Gender Roles and Sexual Pleasure 1 Running Head: GENDER ROLES AND SEXUAL PLEASURE Scripting Sexual Passivity: A Gender-Role Perspective

In two studies, we demonstrate that attitudes towards traditional sexual roles are linked with increased sexual passivity for women but decreased passivity for men. For both genders, sexual passivity is associated with poor sexual functioning and satisfaction. Study 1 showed that endorsement of attitudes toward traditional sexual roles of male dominance and female passivity relates to greater sexual passivity among college-aged heterosexual women but less passivity for college-aged heterosexual men. For both young men and women, greater sexual passivity predicts less overall sexual satisfaction. In Study 2, we surveyed sexually experienced, heterosexual individuals via the Internet. Women, but not men, who endorsed gendered sexual roles were more likely to engage in passive behaviors. For both genders, passive behavior was linked to diminished sexual arousal, ability to reach orgasm, and sexual satisfaction. These relationships were mediated by autonomy and persisted when controlling for multiple potential confounds. Results are discussed in terms of gender role socialization and psychological factors affecting sexual satisfaction. Gender Roles and Sexual Pleasure 3 Scripting sexual passivity: A gender-role perspective Gender roles guide and constrain men and women’s behavior across a wide range of settings, including intimate relationships. The influence of gender roles in professional and academic settings is well studied; less attention has been directed towards their influence within sexual relationships. This lack of attention is noteworthy, because little research has been conducted on sociocultural factors affecting sexual function (Baumeister, 2001) and because gender-based norms and roles likely exert a prominent influence on intimate relationships. Because intimate contexts make gender roles salient, men and women may feel compelled to conform to gender roles during sexual encounters, (Coward, 1985, Rohlinger, 2002; Sanchez, Crocker, & Boike, 2005). Given the importance of sexual satisfaction for overall relationship satisfaction and longevity (see Sprecher & Cate, 2004, for a review), the effects of gender-based sexual roles on sexual enjoyment have potentially broad implications for romantic relationships. In this paper, we contend that gender role conformity indirectly depresses sexual satisfaction for women but not men, because traditional gender-based sexual roles dictate sexual passivity for women but sexual agency for men (e.g., Sanchez, Kiefer, & Ybarra, 2006; Tevlin & Leiblum, 1983). We further propose that although gender roles should differentially affect men and women, men and women should experience less sexual satisfaction to the extent that they are sexually passive. Specifically, sexual passivity should increase sexual problems and lower sexual satisfaction by undermining autonomy during sexual activities (Kiefer, Sanchez, Ybarra, & Kalinka, in press; Sanchez, et al., 2006). Past research has provided preliminary support for this Gender Roles and Sexual Pleasure 4 contention: among college-aged women who reported engaging in passive sexual behavior reported less sexual arousability, a relationship mediated by reduced autonomy (Sanchez et al., 2006). The present research extends these past studies by addressing two important theoretical concerns. First, there are numerous explanations for why women would engage in passive sexual behavior as well as for why sexual passivity would be linked to reduced sexual satisfaction. To rule out alternative explanations for these relationships, we examined the extent to which genderbased sexual roles, relative to other potential contributors (e.g., conservatism in relationships, libido, and partners’ agency), contribute to women’s engagement in passive sexual behavior. We also assess multiple alternative explanations for why sexual passivity might be associated with diminished enjoyment of sexual activities. Second, past research (e.g., Sanchez et al., 2006, Kiefer et al., in press) has focused exclusively on how sexual passivity affects sexual arousal and ability to orgasm among women. This research does not address whether the hypothesized negative relationship between sexual passivity and sexual satisfaction is unique to women. Baumeister (2000; 2004), for example, has argued that women’s sexual behavior and experiences are more influenced by social and contextual forces than are men’s, implying that, sexual passivity should be linked to women’s, but not men’s, sexual satisfaction. However, if both men and women require sexual agency and autonomy for sexual fulfillment, then men’s sexual satisfaction should also negatively correlate with sexual passivity. Thus, we examine whether sexual passivity predicts diminished sexual satisfaction among men as well as women. Gender-Roles and Sexual Satisfaction Gender Roles and Sexual Pleasure 5 For decades, gender-typed roles and behaviors have been theorized to adversely influence sexual functioning and satisfaction (e.g., Tevlin & Leiblum, 1983; MacKinnon, 1982; 1989). During the late seventies, researchers became interested in how psychological androgyny (Bem, 1974), or the presence of both masculine and feminine gender-role personality traits, influences sexual functioning and satisfaction. Studies on psychological androgyny found numerous relationships between androgyny, masculinity and sexual outcomes. For example, compared to couples that were seeking treatment for sexual dysfunction, untreated couples were more likely to have one or more androgynous members (Safir, Peres, Lichtenstein, Hoch, & Shepher, 1982). Androgynous individuals tended to report higher levels of sexual self-esteem (Kimlicka, Cross, & Tarnai, 1983), more liberal sexual attitudes (Johnson, 1989), and greater overall sexual satisfaction (Kimlicka et al., 1983) than sex-typed or undifferentiated individuals. In some of these studies, masculine gender-role self perceptions seem to be the stronger correlate of sexual self-esteem and sexual satisfaction than androgyny per se (e.g., Kimlicka et al., 1983; Willemsen, 1987). Although this research provides evidence that stereotypical masculine and feminine personality traits are related to sexual satisfaction, it fails to explain why androgyny positively correlates with sexual satisfaction. Moreover, despite androgyny—and gender-role selfperceptions more generally—being conceptualized as global personality dimensions, the expression of gender-based traits appears to vary considerably across different social contexts. Women’s gender-role self-perceptions vary across work, social, and sexual contexts in response to context-specific stereotypic cues and role expectations (Rosenzweig & Dailey, 1991). Men’s and women’s reports of their global gender-based traits are frequently at odds with those they report expressing in heterosexual sexual situations (Lawrance, Taylor, & Byer, 1996). Thus, Gender Roles and Sexual Pleasure 6 perceptions of gender-role based expectancies that are specific to sexual contexts may be more important predictors of sexual behavior and satisfaction than global personality traits. In sum, research on androgyny and sexual satisfaction is consistent with our contention that gender-based roles proscribing sexual agency are detrimental to women’s sexual satisfaction; however, past research has not directly tested this hypothesis. To extend this research, we examine whether gender-typed roles relate to sexual satisfaction, because they dictate gender-specific levels of agency in sexual contexts. Gender Roles and Sexual Agency Men and women receive different prescription regarding agentic behavior, with men’s expression of agency favored and women’s expression of agency restricted or met with disapproval in settings ranging from the workplace to the bedroom (Fiske, 1993; Rudman,1998). Heterosexual men are socialized to be sexually empowered, directive, dominant, and assertive, whereas heterosexual women are socialized to be sexually disempowered, to be responsive rather than active (Blumstein & Schwarz, 1983; Schwartz & Rutter, 2000; Sprecher & McKinney, 1993). Men are expected to be more sexually experienced than their partners and to initiate and direct sexual activities. In contrast, women learn to avoid the expression of sexual agency and to adopt a submissive, passive sexual role (Gagnon & Smith, 1973; Schwartz & Rutter, 2000; Tevlin & Leiblum, 1983). A number of societal sources, most notably the mass media, inculcate men and women into gender-appropriate sexual roles. Magazines, television shows, and movies frequently depict female sexual submission to male sexual agency and dominance (Baker, 2005; Dworkin, 1987; Jeffreys, 1990; Jhally, 1995; Kilbourne, 2000a, 2000b; Kitzinger, 1984; Lowry, Love, & Kirby, 1981; MacKinnon, 1989). Magazines targeted to young women promote passivity as a way to Gender Roles and Sexual Pleasure 7 satisfy male partners (Kilbourne 2000a, 2000b; Kim & Ward, 2004); and advertisements in men and women’s magazines frequently portray women as submissive and dependent (Baker, 