"I WW II x 33% llWNllellllWIIUIIHHWIIHWIWHI WWINTIIHHHNIWHITlllHHlHllHHZHHHW 3 1293 10457 This is to certify that the thesis entitled DIMENSIONS OF SELF-ESTEEM AND THE TYPOLOGICAL MEASUREMENT " OF SEX ROLE ORIENTATION presented by Peggy Spiege] has been accepted towards fulfiliment of the requirements for M.A. - degree in__S.)LC.h.Ql9£.YP flew: / l 4 I Myor pr fess I! D t /O"[’Xl W 4/”;ch a e 0-7639 MS U is an Affirmative Action/Equal Opportunity Institution MSU LIBRARIES RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES wil] be charged if book is returned after the date stamped below. DIMENSIONS OF SELF-ESTEEM AND THE TYPOLOGICAL MEASUREMENT OF SEX ROLE ORIENTATION By Peggy Spiegel A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Psychology 1982 ABSTRACT DIMENSIONS OF SELF-ESTEEM AND THE TYPOLOGICAL MEASUREMENT OF SEX ROLE ORIENTATION By Peggy Spiegel There is debate as to whether androgynous individuals are higher in self-esteem than masculine individuals. Inconsistencies across research may reflect a failure to consider dimensions of self- esteem. One hundred and twenty two undergraduates completed the Bem Sex Role Inventory, the Texas Social Behavior Inventory (measuring social self-esteem) and the Tennessee Self-Concept Scale. The TSCS yields - eight dimensional scores and a total score. Analysis of variance and multiple regression techniques were utilized. As predicted, androgynous subjects were found to be highest in social self-esteem, and contrary to prediction, global self-esteem. It was also found that the relation- ship between self—esteem and sex type varied with respect to the dimension of self-esteem. For both sexes, the masculinity ang_ femininity components of androgyny accounted for significant proportions of variance. It was concluded that specifying dimensions of self-esteem clarifies its relationship to sex type and that femininity's contribution to self-esteem may have been underestimated. ACKNOWLEDGMENTS I would very much like to thank the members of my committee, Linda Jackson, Gary Stollak, and Elaine Donelson. To my co-chair, Linda Jackson, 1 want to give my sincere appreciation for her time in discussing and assisting with both theoretical and methodological issues. Her critical comments on earlier drafts of this thesis were especially helpful. My other co-chair, Gary Stollak, was particularly available with encouragement over the course of the project. He helped me to consider the ramifications of this work in a broader context. My discussions with Elaine Donelson, especially at the outset of this project, aided me in formulating my hypotheses and expanded my knowledge of both the androgyny and self-esteem literatures. ii TABLE OF CONTENTS LIST OF TABLES. LIST OF FIGURES Chapter I. INTRODUCTION. The Concept of Androgyny. The Measurement of Androgyny Previous Research on the Relationship Between Sex Role and Self- Esteem. Evaluation of Previous Research The Present Study . Hypotheses 11. METHOD. Subjects . Procedure. Measures . . Method of Analysis. III. RESULTS Hypothesis Testing. . The Analyses of Variance. . M and F's Relationship to Self-Esteem on the TSCS Multiple Regression Analyses . . . M and F' 5 Relationship to TSBI Scores Correlations Between Self- Esteem Measures . IV. DISCUSSION F's Relationship to Self-Esteem Considerations for Future Research Summary APPENDIX A: Consent Form . APPENDIX B: BSRI, 1531. APPENDIX C: TSCS. REFERENCE NOTES LIST OF REFERENCES Page iv Table LIST OF TABLES Distribution of Subjects and Median M and F Scores on the BSRI. . . . Means as a Function of Sex Role Orientation and Analysis of Variance for the T581 and the TSCS. Regressions of Masculinity and Femininity on TSCS Scales for Males and Females . iv Page 17 19 22 LIST OF FIGURES Figure Page 1 Matrix of TSCS Subscale Scores. . . . . . . . . ll CHAPTER I INTRODUCTION The Concept of Androgyny Research has documented the existence of pervasive and persistent sex role stereotypes in our culture (Broverman, Vogel, Broverman, Clarkson and Rosenkranz, l972). Implicit in these sex role stereotypes is the assumption that femininity and masculinity are the opposite poles of a unidimensional continuum (Constantinople, 1973; Gilbert, 1981). The greater one's femininity, the less one's mascu- linity, and vice versa. Thus, according to this bipolar view, an individual cannot be both masculine and feminine; he or she must be one of the other. This conception has been challenged recently (Bem, 1974; Block, 1973; Carlson, 1971; Constantinople, 1973). What is pro- posed instead is that masculinity and femininity are independent, uncorrelated dimensions of personality. According to this view, an individual can be both masculine ang_feminine, or androgynous, in his/ her sex role. Along with her assertion of the androgyny concept, Bem (1974, 1975) explicitly posits a new relationship between sex role orientation and psychological health. Their relationship has historically been an important area of psychological inquiry. Adoption of sex role stereo- types has been considered healthy and desirable (Norell, 1978). Research has upheld this view for men; masculinity is related to psychological well-being. For women, however, the data have been less clear. Heilbrun (1968), for example, found that masculine women were better adjusted than feminine women. 1 2 Bem (1975), on the other hand, claims it is the androgynous individual, not the sex typed individual, who should exhibit the highest degree of psychological health. According to her view, the androgynous person is more flexible and adaptive in interpersonal situations. Because androgynous individuals have both masculine and feminine responses in their behavioral repertoires, they can be effective in a wider range of situations than the traditionally sex typed individuals. That is, androgynous persons can engage in both masculine and feminine behavior depending on situational appropriateness, while sex typed individuals are only effective in situations which demand behavior stereotypically associated with their respective sexes. This conceptualization was supported in a series of four experiments conducted by Bem and her colleagues (Bem, 1975; Bem, Martyna and Watson, 1976). The first was a standard conformity paradigm designed to evoke stereotypically masculine behavior (independence), and the second involved playing with a kitten and was designed to tap traditionally feminine behavior (nurturance). Androgynous subjects of both sexes displayed high levels of indepen- dence and nurturance. Thus, they performed adequately regardless of the appropriateness of the behavior for one sex or another. The sex typed males were adequate in independence but showed deficits in nurturance. The sex typed females were deficient in independence, as expected, but were also deficient in nurturance. which would not be predicted from Bem's position. Bem, Martyna and Watson (l976)hypothesized that the latter finding may have been due to the operationalization of nurturance. Perhaps the feminine women's low nurturance was unique to interaction 3 with animals. Thus, the investigators conducted two further experiments; in the first the subjects interacted with an infant, and in the second they listened to a lonely student. Both studies found that androgynous and feminine subjects did not differ significantly and were higher in nurturance than masculine subjects. Taken together, these four experi- ments demonstrate that the androgynous were most flexible, as they evidenced both feminine and masculine behavior. The sex typed persons, on the other hand, were constricted and limited in situations in which sex inappropriate behavior was required. Bem's formulation has resulted in much research concerning the relationship between androgyny and psychological well-being. Worell (1978) has noted that this research has focused on four aspects of well- being: 1) self-esteem or positive self-evaluation; 2) freedom from obvious pathology such as anxiety and depression; 3) adaptive, flexible and effective interpersonal behavior, such as initiation and maintenance of appropriate social interaction; and 4) broad lifestyle coping variables. The present investigation is concerned with the relation- ship between androgyny and the first aspect of well-being mentioned above, self-esteem or positive self-evaluation. Before the previous literature in this area is reviewed, the measurement of androgyny will be discussed. The Measurement of Androgyny Since the introduction of the androgyny concept, four measures have been developed to tap this construct. These are: the Bem Sex Role Inventory (BSRI) (Bem, 1974, 1977); the Personal Attributes Questionnaire (PAQ) (Spence, et a1., 1975); the PRF ANDRO Scale (Berzins, Welling and Wetter, 1978); and the Masculinity-Femininity 4 Scales of the Adjective Check List (ACL) (Heilbrun, 1976). These scales were constructed differently (Kelly, Furman and Young, 1978; Kelly and Worell, 1977), but each essentially contains separate mascu- linity and femininity scales which are combined to yield sex role category. The method of determining sex role category has been a subject of continuing debate. Bem's (1974) original model defined androgyny as the relative balance between masculine and feminine attributes and was thus a subtractive model. Her typology yielded three categories: masculine, feminine and androgynous. However, Spence, Helmrich and Stapp (1975), proposing an additive model, argued that the term androgyny should only apply to those high in both femininity (F) and masculinity (M). They advocated a median