Introduction

Sexual scripts are guides for men and women that direct and inform their behavior and expectations during sexual encounters. Despite increasingly egalitarian sexual roles for men and women, the predominant sexual script in the United States still prescribes male agency and female receptivity (Gagnon, 1990; Kiefer & Sanchez, 2007; Masters, Casey, Wells, & Morrison, 2013; O’Sullivan & Byers, 1992; Sakaluk, Todd, Milhausen, Lachowsky, & Undergraduate Research Group in Sexuality, 2013). Beginning with people’s first sexual experiences, men are socialized to direct sexual encounters and women are socialized to respond, not too eagerly, to men’s initiation (Martin, 1996; Vannier & O’Sullivan, 2011). Although men and women in long-term relationships may be more likely to deviate from traditional scripts and act in accordance with personal standards, research has found that sexual interactions often follow traditional sexual scripts, affording women less agency even in long-term relationships (Dworkin & O’Sullivan, 2005; Kiefer & Sanchez, 2007; Morgan & Zurbriggen, 2007; Sanchez, Phelan, Moss-Racusin, & Good, 2012b; Vannier & O’Sullivan, 2011). As a potential explanation for gender differences in sexually agentic behavior, the present studies will determine whether women, compared to men, incur (Study 1) or perceive (Study 2) greater costs for sexually agentic behavior.

Gender Differences in Sexually Agentic Behaviors

In the current research, we define sexual agency as the power to initiate sexual intercourse and communicate one’s sexual desires. Much of the extant research on gender differences in sexual agency has focused on initiation patterns, and has found that men tend to initiate sex more frequently than women. In fact, Dworkin and O’Sullivan (2005) found the most common sexual initiation pattern in couples to be male-dominated, even if both partners reported a preference for a more egalitarian pattern. Furthermore, when effect sizes are provided, gender differences in self-reported sexually assertive behaviors are typically large, even among couples (Cohen’s d = .90; Sanchez et al., 2012b). In addition to reporting less agentic sexual behaviors, women may internalize a less agentic role. Women are more likely than men to implicitly associate sex with submissiveness (Kiefer, Sanchez, Kalinka, & Ybarra, 2006) and this association has been found to predict sexually submissive behavior (Sanchez, Kiefer, & Ybarra, 2006). Using a lexical decision task in which participants had to distinguish words from non-words, Sanchez et al. found that women were quicker to categorize submissive words (e.g., comply, submit) following a sexual prime (e.g., climax, sex) compared to a neutral prime. This finding suggests that thinking about sex makes submission more cognitively accessible for many women. The extent to which women associated sex with submission in the lexical decision task predicted women’s actual submissive behavior in their sexual relationships. Moreover, in a set of studies conducted by Hundhammer and Mussweiler (2012), women who had been primed with sex behaved more submissively in the experimental context. Men, on the other hand, associate sex with dominance to such an extent that priming men with sex has been found in laboratory studies to lead to greater aggressive behavior (Mussweiler & Forster, 2000). Thus, women appear to self-report less agentic sexual behavior, compared to men, and have an automatic link between sex and less agentic behavior.

Importance of Sexual Agency

Why does it matter that there are gender differences in sexually agentic behavior? Researchers have found that sexual agency and autonomy are important predictors of both sexual functioning and sexual satisfaction for men and women (Simms & Byers, 2013; Smith, 2007; for a review, see Sanchez, Fetterolf, & Rudman, 2012a). Not surprisingly, women’s sexually submissive behavior predicts lowered sexual autonomy or the belief that one is not free to exert control within sexual contexts (Kiefer & Sanchez, 2007; Sanchez et al., 2006). In turn, a lack of sexual autonomy predicts lowered sexual arousability and satisfaction as well as more sexual problems (e.g., difficulty reaching orgasm) (for a review, see Laan & Rellini, 2011). Submissive behavior is not as detrimental to women’s sexual satisfaction if it is in line with their personal desires (e.g., if they want to be dominated by their partner), but if women engage in sexually submissive behavior when it is inconsistent with their personal desires, their sexual partners experience lower satisfaction as well (Sanchez et al., 2012b). Thus, it is important to examine whether women avoid sexually agentic behavior because they incur penalties related to person perception or whether they expect penalties that do not actually exist.