2005). Female sexual passivity and male sexual agency are prominent themes not only in romance novels (Snitow, 1979), but also in mainstream literature (Millet, 1970; Zilbergeld, 1978). Inculcation of traditional sexual roles may influence men and women’s sexual behavior and enjoyment. Men’s and women’s reports of their sexual behavior suggest a fair degree of conformity to these culturally prescribed roles. For example, heterosexual men are more likely to initiate sex than are their partners (see Impett & Peplau, 2003) and more frequently report pressuring their partners to have unwanted sex; i.e., they seek their partners’ submission to their personal sexual desires (Miller & Benson, 1999). Correspondingly, many women report willingly submitting to unwanted sexual activities. In one recent survey, half of the women interviewed reported this form of sexual compliance (O’Sullivan & Allgeier, 1998). Women’s sexual passivity is evident even at an early age: in qualitative studies, adolescent girls frequently describe their initial sexual experiences as “just happening to them” (Martin, 1996; Tolman, 2002). In addition to the idea that gender-based roles promote sexual passivity among women, numerous other explanations for women’s lack of sexual agency have been offered. Most notably, Baumeister, Cantense, & Vohs (2001) have argued that men’s greater tendency to initiate and direct sexual activities stem from men having innately higher levels of sexual desire than women, a difference these authors believe is largely driven by sex differences in testosterone levels. Similarly, Buss (1989) has proposed that within heterosexual couples, there are large discrepancies in sexual desire, with men typically exceeding their female partners. Buss (1989) argues that these discrepancies in desire lead women to become upset by their male Gender Roles and Sexual Pleasure 8 partners’ strategies to have more sex and men to become upset by their female partners’ strategies to have less sex. These gender-specific strategies could lead men to assume the role of the sexual initiator, a relatively active role, and women to assume the role of the sexual recipient, a relatively passive role. Regardless of one’s gender, having a sexually assertive partner might lead to the adoption of a recipient, passive sexual role. To test these alternative explanations for why women might be more sexually passive than men, we assessed the relationships between men and women’s self-reported libido, their perceptions of their discrepancies of desire between themselves and their partners, and their perceptions of their partners’ sexual agency. Sexual conservatism is another factor that might account for the hypothesized relationships between endorsement of traditional sexual roles and passive sexual behavior. As noted above, previous studies on androgyny and sexual satisfaction have failed to distinguish between the effects of gender-typed behavior and the effects of conservative sexual values. To determine whether conservative sexual values (as opposed to specific gender-based sexual roles) predict women’s engagement in passive sexual behavior, we assessed participants’ beliefs about gender differences in sexual desire and traditional values concerning romantic relationships. Because traditional values and beliefs about gender differences in sexual desire typically imply that men have greater sexual desire than do women, these beliefs might be expected to have opposite effects on men and women’s sexual agency. We therefore also tested for moderation of these effects by gender. Sexual Passivity and Sexual Outcomes Because sexual passivity may undermine autonomy, it has potentially broad implications for intimate sexual relationships (Sanchez et al., 2006). Sexual autonomy refers to feeling that one’s actions in sexual contexts are freely chosen, authentic expressions of the self (Deci & Gender Roles and Sexual Pleasure 9 Ryan, 1995; Sanchez et al., 2005). Sexual autonomy is believed to be a necessary precondition of men and women’s sexual pleasure (Weinberg, Swensson, & Hammersmith, 1983), and empirical research supports this assessment (Haavio-Mannila & Kontula, 1997; Sanchez et al., 2005). The ability to communicate one’s desires is an important precursor of sexual satisfaction for men and women (Haavio-Mannila & Kontula, 1997). Thus, we expected sexual passivity would be linked with less sexual satisfaction for both men and women and that this effect would be mediated by reduced sexual autonomy. As with the hypothesized link between gender-based sexual roles and sexual passivity, there are numerous alternative explanations for why sexual passivity might correlate with reduced sexual satisfaction. For example, individuals who have low levels of sexual desire may be less sexually assertive. In addition, as suggested by Buss (1989), women whose partners exceed them in the desire for sex might frequently submit to undesired sexual activities and therefore experience less enjoyment during those activities. Hence we controlled for disparities in desire, gender beliefs about desire (i.e., endorsement of the idea that women have less sexual desire than men), traditional values, perceptions of partners’ sexual agency, and libido when assessing the relationship between sexual passivity and sexual satisfaction. In summary, we contend that adherence to gender-based sexual roles reduces women’s sexual functioning and satisfaction by promoting sexual passivity. Although women might engage in passive sexual behavior for a variety of reasons, such as having a dominant partner or from a lack of sexual desire, we propose that conformity to gender roles is a primary reason for this behavior. Moreover, because sexual autonomy is thought to be a critical precursor of both men and women’s sexual function and satisfaction, we expected that passive behavior would predict reduced sexual functioning and satisfaction for both men and women. Gender Roles and Sexual Pleasure 10 Study 1 We hypothesized that the more women endorse attitudes toward traditional sexual roles of male agency and female passivity, the more they would engage in passive sexual behavior. The reverse was predicted for men. Moreover, we expected passive behavior to predict lower sexual satisfaction for men and women. To rule out alternative explanations for why women adopt a passive sexual role, we controlled for additional variables believed to contribute to sex differences in sexual behavior (Impett & Peplau, 2003; Baumeister, Cantanese, & Vohs, 2001), including libido, perceptions of partners’ sexual agency, perceived disparities in desire between sexual partners, gender beliefs about desire, and traditional values concerning romantic relationships. Finally, because there is some evidence to suggest that men and women are motivated to respond to certain questions about their sexual behavior in socially desirable ways (e.g., Alexander & Fisher, 2003), we also controlled for social desirability and for possible gender by social desirability interactions. Method Participants. Three hundred ten participants (174 men, 136 women; 170 Caucasians, 15 Blacks, 86 Asians, 19 Latino/Hispanic, 2 Native Americans, and 17 Multiracials) completed the survey for course credit in their introductory psychology class at Rutgers University. The mean age was 18.79 years (SD = 1.26). One hundred thirty nine participants were currently involved in a romantic relationship, 159 were not, and 13 declined to answer this question. There were 285 heterosexuals, 6 homosexuals, 5 bisexuals, and 10 who declined to answer, and 5 who were unsure of their orientation. Two hundred four participants had experienced sexual intercourse, 79 had not, and 28 declined to answer. Because gender roles assume a heterosexual context, only Gender Roles and Sexual Pleasure 11 heterosexual participants were included in the analyses, resulting in a final sample of 285 (160 men, 124 women). Materials. Traditional Sexual Attitudes. Participants rated the following statements on a scale from 1 (strongly disagree) to 7 (strongly agree): "I believe that men should take on the more agentic or active role during sexual activity"; "I believe that women should take on the more passive role during sexual activity"; I believe that men should take on the more dominant role during sexual activities"; “I believe that women should take on the more passive role during sexual activity”; “I believe that men should prefer to take on the more agentic role during sexual activity”; and "I believe that women should prefer to take on the more passive role during sexual activity” (α = .89). Passive sexual behavior. Passive sexual behavior was assessed using the measure developed by Sanchez and colleagues (2006). Participants rated the following statements on a scale from 1 (strongly disagree) to 7 (strongly agree): "I tend to take on a submissive role during sexual activity"; “I prefer to take on the submissive role during sexual activities”; "I tend to take on the passive role during sexual activities"; I prefer to take on the passive role during sexual activities"; “I tend to take on a submissive role during sexual activities”; “I tend to