Predictors of Homophobia in Female College Students

Sex Roles, Vol. 42, Nos. 5/6, 2000 Predictors of Homophobia in Female College Students Susan A. Basow1 and Kelly Johnson Lafayette College Although much research has investigated predictors of homophobia in males, little attention has been given to the predictors of homophobia in females. The current study investigated how self-esteem, self-discrepancy (how much females think they fit others’ expectations of how they should act with respect to gender-stereotyped attributes), and gender-attribute importance (how important gender stereotypes are to their gender identity) related to homophobia in 71 primarily White and middle-class college women. Other predictors evaluated were gender role attitudes, authoritarian attitudes, and extent of contact with lesbians and gay men. Results indicated that unlike for college men, self-discrepancy did not correlate with attitudes toward lesbians. The highest correlations with homophobia for college women were authoritarian attitudes, belief in sex role egalitarianism, degree of contact with gay men and lesbians, and importance of feminine attributes to participant’s femininity. The only significant predictor, however, was authoritarian attitudes, which accounted for 62% of the variance. Numerous research studies have investigated negative attitudes toward homosexuals. Among the most common correlates of homophobia are authoritarianism, traditional gender role attitudes, religiosity, and extent of contact with homosexuals (Herek, 1988; Lance, 1987; Newmann, 1989; Whitley & Lee, 2000). However, most of the research has focused on males’ attitudes toward homosexuals, usually explicitly or implicitly toward gay men. Research that has examined attitudes toward gay men and lesbians separately has found that men’s attitudes toward gay men are more negative than men’s attitudes toward lesbians, while women’s attitudes are equal toward both groups of homosexuals (Kite & Whitley, 1998). However, little 1 To whom correspondence should be addressed at Psychology Department, Lafayette College, Easton, Pennsylvania 18042-1781. 391 0360-0025/00/0300-0391$18.00/0  2000 Plenum Publishing Corporation 392 Basow and Johnson attention has been given to female homophobia except that it seems to be less intense than men’s homophobia (Herek, 1988; Herek & Glunt, 1993; Kite & Whitley, 1998). The present study focuses specifically on predictors of female attitudes toward lesbians. Although research has found that various types of prejudice share some similar correlates (for example, Social Dominance Orientation; Whitley, 1999), homophobia appears to have some unique predictors. This may be because gay men and lesbians are still, for the most part, a socially acceptable outgroup. In addition, homosexuals appear to threaten traditional family and gender-role values more than, say, African Americans. Thus rightwing authoritarianism has been found to be particularly predictive of negative attitudes toward homosexuals (Altemeyer, 1996; Herek, 1988; Whitley, 1999). People who score high on authoritarianism are characterized as accepting traditional values and norms, possessing a willingness to submit to established authorities, and displaying a tendency to aggress against those disapproved of by the authorities (Altemeyer, 1996; Haddock & Zanna, 1998). Because those high in authoritarianism are committed to maintaining the traditional family structure, they feel threatened by liberalization and people who threaten their conventional values, such as homosexuals. Another important and related correlate of homophobic attitudes is belief in traditional gender roles (Kerns & Fine, 1994). Since most people view homosexuals as violating traditional roles (Herek, 1995), those who hold to those roles most strongly tend to be the most negative toward homosexuals, both male and female. Traditional gender role attitudes consist of the beliefs that family labor should be divided based on gender, with men contributing financial support and women providing child care. Nontraditional attitudes consist of beliefs that labor should be divided on the basis of needs, abilities, and family schedules, and career choices and child-care responsibilities should be regarded equally for men and women (Kerns & Fine, 1994). The fact that men tend to hold more traditional attitudes toward gender roles than