo 3039.58 mam: £16m pom 3.95“ .39: find; maggot 838:3”: o5 mo mcoEeoouoe .835? Eon mnogomcsoo 395m 513032300 Ommm 33333.2 353 323 new Edam: SS: 83: Ewe, .cmEoz ecu lawsuzme pom Aommmv mam Nucmnxcmmom 509130 Limo: #355 we c030: ..u:m_5m3.€m.. cm 3883 mums—no.5 H308 ouwgoflmooo 59590.5 .cggohm €353 ouoocmum 32:8 3E. .5603 new cm: new 5de £33305me oguoououm 30m omega. 32533 new $9“. Room: «0 moumocmum ucmuowwwo SEES. 9.3355 933095 txom Saigon oflumfimusumz logoum odomlxom ummmmnofi. mwfipcwm muuofimzm pom: Amvmusmmoz Npsum Lo omflw ~53meng uo no.3 xdmuocuocoxmi 5 $5 xmm co Dumommm 05 no ENE-3m 7N canoe 83 30.0 0000: [00300.6 E02002 0000000000 00000: 000.: 000 000 .0000 000 >0 00030 000.: 0000300000. .000: 953 000.00 000 0000 300000 000.: 00w00n 003 00030 00603 0000030 00.0.00 0000000 000 000000.. £005.00? I030 05. .00>00000 003 00030 000 00 0.00 000 0000:0000 0000000000000 000000 0.00030 9000000 0000 £03000... .3 9.0 0o 33...... 0800...... $0.03 80883:. 5 5.8.. 035.. 0o 00500. 53.50 0202 .05 020.000.... 03.088... .00000 00 >0030> 0 00 203 000.0000 00 .3330 0.00020 0000.00 000 000.: 000 00000 000200 0030030000 000.: 0000 00000300 00000: 0003 0000000000 .0000 000000 00000000 000 000000 00 :03 0.000000 00 3003 0000 000 00000 300000.000 032 0.0030000: 3000003 0003 000 00300000 00 0000000 00 00030 000.000 0000 ”0.2000500 000 000000 .5000 0000 000033 0000 .300: 000.: 00 00030 000.: 000 00303 0000000000 00>0~ 00000000 00300096 00030 00005. 3003 000030 003000000 0000.00 0000 00000300 000003 0000 0000 0000000000 0002 003000000 05¢ 000000w 000 0000 00:53 0000300 00000300 00000: 000.: 0000000000 0.0000050: 0000 00 00000 0003 00003 0 00 0000000000 000050000... 008500000 0 000000 0303000002 333 00.300: .033 0.80 853 0000300. 00030030 0000.00 0000 00003 0000030.... 0000300 00000300 000030000 000.: 0000 00030030 0.0000030: 03: 003000030 0003000000 00 003003 0 00 0000000000 000 03000 00 00.53 0000300 00000300 00 00000 003 000000.003 02 00030030 00300> 0003000 0303000002 $33 0500 000 000000>0o .000030 00003 3000003000 .000030 9330000 00 00300030 030300 0000 0000000 0000000000 0030000000 0000000000 000.00 00 00000000 00 0000000000 0002 00330000 0>300n0< 0303000002 @005 000000 000 00000 3000003 000300.50 00000000000 000010.00 0000000000 0000 0000330 w0000 00300000 0003 000030 000.03 00 9.000000 05. 000030000 Qmmm 000000. 353 00300 30300000 .00000>00 300000.000 003000000 0003 000000000000 .00030 000.: 000 0000 0.0000 003000000 00000350 0030000.: 000.: 00 00030000 30000000000 003 00030 00 00000000 00030 000 000 00 0000 000000 000000 0E. .0008 0003 00000000000 000 0030030 02 00000000 00:3 0000 00005. 355 000:0: 000 0000: 00003000 a .N 0300. 84 £8305 33:03 :0803 :05 00003030 0:9: 00 0033> 0:03 330:: 0:900 0:3 :02 0:00.: 303 00333030 :3333 oz 0:033:20 30.: :0:0 0000cw30 335: 0:9: 0:9: 0: 0:00: 0:03 0:33:20 :0603 0:333:33 0033:0963 :0:: 00:33 :05? 00 3033000 0>300w0c 0:9: 03w 0: 000:3 33:03:30: 03:03:0me .3063 :0 :00?— 00 00:30:03 3:020 00:30w0 030300 003 03: oz $330500 :0::::m 000: 0: ~00m :0300 :33>00: 0:: :33 0:330 0:0 300 3300500 3063 0033:0950 3:0 030:0 :33>00: 0 :0:: 33:00:30 0:9: 00 030:0 :0300 33:58:00 0 0030300 0:300:80 3053 0:0 30.: 0033:0965 J00w :0300 :33>00: 0 :33 0:020 30:0: 0 00 0:300:80 30.: 0033:0983 :0:: 33000000 39: 0:03 0:300:05 30.: 000:3:0axm .0500 300:0 0:330:00 02 0:03:30 330:: :0:: #333003 50:: 00:050m00n :0:0:0: 003000: :85; 933000 38:00 30300: .0:0000000 30.: 50:: 03:00: 30:0 :: ::3:3 0:9: 0:03 0:0000000 :0503 33030:: 10:30: #330030 0:33:30 @303 000:: A305 3 30380033 0:03:003:0>0m 0:0: 00:80 w33 000:0 000300:0:: w:303:00:m 330:0.»00 3:00:80 :3 00:00:00 303030 0:03 000:9: 1:003 0:0: 300:9 . :3 :0: 1:020 0:35 00005030 30:03 5:00 0:030: :33: 300:: .0308 .008: 06:58 $330500 :0:0:3 :0: 000: 000000 .0305 :0300 :0 0:30: £00286 03:833. :080 300m 030m x00 #333095. 00:038. 0009330 3:3 83> 2000:3380 0000 00:03 0500 039:. OH umwdmuhflumz 3:003: 0000 503:5 w0~0:< A000: 3030 .33333 0332500 020:: :50... :300: 33:33: 0000 1:02.": 0:38 503:3 w0~0:< g .0538 0:050:30 33:33: 0000 #0033230: 503:3 modg 80:: 3:5: 0:0 :30: .0503m 80:: 50:38 0:0 33:80:80 .3800. 5:02: Am:0:v 3:550:90. 0:0 3:03:00 333 30303 0:0 00:55. 333 33390::4‘ 0:0 00:50 . 333 u:§::< 0:0 09:00 .Nu:0z:0m .0303 £03534 00:50:00 H.~ 3:09 85 .8880: 888:8 8m 8:838 oz 5:838 85 808 3898 8888.88 .585 5 3838.85 888 :85 8:3: 388888.88 8.83 3838.85 c8603 .88 c3358 8m 8880: 8.83 8888.833 meom 88.58886 8898888 85 so 950w 888.8833 38m 683830 883 8888833 x88 8:838 02 .mdmow 85553 8.8:. 88058 88838858 83: $2828 :8 8m £80m 853889. MO .845: 888% 8 88058 3838828 5:53 .8828 88838888 8 8m :85 8838 838398 :8 .88 380w 85553 8.8:. 8.8058 8:888:30 683830 883 8.838 8983 8:8 5 :33388: umfimwm. 83m Jam—588888 8838.85 :0 8m 8.838 88383 950m 8.83 8888833 .953 8898.838 88 538338 oz 1.5838 3868a 85 80c 83 88.: 85 8m 8398 wagefimfiofigo 85 883 338888 8:8 mo 88.5885. Sufiémma mm :88 883 E8308 8mm... 8.8828 85. 8:838 818.: 85 :85 82388: 83 3 :888 883 88 385833. 838 I853 88.: 8838888 38888.58 8.838 395m 85. 3838.85 139$“. 95.6 8883305838 88838885 we“ 3 888.5 8888888858 locu>ma mcwowuompm mumwuumzozma . mumfiwodozozma . 88283 880m 888 8838.3: #9538889 68895385 .3385. 9.6.5 5:838 8:8 .88 8:3: we 888mg .95888805 #8358 . 8:3 88.8 $85888» 38,5888 . 83 888.55 #835 5 38% >88; 5:88. “:8: I888 .mdmow 85.5888 68.8128 89 3 88:83 .53 A88 988:8 no 5883508 wag—Ema 8.88.5888 . 858 $885 88858 $08883 8888 c8333 moan—2 $88383 8888 c8323 moire $88382 8888 c8883 modg TCQEBm 8885 c8385 wofig 3.8883 8888 £883.83 wo~8c< 33: 8.28 8:8 330:5:2 .8588» .3838 £85 £83589. £88895 .3885. 333 8:00 8:8 8382.888me 33: 3:953 383 CE 35: 83:: 88:5 ”:50 H.~ 838.“. 865 .030: 083 3038 8m 588:0 38m .8308 9.8888 08 0:38 85 mo mwm :33 808805 858m... 8808080 85 :0 mwcfimm .8308 8008009 :mfi 538m 886 mm 803 8.83 95308 808.80 .880 £80 5 38:0 8 3830800 8.8.: 880 8.83 0:0 P889500 Em: :20 00030 8008830100: «:0 mo 9808000 88:. on 00 83830 883 9889500 395m .0830 083088880: mdoulxmm 8:0 00:”. 08:0 08.30888 mdoulxmm 8:0 :33 883 8.8:. 883 9889500 .mmB 8833.3 5 m88m500 8853 00m 886 80303 050m 8.83 8008838 oz .838 395m :05 08:. 85 5000 303 8.86 8.84.8.5» 88m :82 «0 9880500 48.85 8882500 8853 8a 0882.00 83 @5ch 35. 8:80:00 880m 98830 3m: 85 5 :93 8.80000 880m 98030 8853 85 5 08.805 8.85 830% 9883500 082 8:80:00 800839 8m 303 m0 .845: 8:08 8 8.89550 mmxmm 500 m0 3080.500 80080835 8 3038 .838 0830 8 800% 0:030 080050 $5ch 02 6889500 885 :20 .chm 8.8.: mm mmmvam 3 88» 8.83 03 .m88m500 88.: :05 03qu5 08.: mm 0.98920 88» P889500 39.3 £88888 :0de 800m: 8>3 8.8800080 cam 8.80000 . 8%: m .882 88:88 3085080 8>£ m . 80mm: 88:88 30850me 8%: 038.805 mumwwodonuxmm 0080 .5858 8030 $58.“ 3009.0 .0830 00 0000080 808> 800.82; 3988a 8m £080.58 0:0 00830 .8 208 $0095 8.0880 £08058 0:0 008008 :53 H3095 000 8009/ 8w 388.805 .803 83> mo $55 .zwdnoua m.u:mH~0 H0 zmdgoflum m0 acmum Immmmm .muumu mo cowuwcmoomu can ddmumm 0820 0.8300 $583 .0830 00 89808.“ 808> 8303M? 0:030 893 83: mafia A 003 d 0 £088 wo8c< 803 0:50 .38? I805 0830 803253 wofimc< Amamu omcfi> 583805 ”589.500 883:8 wmfiunv modmc< 8:5 0:8 Auhofiv magma Em chaos .zmccme .500: Ashofiv oufiamzm Geo: x305 83$ 8930—; dd 3an 87 00000000000 0300 1x00 0:05 030: 00 0:000 0003 0x30 5503 0:0 :02 .5503 0:0 :05 000 m0000:00m 300x00 0:0 :0503 000 00300 :0 003030 0.03 003 0:030 0:0 0300003 00 00.00030 .3:030 030.: 00 0:03 00 00030 00>0 0000000 :0 :000 0>0s 00 0:000 0003 3:030 030600 .3:030 030500 0:0 030E :003000 0:000 003000030 0:95.003 02 0000000008 0:0mm00000l00c0 0030000 3:030 030.: :03 0300.00 00 0w00:00000 005w} < .3:030 030500 :050 300300 00000030 00300 00: 3:030 0302 $000000 w:0000:0 :05: 03000 3083000005 00 00005 .3350 :0 0:000 0003 00000000030 02 .33000040 03053 000 0:000 003 00:000030 oz .3:030 030:. :050 0000000 00w:03 000 3:030 030500 300 3300005 0302 000303000 0308 :050 0:00.300 000:. 000 E05 300 0:0 0503000 009: 0>0£ 00 3:030 :0053 00300000 000300.500 030500 00:0 30:0300 :05 0:300:05 009: m0 :000 0803000 30500009 3:030 000.500 000 .303000 30:00000000 00 30:000009 c000 0:000095 000:. 00 00300000 0603000 30:0300 3:030 0:0 3300005. 00003000 00 0w:00 00000 0 £003 3:030 000:00 £0300: 300:0: 03:30:00 030500 0:0 0302 3:030 00300000 03:00 £0300: 300:0: 0005550 030500 0:0 030: 3300005 300000000 3300005 03:00 33000500 000000000 0300 .0000 .00 0000005 $000050 00 0.00:3 0>0£0 m0 080m 0003000 05.50000 .3333 0000030 00 0045: .0100 30.000.000.035 0:0 009: .30950 00 mwcfimm 0:00.300 0:303:00 0o b000:2 0:03.000 0:303:00 00 0095: 6000000095 .5380 00 mats. 00 003000002 000003000002 000003000002 70000000 0:033 00033000002 Am:0000m 300000: 0 :0 .30 00 330: 00 :0000003 I000 30000000 000003000002 33: 05300000 88: 800m: 05 0853.0 .3080 08 33: .0380 90 000000 300 .500m 38: 8002 0:0 5:000 £00000 30.50.5000 Jug—0050 83: 2.00: 823:8 0 .~ 03.0. 88 .mpwma xmm mufimoaao cw cmsu mvmzv xmm mumm aw mucmumwmmm whoa cw>wwumu mmzu udmm mucmwdu -< .mumwamumsu amen: nuts wouumn mum“ mucmfldo autos umnu wcficCgu “mumcmo .mumwamumfi 3m: :33 83mm 35:0 985 :93 mocmfiafim ofiusmamuwsu m>fiuwmoq m :mfidnmumw :mumo 98:. mumamumfi 32.3 53 .3305 5 35:0 .mocmwumaxw xamumnu mo uuoaou ucmfifiu cam umflamumsu mo xmm :uw3.uomumuc« cu v:30u mmmocwmwc new mmwuowmumu mzumum mafia “cmwfio .umfiamumsu man mo xmm cu vmumdmu mcofimmwm mamumzu mo ucmucoo cg fiance mmucmummmfic xwflgcpu $0622“? 925:0 53 ~38: m>ms cu n:50m mums mumflamuwsu wdmumw mzmucmumwmmu mocmummmcmuu owuouw: Mucmwfio mfimumm Luvs >u~50wmmwv m>ms cu case“ mme mumfimmumzu mam: .uso: ammumsu mag wcfiusn mucowuwaxo ucwfido cam umvcmm umwampmzu cmmzumn vm>ummno mCOMuumumucfi xmdaeoo .muodmmcsou :85: £33 umflmfimw umoE 9.83 3530 .3530 xwmlmuwmoaao :33 352—5 mm mewwn zudnowMWMn cm; mucmfiuwaxm we m~m>m~ -m um muodmmczoo .muwma xmm muwmoaao aw cmcu mucwwdu xom memm zufi3 mwcfldmmm whoa uwow~m cu manm can «>38 62395 0.85 0.33 3309500 8833935 mucmfifio umucmo @3358 zuwmum>wca mucww~o 59:2; Emcfi mumwamumzu .mucmwfio swag; muqumumsu .m “:33 :95: muodmmcsoo snowuumuq .muofimmcsoo mauofiuomum 38.989: 2.830 .moflumwuwuomumsu $389: 3&2 mo mmcfimm 30w>h¢u¢w “=38 uuoamm cofimmmm xmmposw uuoamm c3 mmmm AQMHQP :OMmmwm mo unmEmmmmmm .uo~mm::oo can ucmfldo we mafiumu .mamu Cavzm xamumsu amnuum .uouu muo«>mzmn nogmmczou mawumm owumwdmusumz Qwumwflmpsumz ofiumfifimpaumz owumfldmusumz Uwumwdmuzumz Aquoav thezmz can mcomumm .mcomumm 353:8 .meouusokwmwumpw Awwofiv Hoaaom cam mason $8: 886: Em 38:5 38: 233 can 6.825 Anwoav dd“: ~.~ odnma 89 .0:o00:o mo 00000000000 000030 5003,0000300000 0033000 0m0£H .0000000050 0300 00000 008095 009: 00 0000030 @0000 000000005 0053 .5920 00006020095 .00 m0000 0300000 :0 .ma000;0_:o0m 0000000000 x330300 w00>0n 00 0000030 030500 0:0 0308 00000 0000000030 030600 0:0 030: $00509: 0350 0003 000000050 010: 0 :03 000000 30 023 00:63 300 05.0. 600030 0305 c050 00080 £03 0033 30 3:300 0080008008. 0:0 0000000000 500080 5003 000030 .0953 .000000300 >30=0u 0:0 0208050500 50000308. 00600000 :0 0:0—0.90095 000000w 000 0000: 003 00000 0030000 < .000000 0000300 000300000 00030 :0 00005006095 000000w 00800 0000030 0300000 .0000030 0305 00 00009.00 m< 00300050 030600 0 £003 0000800000 5 .0053 E00000 000:0 0003 00300000 009: 33000.00w 0003 00603 0000030 0:0 0000000053. mucmw A U HQHCQO £08: 038: CE 000030 000000 0.0000958 3000035 000050 0:0 030: 0000030 000000 20300: 300:8. 00:53 3030005 0:030 .0050030 30005 000 003000 m5 000 .8028“. 03000.93. 0:0—0.05095 0:030 00 09300 00000005. 00005000 0003 8000000000m 0.0030 00005800003 00000000000w 0:030 00 003002002 000003003002 000003003002 00 003003002 383 30003 0:0 00:00 88: 205: Em 32:5 83: 0080: 0:0 3000000 3000:0030 88: :0: 20 008:5 .206: 0085080 00 ~33. CHAPTER III DESIGN OF THE STUDY The purpose of this chapter is to present the plan of operation for the study. The following sections are included: selection and description of the sample, measures used in the study, procedures for data collection, operational definitions for therapist variables, design, research hypotheses, and procedures for quantitative and supplementary descriptive data analysis. Selection and Description of the Sample The sample for this study consisted of both client- subjects and therapist-subjects. Information regarding sex bias in psychotherapy was obtained from client-subjects through a questionnaire survey. Therapist-subjects were used to aid in construct validity testing of the instru- ment developed. 7 Client subjects were recruited from a large University Counseling Center (CC) and from a local Community Mental Health Center 0“flKD. .All client-subjects were female, at least 18 years of age, had cases open at either agency for a minimwm of one month, had an identified therapist, and a current local address. Due to the differing number of 90 91 clients served at the respective agencies, different cut- off dates from termination were used. All Counseling Center clients who met the above criteria and had terminated therapy between December 1981 and December 1982 were included in the initial Counseling Center client—subject pool. All Community Mental Health Center clients who met the above criteria and had terminated therapy between July 1982 and December 1982 were included in the initial Community Mental Health Center client-subject pool. The initial client-subject pool consisted of 210 Counseling Center clients and 275 Community Mental Health Center clients, a total of 485 client-subjects. The final client-subject sample consisted of 79 Counseling Center clients and 113 Community Mental Health Center clients, a total of 192 client-subjects. The client-subject sample was relatively young, with more than half the sample falling between the ages of 18 and 30 years. The sample was also well educated, with 80% having at least some college education, and close to 20% having some graduate education or a graduate degree. The Counseling Center client sample was predominantly single and childless, while the Community Mental Health Center client sample was more evenly distributed across relation- ship status categories,with the majority of these clients having children. 92 Community Mental Health Center client-subjects were more often seen in therapy by female than male therapists. This was opposite for Counseling Center client-subjects, who were more often paired with male than female therapists. The client sample was evenly divided between male and female therapists. Client-subjects were relatively evenly distributed across the number of counseling sessions, ranging between 1 and 20, with half the sample having l-b sessions, 35% having 7-20 sessions, and the remaining 15% having anywhere from 21 to 61+ sessions. The majority of client-subjects were seen in what would be considered brief psychotherapy. The bulk of client-subjects were termi- nated from.therapy between May 1982 and December 1982. Specific percentages for client Life Status Categories and Therapy Experience Variables are found in Table 3.1. Therapist-subjects were also recruited from a large University Counseling Center and local Community Mental Health Center. All therapist-subjects worked in a clinical capacity at one of the above agencies, had a minimum of a master's degree in psychology or social work, or were currently enrolled in a doctoral program in psychology, and saw at least three clients a week in psychotherapy. The therapist-subject sample consisted of 49 therapists representing the disciplines of counseling psychology, clinical psychology, and social work. Senior staff, interns, and practicum students at the two agencies comprised the sample. Close to 70% of these therapist-subjects were male, Table 3.1 Description of Client-Subject Sample by Life Status Categories and Therapy Experience Variables for the Community Mental Health Center. Counseling Center, and Combined Populations CMHC CC Total Life Status Category (N=113) (N=79) (N=l92) Age 1) 18 - 22 years 16.8(%) 46.8(%) 31.8(%) 2) 23 - 30 years 37.5 35.5 36.5 3) 31 - 40 years 27.7 10.9 19.3 4) 41 - 50 years 9.8 5.8 7.8 5) 51 - 60 years 6.3 .9 3.6 6) 61+ years 1.9 .1 1.0 Education 1) Did not finish high school 9.4 3.2 6.3 2) Completed high school/GED 22.2 4.8 13.5 3) Some college 5.0 57.2 53.6 4) Completed college 4.3 10.3 7.3 5) Some graduate education/ graduate degree 14.1 24.5 19.3 Relationship Status 1) Single 34.4 69.7 52.1 2) Married/living with someone 35.3 18.9 27.1 3) Divorced/separated 24.8 10.7 17.7 4) Other 5.5 .7 3.1 Child Status 1) Children 61.0 15.0 38.0 2) No children 39.0 85.0 62.0 Therapy_Experience Variables Therapist Sex Male 34.0 61.8 47.9 Female 66.0 38.2 52.1 94 Table 3.1--Description of Client-Subject Sample by Life Status Categories and Therapy Experience Variables for the Community Mental Health Center, Counseling Center, and Combined Populations-(continued) CMHC CC Total (N=113) (N=79) (N=192) Number of Counseling Sessions 1) 1 - 3 sessions 19.6 30.4 25.0 2) 4 - 6 sessions 27.7 20.5 24.0 3) 7 - 12 sessions 21.4 19.6 20.0 4) 13 - 20 sessions 16.8 13.6 15.1 5) 21 - 30 sessions 7.1 4.5 5.7 6) 31 - 45 sessions 5.2 7.4 6.3 7) 46 - 60 sessions .7 .3 .5 8) 61+ sessions 1.5 3.7 2.6 Date of Last Counseling Session 1) Before December 1981 4.4 14.4 9.4 2) January - February 1982 6.2 10.4 8.3 3) March - April 1982 9.6 16.4 13.0 4) May - June 1982 10.6 19.6 15.1 5) July - August 1982 25.8 11.8 18.8 6) September - October 1982 30.3 7.3 18.8 7) November - December 1982 13.1 20.1 16.6 with a similar percentage being affiliated with the Counsel- ing Center. Therapist-subjects were evenly divided between experienced and inexperienced levels. The experience level of these subjects ranged from one to 20 years, with a mean experience level of nine years practicing psychotherapy. The mean experience level for inexperienced therapists was three years, and for experienced-level therapists, 14 years. On the average, then, these two subject groups were separated by 10 years difference in years practicing psychotherapy. 95 Specific percentages for therapist-subject characteristics are presented in Table 3.2. Measures Used in the Study Two measures were used in the present study. One measure, The Sex Bias in Psychotherapy Questionnaire, was developed by the researcher to directly assess women clients' experience of sex bias in therapy. Another measure, the Stereotype Scale of the Personal Attributes Questionnaire (Spence, Helmreich & Stapp, 1974), was administered to therapist-subjects as a means of determining the therapist's tendency to stereotype others. The results of the Stereo- type Scale were used in construct validation of the instrument developed. Sex Bias in Psychotherapy Questionnaire gSBPQ) For the purpose of this study the Sex Bias in Psycho- therapy Questionnaire (SBPQ) was developed and administered to client-subjects. The SBPQ is a lOO-item self-report instrument that measures the experience of sex bias in psychotherapy. Items for the SBPQ are responded to on a four-point continuum from "definitely false" of my counsel- ing experience to ”definitely true” of my counseling experience, and items are scored on a 0-3-point scale. The lower the score on the item and total scale, the more sex- fair counseling experience reported. 96 Table 3.2 Description of Therapist-Subject Sample (§=49) Therapist Variable Percentage Sex 1) Male 69.4% 2) Female 30.6% Experience Level l) Experienced 53.0% 2) Inexperienced 47.0% Job Role Expectation 1) Agency staff member 55.0% 2) Intern 14.3% 3) Practicum student 30.7% Agency Affiliation 1) University Counseling Center 69.4% 2) Community Mental Health Center 30.6% Professional Discipline 1) Counseling or Clinical Psychology 86. % 2) Social Work 14.0%_ 97 The SBPQ is comprised of four sub-scales developed around the themes of sex bias in psychotherapy outlined by the APA Task Force (1975). The four sub-scales are the Traditional Roles Scale, Bias and Devaluation Scale, Psycho- analytic Scale, and Sex Object Scale. The procedures used for instrument development are presented in Chapter IV. Empirical support for the SBPQ is found in tests of reliability and validity also presented in Chapter IV. The design used to test questions of reliability and validity is presented later in this chapter. The Stereotype Scale The Stereotype Scale of Spence, Helmreich & Stapp's (1974) Personal Attributes Questionnaire (PAQ) was administered to therapist-subjects. The Personal Attributes Questionnaire is an instrument developed from Broverman's, et al. (1970), Sex Role Stereotype Questionnaire (SRSQ), and is composed of two scales: the Stereotype Scale and the Self—Rating Scale. The Stereotype Scale provides a measure of sex-role stereotyping of others and the Self-Rating Scale provides a measure of the respondent's psychological masculinity-femininity. The relationship between the two scales of the FAQ, the Stereotype Scale and the Self-Rating Scale, tends to be low in magnitude and not consistently significant. This suggesus that an individual's sex-role stereotypes have little direct relationship to their perceptions of themselves on the same 98 attributes. The low correlations between scales also suggests that the FAQ is measuring two separate and distinct constructs which supports the use of either scale alone (Spence, 1982). For the purposes of this study only the Stereotype Scale of the PAQ was used. The Stereotype Scale is composed of 55 items adapted from Broverman, et a1.'s (1970) SRSQ. A number of weak- nesses in the SRSQ were identified by Spence, et al. (1974, 1975), including: complicated and cumbersome scoring, order effects due to method of administration,and amount of time required for administration. The Stereotype Scale was designed to: (1) require less time for administration by reducing the number of items; (2) do away with order effects by asking for comparative judgments rather than separate judgments for male and female; and (3) be machine scorable. Even with these above changes, the Stereotype Scale is very similar to the original SRSQ in content, but with fewer items. 0n the Stereotype Scale, one pole of a bipolar stereo- typic adjective is presented for each of the 55 items. The respondent is asked to compare directly the typical male and female on a 5-point scale, running from."much more characteristic of the male" to "much more characteristic of the female.“ For the stereotype ratings, scoring is such that a high score is related to agreement with the stereo- type, indicating the perception of greater differences between the sexes. 99 Research by Spence, Helmreich, and Stapp (1975) pro- vides support for content and construct validity of the Stereotype Scale. In administration of the Stereotype Scale to 550 college students, statistically significant sex-role stereotypes for all 55 items were found for both sexes. This finding is similar to the finding of Broverman, et al. (1970, 1972), when testing the SRSQ for content and construct validity. A cross validation study by Spence, et al. (1975), also found significant stereotypes for all 55 items. Furthermore, Spence, et al. (1974), provide evidence of the Stereotype Scale's validity by demonstrating low corre- lations between scores on the Stereotype Scale and a measure of social desirability. This is an especially important characteristic for an instrument measuring an area as reactive to social desirability as sex-role stereotyping. Initial reliability data for the Stereotype Scale obtained from two groups of college students demonstrated that the Stereotype instrument is internally consistent. Alpha coefficients for the Stereotype Scale are .91 and .90 for men and women, respectively. Additionally, test re-test data on a small sample yielded reliability coefficients of .92 and .98 for men and women, respectively (Spence, Helmreich & Stapp, 1974). Similar reliability estimates were obtained for the Stereotype Scale in this study. The alpha coefficient computed from the 49 completed Stereotype Scales was .92. 100 The Stereotype Scale was chosen for this study as a measure of therapist-subject's tendency to sex-role stereotype men and women. The Stereotype Scale is demonstrated to be a reliable instrument with evidence of content and construct validity. In addition to the Stereo- type Scale measuring the construct needed for the purposes of the research, that of sex-role stereotyping, its demonstrated low sensitivity to social desirability and the fact that it is not a widely used or well known instrument was considered in selection of the instrument. Therapist- subjects are an educated subject group, sensitized to many of the issues related to sex-role stereotyping and familiar with measures used to study the phenomena. Any instrument used with this group needs to be one with minimal social desirability and popularity. Procedures for Data Collection Client-Subjects A survey approach to data collection was used with client-subjects. Following compilation of a client- subject pool, 485 women clients were mailed a Questionnaire Packet. This packet contained a letter of transmittal, with a client-subject consent form, a background informa- tion sheet, a copy of the Sex Bias in Psychotherapy Questionnaire, and a self-addressed, stamped, return envelope. A sample of the Client-Subject Questionnaire Packet is found in Appendix A. lOl Procedures were used to maximize response rate. The letter of transmittal, while providing a brief explanation of the study and assuring the participant's confidentiality, was easy to read and brief enough to maintain the reader's attention. Additionally, an incentive of the chance to win $50 by returning the completed questionnaire by a specified date was used to increase response rate. Use of.this incentive was approved by the University Committee for Research with Human Subjects at Michigan State University, and verbal approval to use this incentive was obtained from the State of Michigan Lottery Board. A few days after the date specified for return of the Questionnaire Packet, a follow-up letter with a new questionnaire packet was mailed to Community Mental Health Center clients1 who had not yet responded. The follow-up letter was designed to catch the client-subjects' attention and capitalized on the cash incentive. A copy of this letter is found in Appendix B. 0f the 485 women mailed at least one c0py of the Questionnaire Packet, 192 responded. The response rate was 40%, slightly higher than the 25% response rate most often observed in social science survey research (Sellitz, wrightsman, & Cook, 1976). Issues of confidentiality and 1A follow-up letter was not sent to non-respondent Counseling Center clients, because the research committee of the Counseling Center did not grant permission to do so. The research committee cited issues of confidentiality and intrusiveness as the reasons for the denial of permission to send a follow-up letter. 102 the client's right to privacy precluded the use of more active methods to increase response rate, such as phone calls and further follow-up letters. All client-subject respondents were offered a copy of summary results, if so desired. Therapist-Subjects A survey approach was also used for data collection with therapist-subjects. Prior to client-subject data collection, 65 therapist-subjects were mailed a Research Packet. This packet included a letter of transmittal, a therapist-subject consent form, a demographic information sheet, a copy of the Personal Attributes Questionnaire and scoring sheet, along with two return envelopes. A sample of the Therapist-Subject Research Packet is found in Appendix C. The full nature of the research was disclosed to therapist-subjects in the letter of transmittal. Addi- tionally, this letter provided therapist-subjects with ample reasons to participate in the research. Anonymity was guaranteed to therapist-subjects through the use of a coding system and a two-envelope return procedure. Therapist- subjects returned their anonymous, completed questionnaire in a separate envelope from their signed consent form, thus assuring that no names would be attached to therapist—subject responses. 103 A follow-up letter was sent to therapist-subjects who did not respond by the specified date. This letter was a brief reminder for therapist-subjects. A copy of this letter is found in Appendix D. 0f the 65 therapist-subjects who received a copy of the Research Packet, 49 responded. The response rate was 75%, slightly higher than that expected for a population of clinicians. A11 therapist- subject respondents were offered a copy of summary results of the research, if so desired. Qperational Definitions for Therapist Variables Therapist Experience Level For the purposes of the present study, therapist- subjects were divided into two levels of experience. Those therapists practicing psychotherapy for five years or less were defined as inexperienced-level therapists. Therapists with more than five years of practice in psychotherapy were defined as experienced-level therapists. A natural split between experienced and inexperienced therapists according to the definition above was observed in the therapist- subject sample. Therapist Agency Affiliation Therapists' agency affiliation is defined by the agency for which they work. Therapists were either affiliated with a university counseling center or a local community ‘mental health center. 104 Therapist Tendency to Stereotype Therapist's tendency to stereotype is defined by his/her Stereotype Scale Score. Design The overall design of this study is descriptive, the purpose of the study being to explore whether or not women clients experience sex bias in psychotherapy and to describe in detail what therapist characteristics, client life-status variables, and therapy variables are associated with the experience of sex bias in psychotherapy. The purposes of descriptive research as outlined by Sellitz, wrightsman & Cook (1976) are to (l) portray accurately the characteristics of a particular individual, situation, or group, and (2) to determine the frequency with which something occurs or with which it is associated with other factors. Descriptive research involves the identification of a phenomena and detailed description of that phenomena (Borg & Gall, 1979). That which is to be studied must be clearly specified and adequate methods for measuring the phenomena must be found or developed. A clear formulation of what and who is to be measured is needed in addition to reliable and valid measurements. A plan of operation consistent with the purposes and procedures outlined for descriptive research by Sellitz, et al. (l976), and Borg & Call (1979) was used for the present study. A phenomena, that of sex bias in 105 psychotherapy, was identified and the area was clearly specified. The Sex Bias in Psychotherapy Questionnaire was developed and empirically tested so that the phenomena of interest could be measured. Sex bias in psychotherapy was then described through client report and factors potentially associated with sex bias in psychotherapy, including: client life-status variables, therapy exper- ience variables and therapist characteristics were explored to further describe the phenomena. In addition to measures , of central tendency, the statistical procedures used to describe the phenomena of sex bias in psychotherapy included: multiple factor analysis of variance, E-test and correlational models. A descriptive research design in the form of a naturalistic field investigation was used in this study because of the flexibility involved with this approach. Flexibility allows for the inclusion of a variety of sources of information and facilitates the study of a com- plex area. This type of research design does not address issues of causality, but rather in description has hueristic value. The results from the present study will begin to build theory, generate further questions, broaden the knowledge base concerned with the definition and measurement of sex bias in psychotherapy and provide direction for future research. 