take on the more dominant role during sexual activity” (reverse-coded); and "I prefer to take on the more agentic or active role during sexual activity"; (reverse-coded; α = .89). Libido. Participants rated the following statements on a scale from 1 (strongly disagree) to 5 (strongly agree): “I have a very strong sex drive”; “I am always in the mood for sex”; “I think about sex almost every day”; “I am not a very sexual person” (reverse-coded); and “If it were up to me, I would have sex at least every day” (α = .86). Gender Roles and Sexual Pleasure 12 Traditional relationship values. Participants indicated their agreement with the following statements on a scale from 1(strongly disagree) to 7 (strongly agree) with 9 items from Burt’s (1980) Sex-Role Stereotyping subscale that assesses conservative relationship beliefs (e.g., “A girl should be a virgin when she marries”; α = .65). Gender beliefs about desire. Participants indicated their agreement on a scale from 1(strongly disagree) to 7 (strongly agree) with the following statements: “Men are more sexual than women”; “Men have consistently stronger sexual appetites than women”; “Women want to have sex more often than men” (reverse-coded); “Women think about sex more often than men” (reverse-coded); and “Men are always in the mood for sex” (α = .70). Disparities in Desire. Participants indicated their agreement on a scale from 1 (strongly disagree) to 7 (strongly agree) with the following statements: “I often feel like I am more sexual than my sexual partners”; “I often feel like I am convincing my partners to have sex or engage in sexual activities”; “I often feel that my partner(s) does/do not really want to have sex or engage in sexual activities with me”; and “I often feel that I need to get my partner(s) in the mood so that we can have sex” (α = .75). Partners’ sexual agency. Participants answered the following questions on a scale from 1 (never) to 7 (always): “Does your current or most recent sexual partner indicate their preferences during sexual activities?” “Does your current or most recent sexual partner tell you what she (he) wants during sexual activities?” (α = .78). Social desirability. Socially desirable responding was assessed using the CrowneMarlowe (1960) social desirability scale. Sexual satisfaction. Participants answered the question, “How satisfied are you with your sex life?” on a scale anchored at 1 (not at all satisfied) and 7 (completely satisfied). Gender Roles and Sexual Pleasure 13 Sexual experience. Participants were asked one question regarding whether or not they had previously engaged in sexual intercourse. Results See Table 1 for zero-order correlations separately for women and men and Table 2 for sex differences for the measured variables. Passive sexual behavior. The data were analyzed using hierarchical linear regressions. At Step 1, all main effects including the control variables were entered: gender, social desirability, partner agency, libido, perceived partner disparities in desire, gender beliefs about desire, traditional relationship values, sexual experience, and attitudes towards traditional sexual roles. At Step 2, the two-way interaction terms of partner agency, libido, perceived partner disparities in desire, traditional relationship values, and traditional sexual scripts with gender were entered. Because several main effects and interactions were not significant, they were trimmed from the regression analyses. Table 3 shows all of the variables retained for the hierarchical regression. These analyses revealed a main effect of social desirability, with greater social desirability predicting less reported passive sexual behavior, β = -.114, p < .03. Greater libido predicted less passive sexual behavior, β = -.324, p < .001. As hypothesized, women reported more passive sexual behavior than men, β = .390, p < .001. The main effect of traditional gender roles was not significant, β = -.017, p = .72, ns. However, the predicted interaction between gender and attitudes toward traditional sexual roles was significant, β = .228, p < .001. Simple slopes were conducted to interpret the interaction, controlling for all significant control variables. The more women endorsed attitudes toward traditional sexual roles, the more likely they were to engage in passive sexual behavior, β = .220, p = .001, whereas the more men endorsed these roles, the less likely they were to engage in passive behavior, β = -.196, p = .01. Gender Roles and Sexual Pleasure 14 Unexpectedly, there was a significant interaction of gender and perceived partner disparities in desire, β = -.132, p < .05. Simple slopes testing revealed that the more men felt that they had more sexual desire than their partners, the more they engaged in passive sexual behavior, β = .173, p < .05, whereas for women, there was no relationship between perceived disparities in desire and passive behavior, β = .088, p = .12, ns. Sexual satisfaction. At Step 1, all main effects including the control variables were entered: gender, social desirability, partner agency, libido, perceived partner disparities in desire, gender beliefs about desire, traditional relationship values, sexual experience, and passive sexual behavior. At Step 2, the two-way interaction term of gender and passive sexual behavior was entered. As in the previous analysis, the non-significant effects were trimmed (see Table 4). Participants who reported having had sexual intercourse were more satisfied than were those who reported not having had sexual intercourse, β = -.246, p < .001. Participants who perceived less disparities in desire between themselves and their partners also reported greater sexual satisfaction, β = -.253, p < .001. As hypothesized, passive sexual behavior predicted less overall sexual satisfaction, β = -.160, p < .05. The interaction of gender and passive behavior was not significant, β = -.052, p = .43, ns, suggesting that passive behavior was linked to lower sexual satisfaction for both men and women. Discussion Results were consistent with the hypothesis that endorsement of gender-based sexual roles enhances women’s sexual passivity but reduces men’s sexual passivity. To our knowledge, this is the first study to demonstrate that endorsement of gender-based sexual roles is linked with men’s sexual agency. Contrary to predictions, traditional sexual attitudes were not directly related to sexual satisfaction. To determine whether we lacked sufficient statistical power to Gender Roles and Sexual Pleasure 15 detect the effects of traditional sexual attitudes on sexual satisfaction, Study 2 employed a larger sample. This study also tested several competing explanations for why women tend to be more sexually passivity than men. As predicted, gender role endorsement correlated with more passive behavior for women, but less passive behavior for men. Providing evidence of the robust nature of the link between gender role endorsement and passive behavior, this relationship persisted when controlling for potential alternative explanations for gender differences in sexual agency. In addition to the effects of gender role endorsement, libido predicted less sexual passivity for both men and women. People with strong libidos may engage in sexually dominant behavior to satisfy their greater desire for sex. We also found a significant interaction effect of gender and perceptions of disparities in sexual desire on sexual passivity. Men who perceived their partners to be less sexually desirous indicated greater sexual passivity. Men may feel uncomfortable initiating sexual activities with partners they perceive to be unwilling. This study was also the first to show that passive behavior is linked to diminished sexual satisfaction for men as well as women. These results suggest that both men and women may need to experience sexual agency for sexual fulfillment. This relationship between sexual passivity and sexual satisfaction persisted when controlling for multiple third variables that could explain the relationship between sexual passivity and sexual satisfaction and that were linked to sexual satisfaction. For example, disparities in desire predicted sexual satisfaction; both men and women who felt their partners had less sexual desire than themselves reported relatively greater sexual dissatisfaction. Gender Roles and Sexual Pleasure 16 We propose that men and women who are sexually passive may be less sexually satisfied because passivity impairs sexual arousability and the ability to reach orgasm by undermining sexual autonomy. In past research, women’s passive sexual behavior correlated with less reported sexual arousability, an effect mediated by reduced sexual autonomy (Sanchez et al., 2006). Because sexual arousal is both a critical precursor of the ability to orgasm (Geer & Janssen, 2000; Masters & Johnson, 1966) and a key component of sexual enjoyment (Laumann, Paik, & Rosen, 1999), we hypothesized that passive sexual behavior would predict less sexual arousability, less ability to reach orgasm, and less sexual satisfaction for men and women in a community-based sample. Sexual autonomy was expected to mediate these relationships. Furthermore, because the