split method for determining sex role orientation, which yielded four categories: androgynous (above median on M and F), masculine (above median on M, below median on F), feminine (above median on F, below median on M) and undifferentiated (below median on M and F). Bem (1977) later endorsed this method, which is currently the most commonly accepted. The debate is still not resolved, however (Worell, 1978); some literature to be reviewed has utilized the original subtractive model. A third model, the multiplicative or interactive model, has been suggested, although support for this conception has been unclear (Harrington and Andersen, 1981; Lubinski, Tellegen and Butcher, 1981). The use of differing models makes interpretation of empirical data difficult. Several other methological issues also confuse results. Due to some criticism of the validity of the BSRI (Locksley and Colten, 5 1979; Pedhazur and Tetenbaum, 1979), Ben (1979) published a short version of her measure. Both the original and the short forms are currently in use. Additionally, comparability of the four afore- mentioned sex role measures is questionable (Kelly, et al., 1978). This too, makes comparison across studies using different sex role measures difficult. Previous Research on the Relationship Between Sex Role and Self-Esteem Nine empirical studies related to the issue of androgyny and self-esteem were located in the literature. Three of these (O'Connor, Mann and Bardwick, 1978; Spence and Helmreich, 1978; Spence, et al., 1975) clearly demonstrate that androgynous subjects had significantly higher self-esteem compared to other groups. Spence, et a1. (1975) found that androgynous subjects were highest in self-esteem, followed by masculine, feminine and undifferentiated subjects. Replications of this study, with a middle aged upper middle class sample (O'Connor, et al., 1978), a college sample, and an adolescent sample (Spence and Helmreich, 1978) found the same results. All of these studies used the PAQ and the Texas Social Behavior Inventory (TSBI) (Helmreich, Stapp and Ervin, 1974), a measure of social self-esteem. A study by Flaherty and Dusek (1980) took a multidimensional approach to self-esteem. Subjects completed the long form of the BSRI and semantic differential self-concept scale containing four factors. These were: adjustment, achievement/leadership, congeniality/ sociability and masculinity/femininity. It was predicted that self- esteem scores would vary according to which aspect of self-esteem was being measured as well as according to sex role category. The adjustment factor (factor 1) was considered the most similar to the 6 self-concept measures used in previous research. The achievement/ leadership factor (factor 2) tended to reflect the traditionally masculine, instrumental role. Factor 3, congeniality/sociability, reflected the expressive social role. Factor 4, masculinity/femininity, reflected traditional descriptions of sex role stereotypes. Thus, according to the flexibility argument by Bem (1975), androgynous individuals would be expected to score high on factors 1-3. Since high scores on factor 4 tend to represent masculinity, the androgynous and the masculine subjects should also score highest on factor 2 and feminine subjects should also score highest on factor 3. Results were as expected for factors 1-3. For factor 4, the masculine groups scored significantly higher than the other groups. The androgynous group had medium scores on this factor, which was interpreted as a lack of rigid sex typing. Overall, then, these results strongly supported the argument of greater flexibility and adaptation and thus higher overall self-concept in androgynous persons. Secondly, they also support the notion that components of self-esteem are differently related to sex role orientation. Three studies (Antill and Cunningham, 1979; Jones, Chernovetz and Hansson, 1978; Wetter, Note 1) clearly found that both androgynous and masculine subjects score higher than feminine and undifferentiated subjects on measures of self-esteem and did not significantly differ from each other. Wetter (Note 1), using the PRF ANDRO Scale and the Self-Esteem Questionnaire, found, for both sexes, that androgynous and masculine subjects did not differ significantly and were higher in self-esteem than feminine and undifferentiated individuals. There were no differences between the latter two groups. 7 Jones, Chernovetz and Hansson (1978) administered the BSRI (long form) and a self-esteem measure by Coopersmith (1967). Using the balance method originally advocated by Ben (1974), they found that androgynous and masculine males were not significantly different in self-esteem. For females, there were no significant differences among any of the sex role groups. When the data were reanalyzed using the median split method (the androgynous category was divided into androgynous and undifferentiated), the results were unaltered. Lastly, Antill and Cunningham (1979) used three sex role measures (BSRI, PAQ and PRF ANSDRO) and two self-esteem measures [Self-Acceptance Scale (Berger, 1952) and the Janis-Field Feelings of Inadequancy Scale (Eagly, 1967)]. For both sexes, androgynous and masculine groups did not differ significantly and were higher on the measures of self-concept than feminine and undifferentiated subjects. Utilizing the long form of the BSRI and the 1581, Bem (1977) reported a main effect of sex role, with feminine and undifferentiated subjects low in self-esteem and masculine and androgynous subjects high in self-esteem. These results have been interpreted by Kelly and Worell (1977) as almost identical to those of Wetter (Note 1). However, Bem did not report any statistical comparisons between the means. Thus, it cannot be completely determined whether Bem's findings actually paralleled those of Wetter. Most relevant to the present study is an investigation by Peterson (Note 2) which utilized the short form of the BSRI and the Tennessee Self-Concept Scale (ISCS) (Fitts, 1965). She found that self-esteem was related to sex type among males, but did not report any further significance tests. Thus, it could not be deduced whether 8 androgynous subjects were in fact the highest in self-esteem. The present investigation replicates and extends Peterson's work. Evaluation of Previous Research Taken together, the results of these studies are inconsistent and inconclusive. They have been interpreted in different ways. The studies cited showing the highest self-esteem in androgynous subjects are claimed by some to support Bem's (1974, 1975) original conception of greater flexibility leading to higher levels of psychological health. Others have argued that the high levels of self-esteem are mainly due to the masculine component of androgyny, since androgynous and_ masculine subjects had higher self-esteem than feminine and undiffer- entiated subjects (Jones, et al., 1978; Kelly and Worell, 1977). Moreover, self-esteem is positively correlated with masculinity scores, whereas femininity scores have weaker, zero-order or negative corre- lations with self-esteem, depending on the measures used (Antill and Cunningham, 1979; Spence and Helmreich, 1978; Spence, et al., 1975). The inconsistencies across studies could also be a function of unreliability in the measures and differing sample characteristics. On the other hand, an alternative explanation is plausible. As Flaherty and Dusek (1980) have argued, it is important to take a multidimensional approach to self-esteem. It appears that self-esteem measures utilized in the literature have been assumed to be equivalent when in fact they are not. The TSBI measures a particular component of self-esteem--social self-esteem (Spence and Helmreich, 1978). Most of the studies that found high self-esteem associated with androgyny utilized the TSBI; hence, it seems that androgyny is related to the highest level of social self-esteem. However, in general, those 9 studies which found no differences between androgynous and masculine subjects in self-esteem appear to have used measures which tapped global self-esteem. Thus, the failure to take differing components of self-esteem into account may explain some of the contradictory results in the literature. This argument also implies that different aspects of self- esteem will have different relationships with androgyny. With the exception of the Flaherty and Dusek (1980) study, it does not appear that the specific components of self-esteem other than social self- esteem have been examined in the literature. A second issue affecting the relationship between self-esteem and sex role orientation, namely, the question of the adequacy of current typological measurements of sex role orientation has not been satisfactorily addressed in the literaure. The four part typology created by the median split method results in gross categorizations. Each category contains a wide range of individuals. The androgynous category, for example, includes individuals with femininity and masculinity scores just above the median as well as those possessing scores approaching the upper limits of M and F. Thus, the variability of scores within each sex role orientation is high. This situation may obscure research results because the composition of the categories is not uniform. The Present Study The present investigation has been designed to shed light on the issues raised above, i.e., the relationship between androgyny and different aspects of self-esteem. The short form of the BSRI, the 10 the TSBI and the Tennessee Self-Concept Scale (TSCS) will be utilized