106 Research‘Hypotheses The purpose of this study was twofold: to develop an instrument that would measure sex bias in psychotherapy, and to investigate the phenomena of sex bias in psycho- therapy with women clients. Research hypotheses were generated for these two areas. The first set of hypotheses were developed to test the reliability and initial validity of the Sex Bias in Psychotherapy Questionnaire. Reliability of the SBPQ Hypothesis I(a): The internal consistency of the total items of’the Sex Bias in Psychotherapy Questionnaire will be sufficiently high to infer homogeneity of the construct of sex bias in psychotherapy. Hypothesis I(b): The internal consistency of each of the four sub-scales of the Sex Bias in Psychotherapy Questionnaire will be sufficiently high to infer that each sub-scale is measuring a separate dimension of sex bias in psychotherapy. Hypothesis II: The correlations between the total scale and the four sub-scales of the Sex Bias in Psycho- therapy Questionnaire will be sufficiently high to infer that the SBPQ is measuring a uni-dimensional construct. Initial Validity of the SBPQ Hypothesis III:2 The content of the items on the Sex Bias in Psychotherapy Questionnaire will be indicative of sex bias in psychotherapy. Hypothesis IV: A linear correlation will be observed between therapist-subject's scores on the Stereotype Scale 2This is not a testable hypothesis in the statistical sense. Support for this hypothesis is found in the pro- cedures used to attain content validity described in Chapter IV. 107 of the Personal Attributes Questionnaire and client- subject's scores on the Sex Bias in Psychotherapy Questionnaire. The second set of hypotheses were developed to test research questions relevant to the phenomena of sex bias in psychotherapy. Phenomena of Sex Bias in Psychotherapy Hypothesis V(a): The total scale scores on the Sex Bias in Psychotherapy Questionnaire will be sufficiently high to infer that women clients experience sex bias in psychotherapy. Hypothesis V(b): The sub-scale scores on the Sex Bias in Psychotherapy Questionnaire will be sufficiently high to infer that women clients experience sex bias in psycho- therapy. Hypothesis VI(a): The total scale scores for the Sex Bias in Psychotherapy Questionnaire will be different for women clients in different life-status categories. Hypothesis VI(b): The total scale scores for the Sex Bias in Psychotherapy Questionnaire will be different for women clients having different therapy experiences. Hypothesis VII: Women clients paired with female therapists will obtain different scores on the Sex Bias in Psychotherapy Questionnaire than women clients paired with male therapists. Hyiothesis VIII: Women clients paired with experienced- level t.erapists will obtain different scores on the Sex Bias in Psychotherapy Questionnaire than women clients paired with inexperienced-level therapists. H othesis IX: Wbmen clients paired with Community Mental fiéaItfi Center therapists will obtain different scores on the Sex Bias in Psychotherapy Questionnaire than women clients paired with Counseling Center therapists. 108 Procedures for Quantitative and Supplementary Descriptive Data Analysis Statistical procedures were used to test the nine research hypotheses. The statistics used to test these research hypotheses included: the E-test statistic, multi- ple factor ANOVA and Pearson product-moment correlations. Additionally, a supplementary descriptive analysis was performed to further describe the phenomena of sex bias in psychotherapy. The analytic procedures used are discussed below for each hypothesis. Instrumentation Hypotheses Hypothesis I (a and b) refers to the reliability of the Sex Bias in Psychotherapy Questionnaire. Test reliability in a broad sense provides an estimate as to hOW‘mUCh individual differences in test scores are attributable to "true" differences in the characteristics under investiga- tion and the extent to which they are due to chance errors (Anastasi, l97b). A more conceptual definition of relia- bility is that it is an estimate of the degree of consistency between two measures of the same thing (Mehrens and Lehmann, 1978). A number of estimates of reliability are attainable. Some of the estimates require data from two testing situations with the same individual. In the present study, two testing sessions were not feasible, due to issues of client confidentiality and privacy, so a reliability estimate made from.one administration was used. 109 A measure of internal consistency was used to estimate the reliability of the SBPQ total scale and sub-scales. The internal consistency of a test is a statement about the homogeneity of test items. It is the degree to which each item correlates with the total test score and the degree to which each item is tapping a similar construct. Cronbach's coefficient alpha was used as a measure of internal consistency for the SBPQ total scale and sub-scales. The coefficient is determined by computing all the possible means of the Split-half coefficients resulting from all possible random pairings (Cronbach, 1951). A value of at least .75 is considered necessary to judge a scale internally consistent and to infer homogeneity of test items. For research purposes an alpha coefficient of .65 is acceptable (Mehrens & Lehmann, 1978). Hypothesis II tests the relationship between the SBPQ sub-scales and total scale. In order to test this hypothesis, a Pearson product-moment correlation was com- puted. A product-moment correlation is used when both variables to be correlated are expressed as continuous scores. The product-moment correlation is the most stable correlation, with the smallest standard error of the correlar tion techniques, which provides a measure of the magnitude and significance of the relationship between two variables. In order to infer that the SBPQ is measuring a uni- dimensional construct, a significant level of correlation between the total scale and sub-scales needs to be observed. 110 Hypothesis III is related to the content validity of the SBPQ and is the only research hypothesis that was not tested statistically. Content validity is defined as the degree to which the sample of questionnaire items and responses represents the domain which the questionnaire is designed to measure, and about which inferences are to be made (Anastasi, 1976, and Mehrens & Lehmann, 1978). Because there is no statistical test of content validity, inferences about the content validity of the SBPQ were made from the procedures used in instrument development. A scale is assumed to have content validity if (1) the content domain was clearly defined, (2) specific objectives were developed and items were generated around these objectives, and (3) high inter-rater reliability was observed among expert judges rating content as reflective of the defined content dmmain. 'Hypothesis IV relates to the construct validity of the SBPQ. The construct validity of a test is the extent to which the test measures a theoretical construct or trait and as such requires the gradual accumulation of information from.a number of sources (Anastasi, 1976). One way to begin gathering evidence on construct validity is to test hypotheses about the characteristics of persons who obtain high scores as opposed to those who obtain low scores on the measure. This support for construct validity is tested in the phenomena of sex bias in psychotherapy hypotheses, the results of which are presented in Chapter V. 111 Another method for obtaining construct validity is to show that the test correlates with other variables with which it is supposed to theoretically (Mehrens & Lehmann, 1978). In the present study, it was hypothesized that a linear correlation would be observed between a client- subject's score on the SBPQ and her/his therapist's score on the Stereotype Scale of the FAQ. Following an analysis of therapist-subject Stereotype Scale scores, a Pearson product-moment correlation was computed on the aggregate SBPQ score and therapist-subject's Stereotype Scale score to determine if there was a relationship between the two variables. Aggregate means were generated by computing a mean scale score for those client-subjects who had the same therapist. A total of 66 aggregate means were generated from the 191 individual scores. Aggregate means were used in computation of the product-moment correlation and in other analyses as well. Phenomena of Sex Bias in Psychotherapy Hypotheses Hypothesis V (a and b) is concerned with testing whether or not women clients experience sex bias in psycho- therapy. Prior to testing of the hypothesis, the mean, standard deviation and range was observed for both the individual client scores and the aggregate mean scores. To test Hypothesis V(a) a one-sample E-test was performed on the aggregate mean for the SBPQ total scale. The E-test statistic is used to determine whether two means, 112 proportions or correlation coefficients, differ signifi- cantly from one another. It is also used to determine whether a single mean, proportion or correlation coefficient differs significantly from a specified population value (Borg & Call, 1979). A one sample E-test analysis is based on the assump- tion of normality. This assumption requires that the dependent variable be normally distributed across the population. The assumption of normality is most often met through the use of a random sampling procedure. Although client-subjects for this study were not randomly selected, it is noted that the E-test statistic is robust to viola- tions of the assumption of normality, particularly with sample n's greater than 30 (Glass & Stanley, 1970). The E-test statistic also relies on the assumption of inde- pendence. In the present study some clients were seen in therapy by the same therapist. Thus, some observations would be considered dependent. The use of aggregate means in the E-test analysis is one method for handling problems with dependency in observations. The SBPQ was designed so that psychometrically a total scale score of 96 or less would be indicative of a relatively sex—fair counseling experience. A mean of 96 is the theoretical mid-point of the distribution which ranges from 0 to 192. This mean was chosen because it represents a hypothetical score of 1.5 (for each item) x 64 (total items) from a possible 0-3 point scoring for each item on 113 the SBPQ. At this point in the development of the SBPQ, use of a theoretical population mean was warranted for lack of a standardization sample or other means of determining the true population mean. A theoretical population mean of 96 is considered to be a conservative estimate,of the true population mean. For Hypothesis V(a) the aggregate sample mean score obtained on the SBPQ total scale was tested against a population mean of 96. The hypothesis that the sample mean was greater than the population mean of 96 was tested using a two-tailed E-test at a signifi- cance level of .05. Similar procedures were used to generate psychometric means for the SBPQ sub-scales and test Hypothesis V(b), whether or not there was evidence of sex bias in psycho- therapy as measured by the four sub-scales of the SBPQ. For the Traditional Roles Scale the aggregate sample mean was tested against a theoretical population mean of 30. The aggregate sample mean for the Bias and Devaluation Scale was tested against a theoretical population mean of 37.5. For the Psychoanalytic Scale the aggregate sample mean was tested against a theoretical population mean of 22.5 and for the Sex Object Scale the aggregate sample mean was tested against a theoretical population mean of 6. All E-tests for Hypothesis V (a and b) were performed at the .05 alpha level. In addition to statistical tests of Hypothesis V (a and b), the frequencies for item endorsement were 114 resented for further description of the phenomena of sex bias in psychotherapy. Observations of these frequencies provides useful information concerning the interpretation of the group means for the SBPQ total and four sub-scales and aids in the continued development of the SBPQ. Hypothesis VI (a and b) is concerned with the effect of client life-status variables and therapy experience variables on the experience of sex bias in psychotherapy. The hypothesis attempts to examine the degree of sex bias experienced by different groups of women clients. In order to test this hypothesis, a number of one-way ANOVAs were performed on individual SBPQ Total Scale Scores for the client life-status variables of age, education, relation- ship status, child status, and for the therapy experience variables, including the number of sessions and length of time since the last session. One-way analysis of variance is an inferential technique used to determine if two or more sample means differ significantly from one another. It is the appro- priate test statistic to use when comparing two or more groups, because of its power and flexibility. Analysis of variance determines if the variability between groups is large enough to justify the inference that the means of the populations from which the different groups are sampled are not the same. If 'the variability between groups is large enough, it is inferred that individuals in these groups came frmm different populations and that there is a 115 statistically significant difference between the two populations (Isaac, 1971). The assumptions for the ANOVA model include those of normality, homogeneity of variance, and independence of observations. The one-way ANOVA is robust to violations of normality._ Although some uncertainty exists over whether or not the assumption of independence will be met, a decision to use the one-way ANOVA on individual scores was made on the intuitive assumption that client life-status variables and therapy experience variables are independent of the therapist. If the therapist variables prove to be insignificant in regard to client's experience of bias, it can be more comfortably assumed that the observations were independent. If this assumption is not met, the error in hypothesis testing will be in a conservative direction. Separate one-way ANOVA's were performed on SBPQ total scale scores for client life-status variables and therapy- related variables to determine if the groups differed in the degree of sex bias in psychotherapy experienced. All one-way ANOVA's were tested at the .05 level of signifi- cance. Following computation of the one-way ANOVA for each variable, select two-way ANOVA's were computed on variables thought to interact with one another. Because the ANOVA model only determines that there is or is not a difference between groups, where a signifi- cant difference was found among groups, a post-hoc test of paired comparisons was performed. The Scheffee test of 116 multiple comparisons was used to determine which groups differed from one another. This is a conservative post hoc test that uses an experimentwise error rate, recommended for use with unequal n's (Glass & Stanley, 1970). A .05 level of significance was used for post-hoc testing. Hypotheses VII, VIIIL and IX are concerned with the effect of therapist variables on women clients' experience of sex bias in psychotherapy. To test these hypotheses three two-way ANOVA's were performed with two levels for each independent variable. The independent variables are therapist sex, experience level, and agency affiliation. Two-factor ANOVA models are based on the same principles as one—factor ANOVA models. In addition to testing for a main effect, two-factor ANOVA models test for interaction effects. Multiple-factor ANOVA models rely on the assump- tions of normality, homogeneity of a variance and inde- pendence. As with the one-factor ANOVA, the two-factor ANOVA is robust to violations of the normality assumption. It is also robust to violations of homogeneity of variance with equal n's or a balanced design. All efforts were made to maintain balanced designs in the two-way ANOVA's. The assumption of independence was met through the use of aggregate mean scores. To test Hypotheses VII, VIII, and IX, two-way ANOVA's were computed on the aggregate mean SBPQ total scale scores for the three therapist variables of sex, experience level, 117 and agency affiliation. A significance level of .05 was established for hypothesis testing. Supplementary Descriptive Data Analysis Certain aspects of the present study did not lend themselves to quantitative analysis and the rigor of test- ing for statistical significance. These aspects have to do with the experience of sex bias in psychotherapy reported by sub-groups of women, including: those women in a relation- ship with a man, women who talked about job and/or career concerns in therapy, working women/women in school with or without children, incest survivors, rape victims, and battered women. A qualitative approach was used to describe the data obtained from these sub-groups of women. Specifi- cally, the response to items in each sub-group category were observed and the mean and range from each sub-group was presented. Those items with means greater than the mid- point of 1.5 were noted for each sub-group category. A similar qualitative approach to description was used to determine which items on the SBPQ were more often acknowledged in a sex-biased direction. The 10 items with the greatest item means were observed and then grouped by area of bias for further description of the phenomena of sex bias in psychotherapy. Finally, some respondents provided anecdotal data regarding sex bias in their counseling experience. This 1LL8 .3333:ng 5 663 66 no 6863688 6.5636 5:53 co 836335. >86w6 6:6 ~6>6~ 6361696 :66 NO 6636561» 539365 65 mo 36636 65 you 566“. 3 66.866 8mm :66... 6umw6umw6 so @6395“. 6.63 6.352 >365 66.5“. .mcomwumnroo 6333: no “.66”. .66ww6nom 3 83:66 5 .665ng 6.63 62992 >63..qu uo6~6m .xdmu655zmd 5 663 X66 mo 6863698 6.3636 :29 co pm: 6636365 6665 36636 5653 65.5366 3 6633.23 .3365 6:6 6636361, 6336 6m: 5:626 65 now 6.866 8mm #656335 65 co pmfiomua 6663 62992 563165 mo 66366 < $966505me 5 6.63 66 6863688 3:636 :95: so: no 65653 565 on 666: 663 636356 .58 6:6 ~33 65 you :66... 6umw6uwwm 65 so ”.66qu 63:66:65 6 .6566 65 63.5666 3 666: 6663 5:66:65 35:66 «0 66.36662 :05ng 56:56:06 6.9mm 65 9366”. 5mg 05 3QO 6930665 65 :o 6366 36336133365 ~633pr 65 pow 65066 3mm :68. 6umw6uwwm 65 56353 6653956 663 8336506 5:688:59qu 5366a < hams, 5:65:06 6.9mm 65 you 68836 6630.5 3 @662 6.63 5:688ng 2555ma now 66.56660?“ .uoaumcoo #5396535 m. 93:666.: 663 gm 65 mo: no 656;: 5665 ob 6.36863...“ 50m 6:6 ~38 gm. 65 .6653 “6639.06 663 «5336.58 ucéufifioud comu66m < $533626“ pom 5:336:06 ~656qu 98$ 65 5665 3 @663 663 236 5:33:68 6.53.6.5 xH mammeuoaxz :9 34.28%: as 33695. 3 + my H> 3.623%: 3 + 3 > mwmgummNm 6666595.— mmfim 6m «0 6:85:65 Z 6.338%: E 528%: S 3858»: 3 + 3 H 3665mm... 26653? :3 53:55.65 6.566% 36365 658 93633630 paw $666605 m.m 6369 3,6650%: 119 information was not analyzed but is described for its added richness to the issues relevant to the present study. Summary The purpose of the present study was twofold. The first purpose was to develop an instrument that would measure sex bias in psychotherapy. The second purpose was to investigate the phenomena of sex bias in psychotherapy. A descriptive research design in the form of'a naturalistic field investigation was used in conjunction with a descriptive approach to data analysis to address the dual purposes of the study. Using a survey approach for data collection, a sample of 192 terminated women clients from a university counsel- ing center and a local community mental health center completed the Sex Bias in Psychotherapy Questionnaire. The SBPQ is a rationally constructed instrument, composed of four sub-scales, developed to measure the phenomena of sex bias in psychotherapy. In addition to client-subjects, 49 therapist-subjects completed the Stereotype Scale (Spence, et al., 1974). The Stereotype Scale provides a measure of one's tendency to stereotype men and women, in general. The results of this scale were used to begin testing the SBPQ's construct validity. Hypotheses related to both instrumentation and the phenomena of sex bias in psychotherapy were developed and 120 procedures to test these hypotheses were outlined. Cronbach's coefficient alpha was used as a measure of internal consistency for the SBPQ total and four sub- levels, and a product-moment correlation was used to deter- mine if the SBPQ was in fact measuring a unidimensional construct. Initial tests of the SBPQ's validity focused on both content and construct validity. Content validity was assessed through an examination of the procedures used for item development. A correlation between a client's SBPQ score and her therapist's Stereotype Scale score was computed to begin testing construct validity. In addition to reliability and validity hypotheses, hypotheses related to the phenomena of sex bias in psycho- therapy were developed and tested. These hypotheses focused on the effects of client life status, therapy experience, and therapist variables on the experience of sex bias in psychotherapy, in addition to studying the extent of sex bias in psychotherapy with women clients. The procedures used for data analysis included: one- and two-factor ANOVA, E-test, and correlational models. Supplementary descriptive analysis for specific sub-groups of women and SBPQ items was also used to aid in descrip- tion of the data. Descriptive research such as this, while not addressing cause and effect, has heuristic value. The results from the present study will begin to build theory and broaden the knowledge base concerned with the definition and measurement of sex bias in psychotherapy. CHAPTER IV RESULTS: PART I DEVELOPMENT OF AN INSTRUMENT TO MEASURE SEX BIAS IN PSYCHOTHERAPY This chapter contains the procedures used in the development and empirical testing of the Sex Bias in Psycho- therapy Questionnaire. The SBPQ is a rationally constructed instrument supported by theory. The theory supporting item generation is presented along with tests of the hypotheses related to the SBPQ's reliability and validity. Instrument Development As demonstrated by the comprehensive review of the literature, there is no measure which directly assesses women clients' experience of sex bias in psychotherapy. For the purposes of this study, such an instrument was developed. The procedure used for questionnaire development involved generation of items around specific objectives, expert judging of items for content and informal piloting of the questionnaire prior to administration. The Sex Bias in Psychotherapy Questionnaire is based on the themes of bias outlined by the APA Task Force on Sex Bias and Sex-Role Stereotyping in Psychotherapy (1975, 1978) and on theories relevant to these themes, including: critiques of traditional psychoanalytic theory, sex-fair personality theories, newer theories on the psychology of 121 122 women and theories of non-sexist and feminist therapy. From these theories a table of specification or "blueprint" was developed around which items were generated. This "blue- print" involved four main themes pertaining to women's treatment in psychotherapy and are as follows: I) Fostering Traditional Sex Roles 2) Bias in Expectations and Devaluation of Women 3) Sexist Use of Psychoanalytic Concepts 4) Responding to Women as Sex Objects (including the seduction of women clients) Objectives for these themes were developed from theory and items for each objective were generated by the researcher and 10 other clinicians sensitive to women's issues in therapy. Initially, 200 items were developed for the SBPQ. These items were then rated by three male and three female expert judges. Expert judges were selected because of their expertise in the area of women's issues in psychotherapy and demonstrated sex-fair counseling style as observed by their peers and clients alike. Expert judges represented the fields of psychology, social work and psychiatry, and have been practicing psychotherapy for eight to 22 years, or an average of 13 years. All of the judges either write, teach, or conduct workshops in areas relevant to sex-role issues in psychology, and are up-to-date with the literature in the area of women's sex-role issues in therapy. All expert judges are currently practicing and see many women clients in therapy. A copy of the instructions given to expert judges is found in Appendix E. 12? For an item to be retained for the SBPQ, five out of six judges had to agree that the item was indicative of sex bias in psychotherapy. Therefore, all items on the SBPQ have an inter-rater reliability of at least .83, with more than 75% of the items having 1.00 inter-rater reliability. Following expert judging and informal piloting, 108 items were retained for the SBPQ. Of the 108 items, 72 are general items, answerable by any woman; the remaining 36 items are written for specific sub-groups of women. The psycho- metrics for the SBPQ are based on the 723 general items, with a supplementary descriptive analysis to interpret the responses to items developed for specific sub-groups of women . Items for the SBPQ are self-report and tap both behavioral and perceptual indicators of sex bias. Items ‘were written in a likert-type format and respondents were requested to answer each item along the following continuum: a) Definitely False b) Mostly False c) Mostly True d) Definitely True . All items were scored as follows: a) = 0, b) = l, c) s 2, and d) = 3. This scoring was reversed for positively worded items. As was discussed previously, a 3In actuality, the psychometrics of the SBPQ, includ- ing the generation of the theoretical population mean, were based on 64 items. Initial revisions of the SBPQ included the removal of eight items. These revisions are discussed later in this chapter. 124 score of 96 on the SBPQ was chosen as the theoretical mid- point of the SBPQ distribution which ranges from O to 192. Four sub-scales make up the total scale of the SBPQ. These subscales reflect the themes of bias outlined by the APA Task Force (1975), and are named the Traditional Roles Scale, the Bias and Devaluation Scale, the Psychoanalytic Scale, and the Sex Object Scale. A copy of the SBPQ is found in Appendix A. Fosterinngraditional Sex Roles Both Broverman (1970) and Chesler (1972) have raised concerns that mental health professionals maintain an adjustment notion of mental health for women. Their research and theorizing suggest that women in therapy are taught to adapt to the traditional female role. Satisfac- tion with the role of wife and mother and little ambition for work and/or career are thus implied. Research also docu- ments that it is at least partially due to these role constrictions that wmmen, more than men, experience depression and other emotional problems which bring them to therapy (Cove and Tudor, 1972; Franks and Rothblum, 1983). Sex-fair therapy allows a woman client flexibility in regard to her role and supports the development of both the expressive and instrumental aspects of her personality. Sex-role stereotyping as evidenced in psychotherapy as fostering traditional sex roles was broken down into the following objectives: 125 l. The therapist's interventions are aimed at teach- ing his/her female client to adjust/adapt to the traditional female role. Lack of tolerance for deviations from this role is demonstrated. 2. The therapist demonstrates the belief that the woman's role is as a wife and that growth is achieved through marriage or a relationship with a man. More concern is shown for the husband/ boyfriend's needs than for the woman client. 3. The therapist expresses the belief that his/her woman client is unfulfilled without being a mother. 4. The therapist lacks sensitivity to his/her woman client's work, career, role diversity, and the conflicts these may raise for her. A fifth objective was originally used, that of: 5. Fostering of traditional sex-roles through the maintenance of a power differential in the therapeutic relationship. But inter-rater reliability for items generated around this objective was insufficient to retain these items for the scale. Nonetheless, the objective was retained as part of the definition for sex—bias in psychotherapy. Additionally, items for these objectives were based on the following Guidelines for Non-Sexist Therapy (APA, * 1978a): l. The conduct of therapy should be free of constric- tions based on gender defined roles, and the options explored between client and practitioner should be free of sex-role stereotypes. 3. The therapist should be knowledgeable about current empirical findings on sex roles, sexism and individual differences resulting from clients' gender-defined identity. * A complete list of these guidelines is found in Appendix F. 126 The following items were developed for the Traditional Roles Scale: General Items l. * 9. 10. ll. 12. l3. 14. I was encouraged by my counselor to accept things as they are. My counselor discouraged my attempts to change things in my life (like ending a relationship, changing jobs, going back to school). My long-range personal and/or professional goals were supported by my counselor. My counselor suggested that it was more important for me to respond to others' needs than to state my own wishes and wants. When I talked about problems in relationships, my counselor suggested that I consider the other person first in an attempt to solve the problems. My counselor suggested that I needed to give in to others in relationships. My counselor suggested that I take a back seat in my relationships with others. Often my counselor focused more on the impact which my problem would have on others than on my own feelings about the problem. My counselor helped me learn ways other than cry- ing, flirting, and/or lying to get what I wanted. I was encouraged to use crying, flirting, and/or lying to get what I wanted. My counselor told me that men should be the decision-makers at home. My counselor told me that men should be the decision-makers at wOrk. My counselor discouraged my interests in activities like sports, car repair, and woodworking. My counselor seemed to devalue my non-traditional women friends. 127 My traditionally feminine women friends seemed to stereotype women as mothers, wives, and/or daughters. roles other than those of mother, daughter, wife/ My counselor suggested that my life would be better develop an ability to be sensitive to others, not feel it was important for my husband/boyfriend to My counselor implied that my marriage/relationship with a man ought to be the most important priority My counselor encouraged me to listen carefully to my husband/boyfriend, so that I would be more able identity, but did not feel this was important for should come second to my husband/boyfriend' 3 needs. 15. be overly valued by my counselor as a support system for me. 16. My counselor discouraged me from reading books showing women in non-traditional roles. 17. My counselor suggested that I read books which *18.- My counselor helped me see that women could have girlfriend. 19. The purpose of my counseling seemed to be for me to learn how to be more satisfied in my role of wife/girlfriend, mother or daughter. 20. My counselor suggested that my problem would be solved by marrying. 21. if I had a baby. Specific Items for Sub-Groups of Women 22. My counselor thought it was important that I develop this for himself. 23. in my life. 24. to make him happy. 25. My counselor seemed to believe that the most healthy option I had once in a relationship was to stay in it and work out any problems. 26. My counselor thought it was important for my husband/boyfriend to develop a sense of his own me. 27. My counselor' 3 comments suggested that my needs *28. My counselor seemed to understand my frustration when my husband/boyfriend dealt with our relation- ship in a logical manner. 128 29. My counselor was accepting of my exploring both male and female dominated fields as career options. *30. My counselor was supportive of my decision to go on to graduate school/return to school/find a full- time job. 31. My counselor seemed to push me towards female- dominated fields (such as teaching, nursing, secretarial work) as a profession. 32. My counselor encouraged me to stay in professions open to women. 33. When I talked about wanting a job, my counselor suggested that it may be better for me to volunteer my time. 34. My counselor felt that I should not place more importance on my career/job education than on my family. 35. My counselor's comments suggested that my career should come second to my children's needs. 36. When I expressed interest in male-dominated work/ academic areas, my counselor reminded me of the difficulty of integrating such a career with a family. Bias in Expectations and Devaluation of WOmen Broverman, et al. (1970, 1972), and others replicating their research note that clinicians maintain a double standard of mental health for men and women reflective of an adjustment notion of mental health for women. The healthy female is perceived as more passive, dependent, and emo- tional and less independent and assertive than the healthy male, possessing more expressive than instrumental qualities. Personality theory, particularly the theories of androgyny (Bem, 1975, Kaplan, 1976, 1979b, and Gilbert, 1981b). Note: Items with an asterisk are positvely worded and require reverse scoring. 