MSU LIBRARIES —_ RETURNING MATERIALS: PIace in book drop to remove this checkout from your record. FINES will be charged if book is returned after the date stamped beIow. mm #1332 ANDROGYNYANDINI'ERPERSONALCH’IPEI‘ENCE By Lisa Sharron Blank A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Psyclnlogy 1984 ABSTRACT ANIBOGYNYANDINIERPERSONALME'IH‘ICE By Lisa Sherron Blank The relationship of sex-type to interpersonal coupetence, defined as behavioral measures of self-acceptance and acceptance of others, was explored in the context of small, interpersonal groups, thought likely to reflect commal and agentic skills and values. Ben's Sex Pole Inventory was administered at two points during each of four term of the course featuring these groups after about 22 and 50 hours of group participation. Near these times self— and peer-ratings of each participant' s interpersonal behavior were also collected. Strong associations of masculinity with self-acceptance, and femininity with acceptance of others, were revealed in both self- and peer-ratings . Androgynous persons were depicted by self and peers as more accepting of both self and others than were their nesmiline, feninine and undifferentiated peers. The four sex-types showed the predicted and some surprising patterns of interpersonal competence. Implications for future research were discussed. AWE First acknowledgements go to the students and facilitators of the Psychology 400 groups who were willing and interested participants in this research. I am also gratefi1l to the friends and fellow graduate students who offered very practical help and warm support through all stages of the process: to Jeff Roach, wtnse enthusiasm and patience helped shape the proposal; to Miriam for her introduction to the computer; to Sallie and Julie for their companionable presence and ready help during long summer days at the computer center; and to Jim, Sallie, Wes and my sister Leslie who shared my anxieties, discouragement and -- finally, the joy of a rite of passage completed, when enough is really, finally, m. For my committee members, in all their varied roles, a special thanks: to Don Grummon, teacher, supervisor and fellow staff member; to Elaine Donelson, whose interest in personality, values, psychology of women and spirituality has fostered and actively supported my own; and especially to John Hurley, who, as chairperson, advisor, teacher, supervisor and real friend, gave me both support and autonomy throughmy experience ofhis confidence inme. A final thanks to Suzy Pavick for careful work and caring. ii TAKEOFCXNI‘ENTS LIST OF TABLES ........................ LIST OF FIGURES ....................... INTHDDUCTION ......................... METHOD ............................ Participants ...................... Procedure ........................ Measures ........................ Ban Sex Role Inventory (BSRI) ........... ARO and ABS Group Behavior Ratings ......... RESUL'I‘S ........................... DISCUSSION .......................... iii Page iv Table LIST OF TABLES Classification of 88 Participants into BSRI Categories by Data Collected Early and late Each Tenn .................... Comparisons of the Masculinity, Femininity and Social Desirability Scores on the Early and late BSRIs ...................... Intercorrelations Among All Measures ......... Means and Standard Deviations (in Parentheses) of Early and late ARS/ARO Ratings by Groups ..... iv Page 16 17 19 22 LISI‘ OF FIGURES Figure Page 1. Patterns of ARS/ARO responding by sex-type using early data .................... 23 2. Patterns of ARS/ARO responding by sex-type using late data .................... 24 INTRODUCTION In the past ten years, the development of the concept of psychological androgyny and instruteits to identify and measure it has contributed to an increased interest in sex-role research. This response has been considered an indication of how well the concept captured and expressed the Zeitgeist in sex-role research, and has resulted in a considerable, and sometimes bewildering, array of 1 As this research comes of age, there has been increased findings. concern about the integration of androgyny and androgyny research within the broader field of social/personality theory and research (Lermey, 1979). Within this larger context, one way of exploring research on the relationship of androgyny to adjustmmt might be through increased attention to criterion and contextual variables: adjustment by what criteria and in what context? These concerns need to be kept in mind when examining the relevant androgyny literature. Early research on sex-role identity and maculinity and femininity essentially assured a bipolar model of masculinity and femininity where those qualities were considered opposite ends of a single contimmmn: individuals were either masculine or feminine but not both (Constantinople, 1973). Tests such as the Terman Miles M—F Test (1936) , the M—F Scale of the Strong Vocational Interest 2 Blank (1943), the Mf scale of the Minnesota Multiphasic Personality Inventory (Hathaway & IVEKinley, 1943), and the Femininity Scale of Gough's California Personality Inventory (1964) , which had been developed to assess masculinity and femininity reflected this bipolar concept of scale construction, and permitted classification of individuals only as masculine males , feminine females , or sex-reversed deviants. Since the primary criterion for item selection on these older tests was differential responses by males and females, items tapped sex-typed behaviors as well as personality traits. This basic model of masculinity and femininity had been supported by the theoretical assumption that mental health and successful adjustment depended upon the internalization of the appropriate sex—role identity along with gender-appropriate behavior. More recent research has criticized the previous lack of conceptual and empirical distinction emng important variables of biological sex, gender identity, sex-role behaviors , and psychological characteristics of masculinity and femininity by researchers and psychometricians, and challenged the bipolar model itself, along with its underlying assunptions. Ben (1974) introduced the concept of androgyny with her development of a new sex-role inventory that treated masculinity and femininity as logically and empirically independent dimensions, making it possible to characterize individuals as masculine, feminine, or androgynous (originally as a function of the difference between the endorsement of masculine and feminine personality characteristics). The Ben Sex Role Inventory (BSRI) had a masculinity scale, a femininity scale, and 3 a social desirability scale. These scales were constructed to reflect consensually validated socially desirable traits for men and women-- the extent to which these traits are endorsed g3}; men and men, rather than by them in everyday behavior. Other scales based on the new model and designed to measure androgyny were the Personality Attributes Questionnaire (PAQ) (Spence, Helmreich, & Stapp, 1974) , the PRF scale (Berzins, Welling, & Wetter, 1978) and the Androgyny measure from the Adjective Check List (Heilbrun, 1976) . Ben's interest in androgyny arose fram a concern with, and awareness of, the debilitating effects associated with strict conformity to sex-stereotypic behaviors (Ben, 1975) . Thus, her view of anch‘ogyny as tlne combined presence of socially valued, stereotypic masculine and feminine characteristics also challenged earlier views of adjustment by predicating that people with anckogynous self-concepts might more freely engage in a wider range of both "masculine" and "feminine" behaviors according to their situational appropriateness, i.e. , be more flexible and adaptively effective in dealing with their world than their sex-typed peers . In contrast to the traditional assumption that strongly sex—typed individuals exenplified mental health, Bem proposed that androgynous individuals may come to define "a more human standard of psychological health."2 A series of studies carried out in laboratory settings (Ben, 1975; Ben & Lemney, 1975; Ben, Martyna, & Watson, 1976; Ben, 1977) supported these notions of greater situationally—appropriate flexibility and broader behavioral reperto ires for androgynous versus sex-types subjects in response to specific tasks. 