in considering this issue. The TSCS yields a global self-esteem score, termed Total P, as well as eight subscale scores. The eight scores are derived from a 3 X 5 matrix of the test items, as illustrated, with examples, in Figure 1. The subscales are: the Physical self (Column A), the Moral- Ethical self (Column B), the Personal self (Column C), the Family self (Column D), the Social self (Column E), Identity (what she/he is) (Row 1), Self-Satisfaction (how she/he accepts herself/himself) (Row 2), and Behavior (how she/he acts) (Row 3). Each of the subscales A through E is comprised of five cells which are summated to obtain the subscale score; each of subscales 1 through 3 contains three cells which are likewise summated. According to the proposition outlined earlier, certain relationships between the measures of self-esteem and sex role category should occur. First of all, the TSCS subscales would be expected to have different patterns of scores for the sex role groups, as each subscale measures a different aspect of self-esteem. That is, the relationship among the four sex role categories will depend upon the aspect of self-esteem being measured. Because previous research has not generally been concerned with different dimensions of self- esteem, the specific patterns these dimensions will yield cannot be predicted. Secondly, the Social self (subscale E) and the TSBI would be expected to have a similar relationship to sex role orientation. Specifically, androgynous individuals should be highest among the groups on both measures of social self-esteem. Lastly, the global ll .mmcoum oHoumnam womb mo ngumz up mgzmHu .zHHoumezn .xHHmmm .HHHsom .uoH a .mmcHsu HHmmzs goH>mcmm mgmspo m>Hmcom HE gqu ccHs HE can zgm> cu we menu go: cc H nggmaa H manage H mmsHumsom H noon mxmp H m 30m .ma on ace: H .wcoe .mgos cugzgu .cwgu oou coHHummmHuMm mm.anmHuom HHHEMH HE .HHmmxs op om go: new oou HH m mm an H “mace quogm H mqummn H o» “gaze H cmsuHmc an H N 3oz .comgwa .HHHEmH .comcmg .comcma quucmuH xHocchw Hana; a do Hzmcmmgu ummco: .xuon AguHmm; m Em H cameos m Em H a Em H cm as H a m>mz H H 3oz HHmm mem HmHuom HHmm HHHEmu HHmm Hegemcmq HmuHsum\HmLoz HHmm Hmumega m cszHou o cszHou u cszHou m cazHou < asaHoo 11 .mmgoum mHmumnsm womb Ho ngpcz "H mgzmHH .xHHmumega .zHHmmm .xHHEmH .uoH m .mmcng wpmmxs LoH>mgmm mgwnpo w>ngoH HE zuH: ccHE HE can Agm> on Ho menu go: on H nggmaa H mmcmsu H mmsmamsom H noon wxm» H m 3oz .ma op Ham: H .mgos .mgoe coczgu .cwsu cop :oHuummmHumm mm mHanuom HHHEMH HE .HHmme op om go: «a» cop HH m mm Em H «mag» onocm H mmHammu H ca usuao H cmzaHmc an H N zoa .comcma .xHHEmw .comgma .comcma quucwuH HHucmHLH Hana; a mo Hamcmmgu ummco; .xeon aguHmm; a Em H Lansms a an H m Em H cm Em H a m>mc H H 3cm mHmm mHmm HmHoom HHmm AHHEmu HHmm Hmcomgma HmuH;HM\Hmcoz HHmm Hmumeca u csaHoo o cszHou u csaHou m cszHou < csaHou 12 measure of self-esteem, Total P, would be highest for both androgynous and masculine individuals. This study will also attempt to improve upon the methodological problems with sex role orientation mentioned earlier. In order to reduce the variability within each category, only the best representations of each of the four sex role categories, the "pure" types, will be selected from a larger subject pool for use in this investigation. Those subjects at the extremes of M and F will thus be utilized. The androgy- nous subjects chosen will be those with the highest M and F scores, masculine subjects will have the highest M scores and the lowest F scores, feminine subjects will have the highest F scores and the lowest M scores, and undifferentiated subjects will have the lowest M and F scores. Hypotheses The specific hypotheses to be tested in this study are as follows: 1. Androgynous subjects will have significantly higher self-esteem than masculine, feminine and undifferentiated subjects as measured by the TSBI. 2. Androgynous subjects will have significantly higher self-esteem than masculine, feminine and undifferentiated subjects as measured by the social subscale of the TSCS. 3. Both androgynous and masculine subjects will score significantly higher than feminine and undifferentiated subjects on the Total P score of the TSCS but will not differ significantly from each other. CHAPTER II METHOD Subjects One thousand seven hundred and eighty students (628 males and 1152 females) from introductory psychology classes at Michigan State University completed the short form of the BSRI (Bem, 1979, 1981) and the short form of the TSBI (Helmreich and Stapp, 1974).1 The test packets were distributed in class and returned by the subjects at the subsequent class time. Consent forms were included in the packets (see Appendix A). Participants received class credit. Using the median split procedure, the subjects were then categorized as androgynous, masculine, feminine or undifferentiated based on their BSRI scores. This procedure was performed separately for each sex, resulting in a total of eight groups. As described earlier, the most extreme members of each sex role group were then selected for inclusion in the experiment. The final sample consisted of 122 subjects, 67 males and 55 females. Procedure Subjects were contacted by phone about participation in the experiment. It consisted of four tasks tapping interpersonal behavior across several different types of situations. This part of the study is not directly relevant to the present study and will not be explained 1The subjects also completed other measures, as the data for this study were collected as part of a larger study. 13 14 in detail. Five questionnaires, one of which was the TSCS (Fitts, 1965), were administered following completion of the tasks. Measures A. Bem Sex Role Inventory--short form: The short form of the BSRI (Bem, 1979, 1981) is a self report inventory which measures sex role orientation (see Appendix B). It contains 30 socially desirable personality traits; ten are stereotypically feminine (femininity score), ten are stereotypically masculine (masculinity score), and ten are neutral. In this study, respondents indicated on a scale of 1 to 5 how well each trait described them. From these scores, classification of sex role orientation was determined by the median split method, as delineated above. Internal consistency and test- retest reliabilities are acceptable for the BSRI. See Bem (1981) for details of the psychometric analyses. B. Texas Social Behavior Inventory--short form A: The TSBI (Helmreich and Stapp, 1974; Helmreich, et al., 1974) is a measure of social self-esteem (see Appendix B). It consists of 16 statements, each having five response alternatives ranging from "not at all characteristic of me" to "very characteristic of me." The items are scored on a scale of 1 to 5, with 1 representing low self-esteem and 5 representing high self-esteem. Reliability and validity indices for the TSBI are reported by Helmreich and Stapp (1974) and Helmreich, et al., (1974). C. Tennessee Self-Concept Scale: The TSCS (Fitts, 1965) also measures self-esteem (see Appendix C). It is composed of 100 self- descriptive statements which the respondent must rate on a scale of 15 1 (completely false) to 5 (completely true). Fitts (1965) reports test-retest reliabilities and validity data. Method of Analysis Consistent with the hypotheses, planned comparisons between sex role groups on the TSBI, the TSCS Social subscale and the TSCS Total P score were performed. A number of further analyses were performed: 1) A series of 2 X 4 (Gender X Sex Role Orientation) analyses of variance were computed for the eight TSCS subscales. Where main effects were demonstrated, t-tests were performed to determine the ordering of the sex role group means within each subscale. 2) For each sex, a multiple regression was performed utilizing the M and F scores from the BSRI and each of the TSCS subscale scores. In doing this, it was determined whether the relative contribution of M and F differed depending on the dimension of self-esteem considered. 3) Separate multiple regressions were completed in the same fashion for the TSBI score and the TSCS Total P score. 4) The correlations between a) the TSBI and the TSCS Social subscale, b) the TSCS subscales, and c) the TSBI and the TSCS Total P were calculated. CHAPTER III RESULTS The distribution of subjects and the median M and F scores for each sex role category are presented in Table 1.1 For males, the mean M and F scores were 3.91 and 3.89, respectively. For females, the scores were 3.65 and 3.98, respectively. Hypothesis Testing It was predicted that androgynous subjects would have greater TSBI scores than all other subjects. Table 2 presents the means on the TSBI for the four sex role groups. Androgynous subjects scored significantly higher than masculine (t(73) = 4.56, p < .oo1)2, feminine (t(36.72) = 7.52, p < .001), and undifferentiated (t(27.8) = 10.59, p < .001) subjects. Thus, the first hypothesis was supported. Androgynous subjects were also expected to score higher then all others on the TSCS Social subscale. The means on the Social sub- scale are also presented in Table 2. The scores show the same pattern as do those on the TSBI. As predicted, the androgynous subjects scored significantly higher than masculine (t(74) = 4.45, p < .001), feminine (t(65) = 3.41, p < .001) and undifferentiated (t(59) = 7.22, p < .001) subjects. 1It proved more difficult to fill some cells than others. Feminine males and undifferentiated subjects of both sexes with extreme scores were relatively rare. Thus, the cell sizes are not equal. 2All tests of significance are two-tailed. 