129 sex-role transcendence (Hefner, Rebecca, & Oleshansky, 1975) and self-actualization (Maslow, 1968) point to the importance of an individual possessing both instrumental and expressive qualities such that flexibility can be used in coping with the demands of one's environment and inter- personal relationships. Bias in expectations would be seen in the clinician's attempt to reinforce feminine or expressive qualities in women clients while ignoring or dis- couraging the development of more masculine or instrumental qualities. Bias would also be observed in a devaluing of the women's strengths, such as the ability to express oneself emotionally and to form affiliations (Miller, 1976, Gilligan, 1982). Support for the recognition and encouragement of women's anger and assertiveness in therapy is provided by Bernardez (1978). She contends that "contemporary women's problems with the expression of anger, rebellion, and protest are central to the understanding of women's difficulties in creative and active pursuits” (p. 215). She theorizes that sanctions against the expression of anger and negative affects in women lead to inhibitions in creative and work endeavors, self-defeating behaviors, pathologic sub- missiveness and dependency, depression, and lack of self- respect. An unbiased counseling experience is one that allows a woman the freedom to experience anger and other negative affects. 130 A sex-fair counseling experience is one that encourages the development of assertiveness, independence, and autonomy while recognizing a woman's need for affiliation with others as a strength. Recognition of the role of attachment in women's lives requires an understanding of women's develop- ment as different from that of men. Gilligan (1979, 1982) identifies the role of care, responsibility and affiliation as central to women's psychological and emotional develop- ment. Recognition of the socialization process for women as contributing to their difficulties with separation and autonomy and strengths in the emotional and affiliative areas is an essential aspect of Gilligan's theory. Lerner (1974) and Bernardez (1978) while incorporating a socio-cultural perspective into an understanding of the devaluation of women and the fear of women's anger, also view the problem as having intra-psychic origins within our culture. Both theorize that this devaluation and in fact the very nature of that which is defined as feminine and masculine stem.from a defensive handling of the affects originating in the early infant-maternal bond. Men and women alike carry with them unconscious fears of women's omnipotence and dread of her potential destructiveness due to beliefs rooted in infant development. These unconscious fears and beliefs create the stereotype of women as devoid of anger and power. In therapeutic interactions these unconscious beliefs become apparent when the therapist makes 131 demeaning statements to or about women and displays more value for men than women. Bias in expectations and devaluation of women is also noted with particular sub-groups of women. Rape victims, incest survivors and battered women have reported instances of bias where their experiences were either not believed or were minimized. Additionally, theoretical constructs such as masochism have been used to condone the victimization of women (APA, 1975, Bograd, 1982). A sex-fair approach to counseling rape victims recog- nizes the crisis of the rape and the resulting trauma which includes a generalized fear of men, fear or uneasiness about all sexuality, guilt, decreased sense of independence, and fears of being alone. This trauma can be dealt with in a supportive counseling relationship, but addressing the issues related to loss of autonomy and power is more easily accomplished through the use of community resources such as feminist self-defense classes and support groups for rape victims (Collier, 1982). Similarly, sex-fair counseling approaches for incest survivors recognize the benefits of group therapy for this sub-group of women. Group therapy for incest survivors allows the survivor to feel less isolated and bound by secrecy. Additionally, finding of commonalities among other survivors in problems of low self-esteem, guilt, passivity, shame, and poor relationships helps clients to externalize these patterns and recognize them as a consequence of the incest rather than as an internal flaw (Gelinas, 1981). 132 The cornerstone of a nonsexist treatment approach for battered women is to first halt the physical abuse by help- ing the client find an alternative living situation. It is only after a woman feels protected from another assault that she can begin to deal with the reality of the situation. The lack of direct intervention to stop the violence is seen as an implicit acceptance of the violence as either normative or understandable in the context of an intimate relationship. Sex-fair therapy with battered women helps the client work out the practicalities of leaving her situation, confronts the client's reluctance to discuss the physical abuse, reallocates responsibility of the abuse to the abuser, and helps the client explore the reasons she remains in a physically abusive relationship (Walker, 1981; Bograd, 1982; Collier, 1982). Additionally, therapists aware of the issues for battered women have an understanding of the psychological and sociological factors contributing to the client's difficulty in separating from an abusive partner (walker, 1980). Sex-fair therapists acknowledge and recognize the reality of crimes against women and take into account the cultural environment in which a woman has developed when dealing with all women, especially those who have been physically and/or sexually victimized (Herman and Hirschman, 1977, Symonds, 1979, and Hilberman, 1980). The following objectives were developed from the above theory regarding bias in expectations and devaluation of women in psychotherapy: 133 The therapist denies the adaptive potential of assertiveness, independence, anger, and ability to identify and experience emotions, and fosters concepts of women as passive, dependent, and "emotional." Sociological/cultural explanations for some of the women client's difficulties are absent. The therapist either directly or indirectly makes demeaning statements about women, often with the use of derogatory labels. Devaluation of women as evidenced by the therapist's interest in the woman client's relationships with men to the exclusion of the therapist's interest in the woman client's relationships with women. The therapist lacks sensitivity and awareness con- cerning the reality of incest, rape, and physical abuse (non-sexual) for women and tends to minimize the impact of these experiences. The therapist uses theoretical constructs (e.g., masochism) to condone the victimization of women. Additionally, items for these objectives were based on the following Guidelines for Non-Sexist Therapy (APA, 1978): 2. 10. ll. Psychologists should recognize the reality, variety and implications of sex discriminatory practices in society and should facilitate client examination of options in dealing with such practices. The theoretical concepts employed by the therapists should be free of sex bias and sex-role stereotype. The psychologist should demonstrate acceptance of women as equal to men by using language free of derogatory labels. The psychologist should avoid establishing the source of personal problems within the client when they are more properly attributable to situational or cultural factors. The client's assertive behavior should be respected. The psychologist whose female client is subjected to violence in the form of physical abuse or rape should recognize and acknowledge that she is a victim of a crime. 134 The following items were developed for the Bias and Devaluation Scale: General Items *1. *2. *3. *4. * 8. *10. ll. 12. *13. 14. *15. 16. My counselor encouraged me to think and act for myself. My counselor helped me to learn to rely more on myself than on others. My counselor understood my anger. My counselor encouraged me to express my angry feelings. My counselor implied that it was okay for me to be angry for others, but not for myself. My counselor dealt with my sadness/depression in an understanding manner, but ignored my anger/rage. It seemed to me that my counselor felt more comfortable with my tears, hurt, or self-blame than with my anger. My counselor was able to accept my anger. I think my counselor saw my anger as unhealthy. My counselor seemed to understand that sometimes when I cried, I was really angry. My counselor criticized me for being irrational and/or illogical when I expressed the emotions I felt. My desire for some power and equality was seen as controlling by my counselor. My counselor saw my ability to identify and under- stand my emotions as a strength. My counselor made me feel that it was not okay for me to compete with others. My counselor saw my ability to express my emotions as a strength. My counselor made me feel that my problems had nothing to do with growing up as a woman in our society. *17. *18. 19. 20. 21. 22. 23. 24. 25. 26. Specific *27. *28. *29. 30. 3l. 135 My counselor pointed out that some of my problems came from being a woman in a society which sees women as unequal to men. My counselor told me that some of my problems were s1m1lar to those of other women. My counselor said that I had done well for a woman. My counselor made comments like that's "just like a woman.” My counselor either directly or indirectly puts women down. My counselor seemed to admire men more than women. My counselor looked at the difficulties I had with men as solely my problem, as if the men had no part in the problem. My counselor was more interested in hearing about my intimate relationships with men than about my work/education/outside interests. My counselor implied that men were more easy to get along with than women. My counselor implied that relationships with men were of more value than relationships with women. Items for Sub-Groups of Women My counselor seemed to understand the frustration I felt when my husband/boyfriend criticized me for expressing my emotions. My counselor seemed to understand my frustration when my husband/boyfriend encouraged me to be more logical. My counselor discussed with me the possibility of joining a group for incest survivors. When I talked about my incest experience, my counselor wondered whether I had unknowingly seduced my father/stepfather/brother/uncle/ grandfather. My counselor blamed my mother for the incest in our family. 136 32. When I discussed the incest with my counselor, s/he asked if I had enjoyed the experience. 33. My counselor did not believe that I was an incest survivor, but called it my fantasy or wish. *34. My counselor discussed with me the option of taking a course in self defense. 35. My counselor would insist that there was no such thing as rape, but that women asked for it. 36. When I discussed the rape with my counselor, s/he asked if I had enjoyed the experience. 37. My counselor asked me what I was wearing when I was raped. *38. My counselor helped me look at the reasons why I stayed in a physically abusive relationship. *39. My counselor helped me to see that women did not have to put up with being beaten by their husbands/boyfriends. *40. My counselor suggested that I look into a home for battered women or move to a safe place. 41. My counselor knew that I was sometimes physically abused, but did not explore this in much depth. 42. My counselor seemed to believe that lots of women in our society are slapped around by their husbands/boyfriends, so it's nothing with which to be concerned. Sexist Use of Psychoanalytic Concepts Freudian theory and practice represents a major component of psychotherapeutic knowledge and as such has had a profound impact on the understanding of female psychologi- cal development. As Robert Jay Lifton noted, "Every great thinker has at least one blind spot; Freud's was women" (Gilman, 1973, p. 10), so have others critiquing the useful- ness of Freud's theory for women (Carlson, 1972, Chessler, 1972, Rice and Rice, 1973, Barrett, et al., 1974, Levine, 137 Kamin & Levine, 1974, and Howell, 1981). His theory on the natural inferiority of women due to the Complexities surrounding their lack of a penis and difficulties in resolving the Oedipal situation have contributed to sex bias in therapy and a view of women as inherently dependent, sub- missive, passive, narcissistic, lower in moral development, more envious, childlike, and incomplete than men (Freud, 1964). In response to Freud's theories on feminine develop- ment, Horney (1932, 1973) and later Thompson (1943) provided less biased insights into women's psychological development. Both reworked the concept of penis envy to account for a socio-cultural understanding. This more socio-cultural view focuses more on women's envy of men's power in the environment and less on envy of their sexual organs. Observa- tions of women's problems with dependency and interest in their appearance are viewed as originating from economic and social factors. Miller (1976) also relies on a socio-cultural under- standing of the development of stereotypic female qualities. She theorizes that women as the subordinate group in our culture developed sex-role stereotypic qualities because of their survival value. It is thus theorized that women are culturally conditioned to be more passive, dependent, and indirect from early childhood and acquire and maintain these qualities as a way to survive as an oppressed minority. 138 Support for Miller's theory is found in the similarity between the sex-role stereotypic characteristics of women and the characteristics of other oppressed minorities in our society. Neo-Freudians, such as Erikson, while paying more attention to the unique aspects of women's psychological development, perpetuated traditional stereotypes of women as mothers and wives whose satisfaction comes from tending to the needs of others. Erikson (1974) as Freud replied on anatomy to explain a woman's orientation to her "inner space" and theorized that a woman's development is centered around her inner organs and ability to bear children. He also theorized that a woman, unlike a man, delays the development of her identity until she "succeeds in select- ing what is to be admitted to the welcome of the inner space for keeps" (p. 310). Erikson maintained that a woman remains unfulfilled without bearing children and that her identity is at least in part developed around that of the man she chooses to marry. Gilligan (1982), in recognizing the importance of relationship for women, explores the hypothesis that women, unlike men, develop a sense of their identity in the context of relationship. Rather than viewing women's development as delayed because it deviates from that of men, 139 Gilligan suggests that for women the stages of identity and intimacy may be more closely woven. A coexistence between these two stages is observed in women's development of identity, implying that for women psychosocial stages of development may not proceed in a hierarchical manner. She theorizes that: From the different dynamics of separation and attach- ment in their gender identity formation through the divergence of identity and intimacy that marks their experience in the adolescent years, male and female voices typically have spoken of the importance of different truths, the former of the role of separa- tion in development as it comes to define and empower the self, the latter of the ongoing process of attachment that creates and sustains the human commity. (Gilligan, 1980, p. 18-19) Out of more traditional psychoanalytic traditions came the practice of psychoanalysis and the concepts of trans- ference and resistance. The negative view of women which comes across in traditional psychoanalytic theory is translated into therapeutic actions. For example, sexist use of psychoanalytic concepts is apparent when insisting that in order to work through a transference neurosis a client must be sexually attracted to her therapist, and in interpreting the need a client has to talk about relevant issues in her life which have nothing to do with being a wife or mother as resistance (APA, 1975, 1978). Because traditional psychoanalytic theory acts as a base and foundation for current-day psychotherapeutic practice, it is likely that the sexism inherent in components of the theory are translated into therapeutic 140 actions during the therapy hour and in the understanding of client issues. A nonsexist therapist is one who is aware of the flaws in traditional psychoanalytic theory for women clients, has knowledge concerning newer theories of women's development, and recognizes the impact of socio- cultural issues on women's lives. Sex-fair counseling maintains flexibility with regard to the use of theory and technique. The following objectives were developed from the above theory regarding sexist use of psychoanalytic concepts in psychotherapy with women clients: 1. The therapist incorporates the negative Freudian view of women into his/her work with clients. 2. The therapist insists on using the concepts of penis envy, Oedipal conflict, transference and resistance, without flexibility or understanding of women's development. 3. The therapist perpetuates myths about motherhood and marriage as all-fulfilling for women, and tends to judge the emotional maturity of women clients on their attitudes towards childbearing and/or rearing and intimate relationships with men. 4. A double standard for sexual experiences of men and women is demonstrated by the therapist. Items for these objectives were also based on the following Guidelines for Non-Sexist Therapy (APA, 1978): 4. Theoretical concepts employed by the therapists should be free of sex bias and sex-role stereotyping. 6. The psychologist should avoid establishing the source of personal problems within the client when they are more properly attributable to situational or cultural factors. 12. 141 The psychologist should recognize and encourage exploration of a woman client's sexuality and should recognize her right to define her own sexual preferences. The following items were developed for the Psycho- analytic Scale: General Items l. *9. 10. 11. 12. My counselor implied that men were better people than women. My counselor said that women had lower moral standards than men. My counselor implied that women need men more than ‘men need women. My counselor implied that women were more envious than men. My counselor seemed to think that women needed another person to make them feel good about themselves. My counselor said I was envious of men when I talked of the inequality between the sexes. My counselor could not understand that it was men's power in the world which I envied, not their sexual organs. My counselor insisted that my problems stemmed from my lack of and/or desire for a male sexual organ. My counselor seemed to understand the frustration I felt about real inequality between men and women. My counselor insisted that my lack of sexual attraction for him/her was delaying my progress. My counselor thought that my talking about my career/job/education was an avoidance of the real problem. My counselor implied that I would not be happy until I was in a relationship with a man. 142 My counselor felt that I would not be fully content My counselor believed that it was natural for men to have sex outside of their primary relationship, My counselor seemed to feel that it was more okay because of my choice to work/go to school, rather My counselor thought that I should be content stay- 13. unless I had children. 14. My counselor called my sexual experiences promiscuous. 15. My counselor seemed to feel that it was natural for men to want to have sex outside of their primary relationship, but not so for women. 16. but not so for women. Specific Items for Sub-Groups of Women 17. for my husband/boyfriend to have sex outside of our relationship than it was for me. 18. My counselor questioned my adequacy as a mother than stay home with my children. 19. ing home and raising children. *20. My counselor was able to help me understand my feelings of sometflmes resenting my children. *21. My counselor helped me question m belief that a "good mother" never resents her c ildren. Responding to Women as Sex Objects (including the seduction of women clients) A well-established sanction exists in the Ethical Princi- ples for Psychologists against sexual intimacies between therapist and client (APAH l98ri anetheless, various studies have found that some clinicians engage in sexual contact with their clients (Dahlberg, 1970, Masters and Johnson, 1970, Chesler, 1972, Belote, 1974, Butler & Kelen, 1977, and Holroyd & Brodsky, 1977). Erotic contact between 143 a clinician and her/his client is viewed as a blatant form of sexism and bias in psychotherapy in at least three ways: a) Nearly all complaints are from women patients regarding male therapists. b) Stereotypic feminine qualities, especially passive-dependence, are exploited. c) The male therapist has considerably more power in the therapy situation than the female patient, a classic situation for the operation of sexual politics. (APA, 1975, p. 1170) Sexual relations between a client and therapist are justified by some as therapeutic for sexually dysfunctional women. Yet, in one study on sexual intimacy between female clients and male psychotherapists, 25 women reported being nonorgasmic before and after therapy with all men including their therapist (Belote, 1974). Masters and Johnson (1970) found an "unfortunately large" number of their patients had had sex with their former therapists. To date, no study has been found in which sexual intimacy between a female client and her male therapist was beneficial. Sexual intimacy between a client and therapist exploits the nature of the therapeutic relationship and the client's trust that the therapist is acting in his/her best interest. Holroyd and Brodsky (1980) found that those thera- pists who engaged in non-erotic touching with only opposite-sex clients were at high risk for more intimate sexual contact with female clients. Differential treatment of men and women around issues of touching, dating, and general seductiveness during the therapy hour can be 144 considered sex-biased therapy with a high probability that this behavior will lead to sexual intercourse or unethical erotic contact. While seduction of women clients is a blatant form of sex bias in psychotherapy, emphasis on a woman's physical appearance in selection of clients, judgments about clients, and in the actual therapy is also a form of sex bias in therapy (APA, 1975). Hobfall and Penner (1978) found that clinicians' ratings of a client's self-concept were unduly influenced by her attractiveness. Likewise, Schwartz and Abramowitz (in press) found that more attractive female clients elicited more social responsiveness from male psychology graduate students and were judged more likely to remain in treatment than did a less attractive female. This form of sex-bias in psychotherapy is observed in the therapist's encouragement of the client to dress or in general to appear more feminine. The following objectives were developed from the above stated ethical code and theory regarding responding to women as sex objects, including the seduction of women clients: 1. Sexual relations between client and therapist. 2. The therapist subtly shows sexual interest in the female client. 3. The therapist responds to the woman client's physical appearance, usually aimed at making the client appear more "feminine.” 145 Items for these objectives were also based on the following Guideline for Non-Sexist Therapy (APA, 1978): 13. The psychologist should not have sexual relations with the client, nor treat her as a sex object. and from the APA Ethical Principles of Psychologists: Psychologists do not exploit their professional relationships with clients, students, employees, or research participants, sexually or otherwise. Psychologists do not engage in sexual harrass- ment. Sexual harrassment is defined as deliberate or repeated comments, gestures or physical contacts of a sexual nature that are unwanted by the recipient. (APA, 1981, p. Principle 7.d) The following items were developed for the Sex Object Scale: General Items 1. I had sexual relations with my counselor sometime during the course of my counseling. I had sexual relations with my counselor after our counseling relationship had ended. My counselor often touched me during sessions. My counselor looked at my body in a way which made me feel uncomfortable. My counselor said that if we hadn't met in a counseling setting, s/he would have liked to ask me out. My counselor made sexual advances. My counselor suggested that I would be more attractive if I spent more time on my makeup. My counselor said that men would be more attracted to me if I appeared more feminine. My counselor suggested that I would be happier if I tried to appear more physically feminine. Table 4.1 provides a list of the objectives by theme used in item generation for the Sex Bias in Psychotherapy 146 Table 4. 1 Objectives Used in the Development of the Sex Bias in Psychotherapy Questionnaire Fostering Traditional Sex Roles 1. 5. The therapist's interventions are aimed at teaching his/her female client to adjust/ adapt to the traditional female role. lack of tolerance for deviations from this role is denonstrated. The therapist denonstrates the belief that the mman's role is as a wife and that growth is achieved through marriage or a relationship with a man. More concern is shown for the husband/boyfriend's needs than for the mman client. The therapist expresses the belief that her/his man client is unfulfilled without being a mother. . The therapist lacks sensitivity to his/her men client's work, career, role diversity and the conflicts these nay raise for her. Fostering of traditional sex roles through the maintenance of a power differential in the therapeutic relationship. Bias in Expectations and Devaluation of Warren 1. The therapist denies the adaptive potential of assertiveness, independence, anger and ability to identify and experience emotions, and fosters concepts of mmen as passive, dependent, and "enotional". Sociological/cultural explanations for some of the wman client's difficulties are absent. The therapist either directly or indirectly makes demeaning statements about mien, often with the use of derogatory labels. Devaluation of women as evidenced by the therapist's interest in the wunan client's relationships with men to the exclusion of the therapist's interest in the womn client's relationships with women. The therapist lacks sensitivity and awareness concerning the reality of incest, rape, and physical abuse (non-sexual) for mnen and tends to minimize the inpact of these experiences. The therapist uses theoretical constructs (e.g., masochism) to condone the victimization of mien. Sexist Use of Psychoanalytic Concepts 1. The therapist incorporates the negative Freudian view < c. women into his/her work with clients. 2. The therapist insists on using the concepts of penis envy, Oedipal conflict, trans- ference and resistance, without flexibility or understanding of mnen's developnent. 147 Table 4.1 continued 3. The therapist perpetuates myths about motherhood and marriage as all fulfilling for women, and tends to judge emotional maturity of women clients on their attitudes towards childbearing and/or rearing and intimate relationships with men. 4. A double standard for sexual experiences of men and women is demonstrated by the therapist. Responding to kbmen as Sex Objects 1. Sexual relations between client and therapist. 2. The therapist subtly shows sexual interest in the female client. 3. The therapist responds to the mmn client's physical appearance, usually aimed at making the client appear more "feminine". 148 Questionnaire. A summary of the four sub-scales composing the total scale of the Sex Bias in Psychotherapy Questionnaire is presented in Table 4.2. Tests of the Research Hypotheses Hypotheses Related to the Reliability of the Sex Bias in Psychotherapy Questionnaire Hypothesis I(a): The internal consistency of the total ifems of the Sex Bias in Psychotherapy Questionnaire will be sufficiently high to infer homogeneity of the construct of sex bias in psychotherapy. The internal consistency of the instrument was determined by computing Cronbach's coefficient alpha. The initial total scale alpha coefficient was .87 Five items were observed to have zero variance and were automatically dropped from.the scale. The itemo with no variance were 67a, 69b, 70c, 71d, and 72a. All except item 70 are from the Sex Objects Scale. Item 70 is from the Psychoanalytic Scale. Further analysis revealed that aItem 67 is "I had sexual relations with my counselor after our counseling relationship had ended." bItem 69 is "My counselor said that if we hadn't met in a counseling setting s/he would have liked to ask me out." cItem70 is "My counselor insisted that my lack of sexual attraction for him/her was delaying my progress." dItem 71 is "I had sexual relations with my counselor sometime during the course of my counseling." eItem72 is "My counselor made sexual advances.” 149 £090 be 00008005 8309500 >2 .3028 :0 :05 “H0mbe :0 0.85 300 00 F82 05 830: 83958 >2 .Hmebe you 80 0:0 :55 00 we 008.5800 uoHomcaoo >2 .m... .N.... .H... 3me 830393 new 003 .33 ... us. a e. .838 B 383 0.: E as: 80800:... 8.858 E .9188 >n @0300 09 3003 9.0303 be emu—u 800038 839.58 >2 £09308 no e050... .oC0HeHHun\0HH3 H0 30» be 5 005.3000 0.8... 09 00 30: EmoH on me .SH on 00 008000 9:00:08 be H0 0088 05. 6:033:33? £00298 50:00.: H0 0005 :05 .0050 030.. 022 3000 00503 005 000 0.: 0030: 830.58 >2 .me0uzwsmo 8%....» 0033 .8059. mm c853 09300.08 2033 9.002 80» H 0020 o0um0mwsm 830500 >2 .038 HESS—00800: 5 00.53 93020 9.002 92800 Eouw 0_e p0w0u8003 uoHomcsoo >2 .0... .SH E0em>m 002300. 0 mm 03000000 be >9 @039 3.850 on 00 @8000 000025 00503 055.5.“ 338333... >2 69.3.5 5.53 10033308 be 030500 00 «08.000 809.000 >2 .wcquozooos 000 .5000.” emu .0803 00.: 0033300 5 300.805 be communeomHo 8380 >2 1.83 or. 000.221.3300“. 05 03 30020 . :05 005 08 300 830500 >2 .060: 00 89.53308 05 0o 3:020 :05 0020 me 300 M03058 >2 60803 H 0023 00m 00 $53 e038 @5933 .918 00: 00 commuoooco 003 H 60803 H 0023 00w on $53 uo\o:m @50qu @530 :05 M050 m>m3 503 05 0030: .3ng8 >2 .60an 05 000%.. 09:03 030 be :0 :05 9.050 :0 023 303 .5303 be 2023 8095 05 :0 0.85 ©0083 uoH0m§oo be :030 .0850 £03 0320:0320.“ be 5 0000 x03 0 0300 H 005 o0umowwzm 8305.8 >2 .qunmcoHumHou 5 0.050 00 5 03m 00 H0800: H 005 o0umowwam 830.58 >2 $030.3 05 0300 00 09.0000 :0 5 000$ 00300 L050 05 83008 H 005 o0um0wwnm uoH0m§8 be 602000330.“ 5 0.5303 000%... @0023 H :05 .3203 new 00:33 :30 be 088 00 :05 800: .3050 00 9.00000 00 0:. HOW 088095 0.88 003 3 00:0 o0um0ww=m 8.00.500 >2 6.300500 be .3 oouuoadsm 00.03 300m Hmconmomoua 83:0 Hmcomuoa 090.793 >2 .3020... 8 x23 mewow .moon wcchmLo .3me38?» m 9.300 053 0a: be :H mwcdt 09020 00 009.030 be o0wmuaoomHo 830.500 >2 0.8 >020 mm 0930 0880 00 .8380 be .3 p0w0u0080 003 H .HN .ON .3 .2 ..._.. .3 .3 .3 .3 .2 .NH .3 38m mmas. 553.8; 0355030930 >dmeofiozo>mm 5 03m xom 05 Ho 8.89%... .30.. 05 .0 page < , N .0 033. 150 .00x00. 000 0003000 33000000 000 m0 09200 H 0003 00e 00 00350 003 H 0000 00300000 >2 00300500 00000 000m 0000 e000 0005 00 000000 0000000 000000 00.53 0000 0.00.00 00 000000 00300000 >2 005 0000 0000.00 0006 0003 0053 0000 00:05 03000000 >2 .0053 0000 00e 0000 0005 006 0000 00.53 0000 00:95 03000000 >2 .00e 0000 000000000 0000.: 0033 000 0053 0000 0000 03000000 >2 .0053 0000 03000 000000 0003 00e 0000 00:95 00300000 >2 0... O. Hmmqmso 030m 00 0300000080.. .0053 0003 003000000000 0000 039 00Qe «0 0003 000 0003 003000000000 0000 00:95 00300000 >2 .0053 0000 000.. M030 00m 00 >000 000e 0003 00.: 0000 00:95 03000000 >2 000000000 00000003000000003093 be 00000 0000 00e 0003 00300000030 0005000 be 00000 $000000 00 0000000000 0005 003 03000000 >2 0.03000 000 00 0000 00 000 00: 05 3 00 0.03000 be >330 00 005 0003 000 H 000030330 000 00 09.03 000000000 >2 .0053 0000 009e 006 000.00 00 00.000 00300000 >2 .0300 0053 00.. 300000000 00 3000000 000000 00000008 >2 €583 0 9.: 00:... 0.000 9.: 00:00.8 000.. 030058 .2 .0053 0 000 :03 0000 000 H 0000 0000 03000000 >2 .0053 00000 00 00000 00 003.00 0003 0003000 be 00 0.50 0000 05 030 03000000 >2 .00.: 00 0000000 00 000.03 0000 0033 500000 0 00 0053 0 w0000 500 00.00 0.03000 be 3 00.00 0000 000 0000000 00300000 >2 .>000000 000 00 00.53 0 00 00 w050w 0003 00 00 M03000 000 0.03000 be 0000 000m 0e 000.: 00300000 >2 .00w00000 0 00 0000090 be 000090 00 >0330 be 300 000000000 >2 .000000 000.3 000950 00 00 000 >000 000 003 00 0000 000m 0.: 0005 000000000 >2 .00w00000 0 00 0003000 be 0000000000 000 3000000 00 >0330 be 300 02000000 >2 03000000 be >0 93300000 00 0000 003 5300.00 000 00300 05.0 000 000000 >2 .0000 H 0000 0000 000 00000098 H 0003 0003030 .8300 300000000 w0000 0cm 0: 0000000000 000000000 >2 .>0w00 >2000 003 H .0030 H 0003 0050050 0000 0000000000 00 00.000 000000000 >2 >00ng 00 0090 be 300 03000000 be 0030 H .00w00 be 000000 00 030 003 03000000 >2 .00w00 be 0003 0000 00031300 00 0000 .0000 be 0003 030003000 0008 0000 00300000 be 0000 0.: 00 00.000 0H .0w00\00w00 be 000003 000 .000005 w000000000000 00 00 03000000300800 be 000.. 