4 Investigating the hypothesis that androgynous flexibility makes for better adjustment, Jones, Chernovetz, and Hansson (1978) looked at the relative adjustment levels of 1404 subjects as a function of their psychological sex. Their findings across measures of five areas of psychological fmnctioning showed that flexibility and adjustment were generally associated with masculinity, rather than androgyny, for botln males and females. Specifically, masculirne males were more confident and competent than androgynous males, androgynous females were better adjusted than feminine fenales (seemingly supporting Ben's theory, but possibly due to the addition of the masculine elenent which is highly valued, not the dual posssession of both masculine and feminine characteristics), but masculine females were even more competent and secure than androgynous females . It seems useful to consider these results in terms of the effects of social context and the enployed research method. The authors cited Bakan's (1966) contirnmm of agency-commonality and speculated that in a culture valuing agentic over communal attributes and skills , . individuals high in agentic qualities will be more successful within ' the context of that society's values, and also more confident, due to a history of differential applications of social rewards (i.e. , masculine males will be better adjusted than androgynous males). This interpretation is consistent with research showing a differential pattern of results of the effects of androgyny for males and females (Kristal, Sanders, Spence, 6: Helmreich, 1975; Allegeier, 1975; Kamens & Liss-Levinson, 1975; Brooks & Birk, 1975; Silvenn & Ryan, 1979). Given the value base of contemporary society, androgyny may be more strongly associated with adjustment for ferales than for males. 5 Kaplan and Sedney (1980) noted that many of the traits and skills measured in the Jones et a1. (1978) study represented agentic qualities, with an absence of traits consistent with a cammunal model , and suggested that if a more representative range of behaviors had been included, the potential strengths of androgynous men might have been more evident. Thus, in conditions/ settings where interpersonal or communal skills as well as agentic skills are rewarded, androgyny might be expected to be more closely associated with adjustment for both males and females. This present study was designed with these concerns about criteria and context in mnind. Jones et a1. (1978) explained their conclusions about the relationship of androgyny to adjustment in terms of the influence of domninant cultural values. There is additional evidence of the role that these values play in socialization and sex-typing. Bakan (1966) noted the relationship of agency to capitalism, which requires exaggeration of the agentic orientation (self-assertion, self- expansion, etc.) . Block (1973) studied sex-role stereotypes across cultures (Norway, Sweden, Denmark, Finland, England, and the United States) and found that stereotypes surrounding masculine and feminine ideals were similar in general and consistent in tl'eir differential emphasis on agency and communion for men and men. Some interesting differences were noted: fewer sex differences and less emphasis on agency characterized the two countries with long- and well-established camnitments to social welfare (Sweden and Demark) , and both American men and women described their ideal self in more agentic terms than did males and females in the other countries studied. Thus, it may be quite accurate to characterize our culture as one which tends to 6 value agentic over communal qualities, and this has implications for sen-role research conducted in the content of those values. The question of the type of traits and skills used to measure "adjustment" is also relevant. In her research on androgyny, White (1979) realized that what she called "communal competencies" were undervalued and underrepresented both in the thoughts of many psychologists and in various personality tests: she viewed Shostran's (1963) Personality Orientation Inventory and Gough's (1957) California Personality Inventory as being highly agentic- oriented and as lacking cammmal items, These issues prompted the search for a content in which to study androgyny that might minimize the effects of the dominant cultural bias towards agency and for measures that might tap a more representative range of behaviors . Such a ”mini—culture" was readily at hand. Participants in this research were students enrolled in an undergraduate psychology course designed to foster awareness of, and sensitivity to, one' s own interpersonal behavior in tlne content of small groups for enperiential learning. Skills ranging from active initiation of interaction and confrontation to supportive listening and articulation of feelings were taught and practiced within these small groups for the development and enhancement of interpersonal competence. Group facilitators were themselves trained in interpersonal skills to encourage and aid in the creationn of a supportive, caring atmosphere in which self-disclosure and the sharing of perceptions of self and others might occur most constructively. In this environment, clearly both communal and agentic qualities are highly valued. 7 In these groups one standard method of feedback for increased awareness of one's interpersonal style involved the use of a set of ratings of interpersonal behavior on semantic differential measures. At two different points in the life of the group, each group menber rated her/himself and every other group member on scales representing two dimensions and the ratings were slnared with group menbers. One dimension identified the intrapersonal behaviors active, expressive , self-effacing and self—concealing, involved in acceptance/rejection of self (ARS). The other dimension identified the interpersonal enotive behaviors warm, helpful, caring and accepting versus cold, harmful, indifferent, and rejecting, involved in acceptance/rejection of others (ARO) . Hurley (1976, 1980) has reviewed much evidence suggesting that Imman social interaction may be characterized as generally involving two principal dimensions which he has labeled Acceptance versus Rejection of Self (ARS) and Acceptance versus Rejection of Others (ARO) . This evidence comes from such diverse sources as parent-child relationships (Schaefer, 1961) , studies of behaviors in a wide range of small groups (Foa, 1961; Lieberman, Yalom, & Miles, 1973), and studies involving personality assessment (Freedman, Leary, Ossorio, & Coffey, 1950) , psycl'opathology (lorr, Bis'rop, & I‘bNair, 1965; Benjamin, 1977, 1979, 1981) . In a comprehensive review of related works, Wiggins (1982) noted that personality theorists have tended to label these two dimensions dominance and affiliation. Hurley believes that the ABS and ABC labels better reflect the underlying psychological processes involved in interpersonal interaction. 