16 17 Table 1. Distribution of Subjects and Median M and F Scores on the BSRI.a Females Males Androgynous M 4.48 (15)b 4.51 (25) F 4.83 4.59 Feminine M 2.77 (15) 2.80 (11) F 4.87 4.58 Masculine M 4.33 (14)c 4.40 (21) F 2.90 3.10 Undifferentiated M 2.75 (10) 2.85 (10) F 2.95 2.95 Note: M - masculinity, F = femininity, BSRI = Bem Sex Role Inventory. aAll responses were coded l to 5. bCell size is indicated in parentheses. cDue to incomplete data, one subject was not included in all the analyses. 18 The third hypothesis predicted that both androgynous and masculine subjects would be higher than feminine and undifferentiated subjected on the TSCS Total P score but would not differ significantly. Table 2 presents the means for Total P by sex type. T-tests indicated that androgynous subjects again scored significantly higher than feminine (t(65) = 3.89, p < .001) and undifferentiated (t(59) = 5.96, p < .001) subjects. Masculine subjects also scored significantly higher than undifferentiated persons (t(53) = 3.32, p < .01), but contrary to prediction, were not significantly greater in self-esteem than feminine subjects (t(59) = .96, ns). Also contrary to prediction, the androgynous subjects had significantly higher scores than the masculine, subjects (t(74) = 3.27, p < .01). The third hypothesis, then, was partially supported. The Analyses of Variance As is apparent from Table 2, significant main effects for sex role were found for each of the subscales. A significant main effect for sex was found only on the Physical subscale (F(l,]14) = 8.61, p. < .05), and no significant interactive effects were demonstrated. The ordering of the sex role group means on the TSCS subscales and the Total P score was determined by the use of t-tests. With the exception of the Physical subscale, the sexes were combined for this analysis since no main effects for sex had been demonstrated. The ordering was performed to determine if the pattern of the means varied from subscale to subscale, and if it differed from the highest-to-lowest pattern generally found in research of this sort. Means for each of the TSCS scales by sex type are presented in Table 2. '19 .eole 2. teens as a Funczxon :' Sex Role Orientat'on anc Anal/5:5 of Varian;e for the TSBI and the TSCS. 2251 5:39519 55.3 15‘ .4 m. J. . .U. _ 5.9.5] 29b Se 1: c x (n=41) (n=26) (n=35) (n=20) SRO iii-:1 11.11.35 4.22 3.31 3.940. 2.84 .03 45.43'“ 1.45 Females ;.11 3.38 3.76 3.16 TSCS Hales “5.24 65.91 72.76 64.40 8.61** 9.19"' .45 30‘5ica1 Females 59.56 64.53 68.86 61.30 Morai- Males 72.72 6‘.36 67.95 66.30 .20 2.80' 1.77 {:01Cel Females 53.38 70.20 70.21 63.70 Ferzcral “0105 ‘5.24 €0.82 67.81 59.50 .21 17.22"' 1.99 Females 71.25 55.53 65.64 61.20 FdWIIy Hales 74.40 56.55 68 95 66.10 .44 3.58' .72 Females 71.00 69.87 67.21 64.90 Social Males 75.60 67.64 68.14 58.50 1.24 19.17"* .77 Fecales 75.37 73.00 68.57 63.90 Identity Males 134.20 124.18 123.29 118.3 .31 12.21"' .59 Ferales 131.63 127.60 124.57 115.60 Self- Males 113.56 101.82 111.19 96.80 1.87 7.02"' .51 Satis‘action Females 139.75 104.20 106.29 96.60 Behavior Males 121.44 104.18 111.14 99.70 .56 l4.65"' 2.03 Females 114.19 108.33 109.64 102.80 TOCuI P Males 371.20 330.18 345.62 314.80 .97 13.05"' 1.21 Females 353.94 312.13 340.50 315.00 Note; TSCS = Tennessee Self-Concept Scale, 1552 = Texas Social Behav1or lrventory. A = encrog/nOos. F = feminine, M = masculine, U = undifferentiated, SRO = sex role orientation. a11F = 1.114. DDF = 3.114. CDF = 3.114. dOne Subject was not incluced in this analysis doc to mlSSlng data. *p < .05 *'p < .01 ""0 < .001 _. 20 Turning first to the Physical subscale scores, androgynous males scored higher than feminine (t(34) = 3.49, p.g .001) and undiffer- entiated (t(33) = 4.27, p < .001) males; masculine males also scored higher than feminine (t(30) = 2.22, p < .05) and undifferentiated (t(29) = 3.08, p < .01) males. For females, only one comparison was significant; androgynous women scored higher than undifferentiated women (t(24) = 2.14, p < .05). On the Moral-Ethical scale, androgynous persons scored higher than undifferentiated persons (t(59) = 2.82, p < .01); no other comparisons were significant. Androgynous individuals scored higher than masculine (t(74) = 3.90, p < .001), feminine (t(65) = 6.12, p < .001), and undifferentiated (t(59) = 7.26, p < .001) individuals on the Personal subscale. Masculine subjects also were greater in self-esteem than undifferentiated subjects (t(53) = 2.72, p < .01). Androgynous persons were higher on the Family subscale than masculine (t)74) = 2.30, p < .05), feminine (t(65) = 2.11, p < .05) and undifferentiated (t(59) = 3.35, p 5_.OOl) persons. On the Social subscale, androgynous subjects scored higher than masculine (t(74) = 4.45, p < .001), feminine (t(65) = 3.41, p 5_.001) and undifferentiated (t(59) = 7.22, p < .001) individuals. Masculine (t(53) = 3.87, p < .001) and feminine (t(44) = 3.56, P.$ .001) persons were higher than undifferentiated persons. With regard to the Identity scale, again androgynous subjects scored higher than masculine (t(74) = 4.12, p < .001), feminine (t(65) = 2.95, p < .01) and undifferentiated (t(59) = 5.62, p < .001) subjects. In addition, masculine (t(53) = 2.22, p < .05) and feminine (t(44) = 2.89, p < .01) subjects were higher in self-esteem than undifferentiated individuals. On the Self-Satisfaction scale, 21 androgynous subjects scored higher in self-esteem than feminine (t(65) = 3.00, p < .01) and undifferentiated (t(59) = 4.54, p < .001) subjects, and masculine subjects also were higher than undifferentiated persons (t(53) = 3.10, p < .01). Lastly, androgynous persons scored higher than masculine (t(74) = 3.55, p 5_.OOl), feminine (t(65) = p < .001) and undifferentiated (t(59) = 5.88, p < .001) individuals on the Behavior subscale. Masculine subjects also scored higher than undifferentiated persons (t(53) = 3.23, p < .01). M and F's Relationship to Self-Esteem on the TSCS Correlations for each sex between TSCS scores and M and F values are presented in Table 3. For males, except for one instance. (the Physical subscale), all dimensions of self-esteem were significantly correlated with both M and F scores. Significant correlations with M and F scores. Significant correlations with M ranged from .29 to .66 (p < .01) and with F from .21 to .53 (p < .05). For females, in all cases but one (the Physical subscale), F was significantly correlated with the respective dimensions of self-esteem, whereas M was significantly correlated with self-esteem on five of the nine scales. Thus, there were fewer significant correlations for females than males. The significant correlations were F ranged from .23 to .46 (p < .05); with M, they ranged from .24 to .32 (p < .05). Multiple Regression Analyses The relative contribution of M and F to dimensions of self-esteem were corroborated by the multiple regression analyses, presented in Table 3. For males, masculinity always entered the regression equation first. For females, except for the Physical subscale, femininity 223 Table 3. Regressions of Masculinity and Femininity on TSCS Scales for Males and Females. Malesa 2 2 R Scale Step Variable 5_ .3 R Change Social 1 M _53itt '53ttt .28**' .zgeoo 2 F .53-“ .70... .50*** .21'** PurSOnal 1 M .66‘*' .66'** .44m .44*** 2 F .26' .68**' .47'*‘ .03 Behavior 1 M ,6?t** ,opttt .3gtat _39ga. 2 F .35" .68'** .46'** .07" Total P 1 M .59"' .59*** .35**a 35... 2 F .36*** .66*‘* .43‘** .08** Identity 1 M .50*" .50"* .25**' 25... 2 F .46*" .64tti .4]... .16*'* Physica1 1 M .58**' .58**‘ .34m _34... 2 f .19 .59“' .35"' .01 Self— 1 M .49*" ,49‘** .zqatt _24te* Satisfaction 2 I .2?‘ .51"* .26“‘ .02 Fam11y 1 M .35** .35.1, .12" .12-~ 2 F .21“r .39" .15" .03 Moral- 1 M .29** .29“ .08' .08. Ethical 2 F .24' .35* .12' .04 Note: TSCS = Tennessee Self-Concept Scale, M = masculinity, F - femininity. 6M = 57 “M = 54 'p < .05 tip < .0] **‘p §_.001 Scale Social Identity Personal Total P Behavior Family Physical Self- Satisfaction Moral- Ethical Femalesb Step Variable g 5 32 1 F .46"' .46"‘ .21"' 2 H .28' .56"‘ .32"‘ 1 r .42... _azooo .18.. 2 M .20 .48"' .23“‘ 1 F .16“ .36“ .13" 2 M .24' .46*‘ .21" 1 F .35“ .35“ .13“ 2 M .24‘ .45“ .20" 1 f .33" .33" .11' 2 M .27' .45" .20" 1 F .37*' .37‘* .14" 7 M .06 .38‘ .10‘ 1 M .32" .32“ .10‘ 2 i .13 .36‘ .ll‘ 1 F .23‘ .23' .05 2 M .20 .32 .10 1 f .16 .16 .01 2 M .13 .22 .05 23 accounted for most of the variance in the TSCS scores. The amount of variance accounted for by M and F independently substantially differed depending on the component of self-esteem. M accounted for between 8% and 44% of the variance for males and between 1% and 11% for females. Between 1% and 21% and between 3% and 21% of the variance was accounted for by F for males and females, respectively. Furthermore, the relative contribution changes; in some cases one variable accounted for most of the variance; in others, the contributions of M and F were more equal. For example, on the personal subscale for men, M and F respective- ly accounted for 44% and 3% of the variance, while on the social scale, M and F respectively explained 28% and 21% of the variance. Table 3 also presents the amount of variance together accounted for by M and F on each of the TSCS scales. For both sexes, the scales are arranged in descending order from the most to the least total variance explained. For men, in descending order the scales were: Social, Personal, Behavior, Total P, Identity, Physical, Self-Satisfaction, Family, and Moral-Ethical. For women, they were Social, Identity, Personal, Total P, Behavior, Family, Physical, Self-Satisfaction and Moral-Ethical. The ordering of the scales was similar for both men and women. For both sexes, M and F together accounted for the greatest amount of variance on the Social subscale (50% and 32%, respectively). The Personal subscale was second and third highest for males and females, respectively. The first five scales were the same for both sexes, as are the last four, although the scales are sequenced somewhat different- ly for the two sexes. 