control variables tested in Study 1 failed to explain the relationships between gender role endorsement and passive sexual behavior and between passive behavior and sexual satisfaction, we examined one additional alternative explanation in Study 2. Men and women who find their partners unattractive may desire sex less and therefore be less sexually assertive. We therefore tested whether perceived attractiveness of sexual partners would explain the hypothesized relationships between sexual passivity and sexual problems. Because Study 2 employed a community-based convenience sample, we also controlled for age, ethnicity, and income. Study 2 Confirmatory structural equation modeling was performed using EQS software on survey data collected via the Internet. The model described below was first tested for heterosexual, sexually experienced men and women and then tested separately for men and women. The following hypotheses constituted our structural model: 1) Endorsement of traditional roles will predict greater engagement in passive sexual behavior for women but less engagement Gender Roles and Sexual Pleasure 17 for men; 2) Passive sexual behavior will predict reduced sexual arousability, ability to achieve orgasm, and sexual satisfaction for men and women; 3) The effects of passive sexual behavior on sexual outcomes will be mediated by reduced sexual autonomy. Method Participants and Procedure The Internal Review Board at the University of Michigan approved the survey instrument and recruitment procedures. Participants were recruited over the Internet via postings on message boards for 150 different Yahoo and MSN groups (please see Appendix A for the full text of the recruitment message). Participants who were interested in participating accessed the survey through an online survey website maintained by the university. The survey was open to all visitors and used unique computer identifiers, cookies, to discourage individuals from completing the survey multiple times. The cookie was read and set at the opening page of the survey (i.e., the informed consent page). Secure Socket Layer (SSL) was used for data encryption. Participants received the questionnaire in 2 random orders, in which sexual behavior questions either preceded or followed sexual functioning and satisfaction items. Because order did not affect the results, all results are reported collapsing across order. Questions were presented on 11 different pages with the number of questions on each page ranging from 3 to 20 items. Participants were free to change any responses prior to submitting their survey. Following survey submission, participants were led to a debriefing page, which thanked them, briefly described the purpose of the study, and provided contact information. Five hundred fifty-three participants (189 males, 364 females, and 1 participant who failed to specify their gender) completed our survey on the Internet during a 9-month period (December Gender Roles and Sexual Pleasure 18 2004-August 2005). Analyses were conducted on sexually experienced heterosexual participants only. After excluding participants who had not experienced sexual intercourse, indicated a sexual orientation other than heterosexual, were under the age of 18 or failed to answer these questions, a total of 398 participants were retained for data analyses. Participants (314 White/Caucasian Americans, 9 Asian/Asian Americans, 31 Black/African Americans, 17 Hispanic/Latino Americans, 6 Native Americans, 18 Multiracial Americans, and 3 missing) ranged in age from 18 to 71 years (M = 28.97, SD = 11.51). Participants’ mean reported personal income was between $20,000-25,000. Fifty-one percent were unmarried, 28% married, 11% divorced, 8% engaged, .5 % widowed, and 1.5% failed to provide their marital status. Materials. Traditional Sexual Attitudes. The measure of attitudes toward traditional sexual roles was identical to that of Study 1. This measure was reliable for men (α = .86), women (α = .86), and the overall sample (α = .86). Passive sexual behavior. The measure of passive sexual behavior was identical to that of Study 1. This measure was reliable for men (α = .78), women (α = .77), and the overall sample (α = .81). Sexual autonomy. To assess autonomy during sexual activities, participants rated 2 items based on the autonomy scale developed by LaGuardia, Ryan, Couchman, and Deci (2000) and adapted to the sexual context (Sanchez et al., 2005; 2006). The following statements were rated on a scale anchored at 1 (not at all true) and 7 (very true): “When I am having sex or engaging in sexual activities with someone, I feel free to be who I am”; and “When I am having sex or engaging in sexual activities with someone, I have a say in what happens, and I can voice my Gender Roles and Sexual Pleasure 19 opinion.” The scale was reliable for men (α = .73), women (α = .71), and the overall sample (α = .72). The two items served as our indicators. Sexual arousability. Fourteen items from the Sexual Arousability Index (SAI) developed by Andersen, Broffitt, Karlsson, and Turnquist (1989) were used to assess sexual arousability. Survey items described specific sexual situations that were rated on a 7-point scale anchored at 1 (adverse effect) and 7 (always causes sexual arousal). The measure contains five subscales assessing arousability from seductive activities, body caressing, oral-genital and genital stimulation, intercourse, and erotica/masturbation. To assess overall arousability, responses on these items were averaged without the erotica subscale. This scale was reliable for men (α = .89), women (α = .88), and the overall sample (α = .88). Ability to reach orgasm. To assess ability to orgasm with one’s partner, participants rated the following two statements using a 5-point scale anchored at 1 (never/almost never) and 5 (always/almost always): “How often do you reach orgasm during sexual activities with your partner(s)?”; “How often do you have difficulty reaching orgasm with your partner?” (reversecoded.” Reliabilities were low for men (α = .55), satisfactory for women (α = .75) and the overall sample (α = .74). Although the reliability of this scale for men is somewhat low, structural equation modeling allows for latent variables with low reliabilities because it corrects for measurement error (Jaccard & Wan, 1996; Osborne, 2003). In addition, low reliabilities are expected with measures with few items, because coefficient alphas are sensitive to the number of items (Clark & Watson, 1995; Streiner, 2003). Higher scores on this measure indicated greater ability to orgasm. These two items served as our indicators. Sexual satisfaction. Participants were asked two questions comprising the indicators of sexual satisfaction: “How often do you feel satisfied after sex?”; “How often do you find sex Gender Roles and Sexual Pleasure 20 pleasurable?” Participants indicated their responses on a scale from 1 (never) to 5 (always). The measure was reliable for men (α = .74), women (α = .82), and the overall sample (α = .79). The two items served as our indicators. Perceived attractiveness of partner. In three separate items, participants were asked to rate the attractiveness of their partners’ face, body, and overall physical appearance (e.g., “I find my partner’s face/body/overall physical appearance very attractive”) on a 5-point scale anchored at 1 (disagree) and 5 (agree). The measure was reliable for men (α = .89), women (α = .85) and the overall sample (α = .86). Results Table 5 presents zero-order correlations among the indicators of each hypothesized underlying factor and the dependent variables by gender and Table 6 shows the mean gender differences. Factors loadings for the item parcels are shown in Figure 1. On average, men reported greater sexual arousability, ability to orgasm, and sexual satisfaction, and less passive behavior than did women. We tested the hypothesized model with confirmatory latent-variable structural analyses using EQS computer software, which allowed for testing of paths between the predictor variables and multiple dependent variables simultaneously (Klem, 2000). We tested the structural model on heterosexual, sexually experienced men and women (N = 475). We also performed multiple group comparisons between sexually experienced heterosexual men (N = 166) and women (N= 308). In accordance with standard structural equation modeling with EQS software (Raykov, Torner, & Nesselroade, 1991), we report the following goodness-of-fit indices: χ2/df, non-normed fit (NNFI), and comparative fit (CFI). Acceptable fit indices exceed .90. We also report the root mean square error of approximation Gender Roles and Sexual Pleasure 21 (RMSEA) as well as the 90% confidence interval of the RMSEA. RMSEA misfit indices should be at or below .06 (Hu & Bentler, 1999). Although χ2 is not considered a good index for tests of fit because of its sensitivity to sample size, we report χ2 to make comparisons between nested models (Klem, 2000). Preliminary Analyses. To assess whether demographic variables or perceived attractiveness of one’s sexual partner explained part of the variance in passive behavior beyond the effect of endorsing attitudes toward