8 Construct validity for the ARS and ARO measures used in the groups has been strongly supported by correlations among their subscales (Hurley, 1976) , and througln comparisons with other prototypical measures of daninance and affiliation dimensions (Gerstenhaber, 1974; Hurley, 1983) . The ARS and ARO dimensions are functionally independent (Hurley, 1976), and can be depicted orthogonally, producing a grid on which individuals can be located according to their relative capacities for self-acceptance and acceptance of others. From this juxtaposition of ARS and ARO dimensions emerges an important theoretical construct, interpersonal competence, which is represented as a diagonal fram the low ARS-low AROquadrant to the high ARS—high ARO quandrant, and is defined in terms of increasing acceptance of self and of others. Thus, sound measures of these two dimensions should be useful in assessing the level of interpersonal competence of individuals in social interaction. These measures also tap both agentic, self-assertive skills (e.g. , Dominant--Sl.nbmissive) and communal qualities (e.g. , Accepts 0t1ners--Rejects Others) in the measureIent of interpersonal competence. This study was designed to enplore relationships between sen— type and self-perceived acceptance of self and acceptance of others , and between sen-type and acceptance of self and others as rated by others. Chosen as the measure of sen role identity, the BSRI was administered at two points during each of four consecutive terms, initially after approximately 22 hours of group participation and near the groups' and (after about 50 tours of participation). Ratings for each person on the ARS and ABC scales by both self and 9 by other group members were collected at or near these sane two times, as measures of interpersonal competence, both as self- perceived and as judged by others. It was hypothesized that individuals with androgynous sen-role self-concepts mild be seen by themselves and others as relatively high in acceptance of self and of otlners. Individuals with masculine sen-typing were enpected to be viewed by both self and others as high on acceptance of self, but low on acceptance of others. The reverse pattern was anticipated for feminine sen—typed individuals, who would be seen as being very accepting of others but less self-accepting. Individuals who indicated little sen-role differentiation (low endorsement of both masculine and feminine characteristics) were enpected to be seen as low on both dimensions. Thus , androgynous individuals were enpected to demonstrate more interpersonal competence as a measure of positive adjustment/ functioning in comparison to their sen-typed peers, where both measure and content/setting were thought likely to reflect comnmnal and agentic values and skills. Particgnants Eighty-eight junior and senior college students at Michigan State University (MSU) participated. All were enrolled in an upper-level psychology course (PSY 400) for one of four terms: spring or fall terms of 1981, and winter or spring terms of 1982. The course entailed one 50-minute classroom meeting each week, the completion of an assigned tentbook on interpersonal skills within small groups, and a total of about 50 hours of participation in small groups for enperiential learning. The 54 fenales and 34 males were assigned to mixed-sen groups of 4-8 members primarily on the basis of scheduling convenience, although efforts were made to balance the proportion of men and wonnen in each group. The only further stipulation was that close friends could not be in the same group. All groups met for two 90-minute sessions weekly and for two continuous 12-hour sessions (maratlnons) around the third and seventh weekends of the term. Each group had one or two facilitators who were either (a) undergraduates who lad taken the course previously and undergone special training to facilitate or (b) graduate students in clinical psychology. Members and facilitators kept logs / journals tlat detailed their interactions and related thoughts and feelings about each other grOLp menber separately for each session and also carpleted rating packets near each term' 3 midpoint and end. 10 11 Procedure In their introduction to the course, these students were routinely informed by the instructor that they might be asked to fill out various formns as part of continuing studies of these groups, and advised ttat their grade was not contingent upon their participation in the studies. I came to the weekly class meetings after the first marathon each of the four terms and requested the students to fill out a short questionnaire for research purposes connected with a study of personality and to allow their post-maratl'on group ratings to be used in the same study. Rsearch consent forms were handed out and the Ben Sex Role Inventory was admninistered. I returned to the classes on the last meeting of each term and repeated the procedure, enplaining the pnrpose and hypotheses of the study after collecting the second BSRI. Mid- and final-group ratings were obtained from the instructor. Two students declined to participate and several students were not present at both class meetings. Measures Ben Sen Role Inventory (BSRI). Ben's (1974) BSRI consisted of sixty adjectives arranged in a seven-step Likert format which yield three scales of twenty itene each: Masculinity (M), Femininity (F) , and Social Desirability (SD). Respondents indicated how accurately each characteristic applied to him/herself, and received M, F, and SD scores derived from the means of the responses on each scale. The M and F scores were used in a median-split procedure to classify androgynous, masculine, feminine, or undifferentiated. Persons who scored above both the median Masculinity and median Fenininity score 12 of this sample were classified as "androgynous" in their sen-role identity. Persons scoring above the median Masculinity score and below the median Femininity score were classified as 'tnasculine;" persons with the reverse pattern were classified as "feminine." Respondents with scores below both medians were designated a "undifferentiated" in their sen—role identity. The BSRI evolved from a rating procedure where 100 student judges were asked to rate 400 personality characteristics as to their appropriateness for men or women. An item was judged masculine (M) or feminine (E) if both male and fenale judges independently considered that characteristic to be significantly more desirable for one sen than the other. Ten positive and ten negative personality characteristics were not found to be differentially desirable according to sen, and were used as "neutral" items for the social desirability scale. The M and F scores were obtained by finding the mean rating for all items in each scale; SD scores were computed in the same way after the socially undesirable items were corrected for their scored direction (by subtracting each such score from six). Ben's original method of determining the androgyny score was to subtract the F score from the M score, than to multiply tl'e resulting difference by a standard score of 2.322 to get a _t_-ratio. Using a normative sample of two-thousand undergraduates, Ben (1974) found test—retest stability coefficients of .89 to .93 for a four-week interval. Ben has since abandoned l'er subtractive method of defining androgyny for an additive method. Concnn'ring with criticismn (Spence, Helmreich, 6: Stapp, 1975) suggesting that ”androgynous" should refer only to persons with high scores on b__o_t£ masculinity and femininity (separating 13 out as "undifferentiated" those persons endorsing few masculine and feminine characteristics), she now advocates use of the median-split method described