do women may account for men’s higher homophobia scores. However, such attitudes should still predict homophobia in women. A third variable related to homophobic attitudes is degree of actual contact with lesbians and gay men. Heterosexuals who personally know a lesbian or gay man appear to hold more positive attitudes toward homosexuals as a group than do individuals without such contacts, and the more contacts a person has, the more favorable the attitudes (Herek, 1988; Herek & Capitanio, 1996; Herek & Glunt, 1993). Although those with favorable attitudes may be more likely to become friendly with gay men and lesbians, Allport’s (1954) contact hypothesis suggests that contact itself may reduce prejudice. Heterosexual women appear to have more contacts Homophobia in Females 393 with known lesbians and gay men than do heterosexual men, another possible mediator of women’s lesser homophobia (Herek, 1995). However, variability among women in relation to the number of gay men or lesbians known should still predict homophobic attitudes. Homophobia, at least in men, also has been viewed as a reaction to self-fears about one’s masculinity. Herek (1986) proposes that the pressure on men to conform to heterosexual masculinity causes male heterosexuals to develop anxiety over not fulfilling these expectations. According to this analysis, men who are most insecure about their own masculinity are most likely to hold negative attitudes toward homosexuals, especially toward gay men. Theodore and Basow (in press) tested this suggestion by assessing how much college men felt they conformed to the gender-stereotypic attributes expected of them (ought self-discrepancy) along with how important it was for them to do so (attribute importance). Their results showed that college men who felt it important to have masculine attributes and not to have feminine attributes and who felt that they did not fulfill masculine stereotypes were the most homophobic, supporting Herek’s hypothesis about heterosexual masculinity. This research also supports Higgins’ (1987) and Pelham and Swann’s (1989) research on self-discrepancy. These investigators found that self-discrepancy along dimensions of importance to the individual might lead to low self-esteem, anxiety, and threat, which might then result in defensive attitudes. Men concerned about their masculinity will fear and avoid circumstances that may lead to their masculinity being questioned, such as interaction with and acceptance of homosexuals. Thus, Theodore and Basow’s (in press) findings would support the view that homophobia sometimes is a reaction to threat to one’s self, at least in men. But would the same dynamic occur in women? Heterosexual femininity, although based on attracting men as sexual partners, appears more variable than heterosexual masculinity with respect to gender-related traits (Basow, 1992). According to Rich (1980), heterosexuality is not an aspect of femininity that must be affirmed and proven, but is instead an aspect of femininity that is assumed and compulsory. Therefore, females may not need to assert their heterosexuality as much as males do. Furthermore, women’s gender role socialization is more flexible than is men’s (Basow, 1992). Heterosexual femininity does not seem as tied to possessing genderappropriate traits and not possessing ‘‘inappropriate’’ ones as does heterosexual masculinity. Thus, homophobia in women may not be a reaction to threat to one’s self. The current study tests Theodore and Basow’s (in press) model on a sample of college women. In addition to measuring gender-related self-discrepancy and attribute importance, Theodore and Basow (in press) also assessed self-esteem. Relying upon Wills’ (1981) downward comparison theory of self-esteem and 394 Basow and Johnson research on low-self-esteem individuals (e.g., Hogg & Sunderland, 1991), Theodore and Basow (in print) predicted that self-esteem and homophobia in males would be negatively correlated. The downward comparison theory proposes that an individual’s self-esteem can be enhanced and protected through direct comparison of oneself with a less fortunate other. Although it seems to follow that individuals with lower self-esteem would hold more negative attitudes toward homosexuals than those with higher self-esteem, Theodore and Basow (in press) did not find support for the prediction. In this study, we use