00000 03000000 >2 .300be 000 000 000 .000000 00“ >0wq0 00 00 0e 00w >000 003 00 0000 00:95 03000000 >2 .0w00H00u >0w00 be 00009.0 00 0e 00w0000000 03000000 >2 .0N .mm .0m .3 .NN .3 .ON .2 .3 a. .5 a. .00 .2 .0 .3 .mH 0. .NH .HH .2 ..... o .m .0. K .0 .m .0 .0. 008.080 0... 030.. 151 8.8: 3mm Hmuamw 05 >30 33300 b.8550 35. .302 $530.... 3033.. 9:23.. 0:0 3033 335038 33 0.333: :m 20H: 30H «. 35:3“ 3303?: 33... 330% 00 330 H NH .3302 3 3:03 H 35 303mg... 330500 >2 35:33 33... 33300.. H 3 8. 00 3009.00... 33:. mp 359.. :8. 35 3mm 33.3500 >2 .983: >E :0 8.3 33:. 38.... H HH 8,309.03 33... up .0303 H 320 303%.... 338500 >2 53:93 338m 308. 339500 >2 .50 8. xmm 00 3x: 33: 3:03 023 .9303 9338500 m :H 08. 0.50: 33 HH 08... Sam 338500 >2 .3303395 33 8. 308. :02: >3 m 5 >000 >5 3 30.03 3330 >2 6:03.30 $.32. 8. 320500 :33 330500 >2 .303 3: 0230.33: 9330.58 50 303 330500 >... 50.3 3333.. 30x3 um: H $538500 >6 3 3500 3:0 @530 8.38.8 330500 >... 5.3 30.33.. 338m 3: H 38m .830 x8. .:8.03 3. 0m 3: 0:0 £33333: 38:3 :35 3 30350 3m 96: 00 :8. 3w 3.53: 2.3 0H 35 3333 339500 >2 .2 £8.03 3m 0m 3: 30 .0233330 >333 :320 MO 30330 .80 3.6: 00 0:3 00 :8. 3w 353: mm... 0H 320. 30a 00 3.53.... 330.500 >2 .9 65300300 30:3qu.8 35.80 >... 3:00 3330 >2 .3 .3333 08. H 33:: 0:30:00 >23 3 3: 3503 H 35 03m 338500 >2 .3 an: m 5.3 .2235... a 5.. mm. H .35 .82 B .8 283 . .2. 8.35 .2358 .2 .N. .3303 33.. 3;... H0 30:33.6 :m mm... 338333033330 >E 0:00... @330 >... 320 029020 330.500 >2 .3 .mmmuwouq >E w:H>m3_u mm: 3582 3w 330303 33.3 .3 x03 >E 320 30035 338500 >2 .3 £8.03 0:... :8. 33330 53963:. 33.. 500.. 03w H 339.30.... 30 33395 00 08.33 338500 >2 .0 u... .330 353m 38. m 3m 3330 33.5 H0 x02 >... 50.3 08:53 3H3ua >E 320 30035 3330 >2 .w .0330 338m :350 3: .333 H 2033 333 83 :H 3300 9:8. 3... 0H 35 0:85.395 3: 0300 338500 >2 .H. 335300 N... 333. 152 A coefficient alpha of .88 is a high coefficient, especially for a self-report instrument. An inference can be made that the SBPQ is measuring a homogenous construct, that of sex bias in psychotherapy. The SBPQ is an internally consistent measure of sex bias in psychotherapy. Hypothesis I(b): The internal consistency of each of the four sub-scales of the Sex Bias in Psycho- therapy Questionnaire will be sufficiently high to infer that each sub-scale is measuring a separate dimension of sex bias in psychotherapy. An alpha coefficient was computed for each sub-scale of the SBPQ. The alpha coefficients for the sub-scales were lower than that of the total scale. Alpha coefficients computed prior to the dropping of items 1, 20, and 42 were as follows: .66 for the Traditional Roles Scale, .79 for the Bias and Devaluation Scale, .61 for the Psychoanalytic Scale, and .69 for the Sex Objects Scale. The sequential removal of items 1, 20, and 42 improved the reliability of each sub-scale in addition to that of the total scale. Further inspection of each sub-scale revealed that Item 50 was adversely affecting the reliability of the Psycho; analytic Scale, but_at the same time was contributing to the reliability of the total scale. A decision was made to drop item 50 from the sub-scale but retain it for the total scale. 153 To summarize, Item 1 was dropped from the Total and Traditional Roles Scales, Item 20 was dropped from the Total and Sex Objects Scales, Item 42 was dropped from the Total and Bias and Devaluation Scales, and Item 50 was dropped from the Psychoanalytic Scale but retained for the Total Scale. Alpha coefficients for the revised sub- scales are .68 for the Traditional Roles Scale, .80 for the Bias and Devaluation Scale, .73 for the Psychoanalytic Scale, and .79 for the Sex Objects Scale. All coefficients with the exception of that for the Traditional Roles Scale are high enough to infer homogeneity of sub-scale items; thus the measurement of a single dimension of sex bias in psychotherapy, While an alpha of .68 on the Traditional Roles Scale is sufficient for research purposes, predictions from that scale alone should not be made (Mehrens and Lehmann, 1978). All further analyses on the SBPQ were performed with Items 1, 20, and 42 removed from the total and respective sub-scales, and with Item 50 removed from the sub-scale only. The revised form of the SBPQ has 64 items and a total scale coefficient alpha of .88. The results are presented in Table 4.3. Table 4.3 Internal Consistency of the Sex Bias in Psychotherapy Questionnaire for the Total Scale and Four Sub-Scales: Cronbach's Coefficient Alpha (gél40) Scale Reliability Total Scale (67a items) Total Scale (67 items) Traditional Roles Scale Bias and Devaluation Scale Psychoanalytic Scale Sex Object Scale .87 .88 .66 .79 .80 .61 .73 .69 .79 with Items lb, 20C, and 42d removed with Item I removed ‘with Item 42 removed with Item 50e removed with Item.20 removed ‘aFive items with no variance were automatically dropped from the instrument. bItem 1 is "I was encouraged by my counselor to accept things as they are." cItem 20 is "My counselor often touched me during sessions." dItem 42 is ”My counselor pointed out that some of my problems came from being a woman in a society which sees women as unequal to men." eItem 50 is "My counselor seemed to understand the frustration I felt about real inequality between men and women." l55 Hypothesis II: The correlations between the Total Scale and the four sub-scales of the Sex Bias in Psychotherapy Questionnaire will be sufficiently high to infer that the SBPQ is measuring a unidimensional construct. The correlations between the SBPQ Total Scale and four sub-scales were determined by computing a Pearson Product Moment Correlation Matrix. The inter-scale corre- lations ranged from .23 between the Sex Object and Bias and Devaluation Scales to .92 between the Bias and Devaluation and Total Scales. The Sex Object Scale is the 33,3 '43” Nonetheless, all interscale correlations were significant Ifl‘i i ('1’ Y I I - 60 at a .05 alpha level, suggesting that the SBPQ is measur- a scale with the lowest inter-scale correlations. ing a unidimensional construct. In light of the fact that the Total Scale has good reliability and is measuring a unidimensional construct, all further analyses were perforned on the Total Scale, rather than on individual sub-scales. The results are presented in Table 4.4. Hypotheses Related to the Validity of the Sex Bias in Psychotherapy Questionnaire Hypothesis III: The content of the items on the Sex Bias in Psychotherapy Questionnaire will be indicative of sex bias in psychotherapy. This hypothesis relates to the content validity of the SBPQ. No formal statistical test of the hypothesis was used to infer content validity. Rather, the three methods outlined by Mehrens and Lehmann (1976) and Anastasi (1976) were used to gain content validity. It can be inferred 156 Table 4.4 Correlation Matrix of the Total Scale and Four Sub-Scales of the Sex Bias in Psychotherapy Questionnaire Traditional Bias & Sex Scale Total Roles Devaluation Psychoanalytic Object Total ** .85 .92 .70 .38 Traditional Roles .85 ** .65 .54 .30 Bias & Devaluation .92 .65 ** .52 .23 Psycho- analytic .70 .54 .52 ** .41 Sex Object .38 .30 .23 .41 ** Note: All correlations are significant at the pg<2p01 level. that the SBPQ has sufficient content validity in that (l)the content domain was clearly defined, (2) specific objectives were developed for the areas of content to be sampled, and (3) very high inter-rater-reliability of .83 to 1.00 was obtained on item content from a group of six expert judges in the field of women's issues in counseling and psychotherapy. 157 Hypothesis IV: A linear correlation will be observed between therapist-subjects' scores on the Stereotype Scale of the Personal Attributes Questionnaire and client-subjects' scores on the Sex Bias in Psycho- therapy Questionnaire. 59: P=0 El: PfO Therapist-subjects averaged a mean of 136.35 with a standard deviation of 13.60 on the Stereotype Scale. The range of scores was from 110 to 175, the possible range being 110-220. Therapist-subjects' scores on the Stereotype Scale were compared for the three therapist variables: sex, experience level, and agency affiliation. Three two-way ANOVA's were computed on the therapist variables. Therapist sex was the only therapist variable for which significant main effects were found. WOmen therapists had significantly higher scores than male therapists on the Stereotype Scale of the FAQ. WOmen therapists averaged a mean score of 143 and male therapists a score of 133 on the measure. No main effects were found for the experience level or agency affiliar tion of therapist-subjects nor were any two-way interactions observed. The results of the two-way ANOVA's on therapist variables are found in tables 4.5, 4.6, and 4.7. Of the 49 therapists who completed the Stereotype Scale measure, 38 had clients who responded to the Sex Bias in Psychotherapy Questionnaire. In order to test Hypothesis IV, that a relationship exists between a thera- pist's score on the Stereotype Scale and his/her client's SBPQ score, a Pearson product-moment correlation was 158 Table 4.5 Two-Way Analysis of Variance of the Stereotype Scale for Therapist Sex X Experience Level Source df SS MS F Therapist Sex 1 1118.45 1118.45 6.198* Therapist Experience 1 133.36 133.56 .739 Sex X Experience 1 115.612 115.61 .641 Within-groups 45 8120.481 180.455 Total 48 9430.77 196.47 7" p <.02 Table 4.6 Two-Way Analysis of Variance of the Stereotype Scale for Therapist Experience Level X Agency Affiliation Source df SS MS F Therapist Experience 1 76.275 76.275 .368 Therapist Agency 1 .000 .000 .000 Experience X Agency 1 24.698 24.698 .119 Within-groups 45 9329.802 207.329 Total 48 9430.776 196.474 159 Table 4.7 Two-Way Analysis of Variance of the Stereotype Scale for Therapist Sex X Agency Affiliation Source df SS MS F Therapist Sex 1 1062.878 1062.878 5.716* Therapist Agency 1 1.555 1.555 .008 Sex X Agency 1 .063 .063 .000 Within-groups 45 8367.833 185.952 Total 48 9430.776 196.474 *PSO3 computed between the therapist score on the Stereotype Scale and the aggregate mean client—subject score on the SBPQ total scale and sub-scales for each of the 38 therapists. The correlation between the Stereotype Scale scores and the SBPQ aggregate mean scores for the Total Scale was —.13, for the Traditional Roles Scale was -.10, for the Bias and Devaluation Scale was -.l8, for the Psychoanalytic Scale was .08, and for the Sex Object Scale was .03. None of these correlations are significant at the .05 alpha level. No relationship was observed between a therapist's tendency to stereotype and her/his client's experience of sex bias 160 in psychotherapy. The null hypothesis is retained. The results of the correlation are presented in Table 4.8. Summary The Sex Bias in Psychotherapy Questionnaire is based on the themes of bias outlined by the APA Task Force on sex bias and sex-role stereotyping in psychotherapy (1975). These themes include: (1) fostering traditional sex roles, (2) bias and devaluation of women, (3) sexist use of psycho- analytic concepts, and (4) responding to women as sex objects. The procedures used to develop the SBPQ involved generation of items around specific objectives, expert judg- ing of items for content and informal piloting of the instrument prior to administration. Four sub-scales comprise the SBPQ total scale. These sub-scales are the Traditional Roles Scale, the Bias and Devaluation Scale, the Psychoanalytic Scale, and the Sex Object Scale. Four hypotheses related to the reliability and initial validity of the SBPQ were tested. Hypothesis I(a) was con- cerned with the SBPQ total scale reliability. An alpha coefficient of .88 was obtained as a measure of internal consistency for the SBPQ total scale. Hypothesis I(b) tested the reliabilities of the four sub—scales. Alpha coefficients of .68, .80, .73 and .79 were obtained for the Traditional Roles, Bias and Devaluation, Psychoanalytic and Sex Object Scales, respectively. The SBPQ has demonstrated 161 Table 4.8 Correlation Matrix of the Stereotype Scale and the Sex Bias in Psychotherapy Scale (g738) Traditional Bias and Psycho- .3 Total Roles Devaluation analytic Sex Object 8 a: crcu 35 a -.13 -.10 -.18 .08 .03 U o m H m U U) Note: No correlation was found to be significant at the .05 alpha level. reliability and homogeneity of the construct of sex bias in psychotherapy can be inferred. Hypothesis II tested whether or not the SBPQ was measuring a unidimensional construct. A product-moment correlation between the SBPQ total and four sub-scales demonstrated significant levels of correlation among the scales. These findings suggest that the SBPQ is measuring a unidimensional construct. Hypotheses III and IV were concerned with the content and construct validity of the SBPQ. The SBPQ is assumed to have content validity because (1) the content domain was clearly specified, (2) specific objectives were developed for the areas of content to be samples, and (3) very high inter-rater reliability among a group of expert judges rating items for content was obtained. 162 Hypothesis IV begins the testing of the SBPQ's construct validity. The correlation between a client's SBPQ score and her therapist's Stereotype Scale score was not significant. No relationship was observed between a therapist's tendency to stereotype and her/his client's experience of sex bias in psychotherapy. Judgments about the construct validity of the SBPQ are reserved until the results of the phenomena of sex bias in psychotherapy hypotheses are presented in Chapter V. 163 CHAPTER V RESULTS: PART II The remaining results of the study are presented in this chapter. Chapter IV contained the results related to instrumentation. This chapter contains the results specifi- cally related to the phenomena of sex bias in psychotherapy. A supplementary descriptive analysis of the data obtained from specific sub-groups of women is presented, along with a brief qualitative analysis of item endorsement. Anecdotal accounts of sex bias in therapy are also presented in this chapter. Tests of the Research Hypotheses Hypotheses Related to the Phenomena of Sex Bias in Psychotherapy Hypothesis V(a): The total scale scores on the Sex Bias in PSychotherapy Questionnaire will be sufficiently high to infer that women clients experience sex bias in psychotherapy. H2:ui96 (total) flzu >96 (total) The total scale SBPQ individual scores ranged from 0 to 78. The overall mean for the individual total scale scores was 21.90 with a standard deviation of 13.69. The median was 164 20, the distribution was bimodal with modes occurring at 19 and 21 and positively skewed. Minor modes were observed at scores of 4 and 24. The measures of central tendency for the aggregate mean scores were slightly different from those for the individual scores. The aggregate means for the total scale SBPQ scores ranged from 2 to 62. The overall mean was 20.82 with a standard deviation of 11.44. The median was 20, the distri- bution was bimodal with modes at 13 and 15 and positively skewed. In order to test Hypothesis V(a) a one-sample p-test statistic was used. The aggregate mean SBPQ total scale score for women clients was tested against a theoretical population mean of 96. The p-value computed was -51.31. This value is not significant at a .05 alpha level. Women clients were not observed to experience sex bias in psycho- therapy as measured by the SBPQ total scale score. The null hypothesis is retained. Hypothesis V(b): The sub-scale scores on the Sex Bias in Psychotherapy Questionnaire will be sufficiently high to infer that women clients experience sex bias in psychotherapy. 39‘ “(Traditional Roles) fi 30 :13 _17 u(Traditional Roles) > 30 I]: O : u(Bias and Devaluation) : 37.5 :1: N __= u(Bias and Devaluation);> 37.5 165 39‘ u(Psychoanalytic) :322.5 H : u . _3 (Psychoanalytic) > 22.5 Eg‘ (Sex Object) <_ 6 H4: (Sex Object) > 6 Separate one sample p-tests were performed for each of the four SBPQ sub-scales. In each case the mean aggregate mean sub-scale score was tested against a theoretical popu- lation mean. For the Traditional Roles Scale this value was 30, for the Bias and Devaluation Scale it was 37.5, for the Psychoanalytic Scale it was 22.5, and for the Sex Object Scale it was 6. The mean aggregate mean for the Traditional Roles Scale was 6.37 with a standard deviation of 3.72 and a range of 0 to 20. The p-value for the Traditional Roles Scale was -51.44. This value is not significant at a .05 alpha level. WOmen clients were not observed to experience sex bias in psychotherapy as measured by the Traditional Roles Scale of the SBPQ. The null hypothesis is retained. The mean aggregate mean score for the Bias and Devalua- tion Scale was 11.02 with a standard deviation of 6.01 and a range of 0 to 28. The p—value for the Bias and Devalua- tion Scale was -35.79. This value is not significant at a..05 alpha level. WOmen Clients were not observed to experience sex bias in psychotherapy as measured by theBias and Devalua- tion Scale of the SBPQ. The null hypothesis is retained. 166 The mean aggregate mean for the Psychoanalytic Scale was 1.81 with a standard deviation of 2.30 and a range of 0 to 13. The p—value for the Psychoanalytic Scale was -73.89. This value is not significant at an alpha level of .05. Women clients were not observed to experience sex bias in psychotherapy as measured by the Psychoanalytic Scale of the SBPQ. The null hypothesis is retained. The mean aggregate mean for the Sex Object Scale was .32 with a standard deviation of .69 and a range of 0 to 3. The p-value for the Sex Object Scale was ~66.82. This value is not significant at an alpha level of .05. Women clients were not observed to experience sex bias in psychotherapy as measured by the Sex Object Scale of the SBPQ. The null hypothesis is retained. The p-test results for Hypothesis V (a and b) are presented in Table 5.1. In addition to statistical tests of Hypothesis V (a and b), the response frequencies to the SBPQ general items were observed. Observation of these frequencies aids in the interpretation of an SBPQ total scale mean of 20.82. Although the statistical test for Hypothesis V(a) proved to be insignificant, the frequency distribution for item responses suggests that on slightly more than a quarter of the items at least some women were acknowledging sex bias in their therapy. Approximately a quarter of the SBPQ items, then, discriminated between those women clients who experienced sex bias and those who did not. This finding 167 Table 5.1 One Sample t-test Results for the SBPQ Total and Four Sub-Scales Standard Scale N Mean Deviation p-value Total 66 20.82 11.44 -53.31 NS. Traditional Roles 66 6.38 3.72 -51.44 NS. Bias & Devaluation 66 11.02 6.0 -35.79 NS. Psychoanalytic 66 1.81 2.30 -73.89 NS. Sex Object 66 .32 .69 -66.82 NS. suggests that although women in general did not experience a statistically significant amount of sex bias in therapy, some women did report sex-biased experiences in psycho- therapy. Item means and responses frequencies for the SBPQ general items are presented in Table 5.2. Hypothesis VI(a): The total scale scores for the Sex Bias in Psychotherapy Questionnaire will be different for women in different life status categories. Hypothesis VI(b): The total scale scores for the Sex Bias in Psychotherapy Questionnaire will be different for women clients having different therapy experiences. For the purposes of hypothesis testing, Hypothesis VI (a and b) was broken down into sex separate hypotheses reflecting the six independent variables to be tested. These independent variables are client age, education, rela- tionship status, child status, number of therapy sessions, and length of time since termination of therapy. 168 Table 5.2 Item Means and Response Frequencies for the SBPQ General Items Prior to Instrument Revision a ._ Definitely Mostly Mostly Definitely Item E, (N) False False True True 1 1.46 (191) 25.1(%) 22.5(%) 34.0(%) 18.3(%) **2 .56 (191) 3.6 6.3 32.3 57.8 **3 1.44 (191) 25.0 18.2 32.8 24.0 4 .72 (187) 51.3 30.5 13.4 4.8 **5 .68 (189) 6.3 7.4 33.9 52.4 **6 .87 (191) 11.0 10.5 33.0 45.5 **7 .76 (191) 8.3 7.8 34.9 49.0 8 .09 (191) 91.6 7.3 1.0 -- 9 .18 (191) 83.9 14.6 1.6 -- 10 .16 (191) 87.5 10.4 1.0 1.0 11 .15 (191) 87.5 10.9 1.0 .5 12 .17 (188) 86.7 10.6 2.1 .5 13 .08 (191) 91.6 8.4 -- -- 14 .39 (191) 69.8 22.9 5.7 1.6 15 .22 (191) 84.4 10.9 2.6 2.1 16 .07 (191) 94.3 4.7 .5 .5 17 .44 (191) 68.6 22.5 5.2 3.7 18 .07 (191) 94.8 4.2 .5 -- **19 .41 (191) 4.7 2.6 21.9 70.8 *20 .15 (191) 88.5 8.3 2.6 .5 21 .19 (191) 83.8 13.6 2.1 .5 22 .33 (189) 79.4 12.7 3.7 4.2 23 .06 (191) 95.8 2.6 1 O .5 24 .01 (191) 99.0 1.0 -- -- 25 .12 (190) 91.6 5.8 2.1 .5 26 .32 (189) 76.2 17.5 4.8 1.6 27 .12 (189) 91.5 6.3 .5 1.6 28 .42 (191) 68.6 22.5 6.8 2.1 29 .15 (191) 88.5 8.9 2.1 .5 3O .10 (188) 92.0 6.4 1.1 .5 31 .79 (190) 58.9 16.3 13.7 11.1 32 .66 (189) 65.1 12.2 14.8 7.9 33 .10 (190) 94.2 3.7 .5 1.6 34 .04 (191) 97.4 1.6 1.0 ~- 35 .21 (190) 85.3 10.5 2.1 2.1 36 .58 (190) 60.5 25.3 9.5 4. 37 .07 (190) 93.7 5.8 .5 -- aFor item content refer to Appendix B t ** Positively worded items Item removed in instrument revision 169 Table 5.2-~Item Means and Response Frequencies for the SBPQ General Items Prior to Instrument Revision (continued) a __ Definitely Mostly Mostly Definitely Item §_ (N) False False True True 38 .21 (191) 84.3 11.5 3.1 1.0 **39 .67 (191) 6.3 6.8 33.9 52.9 40 .08 (191) 94.2 4.2 .5 1.0 *41 .22 (191) 81.3 16.1 1.6 1.0 **42 2.21 (191) 51.0 24.5 19.3 5.2 43 .09 (191) 93.2 4.7 1.6 .5 44 .16 (191) 87.0 10.4 2.6 -- 45 .09 (191) 91.7 7.8 -- .5 46 .17 (191) 88.0 7.3 4.2 .5 47 .14 (191) 88.5 9.9 .5 1.0 48 .27 (191) 79.6 15.7 3.1 1.6 **49 .75 (190) 7.8 6.3 38.0 46.9 **50 1.41 (184) 29.3 8.2 37.0 25.5 51 .48 (181) 64.6 25.4 7.7 2.2 52 .14 (191) 89.6 7.8 2.1 .5 **53 .71 (191) 9.9 4.7 31.9 53.4 54 .23 (190) 81.6 15.3 2.1 1.1 55 .08 (191) 94.3 4.2 .5 1.0 56 .15 (191) 89.1 7.8 2.1 1.0 57 .10 (191) 93.8 4.2 .5 1.6 **58 .67 (190) 9.5 3.2 32.6 54.7 59 .45 (191) 70.8 17.2 8.3 3.6 **60 .90 (191) 12.5 7.8 36.5 43.2 61 .17 (190) 85.3 12.6 1.6 .5 62 .29 (190) 77.4 17.9 3.2 1.6 63 .13 (189) 89.4 8.5 1.6 .5 64 .13 (188) 91.0 6.4 1.1 1.6 **65 1.68 (186) 39.2 11.3 28.0 21.5 66 .11 (185) 91.4 7.0 1.1 .5 *67 O (184) 100.0 -- -- -- 68 .08 (184) 92.9 7.1 -- ~- *69 O (184) 100.0 -- -- -- *70 0 (184) 100.0 -- -- ~- *71 O (184) 100.0 -- -- ~- *72 O (184) 100.0 -- -- -- 170 Hypothesis VI(a): Women clients in different age groupings will obtain different SBPQ Total Scale scores. H0: ul - u2 - u3 - u4 - u5 - u6 = 0 H1: 111 u5 - u6 # 0 This hypothesis tests whether or not the client's age has an effect on her experience of sex bias in psycho- therapy. A one-way analysis of Variance was computed on the mean SBPQ score obtained for each age grouping. No significant differences were found among the six age group- ings. Age of the client was found to have no effect on the experience of sex bias in psychotherapy. The null hypothesis is retained. Hypothesis VI(a); Women clients with differing amounts of formal education will obtain different SBPQ Total Scale Scores. H0: u1 - u2 - u3 - u4 - u5 = 0 H2: ul - u2 - u3 - u4 - uS # 0 This hypothesis tests for the effect of formal educa- tion on women clients' experience of sex bias in psycho- therapy. A one-way Analysis of Variance was computed on the mean SBPQ score obtained within each educational group- ing. A significant difference was found among groups at the .05 alpha level. The null hypothesis rejected in favor of the alternative hypothesis. 171 A post-hoc comparison using the Scheffee method of multiple comparisons with a .05 alpha level determined that the difference was between those women clients who had not completed high school and those who had some graduate educa- tion, or a graduate degree. Women clients who had not completed high school ( i = 32.9) experien-ed significantly more sex bias in their psychotherapy than women clients with a graduate education or graduate degree (i = 19.5). No differences were observed among the other groups. Hypothesis VI(a)3: Women clients in different rela- tionship status categories will obtain different SBPQ Total Scale Scores. 110—: ul H3: “3’“47‘0 This hypothesis tests the effect of a client's rela- tionship status on the experience of sex bias in psycho- therapy. Four categories were compared: (1) single, (2) married/living with someone, (3) divorced/separated, (4) other, using a one-way analysis of variance. No signi- ficant differences were found among the groups. Relationship status was found to have no effect on women clients' experience of sex bias in psychotherapy. The null hypothesis is retained. 172 Hypothesis VI(a); Women clients with children will obtain different'SBPQ Total Scale Scores than women clients without children. :13 O C f—l :1: 4:ul-uziéO This hypothesis tests whether women clients with children experience different amounts of sex bias in psycho- therapy than women clients without children. The mean SBPQ scores for the two groups were compared using a one-way analysis of variance. A significant difference between the two groups was observed at the .05 alpha level. WOmen clients with children (K = 25.0) experienced significantly more sex bias in their therapy than women clients without children (X'= 20.2). The null hypothesis is rejected in favor of the alternative hypothesis. A twoeway analysis of variance computed for the variables of child status and therapist agency affiliation revealed a significant two-way interaction. WOmen with children who were seen at the Community Mental Health Center Y = 27.0) experienced more sex bias in psychotherapy than women with children seen at the Counseling Center (R's 14.8). 173 Hypothesis VI(th: Women clients having had a different number of psychotherapy sessions will obtain different SBPQ Total Scale Scores. HO: ul - uz - u3 - u4 - u5 - U6 - u7 - us = 0 H2: ul - u2 - u3 - u4 - u5 - u6 - u7 - U8 # 0 This hypothesis tests the effect of number of therapy sessions on the experience of sex bias in psychotherapy. A one-way analysis of variance was computed on the mean SBPQ score obtained for each number of sessions groupings. A significant difference in the experience of sex bias in psychotherapy was observed at the .05 alpha level. The null hypothesis is rejected in favor of the alternative hypothesis. Post-hoc comparison testing using the Scheffee method of multiple comparisons at the .05 level of significance indicated that the difference occurred between those women who had the least number of sessions and those who had the most. Women clients having 1-3 therapy sessions (i'- 27.7) experienced more sex bias in their psychotherapy than women clients having more than 45 therapy sessions (R's 13.4) No differences were observed among the other groups. Hypothesis VI(b) : Women clients at different points in time ffbm te ination of psychotherapy will obtain different SBPQ Total Scale Scores. H0: ul u5 U6 H6: ul‘uz-u3-u4'u5'u6'u7#o 174 This hypothesis tests the effect of the length of time from termination of psychotherapy on the experience of sex bias in psychotherapy. A one-way analysis of variance was computed on the mean SBPQ score for the seven groupings of time since termination. No significant differences were found among the groups. Length of time since termination of therapy was found to have no effect on the experience of sex bias in psychotherapy. The null hypothesis is retained. The results of the above one-way ANOVA's and the one two- way ANOVA for Hypothesis VI (a and b) are found in Tables 5.3 - 5.9. Table 5.10 contains a summary of the group means for the above one-and two-way ANOVA's. Hypothesis VII: women clients paired with female therapists will obtain different scores on the Sex Bias in Psychotherapy Questionnaire than women clients paired with male therapists. H0: u (female) = u (male H : u (female) # u (male) 175 Table 5.3 One-Way Analysis of Variance of the SBPQ for Client Age Source df SS MS F Between-groups 5 1275.25 255.05 1.371 (NS) Within-groups 185 34354.66 185.71 Total 190 35629.91 Table 5.4 One-Way Analysis of Variance of the SBPQ for Client Education Source df SS MS F Between-groups 4 1826.17 456.54 2.512 * Within-groups 186 33803.74 181.75 Total 190 35629.91 176 Table 5.5 One-Way Analysis of Variance of the SBPQ for Client Relationship Status Source df SS MS F Between-groups 3 303.54 101.18 .536 (NS) Within-groups 187 35326.36 188.91 Total 190 35629.91 Table 5.6 One-Way Analysis of Variance of the SBPQ for the Client Child Status Source df 88 MS F Between-groups 1 1024.73 1024.73 5.59 * Within-groups 187 34267.08 183.25 Total 188 35291.81 *p '<.02 177 Table 5.7 Two-way Analysis of Variance of the SBPQ for Client Child Status X Therapist Agency Affiliation Source df 'SS MS F Children 1 159.32 159.32 .869 Agency 1 328.40 328.40 1.791 Children X Agency 1 943.50 943.50 5.145* Within-groups 179 32698.01 183.40 Total 181 34129.22 p < .05 Table 5.8 One-way Analysis of Variance of the SBPQ for the Number of Therapy Sessions Source df SS MS F Between-groups 7 3229.18 461.31 2.60* Within-groups 183 32400.73 177.05 Total 190 35629.91 178 Table 5.9 One-Way Analysis of Variance of the SBPQ for the Length of Time Since Termination of Therapy Source df SS MS F Between-groups 6 1258.51 209.75 1.123 (NS) Within—groups 182 34000.06 186.81 Total 188 35258.57 Hypothesis VIII: Women clients paired with experienced- level therapists will obtain different scores on the Sex Bias in Psychotherapy Questionnaire than women clients paired with experienced-level therapists. H0: u (experienced) = u (inexperienced) H1: u (experienced) # u (inexperienced) Hypothesis IX: Women clients paired with community mental healEh center therapists will obtain different scores on the Sex Bias in Psychotherapy Questionnaire than women clients paired with counseling center therapists. H_o‘ ‘1 (CMHC) = u(CC) El: ‘1 (CMHC) 7‘ “(cm 179 Table 5.10 Group Means for the SBPQ for Hypothesis VI (a and b) Variable Mean Agg (F, 1.371 NS) 1) l8 - 22 years 19. 9 2) 23 - 30 years 24.3 3) 31 - 40 years 19.1 4) 41 - 50 years 22.1 5) 51 - 60 years 27.6 6) 61+ years 29.5 Education (F, 2.512 p .05)a 1) Did not finish high school 32.9 2) Completed high school/GED 23.7 3) Some college 21.2 4) Completed college 20.6 5) Some graduate education/graduate degree 19.5 Relationship Status (F, .536 NS) 1) Single 20.9 2) Married/living with someone 22.2 3) Divorced/separated 23.7 4) Other 25.7 Child Status (F, 5.59 p .02) 1) Children 25.01 2) No children 20.21 Child Status & Agenpy (F, 5.145 p .05) l) CMHC & children 27.01 2) CMHC & no children 21.01 3) CC & children 14.81 4) CC & no children 24.5 aPost-hoe comparisons revealed that the significant difference was between groups 1 and 5. 180 Table 5.10--Group Means for the SBPQ for Hypothesis VI (a and b), continued Variable Mean Number of Sessions (F, 2.60 p .02)b l) 1 - 3 sessions 27.7 2) 4 - 6 sessions 23.1 3) 7 - 12 sessions 19.6 4) 13 - 20 sessions 20.6 5) 21 - 30 sessions 16.2 6) 31 - 45 sessions 15.1 7) 46 - 60 sessions 14.0‘ 8) 61 + sessions 13.4 Length of Time Since Termination of Therapy (F, 1.123 NS) 1) Before December 1981 26.3 2) January - February 1982 22.8 3) March - April 1982 21.1 4) May - June 1982 17.3 5) July - August 1982 23.1 6) September - October 1982 23.9 7) November - December 1982 20.2 b Post-hoc comparisons revealed that the significant difference was between groups 1 and 8. 