above (Bem, 1977) . Bernard (1980) included both the E—ratio and median-split methods in an evaluation of the impact of various BSRI scoring procedures and concluded that the median-split procednn‘e was the preferred alternative for research purposes . In this study, median scores for the Masculinity and Fenininity scales of the 22-hour BSRIs were 5.03 and 5.04, respectively. radian scores for the M and F scales of the late BSRIs were 5.1 and 4.91, respectively. An early concern that due to tIe very atypical nature of this course, the present sanple might consist of higher proportions of androgynous individuals than held for the general population of college students (which might produce elevated scores , misclassifications , and limited generalizability) proved unfounded. The sample medians were found to be very similar to medians from past and current research involving large samples of MSU undergraduates (ex. previous median scores on M and F were 4.26, 4.23, from Mirman, 1982; and 4.8, 5.38, from Jackson, Ialongo, & Stollak, 1983). At least in termns of median scores , the present sanple proved very representative of the larger PBU population. ARO and ARS Group Behavior Ratings. These semantic differential scales included four items each. The Acceptance versus Rejection of Others (ARO) scale included the following items: Warm-—Cold; Helps otters--Harms others; Accepts others--Rejects others; and Gentle-Harsh. Items from the Acceptance versus Rejection of Self (ARS) scale included Shows feelings--Hides feelings, Active--Passive, Expressive—Guarded, and Dominant--Suhmissive. Each person rated her/himself and every other l4 gronp member on a continuum of 0-9 for each item. These ratings were routinely collected after each gronp's initial postunarathon session, and members were asked to base their ratings solely upon their impressions of the behaviors of self and others over the entire series of prior group sessions, not just during the marathon. These ratings were subsequently shared in the groups to encourage communication about the interpersonal process itself, as well as for individuals to learn how their interpersonal styles were perceived by others . The ratings were made in booklets, with the ratings for all group members, including facilitators , for one scale per page. After instructions, a 10-point Like-Dislike scale preceded the other scales. This early presentation was assured to give raters the opportunity to ventilate strong feelings that might otherwise produce less cognitive and more enotionally—laden ratings on the following scales (Smith, 1979) . The Like—Dislike scale ratings were not used in the present study. Within a week after making these ratings, each gronp member received a matrix of all scale ratings for his/her gronp and a graphic summary of the discrepancies between each person's self-rating and the ratings given to him/her by all the others. These ratings were usually reviewed during the rent session of each gronp. All measures enployed in this study were derived from these matrices of ratings. For each participant , four scores were cumputed from the midternm ratings, and four from the final ratings. ARS and ARO self-ratings were obtained directly from the matrix, and other- rated ARS and ARO scores were computed as mean ratings received fran gronp members . RESULTS The median-split procedure was used to classify participants according to their sen-role self-concept as measured by tlne BSRI, administered after approximately 22 hours of gronp interaction and again near each gronp ' s end. Participants were classified into four groups (Androgynous , Masculine, Feminine, and Undifferentiated) according to early BSRI scores, and reclassified at the end of the term, using the late BSRI scores. Gronp composition is reported in Table 1. There were some readily apparent shifts in individual sen-role classification over time. These shifts seemed to be an artifact of the median-split scoring procedIre, where an increase in the m_ed_i_§_n_ Masculinity score (from 5.03 to 5.1) and a decrease in the med____1£n_ Fenininity score (from 5.04 to 4.91) on the later BSRI affected the sen-role classification of individuals with scores at or near the median of one or both scales. Since E—tests using the mefl early and late scores on the M and F scales showed no significant differences in participants' mean BSRI scores over time (see Table 2), the shifts in sen-role classification do not seen to reflect meaningful changes in how participants used the scales to describe themselves. The early and late F scores were also highly correlated (E = .72, p_ < .001) , as were the early and late M scores (1; = .81, p_ < .001) . In subsequent aalysis , the early BSRI scores and resulting group classification 15 l6 me an am we an em 23% 95 fl and a some 0 95 5 sea 3 Gas 2 Refinance? sec 8 fine 3 $5 a 98 om abs 3 Qflv a game use on and S 9.3 S. 3.3 8 93 o ans 3 madame; ea: 5 $8 5 e3 m eke em 3.3 8 mac 0 $930. new IIIE Jim a lilies We 33 3.8m Ema some 33 an! and 83230 38 .3 mmflommnmo one 85 Baggage we no Bandage H meme Table 2 17 Comparisons of the Masculinity, Fenininity and Social Desirability Scores on the Early and late BSRIs Than SD E-value Df 2-tail probability Masculinity Early 4. 98 . 74 -l . 40 87 . 16 late 5 . 05 . 67 Fenininity Early 4.88 .65 -0.01 87 .99 late 4 . 88 . 60 Social Desirability Early 4 . 99 . 42 -1 . 42 87 . 16 late 5 . O4 . 41 18 were exployed with the early ARO and ARS ratings by self and by others, the late BSRI scores and second gronp classifications with the late ARO and ARS ratings by self and by others. Participants ' ARS and ARO scores shifted more than BSRI scores during the second half of the gronp enperience. I’han rating received fromn others increased quite significantly fran about 22 bonus of interaction to the 44-hour mark (ARS, E= 4.78, 87 df, p_ < .001; ARO, E = -2.83, 87 df, p_ < .006) . Self-ratings also increased, significantly on ARS (_t_ = -2.32, 87 df, p < .02) but onnly slightly on ARO (£= —l.62, 87 df, p_ < .11). Participants ' ARS and ARO self-ratings were also compared to ratings on those same dimensions given by other group members. For both middle and late ratings , participants rated tlnemselves as significantly more self-accepting than other gronp members perceived them (midtM, £= 4.71, 87 df, p_ < .001; final, E= 4.85, 87 df, p_ < .001) . Mean ARO self-ratings enceeded mean ARO ratings by others on both occasions, but not significantly, (midterm, E = -.74, 87 df, p_ < .46; final, 1_:_= .47, 87 df, p_ < .63): self-perceptions and behavioral observations by others more easily concurred on the dimension of acceptance of others than on the dimension of self-acceptance. Relationships between sen-type and acceptance of self and of others were initially enplored by enamining Pearson product-moment correlations computed for all possible appropriate pairs of variables. These correlations are reported in Table 3 . Using the early BSRI scores with early ARS and ARO ratings , Fenininity was significantly correlated with acceptance of others , whether self-rated (E = . 25 , p < .01) or by others (E = .32, p_ < .001). Masculinity was significantly 19 :3 ESSA 8. v m l. and Base 8. v m i 23 Base So. v mm; Bones 3288 H2 + marquee non/mama HmcomuwmuoucH - a race 8 - ES 8 £2 2 - to». an”? *3 - $8 «.8 2 - W925 -3 - co #2 *2 £8 #8 «.12 8 - 8 *8 no 8 -28 .92 as; - B from an - £8 *8 - 8 crime 2 - S $2. 8 - Woman”a -NN - 8 £3 8 lion 8 as $3 3 2 £3 .92 .52 g - in? «.18 2 fiom ER 8 of”? a2 8 ”gumg-3 - am 2 mo 3 - 9,. S 8 - mm "92 a...» E in or» . mg - anon 2 mm? 8 - 2 than 8 - ”$253-2 - No - 8 {new 8 No “ism a2 - «3 a3 £3 8 3 “am - 2- S. 2 S- “a 33 wot.» - -2 8 - 8 ”z - 2 ram 8 - 8 ”a and I ”z 92 $2 92 a2 92 m2 92 92 am a 2 am. a 2 952-3 mannZN «defies BEEN 33 Nine mamommdaom mwfinmm Ema .wumd bang oHomuxom Ewe new u m.z Hdfi mmhfimmz :4 5 9.3339335 m 3pm“. 