Perceptions of Female Agency and Sexuality

Although researchers have examined women’s sexually agentic behaviors and their link to sexual satisfaction, there is very little research regarding people’s perceptions of female sexual agency. However, a rich literature examining people’s perceptions of female agency outside of the bedroom suggests that women are consistently stigmatized or penalized for agentic behavior (e.g., the backlash effect) (Rudman, 1998; Rudman & Fairchild, 2004; Rudman & Glick, 1999). Even in situations in which agency would seem beneficial, people view agentic women as too aggressive and dominant and thus unlikable (and, subsequently, people are less likely to recommend them for hiring compared to agentic men) (Phelan, Moss-Racusin, & Rudman, 2008; Rudman, 1998). Furthermore, agency in general is associated with selfish behavior; people attribute agency to others who behave in line with their own desires and therefore people high in agency may be viewed as more selfish than others low in agency (Abele & Wojciszke, 2007; Cislak & Wojciszke, 2008). With respect to gender, women are supposed to be sensitive to others’ needs and selfishness is proscribed for women, but not for men (Rudman, Moss-Racusin, Phelan, & Nauts, 2012). Thus, women may be perceived more negatively overall when they are agentic, compared to men. Moreover, women may want to avoid acting with agency if they believe agentic behavior is perceived as selfish.

Research that focuses on perceptions of women’s sexual behavior in general may also be informative with respect to people’s perceptions of sexual agency specifically. The sexual double standard suggests that there are differing standards of sexual permissiveness for men and women. Women are more likely than men to suffer negative consequences for (1) having a higher number of previous sexual partners (Marks, 2008), (2) engaging in casual sex (Fromme & Emihovich, 1998), and (3) having more permissive sexual attitudes (e.g., attitudes which deviate from those expected of women) (Sprecher & Hatfield, 1996; Sprecher, McKinney, & Orbuch, 1987). Furthermore, people—especially women—who are more sexually permissive or experienced tend to be viewed as less desirable relationship partners although they are considered more desirable sexual partners (Conley, 2011; Fromme & Emihovich, 1998; Oliver & Sedikides, 1992). For example, Conley (2011) used the Clark and Hatfield (1989) paradigm to measure perceptions of men and women who approached strangers and expressed a desire to engage in casual sex with them. Both men and women rated the female targets attempting to initiate casual sex as more skilled and, therefore, more desirable sexual partners.

Overall, research on the sexual double standard has produced inconsistent results and many studies do not find evidence for it (Gentry, 1998; Marks & Fraley, 2005, 2006; O’Sullivan, 1995; for a review, see Crawford & Popp, 2003). In line with these mixed results, it is possible that people may be hesitant to report differing standards for men’s and women’s sexual agency. However, studies that examine people’s beliefs about the sexual double standard find that the majority of men and women believe it still exists (Marks & Fraley, 2006; Milhausen & Herold, 1999, 2001; Rudman, Fetterolf, & Sanchez, 2013) and the belief that they may be stigmatized for their sexual behaviors may very likely limit women’s sexual freedom. For example, women will report lower levels of sexual experience if they believe others are likely to see their responses (Alexander & Fisher, 2003) and women who anticipate social repercussions (e.g., being labeled a “slut”) are less likely to accept hypothetical or real offers of casual sex (Conley, Moors, Matsick, Zieglar, & Valentine, 2011). Thus, women may base their sexual behavior on the consequences they expect, rather than actual experiences of negative consequences. With respect to sexual agency, women have been found to underreport the frequency with which they initiate sex in their romantic relationships (Anderson & Aymami, 1993; Anderson & Sorenson, 1999). Women also report less comfort initiating sex to the extent that they believe there are differing sexual standards for men and women (Greene & Faulkner, 2005) or that others would not approve (Simms & Byers, 2013). Similar to reporting the number of sexual partners or accepting offers of casual sex, women may fear negative consequences for their sexually agentic behavior.

Overview of the Research

Because lowered sexual agency has negative consequences for heterosexual women’s sexual satisfaction, as well as their partners’, it is important to consider why women do not act with levels of sexual agency that match their male counterparts (e.g., Sanchez et al., 2012b). In the following studies, we examined potential reasons for the gender difference in sexual agency, focusing specifically on reasons related to person perception. In Study 1, participants viewed a fabricated dating profile of a man or woman who was either high or low on sexual agency (i.e., initiating sex and communicating their desires) and rated them as a potential sexual and romantic partner. To consider how perceptions of sexual agency may be linked to perceptions of other, more frequently studied sexual behavior, we asked participants to indicate the frequency with which the target used safe sex practices and the number of previous partners the target was likely to have. In Study 2, we asked men and women to rate their own reactions to being perceived as sexually agentic.