181 Hypotheses VII, VIII, and IX were tested by computing three two-way Analyses of Variance. This type of analysis allowed for the investigation of main effects for each of the three therapist variables in addition to testing for two- way interactions between the variables. Aggregate means were used in the analysis to assure that the assumption of independence was met. Additionally, all two-way ANOVA's were performed on balanced designs. A two-way ANOVA to test the effect(s) of therapist sex and therapist experience level on the Total SBPQ scores was performed. Neither variable showed a significant main effect nor was there a significant interaction between the two variables. A second two-way ANOVA on Total SBPQ Scores to test the effects of therapist experience level and agency affiliation was performed. Neither variable showed a signi- ficant main effect nor was there a significant interaction between the two variables. The third two-way ANOVA performed tested the effect(s) of therapist sex and agency affiliation on Total SBPQ Scores. Neither variable showed a significant main effect nor was there a significant interaction between the two variables. The results indicate that for Hypotheses VII, VIII, and IX the null hypothesis is to be retained. No differences were found between groups on the three therapist variables of sex, experience level and agency affiliation, suggesting 182 that the therapist variables have no effect on a woman client's experience of sex bias in psychotherapy. The results of the three two-way ANOVA's are found in Tables 5.11, 5.12, and 5.13. Table 5.14 contains a summary of the aggregate group means for the above two-way ANOVA's. Table 5.11 Two-Way Analysis of Variance on SBPQ Scores for Therapist Sex X Experience Level Source df SS MS F Sex 1 70.55 70.55 .525 (NS) Experience 1 71.26 71.26 .531 (NS) Sex X Experience 1 14.24 14.24 .106 (NS) Within-groups 62 8324.60 134.27 Total_ 65 8505.93 130.86 183 Table 5.12 Two-Way Analysis of Variance on SBPQ Scores for Therapist Experience Level Agency Affiliation Source df SS MS F Experience 1 88.80 88 80 .673 (NS) Agency 1 106.32 106.32 .806 (NS) Experience X Agency 1 124.97 124 97 .947 (NS) Within-groups 62 8178.11 131.91 Total 65 8505.93 130.86 Table 5.13 Two-Way Analysis of Variance on SBPQ Scores for Therapist Sex Agency Affiliation Source df SS MS F Sex 1 146.061 146.061 1.118 (NS) Agency 1 164.291 146.291 1.258 (NS) Sex X Agency 1 147.506 147.50 1.129 (NS) Within-groups 62 8098.31 130.62 Total 65 8505.931 184 Table 5.14 Group Means for the Aggregate Mean SBPQ Scores for Hypotheses VII, VIII, and IX Therapist Variables Mean Sex & Experience (F, .106 NS) Male Experienced Therapists 21.8 Male Inexperienced Therapists 23.5 Female Experienced Therapists 21.0 Female Inexperienced Therapists 17.5 Experience & Agency (F, .947 NS) Experienced CC Therapists ' 20.3 Experienced CMHC Therapists 22.2 Inexperienced CC Therapists 16.4 Inexperienced CMHC Therapists 24.6 Sex & Agency (F, 1.129 NS) Male CC Therapists 21.0 Male CMHC Therapists 24.4 Female CC Therapists 15.7 Female CMHC Therapists 22.3 Supplementary Descriptive Analysis Phenomena of Sex Bias in Psychotherapy as Described by Specific sub-Groups of Women The potential sex-biased treatment afforded to women sub-groups was of interest in the present study. The sub- groups of interest were women in intimate relationships with men, women who talked about job/career concerns in therapy, women with children (working and nonworking), incest survivors, rape victims, and battered women. The growing 185 body of literature concerned with women in psychotherapy suggests that the above listed groups of women often bring special needs to treatment. Recognition and understanding of these needs is an essential component of sex-fair counseling for women (Hill, 1979, Brodsky, & Hare-Mustin, 1980, Howell & Bayes, 1981, Collier, 1982). A limited descriptive analysis of the responses to items written for these sub-groups of women follows. As with the general items, items for the sub-groups were rated on a scale of 0 to 3, 0 being the least and 3 being the most sex-biased treatment in psychotherapy. A theoretical mid-point of an item mean of 1.5 was used as a criteria in judging whether or not women in these sub-groups experienced sex bias relative to their particular treatment needs. Women in Intimate Relationships with Men (N = 121) WOmen in this sub-group responded to 10 items written specifically to address bias in therapy as demonstrated by a double standard for identity development and sexuality for men and women and a lack of understanding of the conflict for women between self and family needs. The mean for the 10 items in this group was 6.8 out of a possible score of 30, item means ranging from .09 to 1.78. Minimal sex bias in psychotherapy was observed relative to the issues for this sub-group of women. In general, women clients in intimate relationships with men felt encouraged by their counselors to develop a sense of their own identity 186 and to satisfy their own needs. Relatively more bias was experienced around the therapist's lack of understanding regarding male and female differences in the affective areas, suggestive of a devaluation of the woman client's strengths in the emotional and affiliative areas. WOmen Who Talked About Job/Career Concerns with Their Counselor (N = 107) WOmen in this sub-group responded to six items which focused on issues related to keeping women in traditional work roles. Bias was not observed for any of the items. A mean of .21, out of a possible score of 13, with a range of .20 to .51 observed for the item means,was obtained for the items in this group. Women clients in this sample who talked about job/career concerns with their counselor were responded to in a nonsexist manner and were encouraged to pursue careers in a broad range of areas. WOmen with Children (Working and Non-WOrking) (N,- 70) Women in this sub-group responded to six items reflective of issues relevant to career vs. family conflicts and myths of the perfect mother. The mean for the six items in this group was 4.61 out of a possible score of 18, with item means ranging from .20 to 1.64. Minimal bias was observed in regard to a woman's conflict regarding her work and family roles. Evidence of sex bias was noted in the area of the myths of the perfect mother. Bias was evident in not giving the client permission to 187 acknowledge resentment of her children at times and in not helping to alleviate guilt regarding this resentment. Item 94, ”My counselor helped me question my belief that V a 'good mother' never resents her children,‘ was as often positively endorsed as negatively endorsed by women clients in the sample. Incest Survivors (N = 27) WOmen in this sub-group responded to five items developed around the special treatment needs of incest survivors, sex-biased perceptions of family dynamics in incest, and myths surrounding the occurrence of incest. The mean for this group of items was 3.1 out of a possible score of 15, with item means ranging from .07 to 2.45. While incest survivors report relatively little bias in the counselors' understanding of family dynamics in incest or adherence to myths surrounding incest, a bias is apparent in an act of omission.4 In only two out of 27 cases did the therapist discuss with the client the possibility of joining a group for incest survivors. The mean score for Item 95 - "My counselor discussed with me the possibility of joining a 4Acts of omission are statements or procedures which are left out of the therapy process, whereby omission nega- tively impacts the client's growth. Acts of commission, as opposed to acts of omission, are statements or procedures included in the therapy process which are deemed to be detrimental to the client's growth in therapy. 188 group for incest survivors” - was 2.45 on a 3-point scale, with 3 being the most sex-biased. This act of omission suggests a lack of recognition of the special needs incest survivors have to deal with the feelings of isolation, guilt and shame, which are functions of the incestuous relationship, in a group with other survivors. Rape Victims CE,” 19) Women in this sub-group responded to four items developed around treatment issues for women who had been raped and sex-biased interpretation or understandings of the experience of rape. The mean score for items in this group was 2.6 out of a possible score of 12 with item means ranging from .05 to 2.34. As with incest survivors, bias was evident in an act of omission. In only three of the 19 cases was the option of taking a self-defense course discussed with the client in therapy. Yet, this is a highly recommended adjunct to treatment for rape victims in that it addresses fears of further abuse and aids in empowering the victim. Aside from this, rape victims in this sample did not acknowledge sex bias on the other items. Battered Women (H = 21) WOmen in this sub-group responded to five items developed to explore counselors' recognition and treatment of battered women. The mean score for items in this group 189 was 5.4 out of a possible score of 15, with item means ranging from .15 to 1.95. Relatively more sex bias was observed with this group of clients, relative to their particular treatment needs, than any other group. As with incest survivors and rape victims, acts of omission were the common form of sex bias perceived. Very few women in this group were supported or encouraged to move to a safer home environment. Few women were helped to look at the reasons why they remained in a battering relationship, nor were many helped to realize that they did not have to stay in a physically abusive relationship. Responses to the items in this grouping suggest that counselors in this sample did not deal with the central issues for their battered women clients. Phenomena of Sex Bias in Psychotherapy as Described by the Ten Items Most Often Acknowledged in a Biased Direction In addition to the items from the sub-groupings discussed in the previous section, certain items on the SBPQ were more frequently acknowledged in a sex-biased direction than other items. The ten items with the highest _item means were observed and common themes among them identified. Only two of the item means were equal to or greater than 1.5, the theoretical mid-point for the item. Thus, the items were not as often acknowledged in a sex- biased direction as in a non-biased direction. Nonethe- less, the fact that these items are detecting the most bias of all the SBPQ items is significant and worthy of some exploration. The 190 10 items with the greatest item means are listed below in descending order of acknowledgement in the sex biased direction: 1. 10. My counselor helped me learn ways other than crying, flirting and/or lying to get what I wanted (Item 65, X = 1.68). My counselor told me that some of my problems were similar to those of other women (Item 3, X = 1.50). My counselor seemed to understand the frustration I felt about real inequality between men and women (Item 50, X = 1.41). My counselor seemed to understand that sometimes ‘§hen I cried, I was really angry (Item 60, = .90). My counselor helped me see that women could have roles other than thoge of mother, daughter, wife/ girlfriend (Item 6, X = .87). The purpose of my counseling seemed to be for me to learn how to be more satisfied in my role_of wife/ girlfriend, mother or daughter (Item 31, X - .77). My counselor helped me lgarn to rely more on myself than on others (Item 7, X = .76). My counselor saw my ability to express my emotions as a strength (Item 49, X = .75). My counselor pointed out that some of my problems came from being a woman in a socigty which sees women as unequal to men (Item 4, X = .72). My counselor encouraged me to express my angry feelings (Item 53, X - .72). Observation of these items reveals that the items are from either the Traditional Roles or Bias and Devaluation Scales with the majority of items coming from the Bias and Devaluation Scale. Additionally, seven of the ten items are positively worded. The conclusions to be drawn from these observations are that most of the sex bias experienced 191 by this sample of women clients was in the area of bias in expectations and devaluation of women and as with the sub- groups of women sex bias was more evident in acts of omission than in acts of commission. The three broad areas that these 10 items cover are (l) exploration of the traditional feminine role, (2) anger expression, and (3) socio-cultural explanations for some of women's emo- tional difficulties. Phenomena of Sex Bias in Psychotherapy as Described by Anecdotal Data A few respondents provided unsolicited descriptions of sex bias in their therapy experience. Select verbatim accounts from these clients are presented. A mid-thirties woman currently in graduate school reported: A few years ago I had a sexist counselor. This man suggested to me that I become an AVON Lady. He asked if my husband had ever thrown me on the floor and raped me, and when I said he hadn't, he was disappointed. This counselor saw my husband's failure in this area as part of our marital problem. He also asked why I was so masculine and aggressive. Another woman, in her late 20's, with some graduate- school experience, and in therapy with an experienced male therapist, related these details: This counselor infantalized me when I expressed emotion of any sort. He values affect and seems to view its expression as a strength, but responds as though the expression of emotion implies regression to a e two, therefore implying to me that women's amot ons are immature. . . . One of the many factors which contributed to my decision to terminate therapy was that toward the end of our therapy this counselor told me that he believes that women and men are 192 designed by nature to serve different roles-- that psychological gender differences are biologically determined. A third woman provided insight into the difficulty she had in psychotherapy related to a role-reversed relationship in which she was involved. I was in a relationship that was traditional role- reversed. My male significant other was highly expressive, sensitive, emotional, and nurturing. He was a working-class person, while I am working toward a professional degree. My counselor under- mined my valuing of this man and stated that when I could learn to carry the expressive functions, I would bring these to a relationship--and then find traditional relationships gratifying and high achieve- ment important for my partner. . . . My counselor never for a moment took seriously my valuing of emotional expression over achievement in a man. The above anecdotal accounts of sexism in these clients' psychotherapy parallel the themes of bias outlined by the APA Task Force (1975). These verbatim accounts provide support for the existence of sex bias in psycho- therapy, if only for a small number of women clients. Summary The results of the hypotheses related to the phenomenon of sex bias in psychotherapy were presented in this chapter. Six hypotheses were tested. Hypothesis V (a and b) tested whether or not women clients experience sex bias in psychotherapy. Minimal evidence for the existence of sex bias in psychotherapy was found. Women clients in general did not experience statisi- cally significant amounts of sex bias in therapy, though some women reported sex biased psychotherapy experiences. The null hypothesis was retained. 193 Hypothesis VI (a and b) was concerned with whether or not some groups of women or women with different therapy experiences reported different amounts of sex bias in psychotherapy. The independent variables of concern were the client's age, education, relationship status, child status, number of sessions, and length of time since termination of therapy. Three of these variables were found to be of importance relative to the experience of sex bias. Women with minimal formal education were found to experience significantly more sex bias in psychotherapy than women with the most formal education. WOmen with children were found to experience more sex bias in psychotherapy than women without children. Further analysis of this finding revealed a significant interaction between child status and the therapist's agency affiliation. The above finding holds for the community mental health center clients, not for the counseling center clients. An additional finding was that women with one to three therapy sessions experienced significantly more sex bias in therapy than women with more than 45 therapy sessions. The variables of age, relation- ship status and length of time since termination of therapy showed no differences in the experience of sex bias in psychotherapy for women in the different groupings. Hypotheses VII, VIII, and IX were developed to test the effects of the therapist variables of sex, experience level, and agency affiliation on the experience of sex bias 194 in psychotherapy. No significant differences were found for any of the therapist variables, nor were significant inter- actions noted. The null hypotheses in each case were retained. Therapist sex, experience level, and agency affiliation were found to have no effect on women clients' experience of sex bias in psychotherapy. A supplementary descriptive analysis of the data was undertaken to aid in description of the phenomena of sex bias in psychotherapy. The sub-groups of women in intimate relationships with men, and women who talked about job/ career concerns in therapy did not experience sex bias relative to the issues and needs of their particular sub- groups. Generally, women with children were responded to in a sex-fair manner in regard to conflicts around family and work roles. Bias for this sub-group was evident around the myth of the perfect mother. Incest survivors, rape victims, and battered women were found to experience relatively more sex bias in psychotherapy than the other sub-groups of women, reflective of a lack of understanding of these women's particular needs. Bias was observed primarily in acts of omission; most often the appropriate treatment for these women was not discussed as an option. Acts of omission were also the predominant form of sex bias observed in an analysis of the 10 SBPQ items most often acknowledged in a biased direction. The broad areas for which the most bias was observed were (1) exploration 195 of the traditional feminine role, (2) anger expression, and (3) socio-cultural explanations for some of women's emo- tional difficulties. The anecdotal material presented provided some support for the existence of sex bias in psychotherapy, if only for a small number of women clients. CHAPTER VI SUMMARY, DISCUSSION AND IMPLlCATIONS This chapter summarizes the research study. The findings of the study are presented along with a discussion of the results related to instrument development and the phenomena of sex bias in psychotherapy. Implications of the findings are discussed, along with directions for future research. Summary The purpose of this study was twofold. The primary purpose was to develop an instrument to measure sex bias in psychotherapy. To accomplish this purpose, the Sex Bias in Psychotherapy Questionnaire (SBPQ) was developed and initial reliability and validity data for the instrument was obtained. The secondary purpose of the study was to use the Sex Bias in Psychotherapy Questionnaire to investi- gate the phenomena of sex bias in psychotherapy with women clients. This included exploring whether or not women clients experience sex bias in psychotherapy, identifying specific groups of women who experienced more sex bias in therapy than others, and determining the effects of therapist variables on women's experience of sex bias in psychotherapy. A descriptive research design in 196 197 the form of a naturalistic field investigation was used, in conjunction with a quantitative and supplementary descriptive analysis of the data, to address the dual purposes of the research study. The SBPQ is a 100-item self-report instrument developed around the themes of bias outlined by the APA Task Force on Sex Bias and Sex—Role Stereotyping in Psycho- therapy (1975). These themes include: (1) fostering tradi- tional sex roles, (2) bias in expectations and devaluation of women, (3) sexist use of psycho-analytic concepts, and (4) responding to women as sex objects, including the seduc— tion of women clients. Theories used in the development of the SBPQ include: sex-fair theories of personality (Maslow, 1968; Bem, 1974, Hefner, et al., 1975; Kaplan, 1976, 1979a and Gilbert, 1981), sex-fair counseling theories (Williams, 1976; Rawlings and Carter, 1977; Gilbert, 1980, and Collier, 1982), and theories relevant to sex-role issues in women's mental health (Horney, 1932; Thompson, 1943; Cove and Tudor, 1972; Miller, 1976; Bernardez, 1978; Gilligan, 1979, 1980, 1982; Collier, 1982, Franks and Rothblum, 1983, and others). The procedures used to develop the SBPQ involved generation of items around specific objectives and theory, expert judging of the items for content and informal pilot- ing of the instrument prior to administration. Four sub- scales comprise the SBPQ total scale. These scales are named the Traditional Roles Scale, the Bias and Devaluation Scale, the Psychoanalytic Scale, and the Sex Object Scale. The 198 psychometrics for the SBPQ are based on the first 64 items, which are general items, answerable by any woman. The remaining 36 items were developed for specific sub-groups of women and were subjected to a supplementary descriptive analysis. Following the development of the SBPQ, a survey approach was used to gather information about sex bias in psycho- therapy from a large number of former female therapy clients. A sample of 192 terminated women clients from a university counseling center and a local community mental health center completed the SBPQ in addition to providing information regarding their age, education, relationship status, child status, number of therapy sessions, and length of time since termination of therapy. Therapist variables including sex, experience level and agency affiliation were also identified for each client's former therapist. In addition to client participation, 49 therapists from the same counseling center or community mental health center were involved as subjects in the research. These therapists completed the Stereotype Scale of Spence, Helmreich and Stapp's (1974) Personal Attributes Questionnaire, as a measure of their tendency to stereotype others. The results of this measure were used in construct validity testing of the SBPQ. Therapist variables of sex, experience level, and agency affiliation were also obtained from this group of subjects. 199 Nine hypotheses related to instrument development and the phenomena of sex bias in psychotherapy were developed and tested. The results regarding the reliability and initial validation of the SBPQ are as follows: 1. The SBPQ total scale was demonstrated to be an internally consistent and reliable instrument. A coefficient alpha of .88 was obtained for the SBPQ total scale. The four sub-scales of the SBPQ demonstrated sufficient reliabilities for research purposes. Alpha coefficients of .68, .80, .73, and .79 were obtained for the Traditional Roles, Bias, and Devaluation, Psychoanalytic, and Sex Object Scales, respectively. The levels of internal consistency for the Bias and Devaluation, Psychoanalytic and Sex Object Scales suggest that each scale is measuring a single dimension of sex bias in psycho- therapy. A similar assumption is made for the Traditional Roles Scale, but with some reserva- tion due to the reliability of .68 that was obtained. The SBPQ is assumed to be measuring a unidimensional construct. Significant levels of correlation were obsirved between the SBPQ total and four sub- sca es. Initial content validity for the SBPQ was established by clear specification of the content domain, generation of items around pre-determined objectives and high inter-rater reliability among expert judges rating items for content. Construct validation of the SBPQ has not yet been established. a. No relationship was observed between a therapist's tendency to stereotype and his/her client's experience of sex bias in psychotherapy. b. Although not stated as a construct validity hypothesis, one of the phenomena of sex bias in psychotherapy hypotheses (VI a and 6) provides information on the SBPQ's construct validity. The finding that women with children experienced more sex bias in therapy than women without children provides evidence for the SBPQ's construct validity, while the find- ing that women who had one to three therapy 200 sessions experienced more sex bias than those women having had more than 45 sessions raises questions about the construct validity of the instrument. The results related to the phenomena of sex bias in psycho- therapy are as follows: 1. Women clients in general did not experience a statistically significant amount of sex bias in psychotherapy. Some women in the sample, though, reported sex-biased therapy experiences. The client life-status variables of age and relationship status were found to have no signifi- cant effect on women clients' experience of sex bias in psychotherapy. WOmen clients with less than a high school educa- tion were found to experience significantly more sex bias in psychotherapy than women clients with a graduate school education. WOmen clients with children, particularly those who were seen in therapy at the Community Mental Health Center, experienced significantly more sex bias in psychotherapy than women without children. Women clients who had one to three therapy sessions experienced significantly more sex bias in therapy than women clients who had more than 45 therapy sessions. The length of time since termination of therapy had no effect on women's reporting of experiences of sex bias in psychotherapy. The therapist variables of sex, experience level and agency affiliation had no effect on women clients' experience of sex bias in psychotherapy. The results of the supplementary descriptive analysis are as follows: 8. The sub-groups of women in intimate relationships with men and women wno talked about job/career concerns in therapy did not experience sex bias relative to the issues and needs of their particu- lar sub-group. 201 9. Generally, women with children were responded to in a sex-fair manner in regard to conflicts around family and work roles. Bias was evident for this sub-group around the myth of the perfect mother. 10. Incest survivors, rape victims, and battered women were found to experience more sex bias in therapy, relative to the issues and needs of their particu- lar sub-group, than any of the other sub-groups of women. Bias was apparent in acts of omission, reflective of a lack of understanding of the particular treatment needs of these women clients. 11. Acts of omission were the predominant form of sex bias observed in the therapy experience of the total sample. Discussion Throughout the present study two parallel, sometimes intertwined, purposes have been operating. One line is that of instrument development and validation. The other is the study of sex bias in psychotherapy. For consistency and clarity these two lines are followed in the subsequent discussion of the results. Issues relevant to instrumenta- tion are discussed first, those relevant to the phenomena of sex bias in psychotherapy, second. In integrating the find- ings, the lines separating the two purposes become less distinct. Instrument Development Reliability pf the igsggpmgpg. One type of reliability, that of internal consistency, was tested for on the SBPQ total and four sub—scales. A reliability coefficient of .88 suggests a relatively high level of internal consistency, especially for a self-report measure. 202 An understanding of the factors that contribute to the reliability of a scale provides some explanation for the high level of internal consistency observed. Factors contributing to an instrument‘s reliability, that are particularly relevant to the reliability of the SBPQ, are: the length of the test, the heterogeneity of the sample, and whether the items are scored dichotomously or on a Likert-type scale (Mehrens & Lehmann, 1978). The reliability for the SBPQ total scale was based on 64 items. This number of items is relatively high and is assumed to have significantly contributed to the overall reliability of the scale. Additionally, the sample of client-subjects was heterogenous with respect to client life- status, therapy-related and therapist variables. The heterogeneity of the sample positively influenced the SBPQ's overall reliability. Similarly, the use of a Likert- type rather than a dichotomous scoring system appeared to aid reliability by providing for more heterogeneity in responses. In addition to test construction theory another possible explanation for the adequate reliability obtained on the SBPQ is that each item was based on a set of objectives that were supported by theories relevant to sex-role issues in mental health for women. It is likely that the develop- ment of items around stated objectives, reflective of a clear and consistent definition of that which constitutes sex bias in psychotherapy (APA, 1975), aided in the 203 measurement of a homogenous construct and high internal consistency of the measure. Furthermore, it is likely that the SBPQ is measuring a single construct with a number of different dimensions. The same core theme is reflected in all items, but in a number of different ways. The high inter-scale correlations between the SBPQ total and the Traditional Roles, Bias and Devaluation and Psychoanalytic Scales provide evidence that the SBPQ is measuring a unidimensional construct. The exception to this is the Sex Object Scale which, although statistically was significantly correlated with the other scales, demonstrated observably low inter-scale correlations. Most probably, the Sex Object Scale is measuring a separate dimension of sex bias in therapy which has minimal relationship to the other aspects of the phenomena. From an intuitive perspective, behavior which involves responding to women as sex objects, including seduction of women clients, can be viewed as a more obvious and blatant form of sexism in psychotherapy than the forms measured by the other scales. The consequences of this form of bias are more harmful to women than other forms of bias, in that sexual seduction of women clients is considered to be an extreme form of misuse of power in the therapeutic relationship, and as such a sanction against this type of behavior is included in the APA Code of Ethics (1981). Five of the items from the Sex Object Scale had to be dropped from the SBPQ because they elicited no variance in 204 responses. In its revised form, the Sex Object Scale contains only four items, yet has sufficient internal consistency. It makes sense at this point to suggest that the Sex Object Scale be dropped from the total scale in future administrations of the instrument. Three items from b, and 56C, could easily be moved to the scale, items 8a, 27 the Traditional Roles Scale because of the apparent overlap between the two scales reflected in these items. Movement of these items would most probably improve the reliability of the Traditional Roles Scale and thus would accomplish two purposes in improving the instrument. Internal consistency of the four sub-scales was acceptably high for research purposes. Coefficient alphas of .68, .80, .73 and .79 were obtained for the Traditional Roles, Bias and Devaluation, Psychoanalytic, and Sex Object Scales, respectively. The Traditional Roles Scale has the lowest internal consistency of the four sub-scales. The lack of consistency on this scale may be due to ambiguity in the wording of a number of items. For example, ambiguity is apparent in item 44 -- "My counselor suggested that I needed to give in to others in relationships" -- in that aItem 8 is "My counselor suggested that I would be more attractive if I spent more time on my makeup.” bItem 27 is "My counselor said that men would be more attracted to me if I appeared more feminine.” cItem 56 is ”My counselor sug ested that I would be happier if I tried to appear more eminine." 