the same measure for self-esteem ratings to examine if a negative correlation between self-esteem and homophobia occurs for females. In this extension of Theodore and Basow’s (in press) research, a different measure of homophobia was needed since the Attitudes Towards Homosexuality Scale (Millham, San Miguel, & Kellogg, 1976) used by those researchers does not distinguish between the gender of the target in assessing attitudes toward homosexuals. Because we were interested in examining females’ attitudes toward lesbians in particular, since they would be viewed as more threatening than gay men to the female ‘‘self,’’ we used Herek’s (1988) Attitudes Towards Lesbians and Gay Men Scale. This scale yields two subscales: five questions assessing attitudes toward lesbians and five questions assess attitudes toward gay men. The scores may be combined to determine overall attitudes toward homosexuals. The present study is thus both an extension of Theodore and Basow’s (in press) research to women and an examination of more traditional correlates of homophobia in women as well. It is predicted that homophobia in females is related to authoritarianism, traditional gender role attitudes, and little contact with lesbians and gay men, but is not related to feelings of inadequacy about living up to society’s expectations about one’s femininity. METHOD Participants Seventy-one undergraduate middle- and upper-middle-class women at a private liberal arts college in the Northeast volunteered to participate in the study. Most (80%) of the volunteers received extra credit in a psychology course, and 20% were recruited from a campus sorority and received no extra credit. Nearly one half (49%) of the sample were first-year students, 23% were sophomores, 16% juniors, and 12% seniors. Ethnically, 97% of the participants were White, and 3% were African American. The vast majority was heterosexual. Homophobia in Females 395 Materials Self-esteem (SE) Spence and Helmreich’s (1978) Texas Social Behavior Inventory Short Form A was used to assess each participant’s self-esteem. Participants responded to a series of 16 statements along a 5-point Likert scale ranging from 1 (not at all characteristic of me) to 5 (very much characteristic of me). Higher scores indicate higher self-esteem. Internal reliability was .75 (Cronbach alpha) for this sample. Self discrepancy The Theodore and Basow (in press)-modified version of the Self Attribute Questionnaire (Pelham & Swann, 1989) called the Ought Self Questionnaire was used to measure actual versus ought self-discrepancy scores along both masculine and feminine traits. Participants assessed themselves relative to their ought selves, defined as ‘‘the person you would be if you were exactly as you think your peers expect you, as female, ought to or should be,’’ on 16 gender-linked items taken from the Personal Attributes Questionnaire (Spence & Helmreich, 1978). For the 16 items, respondents rated themselves along a 9-point Likert scale ranging from 1 (very different from how I think my peers expect I ought to or should be) to 9 (very much like how I think my peers expect I ought to or should be). One subscale assessed self-discrepancy along masculine traits (OSDMASC), another along feminine attributes (OSDFEM). Higher scores indicate lower levels of self-discrepancy. Internal reliability (Cronbach alpha) was .74 for OSDMASC and .87 for OSDFEM. (In Theodore & Basow, in press, reliabilities were .75 and .79, respectively.) Attribute importance The importance participants place on certain attributes was measured with Theodore and Basow’s (in press) Attribute Importance Questionnaire, modeled after Pelham and Swann’s (1989) measure of the same name. Each respondent rated the extent to which 23 attributes (the 16 from the Ought Self Questionnaire plus 7 fillers) were important to her self-identity as a female. Responses were measured along a 9-point Likert scale ranging from 1 (not at all important to my femininity) to 9 (extremely important to my femininity). The 16 gender-linked attributes yielded two subscales: 396 Basow and Johnson attribute importance along feminine traits (AIFEM) and attribute importance along masculine traits (AIMASC). Reliabilities were .84 and .77, respectively, compared to .91 and .69 in Theodore & Basow (in press). Gender role ideology Gender role attitudes were measured with the short form (Form BB) of the Sex-Role Egalitarianism Scale (SRE; Beere, King, Beere, & King, 1984). The scale consists of 25 items measuring attitudes concerning marital, parental, employment, social-interpersonal, heterosexual, and educational roles of men and women. Participants were asked to rate the extent to which they personally agreed with each statement. Responses were rated along a 5-point Likert scale ranging from strongly agree (1) to strongly disagree (5). Higher scores indicate more egalitarian gender role attitudes. A Cronbach alpha of .86 was obtained. Authoritarian Attitudes Altemyer’s (1996) Right Wing Authoritarianism (RWA) scale was used to assess the extent to which participants believed in authoritarian submission, authoritarian aggression, and conventionalism. The 34-item questionnaire was interspersed with items from the Attitudes Towards Lesbians and Gay Men Scale (described below), and the combined survey was entitled Social Attitudes Questionnaire. (Scales were scored separately, and combined only to disguise the homophobia measure.) Participants were asked to rate the extent to which they agreed with statements concerning topics from child obedience to political ideology. Responses were rated along a 9-point Likert scale ranging from 1 (strongly disagree) to 9 (strongly agree). The first 5 items were considered ‘‘table-setters’’ and were not scored. Higher scores indicate more rigid, authoritarian attitudes. High internal reliability (Cronbach alpha of .93) was obtained. Homophobia The short form of Herek’s (1988) Attitudes Toward Lesbians and Gay Men (ATLG) scale was used. The 10-item questionnaire was disguised in combination with the RWA questionnaire. Participants rated the extent to which they agreed with statements concerning homosexuals and homosexuality in general along a 9-point Likert Scale. The instrument yields subscales Homophobia in Females 397 for attitudes toward lesbians (ATL) and attitudes toward gay men, as well as a combined score. Higher scores indicate attitudes that are more homophobic. Internal reliability was obtained for the ATLG scale and for the attitudes toward lesbians subscale and attitudes toward gay men subscale (Cronbach alphas of .95, .88, and .91, respectively). Personal Contacts The extent of personal contact with homosexuals was assessed by two items included in the Demographics Questionnaire. Participants were asked if they know or have known, closely or casually, any lesbians or gay men. A follow-up question asked how many homosexuals they knew of each gender. The second question was used in the analyses, with ‘‘no contact’’ coded as 0. Procedure Participants were administered the questionnaire by a female experimenter in groups ranging from 2 to 15 participants. After completing a research participation form to receive extra-credit (if applicable), participants read a brief description of the study (an investigation of political attitudes in females) and told they could withdraw at any time. Following their consent, questionnaires were distributed in the following order: selfesteem, ought self-discrepancy, attribute importance, sex-role egalitarianism, RWA/ATLG, demographic (including personal contacts questions). Upon completion, participants were thanked for their participation. A debriefing summary was distributed to each participant via campus mail once all data were collected. RESULTS Correlational analyses (see Table I) indicated significant positive correlations between attitudes toward lesbians (ATL) and attribute importance of feminine traits (AIFEM) and right-wing authoritarianism (RWA). Significant negative correlations occur between ATL and sex role egalitarianism (SRE) and the number of gay men and lesbians known to the respondent. College women who have the most negative attitudes toward lesbians have the highest right-wing authoritarianism scores, place the greatest importance on having feminine attributes, have the least belief in sex role 398 Table I. Correlations Among the Variables (N ⫽ 71) ATL AIFEM AIMASC OSDFEM OSDMAS SE SRE RWA NUMG AIFEM AIMASC OSDFEM OSDMAS .276* .003 .411*** .086 .445*** .207* ⫺.110 .290** .259* .168 SE SRE RWA NUMG NUML ⫺.049 .058 .305** .019 .200 ⫺.385** ⫺.100 .298** .078 .179 .154 .788*** .253* .081 .076 ⫺.141 .023 ⫺.491*** ⫺.324** ⫺.040 .148 .087 ⫺.003 .159 .171 ⫺.355** ⫺.443*** .023 .240* ⫺.131 .168 .156 .328** ⫺.491*** .438* Note: ATL, Attitudes Toward Lesbians; AIFEM, Attribute Importance of Feminine Traits; AIMASC, Attribute Importance of Masculine Traits; OSDFEM, Ought Self-Discrepancy for Feminine Traits; OSDMAS, Ought Self-Discrepancy for Masculine Traits; SE, Self-Esteem; SRE, Sex Role Egalitarianism; RWA, Right-Wing Authoritarianism; NUMG, number of gay men; NUML, number of lesbians. *p ⬍ .05; **p ⬍ .01; ***p ⬍ .001 (one-tailed tests). Basow and Johnson Homophobia in Females 399 egalitarianism, and know the fewest lesbians and gay men. (Identical results exist for correlates of attitudes toward gay men and are not reported here.) In order to examine whether Theodore and Basow’s (in press) model of ‘‘self-discrepancy’’ applies to attitudes toward lesbians, a multiple regression analysis was performed using the Theodore and Basow variables of selfesteem, ought self-discrepancy (OSDMASC and OSDFEM), and attribute importance (AIMASC and AIFEM) as predictors. As seen in Table II, only AIFEM was a significant predictor, accounting for only 7.5% of the variance in ATL scores. Including the other variables in a second multiple regression analysis revealed a more powerful predictor. As shown in Table II, RWA alone predicted 61.6% of the variance in ATL scores. No other variable contributed anything additional, probably because RWA correlated significantly with attribute importance of feminine traits (AIFEM), low belief in sex role egalitarianism, and little contact with lesbians and gay men (see Table I). DISCUSSION The main hypothesis, that homophobia in college women would function differently than it does for college men, was supported. Unlike the findings of Theodore and Basow (in press), homophobia in women is not significantly predicted by feelings of inadequacy about living up to cultural expectations of appropriate gender traits. Negative attitudes toward lesbians and gay men were not significantly correlated with self-esteem, selfdiscrepancy along masculine or feminine traits, or attribute importance along masculine traits. However, a significant correlation was found between homophobic attitudes and attribute importance along feminine traits, Table II. Predictors of Attitudes Toward Lesbians Using Model 1 (Theodore and Basow Variables) and Model 2 (All Variables), Stepwise Multiple Regressions Variable Entered B SE B Beta Model 1, step 1. AIFEM 2.524 .978 .297 (AIMASC, OSDFEM, OSDMASC, SE not in equation) Adjusted R 2 .075 F 6.666* Model 2, step 1. RWA .199 .019 .788 .616 108.442*** (AIMASC, OSDFEM, OSDMASC, SE, SRE, NUMGAY, NUMLES not in equation) Note: ATL, Attitudes Toward Lesbians; AIFEM, Attribute Importance of Feminine Traits; AIMASC, Attribute Importance of Masculine Traits; OSDFEM, Ought Self-Discrepancy for Feminine Traits; OSDMASC, Ought Self-Discrepancy for Masculine Traits; SE, Self-Esteem; SRE, Sex Role Egalitarianism; RWA, Right-Wing Authoritarianism; NUMGAY, number of gay men; NUMLES, number of lesbians. *p ⬍ .05; **p ⬍ .01; ***p ⬍ .001 (one-tailed tests). 400 Basow and Johnson although the correlation was small and predicted only 7.5% of the variance in ATL scores. College women who feel stereotypic feminine traits are important to their femininity are more homophobic than their female peers who believe such traits are less important. Theodore and Basow (in press) found a similar pattern for college men: greater homophobia was related to the importance of stereotypic masculine traits for one’s masculinity. This finding may reflect the traditional belief that homosexuality is related to gender role nonconformity, such that lesbians possess inherently masculine characteristics and gay men possess inherently feminine characteristics (Kite & Deaux, 1987). Thus, heterosexual females who view the possession of feminine traits as important for females may see lesbians as possessing masculine rather than feminine traits and therefore are more likely to view them negatively. Several possibilities exist as to why certain self-variables (self-discrepancy and attribute importance of gender-related traits) were significant predictors of homophobia in males, but not in females. As Herek (1986, 1988) suggests, the male gender role emphasizes the importance of heterosexuality to masculinity, and males feel the need to affirm their masculinity by rejecting men who violate the heterosexual norm. Thus, homophobia in males serves both a defensive function