20 correlated with acceptance of self, both self-reported (E = .45, p_ < .001) and as rated by others (5 = .34, p_ < .001). Fenininity was negatively correlated, altlnough not significantly so , with self-acceptance , as rated by self (_r= -.l4, p_ < .10) and by others (r_= -.03, p < .39), and Masculinity with acceptance of others, for both self-ratings (E = -.03, p < .39) and rating by others (3= -.01, p < .46). Using the late BSRI scores with late ARS and ARC) ratings , Fenininity was again positively and significantly correlated with acceptance of others (self—rated, g = .45, p < .001; rated by others, E = .44, p < .001) . Masculinity was significantly correlated with self-acceptance, although somedat less strongly than previous correlations (self-rated, 3= .20, p_ < .02; rated by others, 5= .31, p_ < .002). Fenininity was negatively correlated (rated by self, r_ = -.004, p < .49; by others, 5 = -.05, p < .33) with self-acceptance, Masculinity negatively correlated only with self-rated acceptance of other (r_‘ = -.19, p_ < .04) . Because Social Desirability, the neutral scale of the BSRI, correlated significantly with Masculinity (g = .28, p < .004) on the early BSRIs, and with both Masculinity (r_= .18, p < .03) and Fenininity (3= .19, p_ < .03) on tlne later BSRIs, correlations were recalculated as partial correlations, with Social Desirability held constant. The strengths of the reported relationships betWeen Masculinity and self-acceptance, and Fenininity and acceptance of others , were not significantly altered with the effect of Social Desirability held constant. On both theoretical and statistical grounds, multivariate aalysis of variance (MADDVA) seered the most appropriate inferential teclmnique for testing the hypothesized patterning of performance on the ARS and ARO dimensions according to sen-role self-concept. Because the construct 21 of interpersonal competence is defined as involving capacities for both self-acceptance and acceptance of others , both measures of tiese capacities, the two dependent variables ARS and ARO, need to be teld together in aalysis. Also, the specific hypotlneses relating sen-role self-concept to interpersonal competence concerned enpected Etterns of perfonnance (en. high-high, high-low, etc.) on the two dimensions. Statistically, using more than one ANNA on a single set of data increases the probability of obtaining significant results as the number of statistical tests increases. MAMVA allowed for the simultaneous testing of the two dependent variables . Means and standard deviations for ARS and ARO ratings by self and by others at the midpoint of the term were computed for the four gronps defined by the early BSRI. Similar computations were made for the four groups defined by the late BSRI, using the final ratings. These descriptive statistics appear in Table 4. Graphic representations of the pattern of score according to sen-type are presented in Figures 1 and 2. MADDVA results are presented in Appendix A. By design, the analyses tested tte effects of sen-type and sen, sen, and specified plamedcontrastsbetweentleandrognnousgronpandeachoftlneotler three gronps. Neither the interactive effect of sen-type by sen or gender sen itself were significant, therefore allowing testing of the effect of sen-type alone . Each contrast was tested separately to prevent significant contrasts frum confounding tie following contrasts. The androgynous grep showed the enpected pattern (see Figure 1A) of high performance on both ARS and ARO measures as self-rated at midterm (22-hour data). In contrast to the androgynous grep, undifferentiated subjects rated themselves as being significantly 22 Table 4 mans and Standard Deviations (in Parentheses) of Early and late ARS/ARC Ratings by Groups ea are Early Self—ratings Androgynous 27.54 (5.79) 27.66 (5.27) Masculine 28.25 (4.45) 24.05 (6.57) Feminine 21.15 (8.33) 27.20 (4.46) Undifferentiated 23.25 (6.69) 24.95 (6.30) Ratings by others Androgynous 23.37 (6.75) 26.43 (3.36) Masculine 23.92 (6.77) 23.74 (4.82) Feminirne 20.27 (5.64) 26.20 (3.96) Undifferentiated 20.28 (7.54) 25.64 (3.74) Self-ratings Androgynous 28.47 (6.33) 29.47 (3.93) Masculine 26.53 (6.39) 24.30 (6.23) Feminine 25.42 (4.70) 29.07 (3.26) Undifferentiated 26.80 (5.15) 24.93 (4.84) Ratings by others Androgynous 25.56 (6.57) 28.66 (2.75) Masculine 25.06 (5.85) 25.35 (3.63) Feminine 22.85 (5.86) 28.08 (3.50) Undifferentiated 23.36 (5.75) 24.08 (4.09) 28 . 27 ‘~\ p 26 s‘ ’I ‘w’ 25 23 4 CI'” / 22 1 ’ 21 a d 20 1A. Self-ratings K__ey 29- 28- 27-4 264 254 24d 234 224 21+ 201 ARS ARC 1B. Ratings by others Figure 1. Patterns of ARS/ARC responding by sen-type using early data. 28- I 27 - ,’ %‘p’~. 26 - ’Q ‘ ~Q~ 25 - ~ ‘. 24 - 22 - 21 - 20 - ARS ARD 2A. Self-ratings lie-‘1 0 = Feminine D = Undifferentiated 28- 27- 26- 25- 24- fl 23- 22- 20 _ ARS ARO 2B. Ratings by others Figure 2. Patterns of ARS/ARC responding by sen-type using late data. 25 less self-accepting and accepting of others (p_ < .02) , with lowered self-acceptance contributing more to the overall difference in interpersonal competence (univariate results, p < . 02). Feminine sen-typed subjects as a gronp were also significantly different (p_ < .006) from the androgynous group, reporting markedly low self- acceptance and almost comparably high acceptance of others . The masculine sen-typed gronp showed the predicted pattern of high self-acceptance and low acceptance of others , although this gronp did not quite differ significantly (p < .072) fromn the androgynous group. This lack of difference is the result of the contribution of self-reported masculine self-acceptance, higher even than arndrogynous self-acceptance, since univariate results furnished along the MADDVA statistics indicate significant differences between the masculine and androgynous groups on the variable of acceptance of others (p < .02) . When rated by other gronp members near midterm (see Figure 13) , masculine, androgynous, undifferentiated, and feminine gronps were seen as decreasingly self-accepting in the same order indicated in self- ratings; androgyrnous , feminine, undiffereniated, and masculine gronps were decreasingly accepting of others in parallel order. Maj or differences between patterns formed by self-ratings and ratings by others centered around the masculine subj ects' parallel performance on ARS and ARO in contrast to the relatively greater acceptance of others (compared to observed self-acceptance) claimed for androgynous, feminine, and undifferentiated gronps. Thus, masculine subjects were perceived as nearly equally accepting of self and of others, while androgynous, feminine, and undifferentiated were viewed as more accepting of others 26 than accepting of self. Althangh the androgynous gap was more self-accepting than either the undifferentiated or feminine gap, all three were similarly high in acceptance of others, such tlat canparisons between the adrogynans versus feminine group, and adrogynans versus undifferentiated group , reflected no significant differences. Androgynous subjects were, however, significantly more interpersonally canpetent than masculine subjects (p_ < .05) , primarily in being seen as significantly more accepting of others (p_ < .02) . At the end of the term/ gap experience, the patterns of self-rated scores on ARS and ARO showed sane interesting shifts (see Figure 2A). The adrogynans gap continued to report themselves as highly self-accepting and accepting of