205 ”give in” to others would not be interpreted in the same way by individuals responding to the item. Similarly, item 30--"My counselor seemed to devalue my non-traditional women friends"--is ambiguous due to the lack of clarity con- cerning "devalue" and ”non-traditional." Again, a number of different interpretations could be applied to understand- ing the content of the item. Ambiguity in items creates a situation whereby the same individual may respond differently to the item on subsequent administrations of the questionnaire. This lowers the consistency of the measure by increasing the error variance. A revision of the SBPQ would need to evaluate items on the Traditional Roles Scale for ambiguity, making necessary changes where indicated. Another possible explanation for the low internal consistency on the Traditional Roles Scale has to do with the individual respondent's conformity to traditional sex- role expectations. It is assumed that a certain percentage of the clients in the sample are following traditional sex- role prescriptions, thus responding to items from this perspective and not in terms of the therapist's behavior. Rawlings and Carter (1977) and Maracek, et al. (1979), have noted that feminist therapy is threatening to clients who are not attuned with the philosophy of the Women's Movement and to those women who are more traditional in their roles. Consequently, traditional women responding to the SBPQ may have felt threatened by the item content, especially by 206 items on the Traditional Roles Scale, which would have contributed to inconsistencies in their responses. To summarize, the SBPQ demonstrates sufficiently high reliability and internal consistency. The instrument is measuring a unidimensional construct with at least three separate dimensions. Revisions of the scale that are likely to improve the scale's reliability include: elimination of the Sex Object Scale and evaluation of items on the Tradi- tional Roles Scale for ambiguity, making word changes where necessary. Validity of the instrument. Content validity for the SBPQ was established by developing items around Specific objectives reflective of the themes of bias outlined by the APA Task Force (1975). Further support for content validity was provided by the high inter-rater reliability obtained among a group of expert judges rating whether or not items were indicative of sex bias in psychotherapy. The establishment of content validity for the SBPQ helps to further clarify the definition of sex bias in psychotherapy and as such is an important contribution to the field of research on sex bias in psychotherapy. Initial steps for construct validation of the SBPQ were also undertaken in the present study. Construct validity, defined as the extent to which an instrument measures a theoretical construct or trait, is not as easy to establish as is content validity. Construct validation of an instrument requires the gradual accumulation of 207 information from a number of sources. At this point in the development of the SBPQ minimal evidence of construct validity has been obtained. One research hypothesis (IV) was designed to begin testing the construct validity of the SBPQ. This hypothesis was based on the theories of Bernardez (1975) and Kaplan (1979b). Both theories maintain that a therapist's unconscious beliefs about the differences between men and women, in addition to his/her early sex-role training, are invariably reflected in her/his therapy with women clients. It was assumed that some evidence of construct validity would be provided if a correlation was observed between the therapist's tendency to stereotype and his/her client's experience of sex bias in psychotherapy. In other words, what was tested was the assumption that the more a therapist stereotyped others along sex-role dimensions, the more sex bias her/his clients would experience in therapy. A failure to confirm this hypothesis suggests that a client's experience of sex bias in psychotherapy has little to do with the therapist's sex—role stereotypes. There are two explanations for this finding. One reason for the lack of correlation observed may be due to the fact that therapists' stereotypes really have nothing to do with their actions in therapy and that the theory upon which the hypothesis was based is incorrect. The research which attempts to test this theory presents conflicting results. Kahn (1976) found little support for the notion 208 that a therapist's sex-role stereotypes or attitudes towards women influence therapy actions, whereas Gilbert (1981b) found attitudes towards women's roles to effect counselors' beliefs about male and female clients' abilities to perform well in a variety of roles. Another explanation for the failure to find a signifi- cant correlation between therapists' sex-role stereotypes and clients' experience of sex bias in psychotherapy is that the response set of social desirability more than likely influenced therapist-subjects' responses to the Stereotype Scale. _Minimal variation was observed among the therapist- subjects' scores, with minimal evidence of stereotypes being acknowledged. The tendency to respond in a socially desirable fashion, by acknowledging few if any differences between the characteristics of men and women, created a situation in which therapists' scores were very similar to one another. As a group, women therapists seemed to have an easier time acknowledging differences between men and women on the characteristics than male therapists. Yet, the women therapists in the sample were more often experienced by women clients as less sex biased than their male counterparts. In light of these contradictory findings, the scores obtained from therapist-subjects need to be viewed with some skepticism and are probably less than valid. Less reactive measures and measures much less sensitive to social desira- bility would need to be used to further test this hypothesis and determine if in fact there is a relationship between a 209 therapist's sex-role stereotypes and attitudes towards women and her/his client's experience of sex bias in psycho- therapy. Another method for establishing the construct validity of an instrument is to test hypotheses about those people who score high on the measure as opposed to those who score low. One of the phenomena of sex bias in psychotherapy hypotheses (VI a and b.) provides information on the construct validity of the SBPQ relative to this method. Some evidence for the SBPQ's construct validity is found in the fact that women clients with children experienced signifi- cantly more sex bias in psychotherapy, particularly those seen at the Community Mental Health Clinic, than women clients without phildren. A discussion of this finding is presented later in this chapter. Whereas the finding regarding child status helps to establish construct validity for theSBPQ, a finding for the same hypothesis,but testing a different variable, raises questions about the construct validity of the instrument. The finding that women who had one to three therapy sessions experienced more sex bias in therapy than women who had more than 45 therapy sessions suggests that rather than measuring sexist therapy, the SBPQ may be measuring poor therapy. Clients who left therapy after only a few sessions may have done so because they did not feel understood or initially helped by their therapist. If this was the case, feelings regarding the therapist's ineffectiveness may be reflected 210 in the client's responses to the questionnaire and SBPQ items may not be discriminating between sexist and incompetent therapy. On the other hand, those clients who were defensive about receiving counseling and were resistant to treatment were most likely in the 1-3 session category. These clients may have responded to SBPQ items out of their own dynamics, with their responses having little to do with the quality of the therapy received. The theory of cognitive dissonance (Festinger, 1957) provides yet another explanation for the finding that women who had fewer therapy sessions experienced more bias than women who had a greater number of sessions. Cognitive dissonance is a state of tension that occurs when an individual simultaneously holds two cognitions that are psychologically inconsistent. The occurrence of cognitive dissonance is unpleasant and individuals are motivated to reduce it. One way of reducing this tension is to change one or both of the cognitions so that they are more compatible. It is possible that women clients in this sample who were in therapy for a longer period of time experienced more cognitive dissonance than those clients in therapy for only a few sessions, when asked about negative aspects of their counseling experience. Acknowledgement of sex bias in therapy can be assumed to be much more conflictual for the client who was attached to her therapist, had a positive therapy experience and very much felt helped by her therapist 211 than for the client who did not feel helped by her therapist. Women clients who had been in therapy for more than 45 sessions may have been denying the negative aspects of their therapy experience as a way of dealing with the cognitive dissonance engendered by the questionnaire content. If cognitive dissonance was indeed a factor in the results, the above finding may be more reflective of inaccurate reporting than of the actual experience of sex bias in psychotherapy. Clearly, the present study has just begun the process of establishing some construct validity for the SBPQ. At this point there is only minimal evidence to support the fact that the SBPQ is measuring sex bias in psychotherapy as opposed to incompetent therapy or something unrelated to the construct of sex bias in psychotherapy. Phenomena of Sex Bias in Psychotherapy Quantitative findings. The major finding of the present study, relative to the phenomena of sex bias in psychotherapy, is that women clients in this sample did not experience sex bias in their psychotherapy. WOmen clients who responded to the SBPQ reported relatively sex-fair counseling experiences. In light of this finding, the results related to the effects of life-status, therapy- related and therapistvariables need to be understood as reflective of the amount of bias experienced by women in different groups, rather than interpreted as evidence that 212 one group experienced sex bias in therapy while the other group did not. There are a number of possible explanations for the lack of significant findings in the area of women clients' experience of sex bias in psychotherapy. One explanation has to do with instrument construction and scoring relative to the test used to determine significance. In develop— ment of the SBPQ, a score of 96 was determined as the theoretical mid-point for the distribution which ranged from 0 to 192. Statistically what this means is that those clients scoring above the mid-point of the distribution experienced sex bias in psychotherapy and that those scor- ing below the mid-point did not. Conceptually, though, what a score of 96 on the SBPQ would mean is that women clients were consistently acknowledging that sex bias in their therapy "mostly” did not occur and that for at least several of the items they were endorsing the most sex biased alternative, that sex bias "definitely" occurred during their therapy. A score of 96, although the statistical mid- point, would thus be judged by some to be indicative of a sex biased therapy experience. The items for the SBPQ were developed so that the response ”definitely false,‘ and definitely true for posi- tive items, would be indicative of truly sex-fair therapy experiences. Thus a total scale score of 0 would be indica- tive of a truly unbiased therapy experience. If testing against a theoretical population mean of O was feasible, 213 evidence of sex bias would have been documented in the present study. Quantification of the data, including the establishment of a theoretical population mean at the center of the distribution, did not allow for this possibility. A question is thus raised over meaningful versus statistical significance, suggesting that a more qualified conclusion be drawn about women's experience of sex bias in psychotherapy. A more qualified conclusion is that women clients in this sample did not experience statistically significant amounts of sex bias in their therapy, but did acknowledge a small degree of sex bias in their therapy experience. The results of the item response frequency distribution and supplementary descriptive analysis provide further support for this conclusion. The problems noted in the current scoring system for the SBPQ suggest that instrument revisions may need to include a criteria rather than a normative scoring system. A scoring revision such as this would involve weighting of items for degree of bias and an apriori determination of the number of items that would need to be acknowledged in a biased direction as a criteria for a sex-biased therapy experience. In addition to the above measurement issues, another possible explanation for the lack of significant findings centers around the reactivity of the measure. Reactivity is an unwanted influence on the subject's response created by 214 calling attention to the fact that a particular subject is being studied. (Sellitz, et al., 1976). Originally, it was thought that ”filler" items would be used for the SBPQ to detract from the face validity, thus the reactivity of the measure. Due to the length of the instrument and the pressure for an adequate response rate, "filler" items were discarded prior to the final administration of the SBPQ. The discarding of these items made it clear to respondents that they were being asked about sexist practices in their therapy. Several respondents contacted the researcher following receipt of the questionnaire and expressed anger about the content of some of the items. From.the content of phone conversations with client respondents and short notes enclosed with smme of the return questionnaires, it became clear that the SBPQ was a controversial instrument and a very reactive measure. The trade-off of a relatively high response rate may not have been worth the loss of accuracy in client report. Because of the reactivity involved, caution needs to be used in interpretation of the responses to the SBPQ. Additionally, as was discussed earlier in this chapter for the Tradi- tional Roles Scale, some items on the SBPQ were likely threatening to the more traditional clients in the sample, raising their awareness not only of sex-role issues in therapy but of their own sex-role stereotypic beliefs, caus- ing further problems in reporting. 215 One way to handle the problem with reactivity would be to return to ”filler” items by decreasing the number of total items. The Bias and Devaluation Scale, composed of 25 items, has a scale reliability of .80 and is correlated .92 with the total scale. Use of this sub-scale alone with ”filler" items may be a partial solution to the problem. It may be that in the study of an area as sensitive and controversial as sex bias in psychotherapy the use of inter- views to gather information is more appropriate. At least in an interview situation a client could express her anger and the interviewer could deflect it, allowing for more accurate information to be obtained. Or it may be that actual therapy interviews need to be analyzed for sex biased content to avoid direct client participation and reactivity problems. Finally, it is possible that the inability to detect sex bias in psychotherapy with women clients is a true reflection of minimal bias experienced by this group of former therapy clients. Therapists at both the Counseling Center and Community Mental Health Center may be aware of sex-role issues for women in therapy and having explored their own biases regarding mental health for women act in nonsexist ways in therapy. This may be reflective of a unique group of clinicians with awareness raised by being in a university community or may indicate that over the last 15 years therapists have truly become more sensitive to the needs and issues of women clients in therapy. 216 Life-Status Variables. Two of the four life-status variables were found to have a significant effect on women clients' experience of sex bias in psychotherapy. Women with less than a high school education were found to exper- ience more sex bias in therapy than those women who had some graduate school education, though neither group experienced a significant amount of sex bias in therapy. Additionally, women with children, particularly those seen in therapy at the Community Mental Health Center, were found to experience more sex bias in therapy than women without children. The education finding is a difficult one to explain. One possible explanation is that more educated women, assumed to be more aware of feminist issues and potential sex bias in therapy, were more selective than less-educated women in choosing a therapist. Thus, by a self—selection process these women chose more sex-fair counselors initially. Another explanation is that therapists may behave differently with their less-educated than more-educated clients. Therapists may be less biased with their more-educated clients because these clients are more likely to confront sexism in therapy than are the less-educated and aware clients. A final con— sideration that needs attention is that although the SBPQ had an eighth-to ninth-grade reading level, the less-educated women may have had difficulty understanding or reading some of the SBPQ items. This being the case, the results of the education finding would be rendered invalid. 217 The child status finding is less surprising and more easily explanable, especially if interpreted in conjunction with the post-hoc clinical finding that bias was evident for women with children around myths of the perfect mother. It is probable that this group of women experienced more sex bias in psychotherapy than women without children because therapists have a more difficult time accepting unhappiness, dissatisfaction and anger in women who are also mothers. These difficulties stem, as discussed tut Lerner (1974) and Bernardez (1978), from unconscious beliefs of the powerful, omnipotent mother rooted in early infancy. As noted, in the theory section for instrument development, a defensive handling of these fears and beliefs has helped to perpetuate myths of the all "good" mother, leads to a devaluing of women, particularly mothers, and an infantilizing of women as a means of controlling their power, and has helped to perpetuate myths of the all good" mother. Support for this theory is provided by Etaugh and Kasley (1981) who found that women with children were judged as less competent than women without children and by Orlinsky and Howard (1980) who noted complex interaction between women clients' age, marital status, parental status and satisfaction in therapy with various gender-pairings. Additionally, there is partial support in the literature for the devaluation of mothers by blaming them for their children's emotional problems (Kellerman, 1974, & Abramowitz, 1977); the schizophregenic mother is a classic example of 218 this process (Jackson, Block, Block & Patterson, 1958). It is reasonable to assume that these biases reveal them- selves during the therapy hour and are accurately reflected in women clients' report of sex bias in psychotherapy. Therapyerelated variables. One of the two therapy- related variables was found to have an effect on women clients' experience of sex bias in psychotherapy. Women who had one to three therapy sessions were found to experience more bias in therapy than women who had more than 45 sessions. This finding has already been discussed in the section on construct validity. To briefly reiterate, the difference observed may be reflective of a problem in the instrument's ability to distinguish between incompetent therapy and sexist therapy or may be related to cognitive dissonance. The purpose of including the variable length of time since termination of therapy was as a control for retrospec- tive reporting. No differences were found in the responses to the SBPQ among women who had terminated therapy anywhere from.one to 12 months prior to completing the measure. This suggests that there was no effect in reporting sex bias in psychotherapy due to the length of time since termination of therapy. Therapist variables. None of the three therapist variables, sex, experience level or agency affiliation, were found to have a statistically significant effect on women clients' experience of sex bias in psychotherapy. From the statistical standpoint it is concluded that the therapist's 219 sex, experience level, and agency affiliation have minimal effects on a client's experience of sex bias in psycho- therapy and that, more than likely, it is client variables or therapist variables other than those studied which effect the experience of sex bias in psychotherapy for women clients. Moving away for a moment from the rigor of statistical testing allows for an investigation of trends observed in the data. The flexibility afforded to descriptive research methodology encourages the exploration of trends so that theory may be built and further research questions developed. All trends need to be interpreted cautiously, with a recognition that the primary function of trend observation is to generate hypotheses to test new theory. Both a main effect and three-way interaction trend are noted in the data for the therapist variables. A trend (p<:.25) was observed for the main effect of therapist agency affiliation. Examination of this trend suggests that Community Mental Health Center clients experienced more sex bias in psychotherapy than did Counseling Center clients. A three-way interactional trend (p<=.25) suggests that this agency finding may be related to the other therapist variables. The ascending order of sex bias experienced in psychotherapy by women clients as effected by the three therapist variables is as follows: 220 - Counseling Center inexperienced female therapists - éfihhgeling Center experienced female therapists - Cfihhzeling Center inexperienced male therapists - é§;$3fi11y Mental Health Center experienced male therapists (X=20.2) - Community Mental Health Center inexperienced female therapists (X=20.5) - Cgpnseling Center experienced male therapists - Cghéanity Mental Health Center experienced female therapists (X=24.2) - Community Mental Health Center inexperienced male therapists (X=28.7) The three-way interactional trend is complex. Visual inspection of it suggests that the overriding factor in the trend is the therapist's agency affiliation. This find- ing does not necessarily mean that the Community Mental Health Center therapists were more biased than the Counseling Center therapists, but rather that clients seen at the Community Mental Health Center experienced more bias than those seen at the Counseling Center. ’ A recognition of the inherent differences between the two client populations served at the respective agencies may help to explain the trend observed. It is quite possible that a greater percentage of those clients seen at the Community Mental Health Center than at the Counseling Center are fairly traditional in their sex-role orientation. As has been documented in the literature (Rawlings & Carter, 1977, and Maracek, et al., 1979), more traditional women feel threatened by both feminist and nonsexist approaches to therapy and are more likely to be resistant to socio-cultural interpretations of their difficulties. Additionally, Community Mental Health 221 Center clients, as opposed to Counseling Center clients, come to therapy with very real survival concerns, such as needs for food, shelter, and clothing. These types of clients may be more frustrating for Community Mental Health Center therapists who, although wanting to impact on sex- role socialization, feel overwhelmed to do so. Instead, these therapists probably deal with the day-to-day concerns which their clients bring to therapy and rarely address sex- role issues. Ironically, it is probably these clients who in part have such emotional problems as depression, agoraphobia and eating disorders as a consequence of sex-role socializa- tion and sexual inequality, who are most in need of a feminist approach to treating and understanding their problem(s) (Franks & Rothblum, 1983; and Seidenberg & DeCrow, 1983). As compared with the Community Mental Health Center therapists, Counseling Center therapists are likely to be serving a female client population with greater feminist awareness, making them more open to socio-cultural interpreta- tions of their problems and challenges to their traditional sex-role stereotypes. Although these clients often present in crisis, as do the Community Mental Health Center clients, their physical needs for food, shelter, and clothing are often provided for so that therapy can be more insight oriented and exploratory than it can be for Community Mental Health Center clients. Essentially, then, the agency trend as well as the three-way interactional trend is probably more 222 reflective of the varying needs and differences in feminist awareness between the two client populations served than a true reflection of therapist differences in therapeutic approach. Supplementary findings. A limited supplementary descriptive analysis of responses to items written for specific sub-groups of women and of the ten items most often acknowledged in a sex-biased direction on the SBPQ by the entire sample of women, revealed sex bias primarily in the area of bias in expectations and devaluation of women, most often taking the form of acts of omission. Most therapists in this sample did not engage in blatant forms of sex bias in therapy with their women clients, but rather more subtle forms of bias were noted in the omission of certain state- ments or recommended forms of treatment recognized as growth promotion for women. Bernardez (1975) discusses these acts of omission in therapy as related to sex-role countertransference issues. She suggests that through acts of omission therapists implicitly reinforce behaviors that inhibit a woman's growth and development, out of a fear of dealing with the alterna- tive, a more assertive, powerful, and competent woman. This is a probable cause for some of the bias observed in acts of omission. It is also likely that these acts of omission reflect a lack of knowledge or therapists' inability as of yet to integrate the new knowledge about mental health 223 treatment for women into their therapy practice. The latter implies a need for affective as well as cognitive integration when it comes to women's issues in therapy. Acts of omission were also the predominant form of bias experienced by incest survivors, rape victims and battered women in the therapy sample. Too often an essential adjunct to therapy was not offered to these clients. For example, rarely was the possibility of joining a group of incest survivors or taking a self-defense course discussed with the client. Additionally, the condoning of the victimization of women was noted in the lack of attention to the needs of battered women for a safe home environment. Additionally striking were written responses from incest survivors, battered women and rape victims, sharing that they were never asked about past or present physical or sexual abuse, and thus had never talked about these issues in therapy. While therapists in this sample were able to address the needs of the generic woman client, they appeared to fall short with women who had been victimized. This find- ing suggests a lack of awareness on the therapist's part concerning the prevalence and impact of victimization on women, most likely combined with an affective avoidance of the issues. Additional data on sex bias in psychotherapy was pro- vided by anecdotal accounts of bias submitted by-women client respondents. The richness of these accounts suggest that sex bias does exist in therapy, if only for a small 224 number of women clients. It also supports the direction for future research to be geared towards even more qualitative, descriptive designs, allowing for in-depth interviewing and content analysis of actual therapy sessions. Limitations of the Study One of the limitations of the study is related to the generalizability of the results. A 40% response rate to the questionnaire survey is adequate and higher than anticipated, but the clients who responded did not likely respond in a random manner. This impacts the external validity of the study. Because of the issues of client confidentiality, it was not possible to compare the characteristics of respondents to non-respondents as a means of improving the external validity of the study. An examination of the demographic information provided by respondents suggests that the sample was varied and that respondents were well divided along the life-status variables. The one exception to this is in respect to the education variable. Almost 80% of the sample had at least some college education. This fact is most likely accounted for by the Counseling Center respondents, though a signifi- cant percentage of the Community Mental Health Center client- subjects were well educated. This unique sample charac- teristic needs to be considered in generalization of the results, particularly to community mental health center populations. 225 With this unique characteristic of the sample noted, generalization to a larger client population can be made using the Tukey—Cornfield Bridge argument (Glass & Stanley, 1970). This is a logical argument allowing for inferences to be made from.non-randomized samples to populations of interest, provided that the characteristics of the sample are described in detail. Use of this argument allows for the generalization of these results to the larger Counseling Center and Community Mental Health Center client populations. Generalizations beyond this population need to be made with caution. Another limitation of the study was the use of self- report measures for both the client-subjects and therapist- subjects. Self-report instruments are accurate to the degree that self-perceptions are accurate and to the degree that the individual is willing to express them honestly (Sellitz, et al., 1976). Although the Stereotype Scale of the Personal Attributes Questionnaire was found to have low correlations with a measure of social desirability (Spence, et al., 1974), it is likely that therapist-subjects in this sample responded to the instrument in a socially desirable way. Additionally, the Sex Bias in Psychotherapy Question- naire, though not as sensitive to social desirability, was a very reactive measure, possibly threatening to many of the client-subjects. A number of factors contributed to the reactivity of the measure, including: the client's sex-role 226 orientation, feminist awareness, need to protect the therapist and to maintain the therapy experience as positive. This reactivity most likely detracted from the accuracy of the client's report. In addition to the above limitations, naturalistic field investigations are unable to control for many confound- ing variables. One variable not controlled for, and assumed to be a confounder, was client presenting problem and degree of emotional disturbance. A large enough sample was obtained to allow for the assumption that this lack of control did not seriously bias the results. Nonetheless, interaction effects between sex bias experienced in psycho- therapy and diagnosis or degree of emotional disturbance were not tested. One limitation of the SBPQ itself concerns the area of construct validity. Construct validity is not something established in one attempt; it will take more research to establish the SBPQ's construct validity. Minimal construct validity was established in this initial validation study and this needs to be considered in interpretation of the results. Implications of the Research The results of this research have implications for the practice of therapy with women clients, psychotherapy training and for female consumers of psychological services. Taking into consideration that the overall finding of the 227 study was the failure to document evidence of sex bias in psychotherapy with women clients, the implications of the research center around the refining of relatively sex-fair counseling practices. The clearest implication of the research findings is that, as a group, therapists are in need of factual informa- tion regarding the special needs of women seeking therapy and the relationship between sex-role socialization, sexual inequality and women's emotional problems. The fact that the most sex bias experienced by women clients in therapy was due to acts of omission suggests that more cognitive information needs to be passed on to practicing clinicians through continuing education workshops and to clinicians in training through course work and seminars on women and psychotherapy. This is especially true regarding the special treatment needs relative to the impact issues, of incest survivors, rape victims, and battered women. While providing therapists with cognitive information will help to refine already sex-fair counseling practices, Rieker and Carmen (1983) note that a more affective component to training in gender and psychotherapy, allowing for the encountering of unconscious beliefs about men and women, is essential to training more sex-fair therapists. They also advocate for a balance between men and women in training programs, in that this balance tends to facilitate affective integration in value clarification and makes the effects of sexual inequality for both men and women more 228 real for the participants. Learning in both the cognitive and affective domains are thus implied. The Sex Bias in Psychotherapy Questionnaire can be used as a tool in continuing education and graduate school training, in that it provides a clear definition of that which constitutes sex bias in psychotherapy and can be used as a self-monitoring checklist or in supervision of trainees. Additionally, components of the SBPQ can be used in client- satisfaction surveys as a built-in measure of sex bias in psychotherapy. Finally, the SBPQ can be used by women consumers of psychological services as a means of determining the amount of sex bias in their therapy and as a decision-making tool in the selection of a therapist. Directions for Future Research Instrument Revision Throughout this discussion section possibilities for the revision of the SBPQ have been identified. These revisions include: elimination of the Sex Object Scale, revision of items on the Traditional Roles Scale to correct for ambiguities which may have contributed to problems with the scale's internal consistency, camouflage of SBPQ items so as to make the instrument less reactive, and movement to a criteria rather than a normative scoring system. Furthermore, as the Bias and Devaluation Scale has a scale 229 reliability of .80, a correlation of .92 with the total scale and consists of only 25 items, future research should attempt to replace the total scale with the Bias and Devaluation Scale as a more efficient screening measure of sex bias in psychotherapy. Revisions such as these would require new estimates of reliability and further testing of the scale's validity. In addition to possible SBPQ revisions, further research is needed to establish the construct validity of the measure. An important test of the SBPQ's construct validity would involve a comparison of the SBPQ scores obtained by clients paired with feminist therapists as opposed to those obtained by clients paired with non- feminist therapists. It will also be important to correlate the SBPQ with a measure of client satisfaction and therapist competence to begin testing whether the SBPQ can distinguish between poor and sex-biased therapy. Further evidence of construct validity would be gained by correlating verbal report of sex bias in therapy with a client's score on the SBPQ. Although the initial correlation between therapist tendency to stereotype and client SBPQ score was not signi- ficant, this line of reasoning needs to be followed in future validation studies. Possibly, correlations between a client's SBPQ score and a non—reactive measure of therapists' awareness of women's issues and needs in therapy, as measured by a written test, would aid in the construct 230 validation process. The psychometrics of the SBPQ warrant further development of the instrument, including revisions and research to establish the instrument's construct validity. Sex Bias in Psychotherapy The present study, in being one of the three studies4 to directly ask women clients about sex bias in their therapy, merely begins the process of describing the phenomena. The results of the study indicate the need for further investigation of sex bias in psychotherapy with women clients, pose new questions for researchers in the field, and provide direction for future research. More information is needed about the forms of sex bias experienced in psychotherapy. This is particularly true for women with children. Additionally, the results of this study raised many questions and concerns about the thera- peutic treatment of incest survivors, rape victims, and battered women. The finding that essential components of therapy, relative to these women's particular treatment needs, were consistently omitted raises questions about the therapeutic treatment afforded to other sub-groups of women in our culture, including: lesbian women, minority women, chemically dependent women, and adolescent women. Investi- gation of sex bias towards these groups of women in therapy is warranted. “The two other studies being Fabrikant (1974) and APA (1975). 