by asserting dissimilarity from male homosexuals and a social-expressive function by gaining acceptance from their male heterosexual peers. Femininity in females, however, is less rigid than masculinity in men and less tied to heterosexuality, which is assumed and therefore does not need to be ‘‘proven’’ (Basow, 1992; Rich, 1980). Females have less of a need to ‘‘defend’’ against homosexuality and therefore are more likely to have personal interaction with lesbians and gay men (Herek, 1988). Such interaction is correlated with more positive attitudes toward lesbians and gay men, in this study and elsewhere (Herek & Capitanio, 1996). In both Theodore and Basow (in press) and the current study, the measure of self-esteem used (TSBI) did not correlate with measures of homophobia, although it did correlate with the importance of masculine attributes in both studies. As other research has found, the instrumental traits that make up most measures of stereotypic masculinity are strongly correlated with high self-esteem (Whitley, 1988). However, high ought selfdiscrepancy along masculine traits was associated with low self-esteem in Theodore and Basow (in press), a finding not replicated here, again suggesting that gender role conformity may be more important for males than females. More research is needed, perhaps with different measures of selfesteem, to examine if that variable is indeed related to homophobia, as Will’s (1981) downward comparison theory would suggest. Although homophobia in women does not appear to be a reaction to Homophobia in Females 401 one’s self, as it is in part for men, it is related to other social attitudes, namely authoritarianism and belief in sex role inequality. Indeed, authoritarianism alone accounts for more than three fifths of the variance in ATL scores. This supports a great deal of previous research with college women and men as well as the general population that the best predictor of homophobia is adherence to the type of rigid traditional attitudes exemplified by the Right Wing Authoritarianism Scale (Altemeyer, 1996; Haddock & Zanna, 1998; Whitley, 1999; Whitley & Lee, 2000). Altemeyer (1996) proposes that right-wing authoritarians (RWAs) have three personality characteristics in common: (1) a strong acceptance of and commitment to the traditional norms, values, and conventions of their society, including gender role norms; (2) a general willingness to accept the statements and actions of the established authorities and to comply with their instructions; and (3) a tendency to aggress against or intentionally cause harm to individuals if they believe the proper authority approves of the aggression or that it will help preserve such authority. Since homosexuals typically are viewed as violating traditional gender role and family norms, and negative attitudes toward homosexuals are frequently promoted by many religious and political leaders, it is understandable why RWA scores predict homophobia so strongly. Although there is strong theoretical support for the relationship between RWA and homophobia, there is a slight confound inherent in RWA since two questions directly relate to attitudes toward homophobia. In this study in particular, the questions on the two measures were interspersed, which might further enhance the relationship between them. Still, as Whitley and Lee (2000) found, even when RWA is scored without the two homosexual questions and even when other measures of authoritarian attitudes are used, the strong correlation between authoritarianism and homophobia (measured variously) remains (r ⫽ ⫺.52). The current findings support those of Whitley and Lee despite the fact that the participants in the current sample were both less authoritarian (Ms ⫽ 3.95 compared to 5.24) and less homophobic (ATL Ms ⫽ 3.12 compared to 5.47). The current investigation also confirmed the hypothesis that egalitarian attitudes concerning gender roles would be significantly correlated with more positive attitudes toward gay men and lesbians. Belief in sex role inequality is a component of RWA, as suggested by the strong correlation between the two variables and the fact that it adds nothing to the prediction of homophobia beyond RWA. Previous research that found that belief in traditional gender roles predicted homophobia did not also measure RWA, the more powerful variable. Although past research has