24 Women and men are less similar when the actual values of the total variance explained are considered. For males, six scales (Social, Personal, Behavior, Total P, Identity, Physical) had 35% or more of the variance accounted for by M and F. All of the scales for the females have 32% or less of the variance accounted for by M and F. M and F's Relationship_to TSBI Scores Consideration of the relationship between M, F and the TSBI score was investigated by correlation and multiple regression analysis. The TSBI was correlated .82 (p :_.001) with M and .39 (p 5_.001) with F for males, and .73 (p 5,001) with M and .18 (ns) with F for females. Thus, for men, M accounted for 68% (p < .001) of the variance on the TSBI, while F explained only 8% (p < .001). The pattern is the same for women; 53% (p < .001) was explained by M and 6% (p < .01) was explained by F. Correlations Between Self-Esteem Measures Correlations between the TSBI and the Social and Total P scales of the TSCS were computed. For males, the correlation between the TSBI and the Social score was .73 (p 5_.001) and .68 (p 5_.OOl) between the TSBI and Total P. For females, the correlations were .47 (p §_.001) and .44 (p 5_.OOl), respectively. CHAPTER IV DISCUSSION The first two hypotheses were supported. Androgynous subjects were greater than all other subjects on the two measures of social self- esteem. These results replicate several other studies (O'Connor, Mann, and Bardwick, 1978; Spence and Helmreich, 1978; Spence,et al., 1975). The third hypothesis, that androgynous and masculine individuals would not differ but would be significantly higher in global self-esteem than feminine and undifferentiated subjects, was partially supported. Although the former two groups were higher in self-esteem than undiffer- entiated subjects, only androgynous subjects had higher self-esteem than feminine subjects. Further, androgynous subjects scored significantly higher than masculine subjects. In fact, the entire pattern of results strongly supports Bem's (1975) original prediction that androgynous subjects would have the greatest self-esteem when compared to all other groups. In particular, the results provide evidence that androgynous individuals do have a clear advantage over masculine subjects, at least with regard to social and global self-esteem. Although the fact that androgynous and masculine subjects did significantly differ on Total P does not support my argument for components of self-esteem, other data do support the importance of specifying dimensions of self-esteem. When the sex role groups were ordered for each subscale, it was clear that there was not an outstanding pattern that was consistent across all the subscales. Particularly relevant here is the changing relationship between 25 26 androgynous and masculine subjects. In five of nine cases androgynous subjects were significantly higher in self-esteem than masculine subjects, while in four cases they were not. Thus, the difference (or lack there- of) between androgynous and masculine subjects varied with respect to the dimensions of self-esteem being considered. Additionally, when sex type was defined by its components M and F, the strength and the pattern of the relationship between M and F and particular components of self-esteem varied. Social self- esteem, for example, appears to be one of the more important dimensions of self-esteem, insofar as it can be predicted by masculinity and femi- ninity. For both men and women, the correlations between M and F and the TSCS Social subscale were among the highest. Moreover, M and F together accounted for the most variance on this subscale for both men and women. M and F accounted for the next greatest amount of variance in personal self-esteem for males, and the third highest amount for females. Thus, this component also is an important one. The Behavior and Identity subscales were also among the most significant aspects of self-esteem. For both sexes, the Physical, Family, Self-Satisfaction and Moral-Ethical scales were in the bottom half of the rankings for amount of variance explained by M and F. Although the relative importance of the dimensions was similar for both sexes, it is interesting that, in general, more total variance was accounted for by M and F for men than women. This was true for all the TSCS subscales as well as the Total P score and the TSBI score. Thus, it appears that masculinity and femininity are more important for men in determining their self-esteem. This may be explained 27 by our culture's greater demands for sex role conformity from men than from women. F's Relationship to Self-Esteem It is clear from the TSCS that androgynous subjects' self- esteem is not a result of M alone, as has been argued elsewhere (Antill and Cunningham, 1979; Jones,et al., 1978; Kelly and Worell, 1977; Spence and Helmreich, 1978). Both M and F were correlated with components of as well as global self-esteem. Particularly noteworthy are the correlations for men between F and components of self-esteem. Although they were not uniformly as large as the correlations between the respective component and M, they were far from negligible, nonsignificant or negatively correlated with self-esteem, as past research has found (Antill and Cunningham, 1979; Spence and Helmreich, 1978; Spence,et al., 1975). Thus, femininity does make an important contribution to self-esteem in males. For females, the results were an important departure from past research. F was correlated more than M with self-esteem and thus accounted for more variance than M. For women, then, femininity is more related than masculinity to self-esteem, regardless of whether one is referring to particular dimensions of self-esteem or global self-esteem. Antill and Cunningham (1979) have argued that self-esteem is a function of masculinity because research has found the high masculinity groups (androgynous and masculine) to have higher self-esteem than the low masculinity groups (feminine and undifferentiated). In the present research, masculine subjects were not significantly higher than feminine individuals, across all dimensions of self-esteem. This result is inconsistent with the argument that high levels of self-esteem 28 in androgynous individuals are mainly due to the masculine component. On the TSBI, however, past research contending that self- esteem is mostly a function of masculinity (Spence and Helmreich, 1978; Spence,et al., 1975) was convincingly replicated. For both males and females, M was substantially more highly correlated with and thus accounted for more variance in self-esteem than F. Considering the disparities in the findings with the TSBI and TSCS Social subscale, it therefore appears that the two are not equivalent measures of social self- esteem. This is particularly true for women. The TSBI and the Social subscale were only moderately correlated (.47); moreover, M accounted for most of the variance on the TSBI, while F explained the majority for the Social scale. For males, M accounted for most of the variance on the TSBI; on the TSCS, M accounted for only slightly more variance than F. Considerations for Future Research This study attempted to explain inconsistencies in previous research by improving upon the operationalization of self-esteem. To this end, specific dimensions of self-esteem were measured. The results of this work suggest that the specification of components of self-esteem is useful in clarifying the relationship between sex role orientation and self-esteem. Accordingly, it is recommended that future investigators attend to the complexity of self-evaluation by carefully delineating its components. This study also suggests that the current median split method for determining sex role categorization is in need of refinement. This gross typological system results in very heterogeneous categories. The consequences of this fact are illustrated by the present research. 29 Part of this study exactly replicates Peterson's (Note 2) work except for the procedure used to select subjects. The two studies had different results. Peterson (Note 2) found sex type and Total P scores to be related for males but not females. Here, sex type was related to Total P scores across all subjects. Since the two studies were the same in all other respects, the difference is attributed to the subject selection methodology. The problems related to current sex type categorization have been pointed out by others as well. Bem (1981) has proposed the hybrid method of classification in which sex type is determined by a median Split agg_the difference between the subject's M and F scores. Kelly and Worell (1977) have also suggested a different approach: An issue yet to be addressed is the assessment of sex role orientation using graded continua or interval scales, which should permit more precise behavioral predictions. Even though typologies may be useful for gross validation procedures, none of the present scales is currently used to derive a sggrg_rather than a type. Future research should attempt to refine the procedure for determining sex role orientation, either by type or score. Future investigators should also follow the suggestion to use M and F values and multiple regression techniques in the analysis of data because the information that can be gotten from analysis by sex type is limited (Bem, 1977; Lubinski, Tellegen and Butcher, 1981). In this study, although the analysis of variance resulted in no main effects (except for the Physical subscale) for sex, regression analysis utilizing M and F delineated more completely similarities and_ differences between the sexes. 