others, ad the feminine gap least self-accepting but second-highest in acceptance of others. The masculine gap, however, now saw themselves as less self-accepting and accepting of others than the undifferentiated gap, with their decreased ARS still above their ARD. The adrogynous gap was not significantly different fram the feminine gap, primarily because of these two gaps' high self-rated capacities for accepting others. Androgynous subjects did see themselves as significantly more self-accepting ad accepting of others tlan both masculine (p_ < .003) ad undifferentiated (p_ < .03) subjects, particularly in terms of acceptance of others (masculine, p_ < .001; undifferentiated, p < .01). When rated by other gap members near term and (see Figure 2B) , gaps showed the following pattern: androgynous, high ARS, even higher ARO; masculine, similar balance on ARS/ARC; udifferentiated, lower but similar balance an ARS/ARC; ad feminine, low ARS, high ARO. As in the self-rated contrasts, the androgynans gap was significantly more interpersonally competent than the masculine (p < .005) ad 27 undifferentiated (p < .002) groups, particularly in being more accepting of others (univariate ARO; masculine, p_ < .003, undifferentiated, p_ < .001) . 'I'ne androgynens grep was not significantly different fren the feminine grep in interpersonal competence. DISCUSSION The results of this study clarified the relationship of sex-role identity and interpersonal competence, as manifested in behavioral measures of self-acceptance (ARS) and acceptance of others (ARC) . Strong associations of masculinity witln self—acceptance, and femininity with acceptance of others, were revealed in both self-ratings and peers' ratings. These associations were basic to later tests of the performance of specific sex-types with respect to these same two variables. Androgynous, masculine, feminine, and undifferentiated greps of participants showed both predicted and surprising patterns of interpersonal competence, with the strength or significance of differences between the androgynous grep and the other three greps depending on the sence of perceptions (self versus peers), and other factors to be discussed. These findings lend general snpport for the concept of psychological androgyny. They also appear to validate the ecological or contextual concerns that motivated the particular design of this study. Implications for future research follow frem the noted strengths and limitatiens of measures, sample, and design of this study. The discovery of strong positive correlations of femininity with acceptance of others, and masculinity with acceptance of self, whether self-reported or as rated by others in both early and late data collections, encouraged further exploration of variens combinations of masculinity and femininity of specific sex-types and their corresponding patterns of interpersonal cenpetence. Initial hypotheses attempted to 28 29 predict these patterns for androgynous, masculine, feminine, and undifferentiated participants. Androgynous individuals were predicted to be viewed by self and others as relatively high in both self-acceptance and acceptance of others , demonstrating greater interpersonal competence than their peers, as a measure of positive adjustment/ functioning. Thus, not only specific patterns but comparisons of all geps with the anch‘ogynous grep were of significant interest. The pattern predicted for masculine sex-typed individuals was of greater self-acceptance than acceptance of others; feminine sex-typed subjects should shew the reverse pattern of geater acceptance of others than self-acceptance. Participants indicating little sex-role differentiation were expected to be relatively low on both dimensions. For the four geps formed by the midterm BSRI scores , self- ratings on acceptance of self of others showed the expected patterning with some slight variations. Notable were the masculine gep's high self-acceptance, even higher than the androgynous goup's, and the feminine goup's low self-acceptance, even below that of the undifferentiated gep. Significant differences were femd between the androgynous versus undifferentiated, and androgynous versus feminine goups. The lack of such differences between the androgynous and masculine goups was the result of similarly strong scores on self-acceptance, altlnough these geps differed importantly in terms of acceptance of others. When rated by others , scoring patterns shifted somewhat. While both self-perceptions and ratings of other agreed on the four groups' relative ranking on the two dimensions, other group mnenbers rated _a_l_l_ sex-types as less self-accepting than had their self-reports. 30 Consequently, goup differences in self-acceptance were less extreme than they were for self-ratings . Peers ' ratings of acceptance of others were more similar to self-ratings. Perhaps gep feedback is more influential on self-perceptions when it concerns lnow an individual actually treats others , than when it concerns self-accepting or rejecting behaviors. Given these lower peer-based self-acceptance ratings , the androgynous, undifferentiated, and feminine geps had lower self- acceptance compared to higher acceptance of others , with the masculine goup presenting almost equal scores in both. Because of the less extreme differences in self-acceptance and similarly high acceptance of others, the androgynous goup differed significantly only frem the masculine gep, not from either the undifferentiated or feminine goups. Using the four geps generated by the second set of BSRI scores and the later set of behavior ratings , the self-rated relative patterns for the androgynous and feminine greps were as predicted: high-high and low-high, respectively. The undifferentiated and masculine geps , however , reported similarly high self -acceptance and semewhat lower acceptance of others , with the masculine goup rated slightly below the undifferentiated gep on both dimensions . Undifferentiated persons lnad been predicted to be lower in self-acceptance. The androgynous gep was significantly more interpersonally cenpetent than both the masculine and undifferentiated geps, although not significantly different than the feminine goup, largely because of the latter's similarly high acceptance of others. As before, ratings by other goup members 31 near term and portrayed all sex-types as less self-accepting than self-reported, but used similarly high ratings on acceptance of others. Androgynous, feminine, and undifferentiated subjects showed their predicted high-high, low-high, and low-low patterns , respectively, while the masculine gep showed a relatively balanced performnance on both dimensions. They apparently were not viewed as either so self-accepting and rejecting of otters as they viewed themselves, or as had been predicted (high-low) . By both self-ratings and ratings by others, the androgynous gep was significantly more interpersonally cempetent near term's and than either the masculine or undifferentiated gep, but not the feminine gep, with their similar capacity for acceptance of others. These two sets of findings, serving as a semi-replication, provided broad snpport for the patterning hypotl'eses , particularly in terms of androgynous individuals being seen by both self and others as relatively high in self-acceptance and acceptance of others , and feminine persons being seen as less self-accepting and very accepting of others. Vasculine individuals generally seened to have geater relative diffienlty in accepting others, whatever their levels of self-acceptance. Undifferentiated persons seemed to have geater self-acceptance than feminine individuals and often showed varying degrees of acceptance of others . With masculinity and femininity being I strengly associated with self-acceptance and acceptance of others , respectively, the androgynous individual, claiming both types of traits, was thus more likely to be both self-accepting and accepting of others in an interpersonally competent fashion, since both capacities are necessary for successful social interaction. 