suggested that violations of traditional gender roles may be particularly aversive to heterosexual males because of the perceived threat to male power and privilege (Kerns & Fine, 1994), the current study suggests that variability in homo- 402 Basow and Johnson phobia exists within females according to gender role attitudes. Because individuals tend to assume that a person’s sexual orientation is part of his or her gender role (Lance, 1987), homosexuals may be seen as violating the normative prescriptions for the behavior of males and females by those who hold fixed gender role standards. Also, many traditional views hold that homosexuals suffer from gender confusion because they deviate from ‘‘normal’’ (i.e., heterosexual) sexual feelings (Lance, 1987). Thus, negative attitudes toward homosexuals are an integral part of traditional sex role ideology. Regarding contact experiences with homosexuals, the findings of this investigation are consistent with Herek and Capitanio’s (1996) findings that the relationship between contact and favorable attitudes toward homosexuals was stronger to the extent that respondents reported multiple contacts, more intimate contacts, and contacts that involved direct disclosure of sexual orientation. Herek and Capitanio (1996) have suggested that from interacting with lesbians and gay men, heterosexuals can receive information about what it means to be gay and observe behaviors or characteristics that are inconsistent with cultural stereotypes of homosexuals. The relationship between contact experiences and attitudes toward homosexuals may be reciprocal in that heterosexuals who already possess positive attitudes are subsequently more likely than others to experience contact with gay men and lesbians (Herek & Glunt, 1993). Although the extent of contact with homosexuals, sex role egalitarianism, and authoritarian attitudes were all significantly correlated with negative attitudes toward gay men and lesbians, authoritarian attitudes were found to be the best predictor. All of the significant factors were correlated with RWA scores and did not add to the variance in homophobia scores. The findings suggest the possibility that sex role ideology, the extent of contact with homosexuals, and homophobia are all manifestations of the higher order construct of authoritarian conservatism, at least for the predominantly White, heterosexual respondents in this study (Herek, 1988). Unfortunately, changing such embedded attitudes is very difficult (Eagly & Chaiken, 1998). One promising approach with students high in RWA is to modify their negative stereotypes and symbolic beliefs about gays. Haddock and Zanna (1998) found that doing so resulted in more favorable evaluations, at least of gay men. Several limitations to the current investigation should be noted. First, the number of participants was relatively small and all were from a predominantly White, middle- to upper-class liberal arts college. A more representative sample would be needed to generalize the results to the larger population. In addition, some of the participants may have been homosexual themselves, although the low base rate of homosexuality in the general Homophobia in Females 403 population (e.g., Diamond, 1993) and among the particular student population sampled decreases the likelihood of bias on that account. Finally, a small percentage of the participants may have been aware of the experimenter’s involvement with gay, lesbian, and bisexual issues on campus. This knowledge, along with the social desirability factor, may have biased participants’ responses, perhaps accounting for the relatively low levels of homophobia and authoritarianism found. Nonetheless, the pattern of correlations is unlikely to have been affected. Overall, homophobia in females appears to be structured somewhat differently than homophobia in males. While the attitudes of both females and males are strongly predicted by adherence to right-wing authoritarian beliefs, only males seem to endorse homophobic statements as a defensive strategy against threats to their gender identity. Future research might explore these gender differences further. Finding the most significant predictor of homophobia has important implications for the reduction of negative attitudes toward homosexuals, which may require different strategies for men and women. REFERENCES Allport, G. (1954). The nature