3O Behavioral measurement of self-esteem is rare. Savin- Williams and Jacquish (1981) developed an observational measure of self-esteem and found that the observation of adolescent behavior provided a more accurate assessment of self—esteem than self-report measures. More behavioral measures of self-esteem need to be developed for use in this type of research. This study is limited in its generalizability due to its restricted age range. Accordingly, continuing research on the relation- ship between sex role orientation and psychological health needs to address its permutations over the developmental cycle. For children, androgyny may not be predictive of the best psychological adjustment. Worell (Note 3), for example, cites evidence which suggests that preschool and elementary age children who exhibit gender related play and social behavior may be more socially acceptable to peers and more positively . evaluated by teachers. Hall and Halberstadt (1980), furthermore, found that androgynous children were not more healthy or adjusted than other children. In fact, those children who deviated from traditional sex typing were more isolated by peers and teachers and more negatively evaluated. Longitudinal studies are indicated to further delineate the relationship between sex role orientation and psychological well-being over the life span. The sample is also limited by its locale, socioeconomic status and lifestyle. It would be instructive to research, for example, individuals of the same age who are not college students. O'Connor, Mann, and Bardwick (1978) have begun to address the generalizability issue in their work with a middle-aged upper middle class sample. 31 Summar This research attempted to elucidate the relationship between sex role orientation and components of self-esteem, as previous work was not sensitive to the complexity of self-esteem. The results did lend some support to the existence of different relationships between sex type and self-esteem depending on the dimension being considered. Social and personal self-esteem seemed particularly important components. In addition, high self-esteem was found to be a function of M agg_F. APPENDICES APPENDIX A CONSENT FORM I) O 32 MICHIGAN STATE UNIVERSITY Department of Psychology DEPAR‘ITIEII'ILL RESEARCH CONSENT POEM l have freely consented to take part in a scientific study being conducted by: under the supervision of: Academic Title: The study has been explained to me and I understand the ex« plonation that has been given and what Iy participation will involve. I understand that I am free to discontinue my participation in the study at any time without penalty. I understand that the results of the study will he treated in strict confidence and that I will remain anonymous. Within these restrictions, results of the study will be made available to me at my request. I understand that my participation in the study does not guar- antee any beneficial results to me. I understand that, at my reguest, I can receive additional explanation of the study after my participation is completed. Signed: Date: TITLE Of RESEARCH PROJECT: APPENDIX B BSRI (ITEMS 41 -70) TSBI (ITEMS 71-86) 33 INSTRUCTIONS All the questions in this booklet are to be answered on the printed answer sheet. You must use a #2 pencil. Before you begin, fill in all the information requested on the answer sheet and write your telephone number under the INSTRUCIOR' NAME. Do not fill in any number or letter under FORM. The items below inquire about what kind of a person you think you are. Each item consists of a pair of characteristics, with the letters A - E in between. For example. Hot at all artistic A.....B.....C.....D.....E Very artistic Luch pair decsrihos contradictory characteristics--thot is, you cannot be both at the same time, such as very artistic and not at all artistic. The letters form a scale between the two extremes. You are to choose a letter which describes where you {all on the scale. For example, if you think you have no artistic ability, you should choose A. If you think you are pretty good, you might choose 0. If you are only medium, you might choose C, and so forth. ”on go ahead and answer the questions on the answer sheet. Be sure to answer “VCEZ question, even if you're nob Sure, and use a #2 pencil. Do not be concerned that some items are repeated. REMEMBER TO ANSWER QUICKLY: YOUR FIRST IMPRESSION IS THE BEST. 1. Not at all ugnresuIVe Very ougresoIVo [\oowoobuooooC-coooDIoo-IE 2. Very uh Not at all whiny ...BC..O.C....ID..IO.E 1. IIHL .11. all Ilt'pc‘utlt-nt Very independent ACOOIU IDIUDC lllll D II. E 4 Hot at all arrog Very arrogant 5. Not at all emotional Very emotional 6. Very submissive Very dominant ACIOOOBIODIOCIOIII QIOIE 7. Very boastful Not at all boastful AIOOIIBIIOOOCOOOOODOOOO 8. Hot at all excitable in a major Very excitable in a major crisis crisis. AICOIIBOOIDCCO.IIQDOUOICE 9. Very passive AIOOCIBOOOOOCOOOOODIOO.CE 10. ll. 13. 14. 15. 19. 20. 21. 23. 24. 25. 26. 27. 34 t all egotistical ACCIOOBOIIOICIIIIIDIOOIOE Not ll able to devote self y to others. IIIBIOOOOCOOOOIDOOIIOE Not at complct Not at all spineless AIDCOCBOOCCO others IIDOOICUE Hot at all helpful AIIOCIBOOOOOCOO Hot at all competitive AOOOCIBIOOQCCOOCOO COCO-E Subordinates oneself to o AIIIIIBOIOIOCCOCOIDOI Very home oriented AOIOIOBOOIOICOCOOIDOOOOI Very greedy A.OICIBIIOIOCOOOIODOIOOOE Not at all kind AOOOOOBOOOIOCOIOOLD ..... E Indifferent to other's approval AOOIOOBOOOOICOOIOID ..... E Very dictatorial AIOIOOBIOOOOCOUOCODOOIOIE Feelings not easily hurt A.IIIIBIIIIOCICOCODOIOOIE Doesn't nag AIUCCCBOUOCICCC.C.D.I...E Not at all aware of feelings of others. AUDIOOBOOOIOCOIIIODtcolOE Can make decisions easily AOODCDBODIOOCCOIOOD.IOOOE Very fussy A.OIOOBOOOOICOOIOODIIIOOE Very egotistical Able to devote self completely to others. Very spineless Very gentle Very complaining Very helpful to others. Very competitive Hover subordinates oneself to others. Very worldly Not at all greedy Rig y needful of other's approval Not at l dictatorial Feelings e ily hurt Nags a lot Very aware of fe ings of others Has difficulty makin, decisions Not at all fussy 28. 29. 30. 31. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. ‘04. 45. 35 ves up very easily AOOOOOBI‘ODOCOOOIOD..‘OOE Very ynical .ICOCBIICC‘CKOOOOD ..... E Never at OBOOOOICIIOI.DOOOCOE Not at all 8e f—confident AOCOOIBOO .IC..IOOD..I..E Does not look out self: principled AIOOOIBOIICOC. nly for IOIDOOIIOE Feels very inferior AOOOIOBOIOCICICII. OOOOE Not at all hostile AIIIOOBOCOIICIOIODDOCCO E Not at all understandinp of others ADO-OIBQOUOICOIOIODIDODGE Very cold in relations with others AGO-.08..OIOCOIICODO0.00E Very servile ADIOOIBOOOIICOOIIIDOIOIOE Very little need for security AOOOIIBIOOIOCOIIOIDaOOOIE Not at all gullible AOOIIIBOOCO‘COOOIODOOOOIE Goes to pieces under pressure ADIOIIBOOOOOCIIOOIDIOOOIE Almost never defend my own beliefs. AOOOIOBOOOOOCII0.0DIIOOIE Not very affectionate ADIOOIBIOOOOCOOIIOD.O'CIE Not very conscientious AOOIIOBOOOCOCOOOIODOOOOIE Not very independent AOOIOIBOOOOOCIOOOIDIOCOOE Not very sympathetic ADOUIOBOIIIIC..IOIDIOICIE Never gives up easily Not at all cynical Cries very easily Very self-confident Looks out only for self: unprincipled. Feels very superior Very hostile Very understanding of others Ver warm in relations with others servile Very strong n ed for security Very gullible Stands up well under | 05131er Almost always defend my own beliefs. Very affectionate Very conscientious Very independent Very sympathetic 46. 47. 48. 49, 50. 51. $2. 53. 54. 56. 57. 58. 59. 60. 61. 62. 63. 6’0 . 36 Not very moody AOOOOIB"OOOCOOOIODOIIOOE Not very assertive ADO...“-III'CIIOOODOCOOOE not very sensitive to the needs of others. ADIOODBIOIOICDoOOODIOOD-E Not very reliable AIOIIOBOIIOICIOI.DDIOODOE Not a very strong personality AIOOIIBIOOOCCOIOICDIIIOIE Not very understanding AOOOIIBICIOICIOOOIDOOO-IE Not very jealous ACCI..B ..... C ..... D IIIII E Not very forceful AOCCOOBOOIIIC ..... DIICIOE Not very compassionate l‘oeonBocsoaCIooeoD-noan Not very truthful ADIOOOBCIIIOCIOOOIDOOOOOE Not much leadership ability A...-UBOIOOOCIOIIODOOODOE Not too eager to soothe hurt feelings. AIOOCOBIOIOUCOOOICDIDODGE Very secretive ADCIIIBOIDOOCOtOOODOOIOIE Not willing to take risks AOOIIOBOIIOOCOOOOIDOIto's Not very warm AOOOOOBIIDOOCOtOOIDOIOOOE Not very adaptable AOQCIOBCIOCICOOIOIDCOIOCE Submissive AUOOOIBIOIOOCOIOOODOIOOIE Not very tender ACICCIBIOIOOCOIOOODIIOOOE Conceited AOOOOOBOOOOOCOOOOODOOOOOE Very moody Very assertive Very sensitive to the needs of others. Very reliable A very strong personality Very understanding Very jealous Very forceful Very compassionate Very truthful Huch leadership ability Very eager to soothe hurt feelings Not very secretive Uilling to take risks Very warm Very adaptable Dominant Very tender Not conceited 65. 66. 67. 68. 69. ll. 73. 74. 