231 A combination of quantitative and qualitative modes of inquiry were used in the present study. Although important information was obtained through quantification of the data, some of the subtleties of the phenomena may have been missed. It is recommended that future research in this area move towards more qualificathx1than quantification of data. In terms of methodology, it is recommended that women clients be interviewed about sex bias in their therapy and that actual therapy tapes be analyzed for content indicative of sex bias in psychotherapy. A structured interview could easily be developed from the SBPQ in conjunction with the themes of bias outlined by the APA Task Force (1975) and the SBPQ could be used as a rating scale for sex bias in a content analysis study. The results of the study point to the importance of research and evaluation on psychology training program's and continuing education's efforts to address the training needs of clinicians in the area of gender and psychotherapy. Research on the impact of such training on subsequent therapy practices with women clients is very much needed. Conclusion The results of this research study are optimistic. Therapists in the position of helping women to grow and develop appear to have made strides in the last 15 years towards increasingly more sex-fair approaches to counseling women clients. Similarly, women clients are experiencing 232 their therapy as more egalitarian, thus more promoting of well-being. The results of this study warrant continued investigation in the area, particularly with specific sub- groups of women. Equally important at this point in time is to provide therapists with more cognitive information and an environment in which they can effectively integrate this new material, so that relatively sex-fair therapy practices can become even more egalitarian. APPENDICES APPENDIX A LETTER TO EXPERT JUDGES 233 Robin Sesan 6213 Balfour Lansing, MI 48910 October 14, 1982 As discussed in our recent conversation, I am in the process of developing a questionnaire which will measure women client's experience of sex-role stereo- typing in psychotherapy. My definition of sex-role stereotyping in psychotherapy has come from a number of sources, the primary source being the APA Task Force on Sex Bias and Sex-Role Stereotyping in Psychotherapy (see enclosed article). The APA Task Force has identified four broad areas of perceived sex bias or sex- role stereotyping in psychotherapy. These areas are: 1. Fostering Traditional Sex-Roles 2. Bias in Expectations and Devaluation of Women 3. Sexist Use of Psychoanalytic Concepts 4. Responding to Women Clients as Sex Objects (including seduction of female clients) For each of these four areas I have developed specific objectives around which items for the questionnaire were devised. These objectives, which are enclosed in this packet of information, provide a clear definition as to what constitutes sex bias or sex-role stereotyping in psychotherapy and will help provide content validity for the questionnaire. The other method I am employing to gain content validity is the use of expert judges to determine if items are indeed indicative of sex bias or sex—role stereotyping in psychotherapy. Your task then, as an expert judge, is to rate whether or not in your opinion each item is indicative of sex bias or sex—role stereotyping in psychotherapy, as defined by the stated objectives. There are items which measure acts of commission and items measuring acts of ommission, as well as positively worded items and those worded negatively. More often than not, those items stated positively are items measuring acts of ommission. For example, an item such as: "My counselor encouraged me to be independent," is a positively worded item. If the client responds "false" to this item, i.e., that her therapist did not encourage independence, it would be considered an act of ommission. That is, sex-role stereotyping is apparently due to the therapist's lack of encouraging independence in women clients. Items are also on a continuum from blatantly sexist practices to more subtle forms of sex bias or sex-role stereotyping in psychotherapy. Some of the more subtle items appear to have an almost equal probability of being responded to similarly by men and women clients. 231+ October 14, 1982 Page Two Although for some items, it superficially appears that men and women might answer similarly, it is my theory that these subtle items will differentiate between those clients who have experienced sex—role stereotyping in psychotherapy, and those who have not. It is important then, that you judge the items not by their face validity, but rather that you judge items according to their ability to measure the stated objectives. I realize that some items will be easier for you to judge than others. I ask that you do your best in making your judgements and that you make these judgements independently from other judges. Please return your judged items to me in the enclosed envelope, along with the background information sheet, by November 1, 1982. If you have any questions, please do not hesitate to call me at work (374—8000, ext. 166) or home (393-2715). I thank you for your time and energy. Sincerely yours, Robin Sesan, M.Ed. RS/dc Enclosures APPENDIX B CLIENT—SUBJECT QUESTIONNAIRE PACKET 235 MICHIGAN STATE UNIVERSITY COUNSELING CENTER EAST LANSING ' MICHIGAN ' 48824 Dear Friend: I am conducting a research study with people who have used the Center's services. (If you have never used the Center's services, please disregard this letter and packet of information.) I would like to know what your counseling experience was like. Please complete the enclosed questionnaire and return it to me by February 21, 1983. You will have A CHANCE TO WIN $50.00 by volunteering in this study and returning the questionnaire to me. If you return it, you will be entered in a cash drawing with a prize of $50.00 which I will provide. Your responses to the questionnaire will be kept strictly confidential. I will separate your questionnaire from your Consent Form so that your name will not be attached to the questionnaire. Your responses will not be shared with your former counselor. You are under no obligation to participate in this study and may withdraw your consent to participate by contacting me. The questionnaire should take you at most 30 minutes to complete. A return envelope is provided for you. (For those of you who live on the MSU campus, just drop the return envelope in the campus mail.) If you decide to participate, please read and sign the Consent Form below, and mail this signed form back to me with your Background Information sheet and completed questionnaire. The signed Consent Form will be used for the $50.00 cash drawing. Your help is very much needed and I hope you can find the time to complete the questionnaire. If you have any questions, please feel free to call me at 374-8000, ext. 166. I thank you for your time and energy. Sincerely, KW! 13am) Robin Sesan, M.Ed. 7': 7': '1: 3’: 7’: 2’: 7': 7': 1': 7': 2': 7': 7'.- ‘k 7': CONSENT FORM I have read the explanation of (please print name) the purpose and procedure for this study and agree to participate. I understand that my participation is voluntary, that my results will be kept confidential and that I may withdraw my participation at any time by contacting the researcher. Your Signature I would like a copy of the summary results YES NO PLEASE RETURN THIS FORM WITH YOUR COMPLETED QUESTIONNAIRE MS U is at Affirmative Action/Equal Opportunity Institution 236 cumon oemon - II‘IGHAM COMMUNITY MENTAL HEALTH BOARD THOMAS M ENNIS. J D EXECUIIVE DIRECTOR uuumrv MENIAL HEAL rH WILLIAM J ALLEN 'urv EXECUTIVE DIRECIOR 407 w GRFENLAWN LANSING MICHIGAN 48910 PHONE 3148000 GILBERT OE RAIH PH D CLINICAL DIRECTOR 40} w (JRFENLAWN LANSING MICHIGAN 489’0 PHONE 374.8000 :OARD OF DIRECTORS JAN W LY DDON CHAIRPFRSON PHILIP T BALLBACH LAURIE DOWNES EDGAR FLEETHAM JEAN MCDONALD SAM KINTZER NOLAN OWEN HELEN ROMSEK. PH D ROBERT STEINMAN ENID L WHITE RUDOLPH WILSON LOUISE WIRBEL MASON MENTAL HEALTH CENTER P 0. BOX 2” MASON. MI 48850 PHONE: 676-2001 EATON COUNTY CMHC CENTER 228 S COCHRAN CHMLOTTE. MI 48813 PHONE. 543-5111) INGHAM COMMUNITY MENTAL HEALTH CENTER 407 W. Greenlawn o Lansing, Michigan 48910 0 Phone: 374-8000 Edward A. Oxer, Ph.D., ACSW, DIrector Dear Friend: I am conducting a research study with people who have used the Center's services. (If you have never used the Center's services, please disregard this letter and packet of information.) I would like to know what your counseling experience was like. Please complete the enclosed questionnaire and return it to me by February 21, 1983. You will have A CHANCE TO WIN $50.00 by volunteering in this study and returning the questionnaire to me. If you return it, you will be entered in a cash drawing with a prize of $50.00 which I will provide. Your responses to the questionnaire will be kept strictly confidential. I will separate your questionnaire from your Consent Form so that your name will not be attached to the questionnaire. Your responses will not be shared with your former counselor. You are under no obligation to participate in this study and may withdraw your consent to participate by contacting me. The questionnaire should take you at most 30 minutes to complete. A return envelope is provided for you. If you decide to participate, please read and sign the Consent Form below, and mail this signed form back to me with your Background Information sheet and completed questionnaire. The signed Consent Form will be used for the $50.00 cash drawing. Your help is very much needed and I hope you can find the time to complete the questionnaire. If you have any questions, please feel free to call me at 374—8000,«ext.166. I thank you for your time and energy. Sincerely, (\ flat/0070' Q1414“) Robin Sesan, M.Ed. * * 7‘" 9“ 3': 3’? 3': 7’: a 3’: a“: 7': 7': 3‘: CONSENT FORM I have read the explanation of the (please print name) purpose and procedure for this study and agree to participate. I understand that my participation is voluntary, that my results will be kept confidential and that I may withdraw my participation at any time by contacting the researcher. Your Signature I would like a copy of the summary results YES NO PLEASE RETURN THIS FORM WITH YOUR COMPLETED QUESTIONNAIRE. 237 Code #A BACKGROUND INFORMATION Code AM “*Please answer nl Items. l. Sex. . . . . . . . . . . I) Male 2) Female 2. Age. . . . . . . . . . . l) I8-22 years 2) 23-30 years 3) 3l-40 years 4) 4I—50 years 5) 51-60 years b) 01- + years 1. Education. . . . . . . . I) Did not finish high school 2) Completed high school/CED 3) Some college 4) Completed college 5) Some graduate education] graduate degree 4. Relationship status. . . 1) Single 2) Married/living with someone 3) Divorced/separated 4) Other 5. Do you have children?. . I) Yes 2) No 6. How many counseling sessions did you have with your counselor? (For those pooplv Hh“ have had more than one counselor, answer this Ior your most recent counselor.) I) l-) sessions 2) 4-6 sessions 3) 7-12 sessions 4) [3-20 sessions 5) 21-30 sessions 6) 31-45 sessions 7) 46-60 sessions 8) 6I- o sessions 7. When was the lhst time you saw your counselor? I) Before December 1981 2) January - February 1982 3) March - April 1982 4) May - June l982 5) July - August I902 6) September - October I982 7) November - December I982 8. Was your counselor . . . l) Male 2) Female 9. Counselor's name (if you remember) , GO ON TO THE NEXT PAGE... 238 TITJ'DU'O'U‘U ‘OTJ‘U'O'U‘U'U‘O U DEFINITELY TRUE MOSTLY TRUE DDDDDDD 28.005.013.13 .0 MOSTLY FALSE DEFINITELY FALSE ...mu>o mowm xmzk . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .mco_mmom wrZLDv m6 nmsuaou cmuuo 50—03300 >7. .ON . . . . . . . . . . . . . . . . . . . . . . . . . . . L~omxe ecu Hue one xcsn. cu we vowzunooco Lo_~mc:oo z: .0“ . . . . . . . . . . . . . . . . . . . . . . .coE cesu mnutuceun fleece Loao_ um: can: 91:» v_en Lo_mmc:ou x: .mg . . . . .eMnuxuomcm he vuuocm_ “an .uoccmF mcuvcesmuonc: an an codmmoeauu\naocven As no.) u~emv Lo_umc:ou x: .n. . . . . . . . . . . . . . . . . . . . . . . . . .:0503 can“ o_aooa Lagoon who: :06 easy u~_~QEL no.9ucaou a: .0" . . . .u_ow H mco_soso ago fismmmtaxo a con: _euumo_~i Lo\vce Lacosuausm meuon Lew oE neu_uL._uu uo_~mcaou x: .m. ..................................................L~w:1AE£uu3 :anu ~6e.niugom .0 sea; .muemu x6 5“.) «Longueueoo ouoe u.ou Lo_oncsoo >6 “on. 96 cu uoEoom uu .o~ . . . . . . . . . . . . . . . . . .cocndzno on; H awe—c: .coucou adds“ on so: v~303 H uezu u_ou Lo_uac:ou xx .m~ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .nceuuo _maxun . mecu so: .vu.>co H zoos: vfiuo) one as Luzon m.cue no: a“ once tenements: no: v_:ou Lo_onc:ou x: .- . . . . . . . . . . .:ee o no“: adzmcowuc_ou a an no: a ..uc: amass on no: vflsoz _ gags voudae. Lo_onc30u >t ..~ . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . numououcu ovLu.:o\co_uwosuo\xu03 >5 ozone case cos zen: maggmcouun_uu ou~ELU:_ as “soon mcwuuos :‘ voumououc. «toe no: sodoncaou x: .o. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . eoeoz Lou on no: “so 5:525:20L xcne_ua u_ocu Lo ouLmuao xom o>ns cu new) cu cod sou .eusunc no: nu unzu ~oou ou voEoou uouomcaoo x: .o . . . . . . . . . . gnomes >5 co mews «toe ocean n u. o>~uosuouo once on v.30: a nose vounouman Lo_uncsou A: .o . . . . . . . . . . . . . . . . . . . . . «cocoa co cosy u.oa»a :0 shoe >_ou on sumo“ us van—o: so~ouc:0u xx .5 .ncosLL_Luw\oL.3.uwuzmamo .uoguoe .0 among coco uosuo nodes o>az u_:oo c0503 once eon 0e veg—o; uo_oac:oo a: .o . . . . . . . . . . . . . . .Lo.umcaoo >6 an vmuuoamam ~LoJ m_sou ~aco.muou0um uo\vc~ -c0auon omens mco— at .n we “canoummm “Leanna a we .Lo~wwcaou he an uo:_e> >~Lo>o on on oesoon uncouuu c0503 ocu:_6ou >_‘e:o¢u_vnuu x: .q . . . . . . . . . . . . . cmEos uozso Lo omozu ou u1dqfidn one: nEoLnOLQ xe no oEon once we u_ou uo_umc:ou x: .n . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Lumen >6 coconuoucs LO—omcaou at .N . . . . . . . . . . . . . . . . . . . . . . . .oun zone on mmCLzu uaooun cu uo_om::ou As an unmou30uco no: A .~ .MMmEounun m.;~ .vmnmxm use» :~_3 mmoo :u_:3 muwpuu are moumgu >umz~m .szzuhLou_cawoo Fe noun monk I .lIII. zone 0. noncommuc oLDmeoa LJOL xdco mum oeozw .ouoLQEOo on so» Lou muuac.e noiOM coosuon oxnu u_sozm oLLoCCOLLmusc .oauu >_umos Lo .oouu >~ouMcLuou Lo omde. mfloudcwuuv ..:~Evuaum A.Lo~mmc30u ocuoou umoe use» an >_aa1 xozu me mucoEuueom vzu on ocoanuu umea_a .vocosueaxo w:_~bncsou uco can“ sees no; Lo Logomcaoo 0:0 can“ «see van m>nz 0:3 o.uomq snags LOCO unnamuwum .LOLMm:30u custom to moco_cuqxo mc__omciou Loom Lo ~m_e_ so use“ an) L_ unseen) mvdoyv ob yea so; Loam com .mcL_umc:oo c. mmucmLLoaxo Esau Leona mucoEoueum «to women m:.30__o_ one :0 ”szCLflm 239 UU'O'O'U ‘0 ‘O U‘U‘U'U‘U‘U ‘O LY TRUE U MOSTLY TRUE E DEFINIT 'J 00.0.00 .0 .0 .0 00.00.00 E MOSTLY FALS fl! DEFINITELY FALSE .womm mez mxh OH 20 Du . . . . .mcwmuuee so vo>~0m on v_303 nae—scum >6 Leno noumswwsm Lo_om::oo . . . comcmuum a me meoLUOEo x5 ucmuntovcs can xuwucu9_ o. 2.0..ne >6 Jen Lo.om:30u on :15“ we we w~oL 011:0}. xx At a s o O s a o U o 9 s c o o o O o a e O I O a Caucaz ”ca WQLW" n—Jo %E ”Hwy“ we: .muocso ou v:oamou on we .0» oceuuanL sees ms: nu Lesa voumommam uo~om:30u 1mm Lasso one Lovfimcou . ~c_e_EoL oboe outauaua u «a «5 o» uouoeuuun oboe on 0.30: so: one» vane uo_~ncaoo s o o I e o o o s o e o s e e o e s v s s s o s s o e s e s o s s o a s s o s quduon B30 a: wcsaotm ;.s: co co mcszsoc ea; me~_no.a as use. _oou we sues h23.53 «Lasso sows oonEoo on we new asxo no: no: “u Lego .90» we muse Lo—on:30u . . . . .xnna a on: 0 LL Cocoon on u_:oz onw— me once causeway» Lo_emCJOo . . . . . . . . . .caou c0503 nan >~uuouuveu Lo >~uuou.v Locuuo Lodumcaou . . . . . . . . . . . . . area: a .00 d~o3 anon on; — ans» vsam uO—uncaou sx s: as >x s: x: x: a: s e c s n s e o s s s e e o e s o s s e shVU£uaav Ho hN—AUOE -”Cthudhdw\0uu3 uuwmmwumn ocoe on on Jo: cunod an 08 Law on On beacon wed—omcsou 25 no onoauam och . . . . . . .nvcouuu :oan anc0«uuvauuicoc >6 oa_n>ov on uoSoon Logonesou . . . . . o . o o :08 GOSH I30w>00 OBOE 0.003 COB—OJ 0‘09 UUw—Qau uO—UICDOU s: s: . . . . . . . . . . . . . . «So—AOLQ «no o>_0n ou “damage en :0 uuuqu ceased a “any vounommau hogan:30u as .nauzneo.un—os :. n60~30ua ozone vox~nu ~ cos: .Eo~noLa _mot one we ouccu_o>e cm was coauoosuoxnonxuoouuo he “soon wc_x_au x5 ussu Lawson“ uO—oocaou a Low . . . . . .cmEoz vow: :06 can“ once :05 too: c060: yoga vuL~QEL uo_oncaoo ouwmov uo\ucn no xun— >5 2050 message use—coho 55 Logo vouuuueu uO—mncaou . . . . . . . . . :coEoJ a as“. anon: n.un;u oxuu mucoeeoo one! uo_onc30u a: s: a: s: s: I I I I I 0 O C D I I O O O O O O I I O I I I O IA‘OO‘UW 0U xuqn mafia” .mnoa mcwmcvnu .a_cmcofiom_wu m mcwnco oxfidv wuud x6 cw mwc_nu omcnzo cu nauseous as vommusouu_v Lo_onc30u A: . . . .cueoa so“: mawgmcoiumdou can” m30~> muoe no one: :06 cod: mqucmco_u1_oL Deco uwz_qeg Lo_omc:ou a: .an .On .o~ .mm .e~ .nN .NN .- 240 'O‘U‘OTJT’UU‘U'U‘UU‘UO ‘U ‘D‘UU'U‘U‘O DEFINITELY TRUE MOSTLY TRUE 0000000000000 0000.00 0 MOSTLY FALSE '6 ILY FALSE DEFINITI L. mumm quh uo~omcaou .nmwnu 0 cos: mo6_u~60m “ecu vcaumcovc: on neEmmm Lo_vmcaou >2 >2 «a: u . . . . . . . . . . . . . . . . >5L~mozca an mece >6 3am LodomcsOO >6 xc_zu ~ . . . . . . . . . . . . . . . . . . . Lance >6 snouua 00 «Lee or: . . . . . ... . . . . . .maoson_soun neocoweoaxo _c:xon >6 veg—mu . . MCLcLEou >_Lmu_n>ca «hue sesame on voice 0 a. Leanne: on udaos ~ unzu vounuwmsm . . .mcqxcozvoos nan cannon Leo .muuoam ox“. mowuu>uuun :_ nunououc_ >6 nomeusoumuv on: :06 O .L_mm>6 so» Co: San .mLmLuo Lou >smee on an 06 Law >axo no: a“ yoga uoL~Q6¢ .mm:__oeu >Lwcu >6 aneuaxo o» 96 voweL30oco .mo~0u ~1coLuLnanuicoc u“ :muos wcflsoan axoon mc_onou scum 06 vomnssoon_v coosuun one MMXNQ Lo~omc:0o uO—chaou Lodomcsou Logmmcsoo Lodoncaou .900 meDOU .Lo_mmcwnu >6 a“ mom—~0LLEOu «a zoom on: >uL~naco mes Luzon ~60“ Lou us.mov >u__qscocu ~mu» Leona u~oL H coaumuunauu ozu venunuuoc: ou uoEoon . . numcouun s an «coauOEu >6 uuouaxo cu >uq_dna >6 Jan LL on .EoLAOLa >6 >Mo_0n an ewe so“: on: u nowadauauuao one on uuxoo_ on“ cousuun wuL0aswoc. one «o vox_ou — con: :06 mo n:o_>cu no: u owns . . .cmrux each new: moo—a you an >nno ouoe one: :06 “as“ uo._Q6¢ whence nus: muugmcowundou >6 en anon xonn a oxen — umzu vounommsn . . masnmcodumflou es muozuo by c. o>_u ou covey: 0 ans» ooumomwan wscmann Lo\nca mw>wn.mumnu06 no meson oa>ooououm sums: axoon one. ~ unzu voumomman Lodz: aeoLoOm a an canon a wagon Eouu menu m6o_noLa >6 Lo 0 s e s o co cmzu muozuo :0 when v~son Eu~DOLQ >6 suds: suede“ on. LOLoncaou uO—ymcsoo . 60.nouq one :0 Luau Lo—omcaoo uo—omcnoo uo_onc:oo uo_omcaoo Lo_oac:ou Lodomesou >2 >2 2.. >2 >2 >2 >2 >2 22 O..— >2 >2 >2 >2 >2 >2 . . cue cu .eawocs an c0603 moon macn ussu “so noucwoa uo~omcsou >x .6o~noLa on» ozone mwca~oou :30 >6 :0 ouoe nomaooL Lo_omcaoU >6 cmuwo .oo .om .mm .hm .On .mn .Qm .mm .~n ._n .On .oo .3 .ne .3 .3 :3 .3 .3 .Hc 24]. U‘U‘UU‘O'U‘U‘U‘U‘U 'O‘U DEFINITELY TRUE MOSTLY TRUE .0 0000000000 0 MOSTLY FALSE (V DEFINITELY FALSE ..mo<¢ me2 ~26 DH 20 00 I I I I I I I I I I I I I I I I I I I I I s I I I I I I I I I I I o I I I I -- ’5 :‘ >L_:02L0 L:euuoa62 “006 020 on 00 unmao :06 n :u_1 awnmcoflum~ou\owafluue6 >6 00;: v~_~062 :0_omcsou >2 . . . . Odomemz new mwzu 0020>~v on 0coLLL>on\v:enm:£ >6 Lou accuuoaew me: .2 2000 00: 020 030 .muocuo 00 o>202mcom on ou >UI—mne :0 aoflo>wv _ Lazy ucnuuoae_ as: u“ unwsozu Lu—um:000 >2 I I I I I I I I I I I I I I O I I I I I I I I I I I I I U I I86 “0“ m“) U“ egg” a—zncouuw~0h :30 Lo ovsmoo 20m o>ec 00 vcv.uu>omeucnnmsz >6 Lou >axo shoe we: 02 01:0 .09» 00 vasoom Lodum:300 >2 030.2 50223 moo: ommodm .n> .:06 e no.) mwnm:0200_ou a :_ :oEoJ Lo :0603 00_LL12 .aucoeoueuu mo QJOLN 0x0: 050 00 :0 0m .00: 00 30> L0 .nu:~600nun 0:0 00 vcoamoc .00 00> CH .aaoLm use“ one. 022 30> LL 002000 0:4 ooh 0L0 uncoEooeum one :0603 L0 I I I I I I I I I I I I I I I I I I I I I I I I I I Inouca>u~ -3x~. ~v~e h0flamczou %: . wcd.mm:300 >6 Co omuaou ecu w:~uov 0600060u Logoncaou >6 and) acouun_su 00:20. on; 0 . . . . mmutmoLq >6 mcfi>q-n mm) Luz\6.: new cowuuauuun ~aaxon no sued >6 use» nouo_a:. 00.0m:300 >2 0:0 06 xme on voxud o>ng 0—30u uzxn .u:_uuom m:¢_omcsou a :2 006 0.:0q: a) ad 00;» amen uo—omcaou >2 . . .wvucna H .0_Lm:0201~ou . . . . . . . o_nmuuou600:: _oou 06 «one 200:) >a3 o :0 >000 >6 on noxoo_ hedoncsoo >2 . ounce 0a; awnucouus~ou m:L_omesou 0:0 Looms Lodem::oo >6 nous mcoiuo—ou “mason on: u . . . . . . amuse: a “on: son on m:_>~ uo\v:1 mcuuuudu .m:u>uu on: 00 noueuaou:o ma: 0 own: new on m:2>~ 00.0:m mcwuuu~u .m:0>uo can“ Congo n>o3 :s002 «6 000202 sodoncsou >2 . . . . . . . . . 2:03 on mu~216::0_mwuou 020 on v—aozn :06 easy 06 0200 :0_om::00 >2 I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I CQEOJ how 0' U0: “:9 >un6_u0 Luann Lo ouwmuao xon o>an 00 :06 so» ~0L300: an: “a yoga oo>ow~on 00—mmesou >2 mo>~om6050 ozone 000m _ouu 60:0 0106 00 :Onuon Locuocn 00000: :0603 00:0 xcwzu o» noEoon Lo_om::ou >2 . . . . . . . . . . . . . . . . . . .:0603 coco 0:06 :06 oLLEuu 00 voEoum uo~omzsoo >2 .cmEOJ Lo manoLm :240Lou 00 u:a>o.ou on >~:o ~22: m0:~600.un m:230_~02 och .~> .«s .O> .00 .mo .>o .00 .mo .40 .no .No .20 242 v UHF I N I TIC LY TRUE ‘J MOSTLY TRUE MOSTLY FALSE Q DEFINITELY FALSE 06. ...mmao mafia zmnh . . . . . . . . . . . . . . .c00n~_:0 >6 :00) 060: >000 :220 002000 ..00200 00 0w\x003 00 000020 >6 .0 0036000 002006 a a. >0nawanm >6 00:0_0m0:v 0020m:300 >2 . . . . . . 0000: 0.:00v~_:0 >6 00 0:0000 0600 023050 000000 26 0020 umumvwmaa 00:06600 0.00200:300 >2 >~0Eww >6 :0 can: :000auzv0\00>\000000 >6 :0 00:0000060 0006 00020 00: 0.3020 ~ 01:0 020“ 00200::00 >2 ._o .01 .92 .:000~0nu 0>1£ on: 200200 :_ :0603uc0603 u:MM003 I I I I I I I I I I I I I I I I I I I I I I I I I./~‘E‘\M a IU‘J hmghmu ‘- £U3“ wnhfluabm”UP—d mo AUd‘JU‘WW—fl 0:0 20 06 000:0600 00200::00 >6 .00000 0.60nnuq\1003 0000:0600 0.06 :_ 00000000 v0mma0ax0 _ :02: >6 0000::_o> 00 06 000 000000 0n.>06 00.0020 000m0mman 00200::ou >6 .00> a acuucaz 03001 003200 H :02: . . . . . . . . . . . . . . . . . . . .:0603 00 :0no m:owmm0u00q :0 waum 00 06 00m00300:0 00—0mcnou >2 I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I ICOdnm~U°bQ . “a axhoa “Gdh‘uwhuwm .mc0n0a: .m:_:0~00 mm 5030. «02000 0000:0600 0~060u «000300 06 :03: 00 006000 00.00::00 >2 I I ‘ I I I I I I I I I I I I I I I O I I I I I I I I I I I I I I I I I O I I I I O I I I£°fi ”ea” -3N a 0:00\~oozua 00 :03000\2oo:00 0003000m 00 :0 0m 00 :ownuuoo >6 no 0>000onaan «a: 00.000300 >2 . . m:o_0ao 000000 no 002002 0000:0600 02160u 0:0 0206 :00: mc00o~qx0 >6 no ”:0000000 «a: 0020m0300 >2 .mm .ho .cm .mo .00 .mm .0o_0m:300 0>020 50>: 0:000:00 000000 00\u:a nOfi‘unona 00x_~0 on: :0603 2000w0_ 0006 00 00 06 o0m10300:0 0:000u>0n\v:ann:£ >6 :02: :000000030u >6 0:000000:a 00 00600. 00—00::ou >2 0 I I I I n I o I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I ‘P—Oduoev >E “afifiwohax” how 06 0000000000 0:000u>on\u:anman >6 :02: uduu H 0000000030» 020 v:a00000:: 00 00600. 00~0o:000 >2 . O I I l I I O O I I I I I U C I C I . Uhocc~s ‘QUdflOd . C“ I I I I I I I I I I I I I I I I I I I I I awcmzow00200 030 :0.) 0~000 0:0.0u>on\vcaonac >6 :02: :omum0umn0u >6 0:10m00vcn 00 00600» 00.00::00 >2 . . 0000: m.u:000u>onm0cqnmsz >6 00 0:000m 0600 u—aocn mn00: >6 00:0 00000wman 00:06600 0.00.0n6300 >2 0 I I a u n I I I I I I I I I I l I I I I I I a I I I I I I ”6 he“ “C‘UMOQE‘ “a: “‘5“ -- Ho: UM“ “an 000:000 :30 02: no 00:00 a qo~0>00 00 vcu>0w>002vcmnman >6 00m 0:000006> «a: 00 uamaozu 00_0m:300 >2 I ‘2. I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I IWE®~DOBQ ha“ U30 xuo: flea U“ :‘ I‘Un 0” mm: .002m:0001_00 a :2 00:0 to; 0 :00000 >:0_00: 0006 0:0 0:20 I>0>_0n 00 006000 00~00:300 >2 I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I I xaaqg Ed; ”#15 O“ ”‘91 ”“06 ”n u>303 H 0020 on .nc000w>onxncanmac >6 0: >~_:u00m0 000002 00 06 90m10200c0 00—00::00 >2 .Nm ._o .02 .0n .m> .>n .0> 243 U‘U'OU U‘qu 'Ofi'U‘U B TRUI ULFINITHLY ozm mzw u n a u n a u o m u n a u n m u a a u n a u a a u n c u n m u n m u n a u n a u n a u n a u n a u n a E S L A E F ES ULVI R A L T F E T Y Y I L L N T T I m m T m m m \avcmnnsc LHacu zn nanoLa cmaaaHm mun >u~_uou Lao cH cwEOJ Ho muo— vcm ELoH .cumcou vvcmHa Lao; .wuH.ccoHum9:v cuuoHQEou L30> urn mama—m O a .me o. Ewnu cuauwu new waoHopco chauwu may cH Hausa coHuaELoHcH acaomeutm any .mCHutq_uHuLaa Log :9» xcqcpr . . . . . . . . . . . . . vQCLmucou 92 0“ £0.23 squ mcHLuo: m.uH 0m .mvcu.L.mon “as“ ~>~HH~A ou nyéwyn . sumac zuae cH «qu uLoHaxo go: v.6 ”an .vomanm >HkuHm>2n avEHumEOm on: H ”as“ Jvcx .mumHQ uHmm a on m>oe Lo :9803 vogouuen L9H ~50: « cucH xooH H umzu kumammsm vHLHxan mnzaomsc L_9£u mo cmuavn m:_on suHs a: Ham on o>mz no: uHo cufioa gas» now on we tea—w; QHano.La_mL o>Hmana.xHHauMcha a CH uw>Mum H x23 acomamu mg. ”a xooH ye quch oUd zmHJ Lo Amaucau we UH vaHou ”an LoHachOu A: LcHumcaou >z Lonmcaou xx Lonmcaou >2 LoHam:30u >2 .mc_ .40— udeo: meMHHflm . . . . . . . . . . . .voamu an: H can) mcHLnQJ an: H ”as: we vuxmm LoHon::ou xx .mucoHLumxw on” vmmoaco um; H HH vvxma ~;\m .Lonncaou x6 :UHJ mung v5“ vommaamHu H :95: L0H nuxmm cweoz “any ”on .manL mm m:_n. swan 0: mm: wuwcu uazu umHmcH uHsoz Lonmcaou A: . . wmcuwou HHom cH me30u a mcHxau Ho coHuao on“ we nuHJ nvmm:umHv Lonmcaou a: .moH .NOH .HQH .09— .mEHuuH> wmmM .Lo>H>L:m umuucH ca mm) H Hazy o>oHHun go: vHo Lonmcnou at .uuchquxm mzu vmxoncu van H HH noxmm uzxn .uoHomcaou x5 zuHI “moucH as“ vommaumHu H awn: . . . . . . . . . . . >HHE~H Lao :H amoucH ozu LoH Lozuoe >8 u~EaHn uoHuncsou xz . . . . . . . . . . . . . . . . . Lazuvwncmuu\wHucaxuuzu0un\uo£uaHmvun\uocuaH x6 kuanum »_m:Hzocx:: vac H Lasagna vuuuvcoa LoHomcaou x6 .muchLmaxo unmucH >5 uaonw vva~u H awn: . mLo>H>Lsm .moucH Lou mocha n wchHon Ho quHHanmoq any we LHHJ vwmmaumHv .cuLvHqu Lo; mucamvu Lw>uc :uucuoe voom: a was“ HuHHon he coHumuac we onHu; chu_H£u x5 mcHucwmoL moeHquOm Ho mwcHHmoH as vcaumLmvc: 05 quL on oHnm mm) . .canHHLU mchHmL nus 0&0; wcHxsum ucwucou up UHsocm H umzu unmaocu LoHomcaou x: .00 .mo .h¢ .00 .mo .mLo>H>L:m umvucH LonmCJOU >2 LonnCDOU x: LoHun:30u a: .no .mo .~o .cmuvHqu ALHx :on: APPENDIX C CLIENT-SUBJECT FOLLOW-UP LETTER 244 WANTED: COMPLETED QUESTIONNAIRES" REWARD: A CHANCE TO WM 550 Just complete-the enclosed questionnaire along with the Consent Form below and the Background Information sheet. Return them to me by Friday, March 11, 1983 and you will then be entered in the $50.00 cash drawing. *If you have recently returned your questionnaire to me, please disregard t is note. 3': 'k ‘k 7’: 3': 7‘: 3’: 4k 7': * CONSENT FORM I understand that: (please print name) l. I am volunteering in a research study conducted by Robin Sesan, M.Ed. 2. I am answering questions about one aspect of my counseling experience. 3. I do not have to participate if I do not wish to. 4. My responses will be kept confidential and will not be shared with my former counselor. 5. I can withdraw my participation at any time by contacting the researcher (at 374-8000, ext. 166). Your signature I would like a copy of the summary results: Yes No APPENDIX D THERAPIST-SUBJECT RESEARCH PACKET Robin Sesan, M.Ed. 6213 Balfour Lansing, MI 48910 (517) 393—2715 October 4, 1982 Dear Colleague: I have obtained permission from the Director of the Center to conduct a research study, under the supervision of Dr. William Hinds, for the purpose of gathering information about a number of aspects of women client's perceptions of and experiences in psycho— therapy. An area which I will be looking at is the experience and perception of sex- role stereotyping in psychotherapy. This is an area which has generated a good deal of controversy among clinicians and in the literature. Presently, there is no conclusive evidence regarding the existence of sex—role stereotyping in therapy, but much speculation. No study has directly looked at this area of women's experience and clearly there is a need to do so. For the study, I have developed a self-report questionnaire which will be sent to clients after they terminate their therapy. In addition to client participation, I am asking therapists to concurrently complete a measure regarding self-perceptions and perceptions of others on a variety of characteristics. The measure should take at most 45 minutes to complete. Your anonimity will be protected through the use of a coding system developed by a Center staff member other than myself. After assigning the codes, I will no longer have access to identifying information. I will receive your completed measure with only a code number. Client responses will also be coded and confidential. Therefore even if you request information regarding your former client's responses, this will not be available to you. While I cannot promise you any direct benefits from participation in the study, your responses along with those of other mental health professionals, will provide valuable information concerning women's experiences in therapy. Hopefully, these results will aide therapists in dealing with all clients. Although I will not be able to provide you with information regarding your individual participation, or with information regarding your former client's responses, you can receive a summary of the results at the conclusion of the study. If you would like these, you will be able to indicate this on the consent form included in this packet. Of course you are under no obligation to participate in the study, and you will suffer no consequences should you decide not to participate. Enclosed in this packet is a demographic information sheet, a copy of the measure to be completed, an answer sheet, a sealed envelope with your code number inside, a partici- pant consent form and two return envelopes. If you decide to participate in the study, I am asking that you return to me the demographic information sheet and the answer sheet in one envelope and the participant consent form in the other envelope by October 18, 1982. This two envelope system is another way to insure your anonimity. If you have any questions feel free to contact me at work (374-8000, extension 166) or at home. I thank you for your time and energy. Sincerely Yours, flaw JW Robin Sesan, M.Ed. Enclosures 246 Participant Consent Form.for Therapist-Subjects I understand that this study is being conducted by Robin Sesan, M.Ed, under the supervision of Dr. William.Hinds, for the purpose of gathering infbrmation about a number of aspects of women client's experiences in psychotherapy ‘ I understand that an area being studied is women client's perceptions of sex-role stereotyping in therapy. I understand that I will be completing a measure regarding self-percep- tions and perceptions of men and women on a number of characteristics. I understand that my participation in the study, as well as the participation of my fOrmer clients,will remain strictly confidential through the use of a coding system. I also understand that I will not have access to my individual responses or my former client's responses, but my obtain a summary report of the results at the conclusion of the study. I know that I am.under no obligation to participate in this study and will not be penalized if I refuse to participate. I also know that I may withdraw my consent to participate at any time by contacting the researcher. Signature Date I would like a copy of summary results Yes No Name Address 247 Demographic Information Sheet Please answer the following questions on this sheet and return this demographic information sheet in the same envelope as your completed answer sheet. 