traditional sexual roles, we regressed passive behavior on income, partner’s income, age, ethnicity, marital status, perceived attractiveness of partner, gender, attitudes toward traditional sexual roles, and the two-way interaction of gender and attitudes toward traditional sexual roles. No significant effects were found other than the expected main effect for gender (β = .403, p < .001), attitudes toward traditional sexual roles (β = .215, p < .001) and the two-way interaction (β = .213, p < .001). Separate, trimmed analyses by gender revealed that endorsement of traditional sexual roles predicted greater passive behavior for women (β = .397, p < .001) and less passive behavior for men (β = -.159, p < .05). Measurement model. To assess the fit of the observed variables as indicators of the latent variable, we first tested the measurement model. All paths between latent variables were excluded from this assessment. The measurement model fit the data well (see Table 7). Structural models and mediation by sexual autonomy. The direct effects and full models were first tested on the full sexually experienced heterosexual sample, collapsing across gender. These models tested for mediation of the hypothesized relationships between sexual passivity and sexual outcomes (sexual arousal, ability to reach orgasm, and sexual satisfaction) by autonomy. Tests of mediation by structural equation modeling parallel those using multiple linear regressions (Frazier, 2004). According to Kenny (2006), there are four steps required to Gender Roles and Sexual Pleasure 22 demonstrate mediation. First, the predictor variable (sexual passivity) must predict the outcome variables of interest (sexual arousal, ability to reach orgasm, and sexual satisfaction). Second, the predictor should correlate with the proposed mediator (sexual autonomy). Third, the mediator should correlate with the outcomes, controlling for the predictor variable. Fourth, the relationship between the predictor variable and the outcomes should no longer be significant when the mediator is included as a predictor. The significance of the mediation is then determined using Sobel’s Test (1982). Our tests of the first two steps are provided by the direct effects model shown in Figure 1. In this model, sexual passivity was significantly related to all three sexual outcome variables. Moreover, sexual passivity correlated significantly with the proposed mediator, sexual autonomy. The tests of the third and fourth steps are provided by the indirect effects model shown in Figure 2. As illustrated in Figure 2, sexual autonomy predicted sexual arousal and sexual satisfaction, while controlling for sexual passivity. However, sexual autonomy did not significantly predict the ability to reach orgasm. Furthermore, the relationships of sexual passivity to sexual arousal and of sexual passivity to sexual satisfaction were no longer statistically significant when sexual autonomy was included as a predictor of these outcomes. Sobel’s test (1982) was used to test for statistically significant mediation. Mediation of the path between sexual passivity and sexual arousal by autonomy was significant, Z(402) = -3.50, as was mediation of the path between sexual passivity and sexual satisfaction by autonomy, Z(402) = 3.67. Thus, sexual autonomy mediated the relationships between sexual passivity and sexual arousal and satisfaction. However, autonomy did not mediate the relationship between sexual passivity and the ability to reach orgasm. Overall, the full model provided a good fit to the data and was largely consistent with our hypothesis (see Figures 1 and 2 and Table 7). Gender Roles and Sexual Pleasure 23 Gender difference model To test the comparative fit of the model for both men and women, we tested the fit of the covariance matrices for both heterosexual, sexually experienced men and women constraining all paths, factor loadings, and covariances to be equal (Bentler, 1989; Byrne, 1994). To examine whether the hypothesized model fit the data well for both men and women, we examined modification indices to determine whether equality constraints should be released to improve the fit of the model. The full model analysis provided a decent fit to the data (see Table 7). However, serial examination of the modification indices suggested that two constraints be released. First, replicating Study 1, women’s attitudes toward traditional sexual roles predicted more passive sexual behavior (β = .687), whereas men’s attitudes predicted less passive sexual behavior (β = .176). Second, although autonomy predicted women’s sexual satisfaction (β = .299), it was a significantly stronger predictor of men’s sexual satisfaction (β = .538). The final model was compared to the unconstrained model; the difference in χ2 was non-significant, χ2 (9) = 8.62 (see Table 7), indicating that no other constraints should be released. Please see Figure 3 and 4 for results of the nested gender group comparisons. Discussion Using an Internet-based sample, Study 2 replicated and extended the results of Study 1. Gender role endorsement again predicted greater sexual agency for men, but less sexual agency for women. This was true even controlling for partners’ attractiveness, and participants’ personal income and age. Thus, replicating Study 1 with a broader sample, gender role conformity appears to be linked to men and women’s sexual agency. As in Study 1, sexual passivity, but not endorsement traditional sexual roles, was directly related to sexual satisfaction or arousal and Gender Roles and Sexual Pleasure 24 orgasm difficulties. Thus, the relationship between endorsement of traditional sexual roles and sexual satisfaction appears to be indirect. Although gender roles appear to script sexual behavior quite differently for men and women , sexual passivity was associated with reduced sexual function and satisfaction for both men and women. This finding suggests that sexual agency and autonomy are critical components of both men’s and women’s sexual satisfaction. General Discussion The results of two studies suggest that adherence to gender roles may promotesexual passivity among women but reduces sexual passivity among men. The more women endorse attitudes toward traditional sexual roles, the more they engage in passive sexual behavior, whereas the reverse was true for men. For both men and women, however, passive sexual behavior was linked to diminished sexual satisfaction in Study 1 and to less sexual arousal, ability to reach orgasm, and satisfaction in Study 2. Taken together, these findings suggest that women engage in sexual passivity largely because sexual scripts dictate such behavior and that sexual passivity adversely affects subjective sexual experiences for both men and women. Gendered Sexual Roles For centuries, gender norms and roles have dictated submission and passivity for women and dominance and agency for men (Beauvoir, 1954, Foucault, 1985; Irigaray, 1996). Social constructions of sexuality persist in the modern era. Female passivity and male agency are depicted by movies, magazines, television sitcoms, soap operas, and in mainstream literature (Baker, 2005; Kilbourne, 2000a; 2000b; Kim & Ward, 2004; Lowry et al., 1981; Millet, 1970; Snitow, 1979; Tevlin & Leiblum, 1983); and these norms are reflected in women’s tendency to nonconsciously associate sex with submission (Kiefer et al., in press; Sanchez et al., 2006). Thus, Gender Roles and Sexual Pleasure 25 these beliefs regarding the sexual act and concomitant gendered sexual norms persist at a societal level. As with all societal phenomena, however, individuals vary in their exposure to, and in the ways they grapple with, these norms. As shown in Studies 1 and 2, the extent to which men and women endorse attitudes toward traditional sexual roles influences the extent to which their personal sexual behavior conforms to these roles. Gender Roles and Gender Differences in Sexual Problems Diminished sexual satisfaction was linked with engaging in passive behavior for both men and women. These findings imply that both genders may need to engage in active roles during sexual activities to experience sexual autonomy, i.e., to feel that they can authentically express their desires. These results are consistent with theories suggesting that perceived control (e.g., Hurlbert, Apt & Rabehl, 1993), sexual self-efficacy (e.g., Zamboni, Crawford, Williams, 2000), and autonomy (e.g., Sanchez et al., 2005) foster satisfying sexual experiences as well as safer sexual practices. These findings also suggest that the personality characteristics of androgyny and masculinity correlate with greater sexual satisfaction because these personality traits promoted sexual agency and autonomy. Despite the similarity in men and women’s need for sexual autonomy, these studies imply that traditional sexual roles may indirectly lower women’s but not men’s sexual satisfaction. In Study 1 traditional sexual attitudes did not directly predict sexual satisfaction. In Study 2, traditional sexual attitudes predict women’s sexual satisfaction but not sexual arousal or the ability to reach orgasm. However, women’s gender role adherence predicts sexual