32 In considering direct ions for future research, some discussion of the strengths and limitations of tie measures, sample, and design of this present study seems helpful. Using the BSRI and tl'e ARS and ARD ratings at two points in time alleed for tests of the stability of the two measures. Both self-ratings (mean E. = .57) and ratings by peers (mean E = .70) on tle two dimensions were reasonnably stable over time, with ratings by peers somewhat more stable (see Table 3). Scores on the M and F scales of the BSRI nere also stable, as noted earlier. However, the median-split procedure used to classify participants according to sex-role identity was problenatic. Some individuals shifted classification with tie repeated administration of the BSRI which seened to be artifacts of that scoring procedure and not genuine indicatiens of psychologically meaningful changes in sex-role identity. The use of single-sample medians in research classification also makes cenparisons of results from studies using diverse samples difficult; for example, undifferentiated participants with scores just below the M and F medians for one study's sample might be classified as androgynous in a different study sample with slightly lever median scores. Perhaps a different scoring procedure or method for using the infomation from the M and F scores in data aalysis would address these concerns. By dichotimizing the scores, the current procedure discards valuable information that continuens scores contribute . In studies where the androgynous gep is cempared with "all other subjects," the results often mask the relative contributions of each scale, i.e. , the differences femd may be the result of one scale only, M o__r_ F. This present study recovered some of tl'e information lost in the current scoring procedure through its design and data analysis: 33 threigh correlations between M and F and all other variables and threngh the planned contrasts between the androgynous gep and all of the other three goups separately. These contrasts also indirectly provided support for the theoretical and enpirical distinctions between psyclnological sex or sex-type and biological sex, one of the original concerns which prompted tie development of new measures of sex-role identity such as tie BSRI . Neither biological sex or tlre interaction of sex and sex—type had any significant effects in the variens analyses of these data. Sex-type alone was responsible for the significant differences between geps. In comparing the androgynous gep to tl‘e masculire, feminine, and undifferentiated geps separately, multivariate analysis allowed both the ARS and ARO scores to be treated simultaneeisly. Thus this statistical technique was a very appropriate test for the tleoretical construct of interpersonal competence, defined in terms of acceptance of self and acceptance of others. Univariate results furnished as a part of the analysis also made it possible to see the relative centributions of each dimension to tle total analysis. Using both self-ratings and peers' ratings provided two sets of information: the relationship of sex—role identity to self-perceived interpersonal cempetence and also to interpersonal competence as observed and evaluated by others. As noted before, this study served in part as its own replicatien, since both measures were given together twice , although at different points in life of the gep. Because of tlne shifts in individual sex-role classification, the composition of tl’e ferr geps over time is not constant, and thus the results using tle two sets of ARS and 34 ARO ratings cannot be compared for changes over time. However, the two sets of results __;ca_n_ be cempared in a replicative sense. The sample used here was college students involved in small geps as part of a psychology cenrse. The question of high-score homogeneity of the senple with regard to sex-role identity (i.e. , more likely to produce elevated medians cempared to otter samples) was addressed earlier: these results appear generalizable to at least the larger PBU college population. Because sex-role identity, like otter aspects of personnal identity, may vary developmentally, sampling other developmental stages before and beyond the college years may yield different results. These results may thus be limited in their generalizability, applicable primarily for a particular developmental stage, i.e. , yemg adulthood. This sanple also had semewhat restricted ranges of ARS and ARO scores . Such homogeneity M the chances of significant findings , so the significant relationships discovered here appear likely to be even stronger among geps more heterogeneous in level of interpersonal skills. This study looked at androgynens functioning using a possibly more communal criterion, that of interpersonnal cenpetence, in an ecology where both agentic and communal qualities were highly valued. Perhaps because of the criterion and context, the differential effects of androgyny for males and fennales often fennd in other studies (cited p. 4) were not observed here (sex by sex-type interactions). T‘l'ese mall geps are certainly contrived environments . Even tl'e psycl'nlogy cen'se of which they are part is a unique academic experience, quite different from the larger, more impersonal , basic psychology courses. The nature of these geps and this course, with their enphasis on interpersonal 35 behavior, is what makes them attractive to students and to this researcher. However, the very uniqueness of this setting, particularly in terms of its underlying values (which may differ frem those of tie more agentically-oriented, larger society) limits tie generalizability of results obtained in this context. Mnile this may pose a problem for tle science of personality, it raises larger issues of social values for those advocates of androgyny whose work on tl'e psychological functioning of men and women has led tl'em to offer societal critiques, as well as liberating psychological perspectives and models. Given the results of this study done in the "mini-cultnae" of snnall, interpersonal geps, a fruitful new avenue for future explorations of androgyny might be in tte search for natnaal contexts where communal and agentic competencies are equally valued. These contents may be the seedbeds where androgyny is matnaed, the present l'otl'ouses wlnere it fleaisl'es. If androgyny is indeed "a more human standard of psychological lnealth, "3 (Ben, 1974) as we reformulate ea views on sex-role identity, we may also need to reexamine the underlying values which shape the social contexts in which that identity is lived out . APPENDIX APPENDIXA 36 Re. Home .55 83a 92 N8. 8.2m male Scam $2 m8. 8? 8e exam 8m. Bufiufiuoeflufi adem> gmwog 0?. 8o. «3.8 Ema 92 H8. 39: me»? $993 $2 80. 84K Be «a; m5. «£58m «38> ago? Re. amen «and meme: 92 SN. m2. male mean $2 NB. 8.? 8.~ mfi.~ m8. mfiHBmma may? go? mwfiufiefim 38 Spam m mo in «,2 note é flaming m we Salaam”... 83,. Bang mgeanam >5 She Ea agape wanna Edema 8m .o 38m 83.833 .3 Baden .888 92 as... we. mam: 35.5, no 359% Sigma: Hoe madame same. 9:3 <fi§ 37 ewe. new. momda 80.x. 92. Na . Hoe . N mam . we con . 6: $2 mew . co . ex. 00 . N wNN . H ewe . omufiuaummmg new? :03 mew . emo . men . m.” m3 . 92 mm.” . one . N men . 3 0mm . 3H 9?. 