1) Code number 2) Sex 1) Male 2) Female 3) Agency Affiliation 1) Michigan State University Counseling Center 2) Ingham Community Mental Health Center 4) For how many years have you been practicing psychotherapy? years QUESTIONNAIRE 248 INSTRUCTIONS 0n the following pages are a series of S-point scales which describe a variety of psychological characteristics. For each one, you are to rate yourself on that characteristic. For example, how artistic are you? On the scale below very artistic is indicated at the far right and not at; all artistic at the far left. (The scale is marked with letters rather than numbers to correspond with the answer sheet.) Not at all artistic A ..... B ..... C ..... D ..... E ..... Very artistic If you think you are moderately artistic, your answer might be D; if you are very unartistic, you should choose A, etc. For each scale, select the letter on the scale that best describes you and indicate it on the answer sheet in the same column as the item number. Please be sure to answer every item. 10. ll. 12. 14. IS. 16. 17. Not at all aggressive Net at all independent Net at all emotional A. ..... B ..... C ..... D ..... E Does not hide: emotions at all A. ..... B ..... C ..... D ..... E. NOnconfionming to social expectations A. ..... B ..... C ..... D ..... E NOt at all considerate Not at all easily influenced Very'ungrateful very'submdssive DislikeSImath.and science verylnmfii A. ..... B ..... C ..... D ..... E Poor at sports A. ..... B ..... C ..... D ..... E Not at all excitable in a @919; crisis A. ..... B ..... C ..... D ..... f Not at all excitable in.auminor crisis A. ..... B ..... C ..... D ..... E very passive A. ..... B ..... C ..... D ..... E Not at all able to devote self completely to others A. ..... B ..... C ..... D ..... E very'blunt .A ..... B ..... C ..... D ..... E ‘Weak conscience A ..... B ..... C ..... D ..... E 249 Very aggressive Very independent Very emotional Almost always hides emotions (Ixfikmnfing to sDCial expectations Very considerate very easily influenced very grateful Very dominant Likes math and science‘very“much Good at sports very excitable in a rqflfigicrisis very excitable in.a ndnor crisis very active Able to devote self completely to others Very tactful very strong conscience l8. 19. 20. 21. 23. 24. 26. 27. 28. 29. 30. 31. 32. 33. 35. 36. Very rough Not at all helpful to others Not at all canpetitive Very home oriented Not at all skilled in business Knowsthewayof the world Not at all kind Low mechanical aptitude Indifferent to other ' s approval Feelings not easily hurt Not at all adventurous Not at all aware of feelings of others Not at all religious Not at all outspoka'i Not at all interested inscx Canmakes decisions easily Gives up very easily Very shy Never cries 250 A ..... B ..... C ..... D ..... E A ..... B ..... C ..... D ..... E A ..... B ..... C ..... D ..... E A ..... B ..... C ..... D ..... E A ..... B ..... C ..... D ..... E A ..... B ..... C ..... D ..... E A ..... B ..... C ..... D ..... E A ..... B ..... C ..... D ..... E A ..... B ..... C ..... " ..... E A ..... B ..... C ..... D ..... E ..... B.....C.....D.....E A ..... B ..... C ..... D ..... E Very gentle Very helpful to others Very competitive Very worldly Very skilled in business Does not lmow the way of the world Very kind High mechanical aptitude Highly needful of other ' s approval Feelings easily hurt Very adventurous Very aware of feelings of others Very religious Very outspoken Very interested in sex Has difficulty making decisions Never gives up easily Very outgoing Cries very easily O‘Ia. ‘c'd an, In 9... ,- '-.Ic l I .UJUXNJL- “mm. «3.13 .(1 db a. leaner? ’- o ‘J ..1 5r mat in 1er nits 21.;5. Very (13.23322; 1:“ 751.»: at all intellectual Not at all self~ 2.27: dent 35.-:13 very inferior 3...“. at all creative Always seas self as mamMg me show 1:13. ALB-Jays takes a stand [kphJOO}$OOOOICOOQOIDOOOCQE AOOOOOBOO..ICOIDOQDOQOUQE A.....B.....C.....D.....E AOCOOOBOVOOOCOOOOODODODDE AOOOOOBQOOVOCDOOOODOOOQOE A.JOCOBOOOOOCOOOOODOOOOOE A009.OBQOOOOCOO'OODOOOOOE AOOOOOBOQOOOCOOIOQDO'OOOE £z.-';~.E~1ot at all mderstanding of others -’:-'.'I Ves. cold in relations :91" others 1:2. Very little need for semrlty Ric-t at all ardoitimzs . 1'2 . Dislikes clfilo‘ren’ 377;. Does not exjoy art and music at all .1 feelings : 53. (3:) as re pieces unis: Erasure 5!}. Retiring '1. . list at all timid £BCD£ A.....B.....C.....D.....E A.....B.....C.....D.....E A.....B.....C.....D.....E AOOQIOBOOQOOCOOIOQDODOOOE ix...-oBovoooCoooooDoooooI-rl a. Easily mqn'esses tender AX""'B°.°"C'COOvDOooo0E ABCDL ? 'é‘0~0vaoogo.C.o...D..,,.E AOOOOOBCOC'. Amos}: always acts as a leader Very sloppy in habits Very loud Very intellectual Very self-cmfident Feels very supezi r Very creative “Nam seas self as naming the slow Never takes a stand Very understmdim of otlms Vary warm in relatims with others Very strong need for security Very ambitious Enjoys art and music verymch Boesmtexpresstmder feelingsatall Stands up well mfler pressure Pomerd Very timid 252 Male - Female Comparisons You will now be asked to consider the same attribute again. This time you will be given a description at only one end of the scale. In each case, you are being asked to compare the typical adult male and the typical adult female on that characteristic. For example, how artistic is the typical adult male in comparison to the typical adult female. Indicate the letter which best expresses your judgement on the answer sheet. (Note that the item numbers are continued from the previous set.) Example: Artistic: A B C D. E Midi more Slightly No Slightly more Much more dunecflmfistfl: uorelfide dififinenme Femfle dunecflmfistfl: oflkfle oflhmnle GO<11TO'DHSFEXTI¥£E 1:4” 05: (C: .J I 0 (SO. 61. '- 2 more 8 (ilaracteristic .T. lale A l) i: l. Clxr '.'“,g'fl‘g? Hal-3. Ag? _;;r:(-: 3.51' .13.: Illdszr-a limit L315! ..‘lolldl Hides u on”! ma Quill '.'O‘u’ 31.11,“; L0 130 ('20: 13 idet'a Ct. Easily inflamed Grateful Dardnmxt Likes math and science mry m 103 ‘1 Cmd at a sports writable in a 1.1-137', if»; Ez>tei'tz'xl::1.~:-3 51>... ’ mi: 3’ ‘1‘ -w —.—-—. ‘fia "f, STEVE Able to demote self znlqflrytely i‘a 2t: fill Straw, corsciezu‘ {5:91 11:21:: 1‘ llszlfg‘fifUl to 0t: IBIS (imam BIL-$1198. lime oriental Fkilled in basin. ":3 RI was the way of the “arid Kind High Euchariical a;jrtiiule 1.1.31 azegeztatiolm 25 (I L' J! $3.1: (1.x..‘enxxs: 9'" 3 Q' I '~ (1 J. ,. 8.7. 94 . to or} era 95. '36 _ 93 . I l 99 . ()0 .. 01 . 3.02 .. 103 . ‘ .L ()4... 105. ’rurh W119 Cl 3.14 fintflfl. 1..) 7-1.0. (IF Blew-53:) (.1 U’l Lil: / Mr L-' 3.'.(J'kllr'1..: .C Hes-clifui of ‘:)'§.'."l»:.a.."a3 approval Fuel}; "43";5.’ c.5131 ly in it} i“. AIh-‘Li'l: J prom; r: [4.2-rare of feel in s t l others 11-1-1; gious Ceramic-an Interested in sex ?' Takes decisions easily Gives up vex er}; easily 3} 13' (kills; easily 23.15 Acts a leader Sloppy in habit}; ( 'fuiet Intellectual Self- omfide: ‘.I'. }"<;:els superior Creative Seas self as ruining t. =2 show lib-rays tales a stand Yin-11ers finding of camera Nair. in relatim‘ss with others Strong need for security Inabitious Likes chi ldren 254 A B C Much mre Slightly more No characteristic Male difference of Male 106. Enjoys am: and music very much 107. Express tender feelings 108 . Stands up well under pressure 109. Forward llO. Timid Slightly more Female Much more characteristic of Female APPENDIX E THERAPIST-SUBJECT FOLLOW-UP LETTER r‘ l. f' /_ -.‘o 3- Robin 3. Sesan 6213 Balfour Lansing. MI October 27, 1982 Dear Colleague, Recently. you received a questionnaire packet from me with a request for you to participate in dissertation research which I'm conducting on women client's experiences in psychotherapy. While I have received many completed questionnaires from therapists. I have not yet received one from you.* It is important for my study that a large number of therapists participate: so if my questionnaire has slipped your mind or if it has been set away as a low priority. would you please reconsider participating? Your 30 minutes of involvement will help to provide valuable informa- tion regarding women client's experiences in therapy, as well as direction for the ways we as clinicians can best serve our client's needs. If you've misplaced your packet or need certain items from it. please do not hesitiate to call me at work (374-8000 ext.l66) or home (393-2715) so that I can get the item(s) to you. Thank you for reconsidering. Sincerely yours. [(2 ii: ’4, ’. 171< I; ..t. ‘3' ( \ILcr'll’J/‘J / Robin Sesan. M.Ed. * If you have recently mailed a completed questionnaire back to me, it is most likely that our letters have crossed in the mail. Please disregard the request in this letter and thanks for your participation. APPENDIX F GUIDELINES FOR NONSEXIST THERAPY 10. 256 GUIDELINES FOR NONSEXIST THERAPY (APA, 1978) The conduct of therapy should be free of constrictions based on gender defined roles, and the options explored between client and practitioner should be free of sex- role stereotypes. Psychologists should recognize the reality, variety, and implications of sex discriminatory practices in society and should facilitate client examinations of options in dealing with such practices. The therapist should be knowledgable about current empirical findings on sex roles, sexism, and individual differences resulting from the client's gender-defined identity. The theoretical concepts employed by the therapists should be free of sex bias and sex-role stereotyping. The psychologist should demonstrate acceptance of women as equal to men by using language free of derogatory labels. The psychologist should avoid establishing the source of personal problems within the client when they are more properly attributable to situational or cultural factors. The psychologist and a fully informed client mutually should agree upon aspects of the therapy relationship such as treatment modality, time factors, and fee arrangements. While the importance of the availability of accurate information to a client's family is recognized, the privilege of communication about diagnosis, prognosis, and progress ultimately resides with the client, not with the therapist. If authoritarian processes are employed as a technique, the therapy should not have the effect of maintaining or reinforcing stereotypic dependency of women. The client's assertive behaviors should be respected. ll. 12. 13. 257 The psychologist whose female client is subjected to violence in the form of physical abuse or rape should recognize and acknowledge she is the victim of a crime. The psychologist should recognize and encourage exploration of a women client's sexuality, and should recognize her right to define her own sexual preferences. The psychologist should not have sexual relations with the client, nor treat her as a sex object. REFERENCES 258 REFERENCES Abramowitz, C. Blaming the mother: An experimental investi- gation of sex-role bias in countertransference. Psychology of Women Quarterly, 1977, g, 24-34. Abramowitz, S., Abramowitz, C., Jackson, C., & Comes, B. The politics of clinical judgment: What nonliberal examiners infer about women who do not stifle them- selves. Journal of Consulting and Clinical Psychology, 1973, fil,384-391. Abramowitz, S., Weitz, L., Schwartz, J., Amira, S., Comes, B., & Abramowitz, C. Comparative counselor inferences toward women with medical school aspira- tions. Journal of College Student Personnel, 1975, £9: 128-130. Abramowitz, S., Abramowitz, C., Roback, H., Corney, R., & McKee, E. Sex-role related countertransference in psychotherapy. Archives of General Psychiatry, 1976, 33, 7l-73. Abramowitz, S., Abramowitz, C., & Weitz, L. Are men therapists soft on empathy?’ Journal of Clinical Psychology, 1976, 2;, 434-437. Abramowitz, S., Roback, H., Schwartz, J., Yasuna, A., Abramowitz, C., & Comes, B. Sex bias in psychotherapy: A failure to confirm. American Journal of Psychiatry, 1976, ;;3, 706-709. Abramowitz, S., & Abramowitz, C. Sex-biased researchers of sex bias in psychotherapy and impartial reviewers. American Psychologist, 1977, 1;, 893-894. Abramowitz, S. The sexual politics of sex bias in psycho- therapy research. Micropolitics, 1982a, g, 21-34. .Abramowitz, S. Sex bias research: A game of sexual politics for mature men and women. Micropolitics, 1982b, g, 43-49. .American Psychological Association. Report of the task force on sex bias and sex-role stereotyping in psycho- therapeutic practice. American Psychologist, 1975, ;Q, 1169-1175. 259 American Psychological Association. Guidelines for therapy with women. American Psychologist, 1978a, 33, 1122-1133. ‘_— American Psychological Association. Source materials for nonsexist therapy. JSAS Catalog of Selected Documents in Psychology, 1978b, g, 40. American Psychological Association. Ethical Principles for Psychologists, Washington, D.C.: APA, 1981. Anastasi, A. Psychological testing. New York: MacMillan Pub. Co., 1976. Aslin, A. Feminist and community mental health center psychotherapists' expectations of mental health of women. Sex Roles, 1977, 3, 537-544. Bakan, D. The quality of human existence. Chicago: Rand McNally, 1966. Barrett, C., Berg, P., Eaton, E., & Pomeroy, E. Implications of women's liberation and the future of psychotherapy. Psychotherapy: Theory, Research and Practice, 1974, 11, 11-15. Bart, P. The myth of a value-free psychotherapy. In W. Bell & J. Mann (Eds.) Sociology and future. New York: Russell Sage Foundation, 1971. Bem, S. The measurement of psychological andro yny. Journal of Consulting and Clinical Psychology, 197 , 4g, - 162. Bem, S. Sex-role adaptability: One consequence of psycho- logical androgyny. Journal of Personality and Social Psychology, 1975, 31, 634-643. Bernardez, T. Unconscious beliefs about women affecting péychotherapy. Paper presented at the annual meeting 0 the American Psychiatric Association, Anaheim, CA, May 1975. Bernardez, T. WOmen and anger: Conflicts with aggression in contemporary women. Journal of the American Medi- cal WOmen's Association, 1978, 33,215-219. Belote, B. Sexual intimacy between female clients and male psychotherapists: ‘Masochistic sabotage. Unpublished' Ph.D. dissertation,*Ca1ifornia School of Professional Psychology, 1974. 260 Billingsley, D. Sex-bias in psychotherapy: An examination of the effects of client sex, client pathology and therapist sex on treatment planning. Journa of Consulting and Clinical Psychology, 1977, 45, 250-256. Block, J. Conception of sex role: Some cross-cultural and longitudinal perspectives. American Psychologist, 1973, gg, 512-526. Bograd, M. Battered women, cultural myths and clinical interventions: A feminist analysis. Women and Therapy, 1982, I, 69-77. Borg, W., & Gall, M. Educational Research. New York: Longman, 1979. Bowman, P. An analog study with beginning therapists sug- gesting bias against "activity" in women. Ps cho- therapy: Theory, Research and Practice, 19 , __, 318-324. Broverman, 1., Broverman, D., Clarkson, F., Rosenkrantz, P., & Vogel, S. Sex-role stereotypes and clinical judg- ments of mental health. Journal of Consulting and Clinical Psychology, 1970, 24, 1-7} Broverman, 1., Broverman, D., Clarkson, F., & Rosenkrantz,!h Sex-role stereotypes: A current appraisal. Journal of Social Issues, 1972, gg, 59-78. Brown, C., & Hellinger, M. Therapists' attitudes towards women. Social WOrk, 1975, 39, 266-270. Buczek, T. Sex biases in counseling: Counselor retention of the concerns of a female and male client. Journal of Counseling Psychology, 1981, gg, 13-21. Butler, 8., & Zelen, S. Sexual intimacies between thera- pists and patients. Psychotherapy: Theory, Research and Practice, 1977, 14, 139-145. Carlson, R. Understanding women: Implications for personality theory and research. Journal of Social Issues, 1972, gg, l7-3l. Carmen, E., Russo, N., & Miller, J. Inequality and women's mental health: An overview. American Journal of Psychiatry, 1981, 138, 1319-1330. 261 Chesler, P. Patient and patriarch: Women in the psycho- therapeutic relationship. In V. Gornick and B. Moran (Eds.) Woman in a Sexist Society, New York: Basic Books, 1971} Chesler, P. Women and madness. New York: Doubleday, 1972. Collier, H. Counseling women. New York: The Free Press, 1982. Cooperstock, R. Sex differences in psychotropic drug use. Social Science Medicine, 1978, 12, 179-186. Costrich, N., Feinstein, J., Kidder, L., Maracek, J., & Pascale, L. When stereotypes hurt: Three studies of penalties for sex-role reversals. Journal of Experi- mental Social Psychology, l975, 11, 520-530. Cowan, G. Therapist judgments (xf'clients' sex-role igbigzs. Psychology of Women Quarterly, 1976, 1, Cronbach, L. Coefficient alpha and the internal structure of tests. Psychometrika, 1951, 16, 297-334. Dahlberg, C. Sexual contact between patient and therapist. Contemporary Psychoanalysis, 1970, 6, 107-124. Davenport, J., & Reims, N. Theoretical orientation and attitudes towards women. Social Work, 1978, 23, 306-309. Davidson, C., & Abramowitz, S. Sex bias in clinical judg- ment: Later empirical returns. Psychology of WOmen Quarterly, 1980, 4, 377-395. Del Gaudio, A., Carpenter, P., & Mbrrow, G. Male and female treatment differences: Can they be generalized? Journal of Consulting and Clinical Psychology, 1978, 46, 1577-1578. Delk, J., and Ryan, T. Sex-role stereotyping and A-B therapist status: Who is more chauvinistic? Journal of Consulting and Clinical Psychology, 1975, 43, 589. Erikson, E. WOmanhood and inner space. In J. Strouse (Ed.) WOmen and analysis. New York: Grossman Publishers, 1974. Etaugh, C., & Kasley, C. Evaluating competence: Effects of sex, marital status and parental status. Psychology of WOmen Quarterly, 1981, 9, 196-203. 262 Fabrikant, B. The psychotherapist and the female patient: Perception, misperceptions and change. In V. Franks and V. Burtle (Eds.) Women in therapy, New York: Brunner/Mazel, 1974. Festinger, L. A theory of cognitive dissonance. Stanford: Stanford University Press, 1957. Fidell, L. Sex differences in psychotropic drug use. Professional Psychology, 1981, 12, 156-162. Franks, V., & Rothblum, E. (Eds.) The stereotyping of women: Its effects on mental health. New York: Springer PubliShing Co., 1983. Freud, S. The dissolution of the oedipus complex. In J. Strachey (Ed.) The standard edition of the complete psychological works of’Sigmund Freud) V61. IX}*London, Hogarth Press (1924), 1964. ’ Gelinas,D. Identification on treatment of incest victims. In E. Howell and M. Bayes (Eds.) WOmen and mental health, New York: Basic Books, Inc., 1981. Gilbert, L. The sexist therapistz' An ephemeral species? American Psychologist, 1977, 32, 888-889. Gilbert, L. Feminist therapy. In A. Brodsky and R. Hare- Mustin, WOmen and psyghotherapy: An Assessment of research and practice, New York: GuilfOrd Press, 1980. Gilbert, L. Toward mental health: The benefits of sycho- logical gndrogyny. Professional Psychology, 19 la, 12, 29-3 . Gilbert, L. Perceptions of sex roles: The influences of client gender and student-counselor attitudes. Counselor Education and Supervision, 1981b, 21, 57-62. Gilligan, C. Woman's place in man's life cycle. Harvard Educational Review, 1979, 49, 431-446. Gilligan, C. Restoring the missing text of women's develop- ment to the life cycle theories. In McGuigan (Ed.) WOmen's lives, Ann Arbor, MI, University of Michigan Press, 1980. Gilligan, C. In a different voice. Cambridge, MA, Harvard University Press, 1982. Gilman, R. The feminist liberation case against Sigmund Freud, New York Times Magazine, January 31, 1971. 263 Class, C., and Stanley, J. Statistical methods in education pgdopsychology. New Jersey: Prentice-Hall, Inc., Gomes, B., & Abramowitz, S. Sex-related patient and Egggapéstlefgects on clinical judgment. Sex Roles, Cove, W. Mental illness and psychiatric treatment among gzgegél Psychology of Women Quarterly, 1980, 4, Gove, W., & Tudor, J. Adult sex roles and mental illness. American Journal of Sociology, 1973, 11, 812-835. Greenspan, M. A new approach to women and therapy. New York: McGraw-Hill Book Co., 1983. Hammer, E. Creativity and feminine ingredients in young male artists. Perceptual and Motor Skills, 1964, 12, 414-421. Hare-Mustin, R. What is implied? American Psychologist, 1977, 33, 889-890. Hayes, K., & Wolleat, P. Effects of sex in judgments of a simulated counseling interview. Journal 0 Counseling Psychology, 1978, 33, 164-168. Hefner, R., Rebecca, M., & Oleshansky, B. Development of sex-role transcendence. Human Development, 1975, 33, 143-156. Helms, J. Counselor reactions to female clients: Generali- zation from analogue research to a counseling setting. Journal of Counse ing Psychology, 1978, 33, 93-199. Helson, R. Personality of women with imaginative and artistic interests: The role of masculinity, origi- nality and other characteristics in their creatiVity. Journal of Personality, 1966, 34, 1-9. Helwig, A. Counselor bias and women. Journal of Employ- ment Counseling, 1976, 33, 58-67. Herman, J., & Hirshman, L. Father-daughter incest. SIGNS: Journal of women in Culture and Sociegy, 1977, _2—735”, -756. Hilberman, E. Overview: The ”wife beater's wife" reconsidered. American Journal of Psychiatry, 1980, 137, 1336-1347. Hiler, E. An analysis of patient-therapist compatibility. Journal of Consulting Psychology, 1958, 33, 341-347. 264 Hill, C. Sex of client and experience level of counselor. Journal of Counseling Psychology, 1975, 33, 6-11. Hill, C., Tanney, M., Leonara, M., & Reiss, L. Counselor reactions to female clients: Type of problem, age of client and sex of counselor. Journal of Counselipg Psychology, 1977, 34, 60-65. Hofball, S., & Penner, L. Effects of physical attractive- ness on therapists' initial judgments of a person's self-concept. Journal of Consulting and Clinical Psychology, 1978, 43, 200-201} Horney, K. The dread of women. International Journal of Psychoanalysis, 1932, 33, 348-3602 Horney, K. On the genesis of the castration complex in women. In J. Miller (Ed.) Psychoanalysis and women. New York: Brunner/Mazel, 1973. Howard, K., Orlinsky, D., & Hill, J. The therapist's feelings in the therapeutic process. Journal of Clini- cal Psychology, 1969, 33, 83-93. Howard, K., Orlinsky, D., & Hill, J. Patient's satisfaction in psychotherapy as a function of patient-therapist pairing. Psychotherapy: Theory, Research, and Practice, 1970, 1, 130-134. Howell, E. Women: From Freud to the present. In E. Howell & Bayes (Eds.) WOmen and mental health, New York: Basic Books, Inc., 1981. Holroyd, J., & Brodsky, A. Does touching patients lead to sexugl intercourse? Professional Psychology, 1980, 11, 07-811. Hughes, R., & Brewin, R. The tranquilizing of America: Pill popping and the American way offlife. New York: Harcourt, Brace, Janovich, 1979. Isaac, S. Handbook in research and evaluation. San Diego: Edits,’197l. Jackson, D., Block, J., Block, J., & Patterson, V. Psychiatric conceptions of the schizophrenic parent. AMA Archives of Neurology and Psychiatry, 1958, 19, 448-459. Johnson, M. Influence of counselor gender on reactivity to clients. Journal of Counseling Psychology, 1978, 33, 359-365. 265 Johnson, R. Elementary statistics. North Scituate, MA: __— Duxbury Press, 1980. Jones, E., & Zoppel, C. Impact of client and therapist gender on psychotherapy process and outcome. Journal of Consulting and Clinical Psychology, 1982, 39, 259-272. Kahn, L. Effects of sex and feminist orientation on therapists' clinical judgments. Dissertation Abstracts International, 1976, 3Q, 3613B. Kaplan, A. Androgyny as a model of mental health for women. From theory to therapy. In A. Kaplan & J. Bean (Eds.) Beyond Sex-Role Stereotypes, Boston: Little Brown_ &6Co., 1976. Kaplan, A. Clarifying the concept of androgyny: Implica- tions for therapy. Psychology of women Quarterly, 1979a, 3, 223-230. Kaplan, A. Toward an analysis of sex-role related issues in the therapeutic relationship. Psychiatry, 1979b, 42, 112-120. Kellerman, J. Sex-role stereotypes and attitudes towards parental blame for the psychological problems of children. Journal of Consulting and Clinical Psychology, 1974, 43, 153-154. Kirshner, L., Genack, A., & Hauser, S. Effects of gender on short-term psychotherapy. Psychotherapy: Theory, Research and Practice, 1979, 33, 158-167} Kohlberg, L. A cognitive-developmental analysis of children's sex-role concepts and attitudes. In E. Maccoby (Ed.) The develoPment of sex differences, Stanford: Stanford University Pfess, 19661 Lerman, H. What happens in feminist therapy? In S. Cox (Ed.) Female psychology: The emerging self, Chicago: Science Research Association, 1976. Lerner, H. Early origins of envy and devaluation of women: Implications for sex-role stereotypes. Bulletin of the Meninger Clinic, 1974, 33, 538-553. Levine, S., Kamin, L., & Levine, E. Sexism and psychiatry, Journal of Orthopsychiatry, 1974, 44, 327-336. 266 Maccoby, E. The development of sex differences. Stanford: Stanford University Press, 1966. Maracek, J., & Johnsbn” 14. Gender and the process of therapy. In A. Brodsky and R. Hare-Mustin (Eds.) Women and psychotherapy: An assessment of research and practice, Newaork: Guilford Press, 1980. Maracek, J., & Kravetz, D. Women and mental health: A review of feminist change efforts. Psychiatry, 1977, 49, 323-329. Maracek, J., Kravetz, D., & Finn, S. A comparison of women who enter feminist therapy and women who enter tradi- tional therapy. Journal of Consulting and Clinical Psychology, 1979, 41, 734-742. Maslin, A., & Davis, J. Sex-role stereotyping as a factor in mental health standards among counselors-in- training. Journal of Counseling Psychology, 1975, 33, 87-91. Maslow, A. Self-esteem and sexuality in women. Journal of Social Psychology, 1942, $9: 259-294. Maslow, A. Toward a psychology of being. New York: D. Van NOstrand, 1968. Masters,W., & Johnson, V. Human sexual inadequaoy, Boston: Little Brown &‘C6., 1970. Maxfield, K. The double standard: Keeping female clients in auxiliary roles. Paper presented—at American Personnel andiGuidance Association annual convention, Detroit, MI, March 1982. McNair, D., Lorr, M., & Callahan, D. Patient and therapist influences on quitting psychotherapy. Journal of Consulting and Clinical Psychology, 1963, 31, IU-17. Mehrens, W., & Lehmann, I. Measurement and evaluation in education and psychology. New York: Holt, Rhinehart and Winston, 1978. Miller, D. The influence of the patient's sex on clinical judgment. Smith College Studies in Social WOrk, 1974, 44, 89-100. Miller, J. Toward a newspsychology of women. Boston: Beacon Press, 1976. Orlinsky, D., & Howard, K. Therapy session report: Form P, Chicago: Institute for JuvenileiResearCHj’1966. 267 Orlinsky, D., & Howard, K. The effects of the sex of the therapist on the therapeutic experiences of women. Psychotherapy: Theory, Research and Practice, 1976, 13, 82-88. Orlinsky, D., & Howard, K. Gender and psychotherapeutic outcome. In A. Brodsky and R. Hare-Mustin (Eds.) Women and psychotherapy: An assessment of research and practice. New York: Gfiilfordiness, 1980. Oyster-Nelson, C., & Cohen, L. The extent of sex bias in clinical treatment recommendations. Professional Psychology, 1981, 33, 508-515. Persons, R., Persons, M., & Newmark, I. Perceived helpful therapist characteristics, client improvement and sex of therapist and client. Psychotherapy: Theory, Research and Practice, 1974T11, 63-65. Rawlings, E., & Carter, D. Psychotherapy for women: Treatment towards equality. Springfield, MA: Charles C. Thomas,61977l Rebecca, M., Hefner, R., & Oleshansky, B. A model of sex- role transcendence. In A. Kaplan and J. Bean (Eds.) Beyond sex-role stereotypes, Boston: Little Brown & Co.,‘1976. Rice, J., and Rice, D. Implications of the women's libera- tion movement for psychotherapy. American Journal of Psychiatry, 1973, 130, 191-196. Rieker, P., & Carmen, E. Teaching value clarification: The example of gender and psychotherapy. American Journal of Psychiatry, 1983, 140, 410-415. Rosenthal, R., & Jacobson, L. Pygmalion in the classroom, New York: Holt, Rinehart anHYWinston, Inc., 1968. Schwartz, J., & Abramowitz, S. Value-related effects on psychiatric judgment. Archives of General Psychiatry, 1975, 33, 1525-1529. Schwartz, J., & Abramowitz, S. Effects of female client attractiveness on clinical judgment. Psychotherapy: Theory, Research and Practice, in press. Seidenberg, R., & DeCrow, K. WOmen who marry houses. New York: McGraw-Hill PuinShers, 1983. Sellitz, C., Wrightsman, L., & Cook, 8. Research methods in social relations. New York: Holt, Rinehart and Winston, 1976. 268 Shapiro, J. Socialization of sex roles in the counseling setting: Differential counselor behavioral and attitudinal responses to typical and atypical female sex roles. Sex Roles, 1977, 3, 173-184. Sherman, J. Therapist attitudes and sex-role stereotyping. In A. Brodsky and R. Hare-Mustin (Eds.) Women and psychotherapy: An assessment of research andspractice. New YOrk: Guilforngress, 1980. Smith, M. Sex bias in counseling and psychotherapy. Psychological Bulletin, 1980, 31, 392-407. Spence, J., Helmreich, R., & Stapp, J. The Personal Attributes Questionnaire: A measure of sex-role stereotypes and masculinity-femininity. JSAS Catalog of Selected Documents in Psychology, 1974, 4:643. Spence, J., Helmreich, R., and Stapp. J. Ratings of self and peers on sex-role attributes and their relation to self-esteem and conceptions of masculinity and femininity. Journal of Personality and Social Psychology, 1975, 33, 29-39. Spence, J. Personal communication, April 1982. Stearns, B., Penner, L., & Kimmel, E. Sexism.among psycho- therapists: A case not yet proven. Journal of Consult- ing and Clinical Psychology, 1980, 43, 548-550. Stein, L., Del Gaudio, A., & Ansley, M. A comparison of female and male neurotic depressives. Journal of Clinical Psychology, 1976, 33, 19-21. Stricker, G. Implications of research for the psycho- therapeutic treatment of women. American Psychologist, 1977, 33, 14-22. Symonds, A. Violence a ainst women: The myth of masochism. American Journal 0 Psychotherapy, 1979, 33, 161-173. Thomas, A., & Stewart, N. Counselor response to female clients with deviate and conforming career goals. Journal of Counseling Psychology, 1971, 33, 352-357. Thompson, C. Notes on the psychoanalytic si nificance of the choice of an analyst. Psychiatry, 1 38, _l_._, 205- 216. Thompson, C. Penis envy in women. Psychiatry, 143, s, 123-125. 269 Voss, J., & Gannon, L. Sexism in theory and practice of clinical psychology. Professional Psychology, 1978, 9, 623-632. Walker, L. Battered women. In A. Brodsky and R. Hare- Mustin (Eds.) women and psychotherapy: An assessment of research and practice. New York: Guilford Press, 1980. Walker, L. Battered women: Sex roles and clinical issues. Professional Psychology, 1981, 33, 81-91. Waskow, I. Research on psychotherapy with women: A work- shop introduction. Psychotherapy: Theory, Research, and Practice, 1976a, 33, 64-65. Waskow, 1. Summary of discussion following the workshop. Psychotherapy: Theory, Research and Practice, 1976b, 13, 96-98. Whitley, B. Sex roles and psychotherapy: A current appraisal. Psychological Bulletin, 1979, 39, 1309- 1321. Williwms, E. Notes of a feminist therapist. New York: Dell Publishing C6., 1976. wright, C., Meadow, A., Abramowitz, S., & Davidson, C. Psychiatric diagnoses as a function of assessor profession and sex. Psychology of WOmen Quarterly, 1980, 3, 241-254. , Zeldow, P. Sex differences in psychiatric evaluation and treatment: An empirical review. Archives of General Psychiatry, 1978, 33, 89-93. MICHIGAN smTE UNIV. LIBRARIES 1|V||WI“HIIWIWNIWHIHWIWHIIIWW 31293105929107