of prejudice. Reading, MA: Addison-Wesley. Altemeyer, B. (1996). The authoritarian specter. Cambridge, MA: Harvard University Press. Basow, S. (1992). Gender: Stereotypes and roles (3rd ed.). Pacific Grove, CA: Brooks/Cole. Beere, C. A., King, D. W., Beere, D. B., & King, L. A. (1984). The Sex-Role Egalitarianism Scale: A measure of attitudes toward equality between the sexes. Sex Roles, 10, 563–576. Diamond, M. (1993). Homosexuality and bisexuality in different populations. Archives of Sexual Behavior, 22, 291–310. Eagly, A. H., & Chaiken, S. (1998). Attitude structure and function. In D. T. Gilbert, S. T. Fiske, & G. Lindzey (Eds.), The handbook of social psychology (4th ed., Vol. 1, pp. 269–322). Boston McGraw-Hill. Haddock, G., & Zanna, M. P. (1998). Authoritarianism, values, and the favorability and structure of antigay attitudes. In G. M. Herek (Ed.), Stigma and sexual orientation: Understanding prejudice against lesbians, gay men, and bisexuals (pp. 82–107). Thousand Oaks, CA: Sage. Herek, G. M. (1986). On heterosexual masculinity. American Behavioral Scientist, 29, 563–577. Herek, G. M. (1988). Heterosexuals’ attitudes towards lesbians and gay men: Correlates and gender differences. Journal of Sex Research, 25, 451–477. Herek, G. M. (1995). Psychological heterosexism in the U.S. In A. R. D’Augelli & C. J. Patterson (Eds.), Lesbian, gay, and bisexual identities over the lifespan (pp. 321–346). New York: Oxford University Press. Herek, G. M., & Capitanio, J. P. (1996). ‘‘Some of my best friends. . .:’’ Intergroup contact, concealable stigma, and heterosexuals’ attitudes toward lesbians and gay men. Personality and Social Psychology Bulletin, 22, 412–424. Herek, G. M., & Glunt, E. K. (1993). Interpersonal contact and heterosexuals’ attitudes towards gay men: Results from a national survey. Journal of Sex Research, 30, 239–244. Higgins, E. T. (1987). Self-discrepancy: A theory relating self and affect. Psychological Review, 94, 319–340. 404 Basow and Johnson Hogg, M. A., & Sunderland, J. (1991). Self-esteem and intergroup discrimination in the minimal group paradigm. British Journal of Social Psychology, 30, 51–62. Kerns, J. G., & Fine, M. A. (1994). The relation between gender and negative attitudes toward gay men and lesbians: Do gender role attitudes mediate the relation? Sex Roles, 31, 297–307. Kite, M. E., & Deaux, K. (1987). Gender belief systems: Homosexuality and the implicit inversion theory. Psychology of Women Quarterly, 11, 83–96. Kite, M. E., & Whitley, B. E., Jr. (1998). Do heterosexual women and men differ in their attitudes toward homosexuality? A conceptual and methodological analysis. In G. M. Herek (Ed.), Stigma and sexual orientation: Understanding prejudice against lesbians, gay men, and bisexuals (pp. 39–61). Thousand Oaks, CA: Sage. Lance, L. M. (1987). The effects of interaction with gay persons on attitudes toward homosexuality. Human Relations, 40, 329–336. Millham, J., San Miguel, C., & Kellogg, R. A. (1976). Factor analytic conceptualization of attitudes toward male and female homosexuals. Journal of Homosexuality, 2, 3–10. Newman, B. S. (1989). The relative importance of gender role attitudes to male and female attitudes towards lesbians. Sex Roles, 21, 451–465. Pelham, B. W., & Swann, W. B., Jr. (1989). From self-conceptions to self-worth: On the sources of global self-esteem. Journal of Personality and Social Psychology, 57, 672–680. Rich, A. (1980). Compulsory heterosexuality and lesbian existence. Signs, 5, 631–660. Spence, J. T., & Helmreich, R. L. (1978). Masculinity and femininity: Their psychological dimensions, correlates, and antecedents. Austin, TX: University of Texas Press. Theodore, P. S., & Basow, S. A. (in press). Heterosexual masculinity and homophobia: A reaction to the self? Journal of Homosexuality. Whitley, B. E., Jr. (1988). Masculinity, femininity, and self-esteem: A multitrait-multimethod analysis. Sex Roles, 18, 419–432. Whitley, B. E., Jr. (1999). Right-wing authoritarianism, social dominance orientation, and prejudice. Journal of Personality and Social Psychology, 77, 126–134. Whitley, B. E., Jr., & Lee, S. E. (2000). The relationship of authoritarianism and related constructs to attitudes toward homosexuality. Journal of Applied Social Psychology, 30, 144–170. Wills, T. A. (1981). Downward comparison principles in social psychology. Psychological Bulletin, 90, 245–271.