37 5 hot very willing to take a Very willing to take a steel. stand. AOOUIOBOOIOCCOICOODOOOOOE Not very fond of children Very fond of children AOOOOOBIOOIOCOOODODOOOOIE not Very tactful Very tactful I‘OOIOOBDIIOICOOIOODCOOCOE Not very aggressive Very aggressive A‘oooeeBeaeeeCeoa-eDea-uaE not very gentle Very gentle AUODOIBIUOOOCOICIDDOOOOOE Very conventional Not very conventional AOOOCOBOIOCICIOOIODIIOOOB 1 am not likely to speak to people until they speak to me. A R C D E Nnt at all hot Slightly Fairly Very much charactersitic character- Very ‘of me 713' ‘c Hf mu I wunld describe myself as sell-confident. A B C D E Not at all Not Slightly Fairly Very much characteristic character- very of me istic of me I feel confident of my appearance. A B C D E Not at all Not Slightly Fairly‘ Very much characteristic character- very of me istic of me I am a good mixer. A B C D E Hot at all Not Slightly Fairly Very much characteristic character- very of me istic of me inien in a group of people, I have trouble thinking uf the right thiups to say. A H C D E Not at all Not Slightly Fairly Very much characteristic character- very of me istic of me 7f». 77. 73. 79. 80. 81. 38 6 when in a group of people. 1 usually do what the others want rather than make suggestions. A B C D E Not at all Not Slightly Fairly Very much characteristic character- very of me istic of me Vhen I am in disagreement with other people, my opinion usually prevails. A B C D E Not at all Not Slightly Fairly Very much characteristic character- very of me istic of me I would describe myself as one who attempts to master situations. A B C D E flat at all Not Slightly Fairly Very much characteristic character- very of me isric of me Other people look up to me. A B C D E Hot at all Hot Slightly Fairly Very much characteristic character- Very of me istic of me I enjoy social gatherings just to be with people. A B C D E Hot at all Not Slightly Fairly Very much characteristic character- very of me istic of no I make a point of looking other people in the eye. A B C D E Not at all Not Slightly Fairly Very much characteristic character- very istic of me I cannot seem to get others to notice me. A B C D E Not at all Not Slightly Fairly Very much characteristic character- very of me istic of me 83. 85. 86. 39 7 I would rather not have very much responsibility for other people. A B C D Hot at all Not Slightly Fairly Very character- Very of istic of me I feel comfortable being approached by someone in A B C D Not at all Not Slightly Fairly Very character- very of istic of me I would describe myself as indecisive. A 3 C D Not at all Not Slightly Fairly Very character- very of lot iv of na- 1 have no deubts about my social competence. A n C D Not at all Not Slightly Fairly Very character- very of istic of me E much characteristic me a position of authority. E much characteristic me E much characteristic me E much characteristic ac APPENDIX C TSCS 40 INSTRUCTIONS Please fill in your student number in place of your name at the side of your answer sheet. Leave the remaining ID information blank. The statements in this booklet are to help you describe yourseli as you see ynursell. l'lrasv respond to then; as if you were tlt'bo rihilzjn yourself _t__-_v xonrseli. Do not omit any item. Read each statement carelully; then select one hi the iiye responses listed below, Hn your answer sheet, put a £1151: around the tespousse you t‘lln‘u'. if you want to than“ an answer alter you hayo cin led it, do not erase it but put an X mark through the respOHse and then tirtle the la'ulu‘u‘t' VUII K-Ilil. PLFASE NOTF WELL: The questions in this booklet do not appear in numerical order. Shock and you will see that after question #1 comes #3, #5 and #19. This mums. you must he yrl'y taroful when circlim' your ('holu'r. on th. .msm r slut-t. work in the order the questions appear in the test booklet and he sure that the itrm Hulda! mm .nrv t/ll's'llll)‘ Hll Hut-.Innwvr r-lh‘rl ls Iln- sumo .IJ- llu' llt'l' mmlml' in w-ul l‘|’\'lt!\l. ‘I‘ml can lino ”l' [hr lllllhlu’l‘. .Ili‘lLL ilu- r_i)_',lli uh}; ml Hun l-uwl lot with w...—-—-—_....._.. the I‘:..l‘.l,l‘l'o on the answer sheet. RFSPUNSES: completely mostly partly false mostly completely false false and true true partly true 1 3 l -'o ' ; You will find these responses repeated at the bottom of each page to help you renember them. THANKS 41 Page l [143m I. I have a healthy body ................................................... I 3. I am an attractive person ................................................ 3 5. I consider myself a sloppy person ......................................... 5 19 I9. I am a decent sort of person ............................................. 2i . I am an honest person ................................................... 31 23. I am a bad person ...................................................... '25 37. I am 0 cheerful person .................................................. 37 39. I am a calm and easy going person ........................................ 39 41. lam a nobody ......................................................... 1‘1 55. l have a family that w0uld always help me in any kind of tr0ub|e ............. SS 57. I am a member of a happy family ......................................... 57 59. My friends have no confidence in me ..................................... 59 73. I om a friendly person .................................................. 73 75. lam popular with men .................................................. 75 77. I am not interested in what other people do ................................ 77 91 . I do not always tell the truth ............................................. 91 93. I get angry sometimes ................................................... 93 Completely Mostly Partly false Mostly Completely Responsvs- false false and true true portly true I 2 3 4 5 42 Page 2 film 2. I like to look nice and neat all the time ................................ 2 4. lam full of aches and pains ........................................... f0 6. I am a sick person .................................................... 6 20. I am a religi0us person ................................................ ’20 22. lam a moral failure .................................................. 22 24. I am a morally weak person ............................................ 210 38. I have a lot of self-control ............................................ 38 40. I am a hateful person ................................................. (‘0 42. I am losing my mind .................................................. [‘2 56. I am an important person to my friends and family ......................... 56 58. I am not laved by my family ........................................... 58 60. I feel that my family doesn't trust me ................................... 60 74. I am popular with women .............................................. 7“ 76. I am mad at the whole world ........................................... 76 78. I am hard to be friendly with ........................................... 73 92. Once in a while I think of things too bad to talk ab0ut .................... 92 94. Sometimes, when I am not feeling well, I am cross ........................ 9‘0 Completely Mostly Partly false Mostly Completely uses- false false and true true portly true I 2 3 4 5 4&3 Page 3 iii)?” 7. I am neither too fat nor too thin ........................................ 7 9. I like my looks iust the way they are .................................... 9 . II II. I w0uld lrke to change some parts of my body ............................. 2‘) 25. I am satisfied with my moral behavior ................................... 7 27. I am satisfied with my relationship to God ................................ ‘7 29. laught to go to church more ........................................... 29 . . . lo} 43. I am satisfied to be lust what I am ....................................... 45. lam iust as nice as I shauld be ......................................... 4‘3 47. I despise myself ...................................................... .. 6l. lam satisfied with my family relationships ................................ "I 63. I understand my family as well as I should ................................ M 65. I shOuld trust my family more ........................................... OS 79. lam as sociable as I want to be ......................................... 79 8I. I try to please others, but I don't overdo it ............................... M 83. I am no good at all from a social standpoint. . . . . . . . . . . . . ., ............... an 95. I do not like everyone I know ........................................... 95 97. Once in a while, I laugh at a dirty joke ................................ "7 Completely Mostly Partly false Mostly Completely Responses- false false and true true partly true I 2 3 4 5 44 Page 4 {462m 8. I am neither too tall nor too short ..................................... 3 l0. I don't feel as well as l shOuId ........................................ 10 I2. I shOuld have more sex appeal ........................................ 12 26. I am as religiOus as I want to be ....................................... 26 28. I wish I cauld be more trustworthy ..................................... 23 30. I shOuldn‘t tell so many lies .......................................... 3o 44. I am as smart as I wont to be .......................................... Mo 46. I am not the person I wauld like to be .................................. ‘06 48. I wish I didn't give up as easily as I do ................................. ‘03 62. I treat my parents as well as | shOuld (Use past tense if parents are not living). 62 64. I am too sensitive to things my family say ............................... 64 66. I should lave my family more ......................................... 