passivity, which is linked with greater sexual problems and less satisfaction. In contrast, because gender role adherence promotes sexual agency for men, it may indirectly benefit their sexual Gender Roles and Sexual Pleasure 26 satisfaction. This difference might explain why several studies have found that women report lower sexual interest, arousal, and satisfaction than do men (e.g. Baumeister et al. 2001; Laumann et al., 1999; Murnen & Stockton, 1997). On the other hand, some studies have failed to find gender differences in sexual arousal (e.g., Griffitt, 1987) or sexual satisfaction (e.g., Oliver & Hyde, 1993). The extent to which gender differences exist in measures of sexual arousal may depend heavily on the age of participants, the way in which sexual outcomes are measured, and the experimental setting (Murnen & Stockton, 1997). Nevertheless, if by some indicators women are less sexually satisfied than men, engagement in passive behavior may partially account for that difference. Limitations and Future Directions There are several aspects of this research that limit our conclusions. The most notable limitation is that we did not test the causal relationships between attitudes towards traditional gender-based sexual roles, passive behavior, and sexual outcomes. Thus, alternative explanations for our findings cannot be ruled out. For example, the present findings could be interpreted as an indication that women and men who lack sexual arousal and satisfaction tend to be sexually passive, for why would individuals actively seek sex that is undesired? For several reasons, however, we believe this alternative explanation to be less plausible than the proposal that passive sexual behavior does in fact lower arousability. First, our results strongly suggest that women are motivated to adopt a passive role because they believe that role to be gender appropriate. Second, our studies found that passive behavior predicted reduced sexual satisfaction when controlling for numerous possible confounds, including libido, partner attractiveness, and partner agency. Finally, research on the disconnection between women’s physiological and subjective experiences of sexual arousal suggests that sexual agency plays a Gender Roles and Sexual Pleasure 27 causal role in women’s sexual arousal (Brody, Laan, & Lunsen, 2003). Women’s physiological arousal rarely predicts their subjective experience of sexual arousal (Both, Spiering, & Everaerd, 2004; Heiman, 1977; Steinman, Wincze, Venditti, Barlow, & Mavissakalian, 1981). Even chemically-induced increases in women’s physiological arousability fail to increase their subjective arousability (Harris, 2004). However, compared to other women, women who are more attuned to their physiological arousal reach orgasm more easily (Brody, et al., 2003). Moreover, exposure to images of sexually agentic women reduces the disconnection between women’s physiological and subjective sexual arousal (Both et al., 2004). Taken together, these findings suggest beliefs that women should or can be sexually agentic enhance women’s sexual arousability and ability to reach orgasm. In fact, the ability to control and direct sexual interactions may have a direct, biological influence on sexual pleasure. Using a rodent model, Jenkins and Becker (2003) have shown that female rats experience the greatest release of dopamine in reward centers of the brain when they are allowed to dictate the pace of copulation. Although conducting such research on humans is currently infeasible, future studies should explore the effects of sexual autonomy and agency on sex hormones and on the activation of brain centers known to be involved in reward and pleasure. A second limitation is that, because the research samples in these studies were convenience samples, our findings cannot be said to represent the United States population as a whole. Despite this concern, recent studies suggest that Internet survey research is just as representative as, if not more diverse than, traditional methods using college-based samples (Gosling, Vazire, & Srivastava, 2004; Horswill & Coster, 2001). Notably, there was a wide range of ages and income levels represented across these two studies, and findings were consistent Gender Roles and Sexual Pleasure 28 across the college-based and Internet-based samples. However, our sample was limited to heterosexuals. Future research should assess whether or not these findings generalize to nonheterosexual populations. A final question unaddressed by the present research concerns the root cause of gendered sexual roles of male agency and female passivity. These roles obviously have a longstanding history in Western culture (Beauvoir, 1954; Foucault, 1985). Some (e.g., Baumeister et al., 2001) have argued that gender differences in sexual agency stem from sex differences in hormones. Estrogen, for example, is thought to promote passive sexual behavior, whereas testosterone is thought to promote dominant sexual behavior (Baumeister et al., 2001). This evolutionary perspective is not inconsistent with the suggestion that gender roles promote female sexual passivity. However, our research strongly suggests that gender norms also exert an effect on sexual behavior and, at the very least, exacerbate biologically-driven gender differences in sexual agency. Thus, we believe that the study of the relative contributions of hormones and gender norms in determining sexual agency remains an important avenue for future investigation. Conclusions Women’s adherence to traditional gender-based sexual roles is linked to diminished sexual function and satisfaction. This impairment may stem from the loss of sexual autonomy, which both sexes require for sexual fulfillment (Sanchez et al., 2005). Adherence to gender roles may therefore have broader implications for heterosexual relationships, as sexual satisfaction is believed to be an important component of satisfying romantic relationships (see Sprecher & Cate, 2004, for a review). For example, sexual satisfaction across time predicts likelihood of divorce (Edwards & Booth, 1994), even when controlling for relationship satisfaction (White & Keith, 1990). Thus, the present findings suggest that women’s adherence to traditional genderGender Roles and Sexual Pleasure 29 based sexual roles may pose a significant barrier to the development and maintenance of satisfying sexual relationships. Gender Roles and Sexual Pleasure 30 Appendix 1 At the University of Michigan, we are conducting important research on men and women’s behavior in intimate relationships. If you are a 18 years of age or older please take 15-25 minutes of your time to help us reach better understanding of romantic relationships by filling out a short survey. Men and women of all sexual orientations and ages are encouraged to participate. Please visit our survey @ [link inserted here] *This is a secure website* This work has been reviewed and approved by the University of Michigan Review Board. Our success is dependent on your help. We would be honored if you would participate. Sincerely, Amy Kiefer, Researcher, Department of Psychology, University of Michigan. Gender Roles and Sexual Pleasure 31 ReferencesAlexander, M. G., & Fisher, T. D. (2003). Truth and consequences: Using the boguspipeline to examine sex differences in self-reported sexuality. Journal of SexResearch, 40, 27-36.Andersen, B. L., Broffitt, B. Karlsson, J. A., & Turnquist, D. C. (1989). A psychometricanalysis of the sexual arousability index, Journal of Consulting and ClinicalPsychology, 57, 123-130.Baker, C. N. (2005). Images of women's sexuality in advertisements: A content analysisof blackand white-oriented women and men's magazines. Sex Roles, 52, 13-27.Bandalos, D. (2002). The effects of item-parceling on goodness-of-fit and parameterestimate bias in structural equation modeling. Structural Equation Modeling: AMultidisciplinary Journal, 9, 78-102.Baumeister, R. F. (2000). Gender differences in erotic plasticity: The female sex drive as sociallyflexible and responsive. Psychological Bulletin, 126, 347-374.Baumeister, R. F. (2001). Social psychology and human sexuality: Key readings in socialpsychology. New York: Psychology Press.Baumeister, R. F. (2004). Gender and erotic plasticity: Sociocultural influences on the sex drive.Sexual and Relationship Therapy, 19, 133-139.Baumeister, R. F., Catanese, K. R., & Vohs, K. D. (2001). Is there a gender difference in strengthof sex drive? Theoretical views, conceptual distinctions, and a review of relevant evidence.Personality and Social Psychology Review, 5, 242-273.Beauvoir, S. (1954). The Second Sex. (H. M. Parshley, Trans.). New York: Knopf. Gender Roles and Sexual Pleasure 32 Bem, S. L. (1974). The measurement of psychological androgyny. Journal of Consulting andClinical Psychology, 42, 155-62.Bentler, P. M. (1989). EQS: Structural Equations Program Manual. LA: BMDP StatisticalSoftware.Blumstein, P., & Schwarz, P. (1983). American couples: Money, work, and sex. NewYork: William Morrow.Both, S., Spiering, M., Everaerd, W., & Laan, E. (2004). Sexual behavior andresponsiveness to sexual stimuli following laboratory-induced sexual arousal. Journal ofSex Research, 