3m. code oo.N e34 .Ke. mfifimfiwm noun? ammo? Re . Ho . m mam . ma owo . ex. 92 men. 30. mend.» came $2 08. code oo.N mid owe. wagon—ma may? mag ego he mwfiuom N no «m «m: ache in: 3858a»: m no Safimm 83> 33380 megawawm >5 “Sham >5 memofioehm page 388 < on; 38 0.8 . omo . N. mam . mm omo . «0H 92 awe . «no. moo . qm emu . um 92 omo . oo . em oo . N 03. m mam . 833533on monomer gouge «on . moo . omm . mm New . H 92 woo. Nome moodm momma mm< 3m. 8.? oo.~ we»; mom. «gamm— manor, g5 Hoo. 9.an mam.m~ moaoem 92 mum. m3; moorum mNmoe mm< moo. ooex. oo.~ mead new. 05383 may? magg mwfiwumunwaom 3mm 33 N no «a. «as She an: magnate m no oumhumm 83, mummhaoo aghrmhm Ea hohm an weapon? 898$ :88 <§§ 39 33me 3%? cu mummmhyn So . men . m." Hmm . NH om.» . no.” 92 mom. who; 39mm mmmdm mm< moo. 8.3 oo.~ 89m omm. gufluaummmflg mam? gag mom . mum . Hmm . NH 3m . m 9?. 02. ofl.m 39mm $5.? $2 omm. 8.3 cod 304 m3. mfifimfimm may? «3:835 moo . 93. m Hmm . NH mwm . w: om< an. Ro. «$6.... qmmd mm< moo. 8.3 oo.m 306 «B. wfigomme gang «583065. mumfio ~E mwfiumd m mo «m #2 “ohm in: wagon? m mo mumfimm 83, 3%th mgmflwa >5 Sum an $33; 8:58 388 <finzmmm< REFERENCES lEllen Lenny, "Androgyny: Sane audaciws assertions toward its caning of age." Sex Roles, 2(6), (Decenba, 1979): 29_4_. 2Samdra Ben, "The measuranent of psycl'ological androgyny," Journal of Consulting and Clinical Psychology, {L22 1974: 162. 3Ibid. 40 Allegeier, E.R. (1976, May). Heterosamality and sex typing. Paper presented at the meeting of the Ffichestern Psychological Association, Chicago. Bakan, D. (1966). The duality of human existence. Chicago: Rand NbNally. Ben, S. L. (1974). The measmenent of psychological androgyny. Journal of Consultingand Clinical Psychology, 4__2_,155-162. Bem, S.L. (1975). Sex-role adaptability: One consequence of psychological androgyny. Journal of Personality and Social Psychology, 11, 634-643. Ban, S.L. (1977). Ch the utility of alternative procedures for assessing psychological anch‘ogyny. Jom'nal of Consulting and Clinical Psychology, 52, 196-205. Bem, S.L., & lamey, E. (1976). Sex—typing and the avoidance of cross-sex behavior. Journal of Personality and Social Psychology, Z_3_3_, 48-54. Ben, S.L., Martyna, W., &Watson, C. (1976). Sex-typing and androgyny: Further explorations of the expressive dormin. Journal of Personality and Social Psychology, 1_3_4_, 1016-1023. Benjamin, L.S. (1977) . Structural analysis of a family in therapy. Journal of Consulting and Clinical Psychology, l_+_5_, 391-406. Benjamin, L.S. (1979). Structural analysis of differentiation failure. Psychiatry, fl, 2-23. Benjamin, L.S. (1981). A psychosocial competence classification system. In J. Wine & M. Smye (Eds.), Social competence (pp. 189-231). New York: Guilford Press. Bernard, L.C. (1980). Miltivariate analysis of new sex role formJlations and personality. Journal of Personality and Social Psychology, g, 323-336. 41 42 Berzins, J. I. ,Welling, M. A. 6: Wetter, R. E. (1978). A new measure of psychological androgyny based on the personality research form. Journal of Consulting and Clinical Psychology, 46,126-138. Block, J. H. (1973). Conceptions of sex-role: Some cross-cultural 2:113 gcznggitudinal perspectives. American Psychologist, 2_8, Brooks, L., & Birk, J .M. (1975, August-Septaxber). Do androgynous undels generate androgynous counselors? Paper presentef at the meeting of the American Psychological Association, Chicago. Constantinople, A. (1973) . Masculinity-femininity: An exception to a fannus dictun. Psychological Bulletin, Q, 289-307. Foa, V.G. (1961). Convergmcies in the analysis of the structure of interpersonal behavior. Psychological Review, 68, 341-353. Freedman, M., Leary, T., Ossario, A., & Coffey, H. (1950). Social implications of the group therapy situation. Journal of Social Issues, 6, 44-61. Gerstenhaber, L.M. (1975) . Acceptance versus rejection of other and self in personality self-reports (Doctoral dissertation, Michigan State University, 1974). Dissertation Abstracts International, g6, 443-442B. Cough, H.G. (1964) . California chhological Invaitory: ‘ Marmal. Palo Alto: ConsultTng Psychogists Press . Hathaway, S.R., & NbKinley, J.C. (1943). The Minnesota Miltiphasic Personality Inventory . New York: Psychological Corporation. Heilbrun, A.B. (1976). Measuratent of masculine and faninine sex role identities as independent dimensions. Journal of Consulting and Clinical Psychology, _44, 183-190. Hurley, J.R. (1976) . Helpful behaviors in groups of mental health professionals and undergraduates. International Journal of GrmiLPsychotherapy, 26, 173-189. Hurley, J .R. (1980). Two interpersonal dimensions relevant to group and family therapy. In L.R. Wolberg & M.L. Aronson (Eds.), Group and family therapy, 1980 (pp. 65-78). New York: Brunner/ FEzel. Hurley, J. R. (1983). (Correlations among measures of the principal dimmsions of interpersonal behavior). Unwblished raw data. 43 Jackson, L.A., Ialongo, N., & Stollak, G.E. (1983, August). Another look at parental antecedents of sex-role development . Paper presented at the meeting of the American Psychological Association, Anaheim, CA. Jones, W., C‘retnovetz, M.E., & Hansson, R.0. (1978). The emigre of androgyny: Differential implications for males and females? Jommal of Consulting and Clinical Psycholoy, _4_6_, 298-313. Kamens, L., & Liss-Levinson, N. (1975, August-September). Effect of sex-role on career deveth and aspirations. Paper presenta at the meeting of the American Psychological Association, Chicago. Kaplan, A.G., & Sedney, M.A. (1980). Psychology and sex-roles: An androgynous perspective. Boston: little, Bram. Kristal, J., Sanders, D., Spence, J.T., &Helmreich, R. (1975) . Inferences about the femininity of competait women and their implications for likability. Sex Roles: A Journal of Research, 1, 33-40. Lenney, E. (1979, December). Anchrogyny: Sane audacious assertions toward its caning age. Sex Roles: A Journal of Research, _5_, 703. Lieberman, M.A., Yalom, I.D., & Miles, M.B. (1973). Encounter groups: First facts. New York: Basic Books. Lorr, M., Bishop, P.F., & PbNair, D.M. (1965). Interpersonal types amrmg psychiatric patients. Journal of Abnormal Psychology, E, 468-472. Mirman, M. C. (1982). The effects of level of self-awareness on the self-presentation of individuals varying in re ed sex-role orientation. Ihpublished master 3 t1iesis,.1igan State University, East Lansing. Schaefer, E.S. (1961). Converging conceptual models for maternal behavior and child behavior. In J.C. Glidewell (Ed.), Parental attitudes and child behavior (pp. 124-253) . Springfield, 11.: Charles C. Thanas. Silvern, L. E., 6: Ryan, V. L. (1979). Self-rated adjustmmt and sex-typingontheBemSexRole Inventory: Ismasculinitythe primary predictor of adjustment? Sex Roles: A Journal of Research, _5_, 739- 763. Smith, J .A. (1979) . Use of the dislflced--liked scale in interpersonal ratgés. Unpublished master‘s thesis, Michigan State lhfversity, st ' . 44 Spence, J.T., Helmreich, R., & Stapp, J. (1974). The Personal Attributes Questiormire: A measure of sex role stereotypes and masculinity-femininity. Journal Supplement Abstract Service CataloLof Selected Documents inPsychology, 3, 43, Ms. No. 617. Spence, J.T., Helmreich, R., & Stapp, J. (1975). 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, §_Z_, 29-39. Terman, L., &Miles, C.C. (1936). Sex and personality. New York: PhGraw-Hill. White, M.S. (1979). Pbasuring androgyny in adultlnod. Psychology of Woman 3, 293-307. Wiggins, J .S. (1982). Circunplex models of interpersonal behavior in clinical psychology. In P.C. Kendall & J.N. Butcher (Eds), Handbook of research methods in clinical psychology (pp. 183- 221). New York: Wiley. "1111mm?