66 80. I am satisfied with the way I treat other people .......................... 80 62, I should be more polite to others ...................................... 32 84. I Ought to get along better with other people ............................ 8’0 96. I gossip a little at times .............................................. 96 98. At times I feel like swearing .......................................... 93 Completely Mostly Partly false Mostly Completely Responses - false false and true true partly true I 2 3 4 5 Page 5 I3. I take good care of myself physically ................................. IS. I try to be careful ab0ut my appearance ............................... I7. I often act like I am "all thumbs" .................................... 3l . I am true to my religion in my everyday life ........................... 33. I try to change when I know I'm doing things that are wrong .............. 35. I sometimes do very bad things ....................................... 49. I can always take care of myself in any situation ....................... 5l. I take the blame for things withOut getting mad ......................... 53. I do things withOut thinking about them first ........................... 67. I try to play fair with my friends and family ............................ 69. I take a real interest in my family .................................... 7I. I give in to my parents. (Use past tense if parents are not living) ......... 85. I try to understand the other fellow's point of view ...................... 87. I get along well with other people .................................... 89. I do not forgive others easily ........................................ 99. I w0uld rather win than lose in a game ................................ Completely Mostly Partly false Mostly Completely Responses - false false and true true panlyttue l 2 3 4 5 45 Item Nu. 13 15 17 3| 33 35 49 51 53 b7 69 71 85 87 89 99 46 Page 6 tab?!“ I4. I feel good most of the time ........................................... 14. I6. I do p00rly in sports and games ........................................ 16 l8. lom a poor sleeper .................................................. 18 32. I do what is right most of the time ..................................... 32 34. I sometimes use unfair means to get ahead .............................. 3t. 36. l have trauble doing the things that are right ............................ 35 50. I solve my problems quite easily ....................................... 50 52. | change my mind a lot ............................................... 52 54. I try to run away from my problems ..................................... 54 68. I do my share of work at home ......................................... 68 70. l quarrel with my family .............................................. 70 72. I do not act like my family thinks I shOuld .............................. 72 86. | see good points in all the people I meet .............................. 86 88. I do not feel at ease with other people ................................. 88 90. I find it hard to talk with strangers ..................................... 9O IOO. Once in a while I put off until tomorrow what I ought to do today .......... 100 Completely Mostly Partly false Mostly Completely Responses- false false and true true partly true I 2 3 4 5 REFERENCE NOTES REFERENCE NOTES Wetter, R.E. Levels of self-esteem associated with four sex role categories. In R. Bednar (Chair), Sex roles: Masculine, feminine, androgynous or none of the above? Symposium presented at the meeting of the American Psychological Association, Chicago, August, 1975. Peterson, L.M. Self-esteem and sex type: A function of the scale? Paper presented at the meeting of’the Eastern PsyChdlogical Association, April, 1981. Worell, J. Androgyny and psychological well-being: Some ideological dilemmas. In M.A. Sedney (Chair), Androgyny: Conceptual and methodological distinctions. Symposium presented at the meeting of the Society for Research in Child Development, Boston, April, 1981. 47 LIST OF REFERENCES LIST OF REFERENCES Antill, J.K., and Cunningham, J.D. Self-esteem as a function of masculinity in both sexes. Journal of Consulting and Clinical Psychology, 1979, 41, 783-785. Bem, S.L. The measurement of psychological androgyny. Journal of Consulting and Clinical Psychology, 1974, 58, l55-l62. Bem, S.L. Sex role adaptability: One consequence of psychological androgyny. Journal of Personality and Social Psychology, 1975, 81, 634-643. Bem, S.L. 0n the utility of alternative procedures for assessing psychological androgyny. Journal of Consulting and Clinical Psychology, 1977, 48,196-205. Bem, S.L. Theory and measurement of androgyny: A reply to the Pedhazur-Tetenbaum and Locksley-Colten critiques. Journal of Personality and Social Psychology, 1979, 88, 1047-1054. Bem, S.L. Bem Sex-Role Inventory: Professional Manual. Palo Alto, CA.:’ Consulting Psychologists Press, l981. Bem, S.L., Martyna, W., and Watson, C. Sex typing and androgyny: Further explorations in the expressive domain. Journal of Personality and Social ngchology, 1976, 88, 1016-1023. Berger, E.M. The relation between expressed acceptance of self and expressed acceptance of others. Journal of Abnormal and Social Psychology, 1952, 51, 778-782. Berzins, J.I., Welling, M.A., and Wetter, R.E. A new measure of psychological androgyny based on the Personality Research Form. Journal of Consulting and Clinical Psychology, 1978, 46, 126-138. Block, J.H. Conceptions of sex role: Some cross-cultural and longitudinal perspectives. American Psychologist, 1973, 28, 512-527. Broverman, I.K., Vogel, S.R., Broverman, D.M., Clarkson, E.E., and Rosenkrantz, P.S. Sex role stereotypes: A current appraisal. Journal of Social Issues, 1972, 88, 59-78. Carlson, R. Sex differences in ego functioning. Journal of Consulting_ and Clinical Psychology. 1971, 81, 367-377. Constantinople, A. Masculinity-femininity: An exception to a famous dictum? Psycholggical Bulletin, 1973, 89, 389-407. 48 49 Coopersmith, S. The antecedents of self-esteem. San Francisco: Freeman, l967. Eagly, A.H. Involvement as a determinant of response to favorable and unfavorable information. Journal of Personality and Social Psychology Monograph, 1967, Z, (3. Pt. 2). Fitts, W.H. Tennessee Self-Concept Scale: Manual. Nashville, TN.: Counselor Recordihgs and Tests, 1965. Flaherty, J.F., and Dusek, J.B. An investigation of the relationship between psychological androgyny and components of self-concept. Journal of Personality and Social Psychology, 1980, 88, 984- 992. Gilbert, L.A. Toward mental health: The benefits of psychological androgyny. Professional Psychology, 1981, 18, 29-38. Hall, J.A., and Halberstadt, A.B. Masculinity and femininity in children: Development of the children's Personal Attributes Questionnaire. Developmental Psychology, 1980, 18, 270-280. Harrington, D.M., and Andersen, S.M. Creativity, masculinity, femininity and three models of psychological androgyny. Journal of Personality and Social Psychology, 1981, 81, 744-757. Heilbrun, A.B., Jr. Sex role, instrumental-expressive behavior and psychopathology in females. Journal of Abnormal Psychology, 1968, 88, 131-136. Heilbrun, A.B. Measurement of masculine and feminine sex role identities as independent dimensions. Journal of Consulting and Clinical Psychology, 1976, 55, 183-190. Helmreich, R., and Stapp, J. Short forms of the Texas Social Behavior Inventory, an objective measure of self-esteem. Bulletin of the Psychonomic Society, 1974, 8, 473-475. Helmreich, R., Stapp, J., and Ervin, C. The Texas Social Behavior Inventory (TSBI): An objective measure of self-esteem or social competence. JSAS Catalog of Selected Documents in Psychology, 1974, g, 79. Jones, W.H., Chernovetz, M.E.0., and Hansson, R.0. The enigma of androgyny: Differential implications for males and females? Journal of Consultingyand Clinical Psychology, 1978, 58, 298-313. Kelly, J.A., amd Worell, J. New formulations of sex roles and androgyny: A critical review. Journal of Consulting and Clinical Psychology, 1977, 81,1101-1115. 50 Kelly, J.A., Furman, W., and Young, V. Problems associated with the Locksley, Lubinski, O'Connor, Pedhazur, Savin-Will Spence, J Spence, J Worell, J. typological measurement of sex roles and androgyny. Journal of Consulting and Clinical Psychology, 1978, 58, 1574-1576. A., and Colten, M.E. Psychological androgyny: A case of mistaken identity? Journal of Personality and Social Psychology, 1979, 81, 1017-1031. 0., Tellegen, A., and Butcher, J.N. The relationship between androgyny and subjective indicators of emotional well-being. Journa1 of Personality and Social PSychology, 1981, 59, 722-730. K., Mann, D.W., and Bardwick, J.M. Androgyny and self-esteem in the upper-middle class: A replication of Spence. Journal of Consulting and Clinical Psychology, 1978, 88, 1168-ll69. E.J., and Tetenbaum, T.J. Bem Sex Role Inventory: A theoretical and methodological critique. Journal of Personality and Social Psychology, 1979, 81, 996-1016. iams, R.C. and Jaquish, G.A. The assessment of adolescent self-esteem: A comparison of methods. Journal of Personality, 1981. 52, 324-336. .T., and Helmreich, R.L. Masculinity and femininity: Their psycholo ical dimensions, correlates, and antecedents. Austin, X.: University Of Texas Press, 1978. .T., 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 Psycholggy, 1975, 88, 29-39. Sex roles and psychological well-being: Perspectives on methodology. Journal of Consulting and Clinical Psychology, 1978. fig, 777-791. I‘IICHI 11111111111111“