41, 242-258.Briggs, S. R. & Cheek, J. M. (1986). The role of factor analysis in the development andevaluation of personality scales. Journal of Personality, 54 , 106-148.Brody, S., Laan, E., & van Lunsen, R. H. W. (2003). Concordance between women'sphysiological and subjective sexual arousal is associated with consistency of orgasmduring intercourse but not other sexual behavior. Journal of Sex & Marital Therapy, 29,15-23.Byrne, B. M. (1994). Structural Equation Modeling with EQS and EQS/Windowns: BasicConcepts, Applications, and Programming. Thousand Oaks, CA: Sage Publications.Burt, M. (1980). Cultural myths and supports for rape. Journal of Personality and SocialPsychology, 38, 217-230.Buss, D. M. (1989). Sex difference in human mate preferences: Evolutionary hypotheses testedin 37 cultures. Behavioral and Brain Sciences, 12, 1-49.Clark, L. A., & Watson, D. (1995). Constructing validity: Basic issues in objective scaledevelopment. Psychological Assessment, 7, 309-319 Gender Roles and Sexual Pleasure 33 Coward, R. (1985). Female desires. New York: Grove Press, Inc.Crowne, D. P., & Marlowe, D. (1960). A new scale of social desirability independent ofpsychopathology. Journal of Consulting Psychology, 24, 349-354.Deci, E. L., & Ryan, R. M. (1995). Human autonomy: The basis for true self-esteem. In M. H.Kernis (Ed.), Efficacy, agency, and self-esteem (pp. 31-49). New York: Plenum.Dworkin, A. (1987). Intercourse. New York: Free Press Paperbacks.Edwards, J. N., & Booth A. (1994). Sexuality, marriage, and well-being: The middleyears. In (Ed) A. S. Rossi. Sexuality Across the Life Course (pgs. 223-259)Chicago: University of Chicago Press.Fiske, S. (1993). Controlling other people: the impact of power on stereotyping.American Psychologist, 48, 621-628.Foucault, M. (1985) The Use of Pleasure: A History of Sexuality, Volume 11. London: PenguinFredrickson, B. L., & Roberts, T. (1997). Objectification theory: Toward an understandingwomen's lived experiences and mental health risks. Psychology of Women Quarterly,21(2), 173-206.Gagnon, J. & Smith, W. (1973). Sexual conduct: The social sources of sexuality.Chicago: Aldine.Geer, J., & Janssen, E. (2000). The sexual response system. In J. T. Cacioppo, L. G.Tassinary, & G. G. Berntson (Eds.), Handbook of psychophysiology (pp. 315-341).New York: Cambridge University Press.Gosling, S. D., Vazire, S., Srivastava, S., & John, O. P. (2004). Should we trust web-basedstudies? A comparative analysis of six preconceptions about Internet questionnaires.American Psychologist, 59, 93-104. Gender Roles and Sexual Pleasure 34 Griffitt, W. (1987). Females, males, and sexual response. In K. Kelley (Ed.), Females, males,and sexuality: Theories and research. Albany, NY: State University of New York Press.Haavio-Mannila, E., & Kontula, O. (1997). Correlates of increased sexual satisfaction.Archives of Sexual Behavior, 26, 399-419.Harris, G. (2004, February 28). Pfizer gives up testing Viagra on women. New YorkTimes, p. 10, C1.Heiman, J. R. (1977). A psychophysiological exploration of sexual arousal patterns infemales and males. Psychophysiology, 14, 266-274.Horswill, M. S., & Coster, M. E. (2001). User-controlled photographic animations, photograph-based questions, and questionnaires: Three internet-based instruments for measuringdrivers' risk-taking behavior. Behavior Research Methods, Instruments, and Computers,33, 46-58.Hu, L., & Bentler, P.M. (1999). Cutoff criteria for fit indexes in covariance structureanalysis: Conventional criteria versus new alternatives. Structural EquationModeling, 6, 1-55.Hurlbert, D. F., Apt, C. & Rabehl, S. M. (1993). Key variables to understanding femalesexual satisfaction: An examination of women in nondistressed marriages.Journal of Sex and Marital Therapy, 19, 154-165.Impett, E. A., & Peplau, L. A. (2003). Sexual compliance: Gender, motivational,and relationship perspectives. Journal of Sex Research, 40, 87-100.Irigaray, L. (1996). The sex which is not one. In S. Jackson, & S. Scott (Eds.).Feminism and sexuality: A reader (pgs 79-83). New York: Columbia UniversityPress. Gender Roles and Sexual Pleasure 35 Jaccard, J. and Choi K. W. (1996). LISREL approaches to interaction effects inmultiple regression. Thousand Oaks, CA: Sage Publications.Jeffreys, S. (1990). Anticlimax: a feminist perspective on the sexual revolution. London:Women’s Press.Jenkins, W. J. & Becker, J. B. (2003). Female rats develop conditioned placepreferences for sex and their preferred inverval. Hormones & Behavior, 34, 15-29.Jhally, S. (1995). Video: Dreamworlds 2: desire/sex/power in music video.Northhampton: Media Education Foundation.Johnson, M. E. (1989). Sex-role orientation and attitudes toward sexual expressionPsychological Reports, 64, 1064.Kiefer, A. K., Sanchez, D. T., Kalinka, C., & Ybarra, O. (in press). How women’snonconscious association of sex with submission relates to their ability to reachorgasm. Sex Roles.Kilbourne, Jean (2000a). Video: Killing Us Softly 3: Advertising's Image ofWomen. Northhampton: Media Education Foundation.Kilbourne, Jean (2000b). Can't buy My Love: How Advertising Changes the WayWe Think and Feel. New York: Simon and Shuster.Kim, J. L., & Ward, L. M. (2004). Pleasure reading: associations between youngwomen’s sexual attitudes and their reading of contemporary women’smagazines. Psychology of Women Quarterly, 28, 48-58.Kimlicka, T. M., Cross, H. J. and Tarnai, J. (1983). A comparison of androgynous,feminine, masculine, and undifferentiated women on self esteem, body Gender Roles and Sexual Pleasure 36 satisfaction, and sexual satisfaction. Psychology of Women Quarterly, 7, 291-294.Kitzinger, S. (1984). Woman's experience of sex. New York: Collins.Klem, L. (2000). Structural equation modeling. In Grimm, L. G. & Yarnold, P. R. (Eds.).Reading and understanding MORE multivariate statistics (pp. 227260).Washington, DC: American Psychological Association.La Guardia, J. G., Ryan, R. M., Couchman, C. E., & Deci, E. L. (2000). Within-personvariation in security of attachment: A self-determination theory perspective onattachment, need fulfillment, and well-being. Journal of Personality and SocialPsychology, 79, 367-384.Laumann, E. O., Paik, A., Rosen, R. C. (1999). Sexual dysfunction in the United States:prevalence and predictors. Journal of American Medical Association, 281,537-544.Lawrance, K., Taylor, D., Byers, S. E. (1996). Differences in men and women's global, sexual,and ideal-sexual expressiveness and instrumentality. Sex Roles. 34, 337-357Lowry, D. T., Love, G., & Kirby, M. (1981). Sex on the soap operas: Patterns of intimacy.Journal of Communication, 31, 90-96.Martin, K. A. (1996). Puberty, sexuality, and the self: Boys and girls at adolescence.New York: Routledge.Masters, W. H., & Johnson, V. E. (1966). Human sexual response. Oxford, England: LittleBrown. MacKinnon, C. A. (1982). Feminism, marxism, method and the state: Anagenda for theory. Signs, 7, 515533. Gender Roles and Sexual Pleasure 37 MacKinnon, C. A. (1989). Toward a feminist theory of the state. Cambridge, MA:Harvard University Press.Miller, B. C., Benson, B. (1999). Romantic and sexual relationship development duringadolescence. In W. Furman, W., & B. B. Brown (Eds.) Development of romanticrelationships in adolescence (pp. 99-121). New York: Cambridge University Press.Millet, K. (1970). Sexual Politics. New York: Avon Books.Murnen, S.K., & Stockton, M. C. (1997). Gender and self-reported sexual arousal in responseto sexual stimuli: A meta-analytic review. Sex Roles, 37, 135-153.Oliver, M. B, & Hyde, J. S. (1993). Gender differences in sexuality: A meta-analysis.Psychological Bulletin, 114, 29-51.Osborne, J. W. (2003). Effect sizes and the disattenuation of correlation and regressioncoefficients: lessons from educational psychology. Practical Assessment, Research andEvaluation, 8(11). Retrieved August 23, 2005 fromhttp://PAREonline.net/getvn.asp?v=8&n;=11.O’ Sullivan, L. F., & Allgeier, E. R. (1998). Feigning sexual desire: Consenting tounwanted sexual activity in heterosexual dating relationships. Journal of Sex Research,35, 234-243.Raykov, T., Tomer, A., & Nesselroade, J. R. (1991). Reporting structural equationmodeling results in Psychology and Aging: Some proposed guidelines.Psychology and Aging, 6, 499-503.Rennie, K. M. (1997, January). Exploratory and confirmatory rotation strategies in exploratoryfactor analysis. Paper presented at the annual meeting of the Southwest EducationResearch Association. Gender Roles and Sexual Pleasure 38 Rohlinger, D.A. (2002). Eroticizing Men: Cultural influences on advertising andmale objectification. Sex Roles, 46, 61-74.Rosenzweig, J. M., & Dailey, D. M. (1991). Women's sex roles in their public and private lives.Journal of Sex Education and Therapy. 17, 75-85.Rudman, L. A. (1998). Self-promotion as a risk factor for women: The costs and benefitsof counterstereotypical impression management. Journal of Personality and SocialPsychology, 74, 629-645.Rudman, L. A., & Glick, P. (2001). Prescriptive gender stereotypes and backlash towardagentic women. Journal of Social Issues, 57, 743-762.Safir M. P, Peres Y., Lichtenstein M., Hoch Z., & Shepher J. (1982). Psychological androgynyand sexual adequacy. Journal of Sex and Marital Therapy, 8, 228-40. Sanchez, D. T., Crocker, J., & Boike, K. R. (2005). Doing gender in the bedroom:How investment in gender norms affects the sexual experience. Personality and SocialPsychology Bulletin, 31, 1445-1455.Sanchez, D. T., Kiefer, A. K., & Ybarra, O. (2006). Sexual submissiveness in women: Costsfor sexual autonomy and arousal. Personality and Social Psychology Bulletin, 32, 512-524.Schwartz, P. & Rutter, V. (2000). The gender of sexuality. (2 Edition). Thousand Oaks, CA: Pine Forge Press.Snitow, A. B. (1979). Mass market romance: Pornography for women is different.Radical History Review, 20, 141-161. Gender Roles and Sexual Pleasure 39 Sobel, M. E. (1982). Asymptotic intervals for indirect effects in structural equations models. InS. Leinhart (Ed.), Sociological methodology 1982 (pp.290-312). San Francisco, CA:JosseyBass.Sprecher, S., & Cate, R. (2004). Sexual satisfaction and sexual expression as predictors ofrelationship satisfaction and stability. In J. Harvey, A., Wenzel, & S. Sprecher (Eds.),Handbook of sexuality in close relationships (pp. 235-256). Mahwah, NJ: Erlbaum.Sprecher, S. & McKinney, K. (1993). Sexuality. Newbury Park, CA. Sage.Steinman, D. L., Wincze, J. P., Sakheim, B. A., Barlow, D. H., & Mavissakalian, M.(1981). A comparison of male and female patterns of sexual arousal. Archives ofSexual Behavior, 10, 529-547.Tevlin, H. F., & Leiblum, S. R. (1983). Sex role stereotypes and female sexualdysfunction. In V. Franks & E. D. Rothblum (Eds.). The stereotyping of women:Its effects on mental health (pp. 129-150). New York: Springer.Tolman, D. L. (2002). Dilemmas of desire. Cambridge, MA: Harvard University Press.Weinberg, M. S., Swensson, R. G. & Hammersmith, S. K. (1983). Sexual autonomy and thestatus of women: Models of female sexuality in the U.S: Sex manuals from 1950 to 1980.Social Problems, 30, 312-324.White, L., & Keith, B. (1990). The effect of work-shift on the quality and stability of maritalrelations. Journal of Marriage and Family, 52, 453-462.Willemsen, E. W. (1987). Sex-role orientation, self-esteem, and uniqueness: An exploration ofthe undifferentiated category. Psychological Reports, 60, pp. 859-866. Gender Roles and Sexual Pleasure 40 Zamboni, B. D., Crawford, I., & Williams, P. G. (2000). Examining communication andassertiveness as predictors of condom use: Implications for HIV prevention. AIDSEducation and Prevention, 12, 492-504.Zilbergeld, B. (1978). Male sexuality. Boston: Little Brown. Gender Roles and Sexual Pleasure 41 Footnotes1 We factor analyzed the items for our measure of attitudes towards traditional sexual roles usingprinciple axis factoring with oblimin rotation (Delta = 0) to obtain a simple structure and to allowthe items to be intercorrelated (Rennie, 1997). The factor analysis revealed a single factor witheigenvalue greater than 1, which explained 64% of the variance. All items in the scale loadedhighly on this factor (factor loadings > .70).2 To examine the factor structure of our measure of passive sexual behavior, we factor analyzedall items using principle axis factoring with oblimin rotation (Delta = 0) to obtain a simplestructure and to allow the items to be intercorrelated (Rennie, 1997). The factor analysis revealeda single factor with eigenvalue greater than 1, which explained 64% of the variance. All items inour scale loaded highly on this factor (factor loadings > .70).3 Neither relationships status nor sexual experience moderated the predicted interaction of genderroles by gender on passive sexual behavior.4 Neither relationships status nor sexual experience moderated the predicted relationship betweenpassivity and sexual satisfaction.5 Seventeen participants who terminated the Internet survey before answering the criticalquestions were excluded from analyses.6 For the structural equations analysis, we randomly parceled the scale into two indicators.Parceling is a commonly used procedure to improve the goodness of fit and reduce bias inestimations of structural parameters relative to estimations using single item factors (Bandalos,2002).7 To examine the underlying factor structure of our measures of sexual arousability, ability toreach orgasm, and satisfaction, we conducted a factor analysis with items from all the subscales Gender Roles and Sexual Pleasure 42 of the Sexual Arousability Index and with the items designed to assess sexual satisfaction and theability to reach orgasm. We factor analyzed these items using principle axis factoring withoblimin rotation (Delta = 0) to obtain a simple structure and to allow the items to beintercorrelated (Rennie, 1997). The factor analysis revealed three factors with eigenvalue greaterthan 1. All items from the different subscales of the SAI—save the erotica subscale—loadedhighly on the first factor (factor loadings > .50), which explained 33% of the variance. Theerotica subscale, orgasm questions, and sexual satisfaction questions did not load highly on thisfirst factor (all factor loadings < .20). Similarly, only the items from the erotica subscale of theSAI loaded highly on the second factor (all factor loadings > .80), which explained 13% of thevariance. These findings replicating past work (Kiefer et al., 2006; Sanchez et al., 2006), whichfound that the erotica subscale did not correlate strongly with the other SAI subscales. In thepresent study, the erotica subscale was weakly correlated with the other subscales for men (r =.238, p < .001) but was uncorrelated for women (r = .051, p > .4, ns). Hence we excluded itemsfrom the erotica subscale from subsequent analyses. Items from all the other SAI subscales wereparceled and were entered as observed variables that reflected one underlying latent variable ofsexual arousability in our structural models. Finally, only the orgasm and sexual satisfactionitems loaded highly on the third factor (all factor loadings > .70), which explained 11% of thevariance. Because there are important theoretical distinctions between the ability to reach orgasmand sexual satisfaction, we treated these items as reflecting two separate underlying latentvariables (i.e., the ability to reach orgasm and sexual satisfaction) in our structural models.8 Relationship status did not moderate any of the observed relationships in our structural model. Gender Roles and Sexual Pleasure 43 Table 1. Zero Order Correlations by Gender of Participant in Study 1 ____________________________________________________________________________________________________________1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 1. Traditional sexual roles--.365*** -.140 -.452*** .232* .150 -.007 .103 .127 -.135 2. Passive Behavior-.151+ ---.357*** .152 .224* -.079 .081 -.056 .084 -.213*** 3. Libido-.055 -.328*** ---.125 -.403*** .090 .131 -.190* -.394*** .113 4. Traditional Relationship Values .410*** -.081 -.107 --.321*** .050 .176+ .189* .195* -.026 5. Gender Beliefs about Desire -.122 .018 .231** .030 ---.205* .016 .211* .138 .0886. Disparities in Desire.092 .178* .094 .181* .389*** ---.198* -.212* .144 -.183+7. Partner’s Sexual Agency.050 -.018 .117 -.032 -.023 -.125 --.006 -.324*** .1598. Social Desirability-.011 -.087 -.156+ -.038 -.081 -.186* .080 --.023 .0939. Sexual Experience-.095 .162+ -.411*** .032 -.097 .057 .007 .196* ---.348**8. Sexual Satisfaction-.031 -.190* .169* .005 -.185* -.246** .000 .158+ -.289** ----____________________________________________________________________________________________________________Note: Men are represented below diagonal, women above diagonal. * p < .05 ** p < .01 *** p < .001, “Disparities in desire”refers to participants’ perceptions that their sexual partner has less sexual desire than they do. “Gender beliefs about Desire” refer tothe tendency to belief that men have greater sexual desire than women. “Sexual experience” refers to whether or not the participant Gender Roles and Sexual Pleasure 44 has engaged in sexual intercourse. Sexual experience was coded as 1 = experienced sexual intercourse, while 2 = not having experienced sexual intercourse. Gender Roles and Sexual Pleasure 45 Table 2. Gender Differences in Gender Roles and Sexual Attitudes in Study 1 MeasureMen Women DifferenceCohen’s M (SD) M (SD)td Traditional sexual roles 3.96 0.93 3.90 1.12 0.52, ns0.09 Passive Behavior3.04 0.94 4.20 0.99 -9.76*** -1.20Libido4.43 0.77 3.68 0.89 7.46***0.90Traditional Relationship Values 3.53 0.90 3.03 0.96 4.43***0.54Gender Beliefs about Desire 4.65 1.07 4.56 1.03 0.75, ns0.09Disparities in Desire3.33 1.08 2.23 0.83 9.33***1.14Partner Sexual Agency 2.92 0.79 3.27 0.86 -3.49*** -.0.44Social Desirability3.36 0.19 3.34 0.21 0.74, ns0.10Sexual Experience1.21 0.41 1.34 0.48 -2.43***-0.29Sexual Satisfaction4.36 1.80 4.71 1.93 0.43, ns-0.19 Note: “Disparities in desire” refers to participants’ perceptions that their sexual partner has lesssexual desire than they do. “Gender beliefs about desire” refer to the tendency to belief that menhave greater sexual desire than women. “Sexual experience” refers to whether or not theparticipant has engaged in sexual intercourse. Sexual experience was coded as 1 = experiencedsexual intercourse, 2 = not having experienced sexual intercourse. Gender Roles and Sexual Pleasure 46