Study 1

In Study 1, participants viewed a dating profile of either a man or woman who indicated that they were highly sexually agentic, not sexually agentic, or did not report their sexual role. Participants then rated the extent to which they viewed these targets as desirable sexual partners, selfish partners in general, and good romantic partners overall. Based on previous research, we hypothesized that sexually agentic targets, compared to those low on agency or controls, would be perceived as more desirable sexual partners (Conley, 2011; Fromme & Emihovich, 1998; Oliver & Sedikides, 1992; Sanchez et al., 2012b), more selfish partners, both sexually and romantically (Abele & Wojciszke, 2007; Cislak & Wojciszke, 2008), and worse romantic partners overall (Oliver & Sedikides, 1992). We also expected target gender to affect participants’ ratings; we predicted that participants would rate the sexually agentic woman as the most selfish target and the worst romantic partner because agency and selfishness are proscribed for women (Rudman et al., 2012).

Finally, we examined the relationship between sexual agency and perceptions of other potentially risky sexual behaviors (i.e., infrequent use of safe sex practices and a higher number of previous sexual partners). Communicating sexual desires and initiating sexual encounters may suggest knowledge learned through experience. Thus, participants may believe highly agentic targets have had more previous sexual partners than targets who are low on sexual agency. Increased sexual communication (an aspect of sexual agency as it is defined in the current study) and assertiveness both correspond with a greater likelihood of condom use (Baele, Dusseldorp, & Maes, 2001; Farmer & Meston, 2006; Sterk, Klein, & Elfison, 2003; van Anders, Goldey, Conley, Snipes, & Patel, 2012); thus, participants may believe that sexually agentic targets use contraception more frequently compared to other targets. However, we did not find any research on sexual agency and perceived contraception use. It is possible that participants may view sexually agentic targets as indiscriminating and risky and therefore not only having more previous partners, but also using safe sex practices less frequently compared to less sexually agentic targets or controls.

Method

Participants

Undergraduate students were recruited to participate in exchange for partial credit toward their introductory psychology research requirement. Students who indicated in a prescreen questionnaire that they were under 18 years of age, in a committed relationship, or not heterosexual were excluded from data collection. We decided to only include heterosexual participants because the vast majority of research on sexual roles and scripts focuses on heterosexual relationships. Two people whose reported sexual orientation changed before participating in the experiment were dropped from data analysis, for a total sample size of 235 (115 men, 120 women). The mean age was 19.02 years (SD = 1.57). The ethnic composition of the sample was as follows: 108 (46.0 %) were White, 30 (12.8 %) were East Asian, 21 (8.9 %) were South Asian, 18 (7.7 %) were Latino, 31 (13.2 %) were Black, 1 (0.4 %) was American Indian, 7 (3.0 %) were Middle Eastern, and 19 (8.1 %) indicated another ethnicity.

Procedure and Measures

Participants completed the approximately 20-min study in a psychology laboratory. They were told that the study was about the effects of peripheral cues (i.e., advertising) on perceptions of dating profiles. Participants viewed a print out of a fictitious dating profile for a target individual named either Anna or Jason. Each profile contained the same two ads, created for the purposes of this study: a cell phone recycling service and a college travel website. In addition to information about their favorite type of food, ideal dating activities, and their relationship style, which was held constant across all conditions, participants also read about the targets’ sexual role. To place the information about targets’ sexual role in the context of a dating site, participants were told that sexual compatibility is an important part of a successful relationship. In the high agency condition, targets’ profiles read: “I don’t know whether I have a ‘sexual style’ but, during sex, I typically take the lead. I usually initiate sex and I tend to tell my partner what I like in bed.” In the low agency condition, they read: “I don’t know whether I have a ‘sexual style’ but, during sex, I typically let my partner take the lead. I don’t usually initiate sex. I just let things happen.”

In addition to the high and low agency sexual role conditions, this study also included a control condition in which no information about the targets’ preferred sexual role was presented. Additionally, participants were randomly assigned to both a target gender (i.e., all participants could view the dating profile for either Jason or Anna) and sexual role condition. Thus, the design was a 2 (Participant Gender) × 2 (Target Gender) × 3 (Target Sexual Role: High Agency, Low Agency, Control) between-subjects factorial. Because all of the participants were heterosexual, participants who viewed a same-sex target’s profile responded to the measures in the way they believed a potential romantic or sexual partner would (e.g., men indicated what they believed the average woman would think of Jason).

After viewing the dating profile, participants rated the extent to which they believed different traits related to sexual and romantic relationships (e.g., selfish romantic partner, sexually alluring partner) would be descriptive of the target on a scale of 1 (not at all) to 7 (very). Before analysis, the individual traits were aggregated into the scales presented below. As part of a manipulation check to confirm whether the low and high agency profiles were perceived as differing in agency, participants indicated how sexually dominant, assertive, aggressive, passive (reverse scored), and submissive (reverse scored) they believed the target to be. These items were averaged to create the agentic sexual partner index (α = .92).

Perceptions of Targets as Sexual and Romantic Partners

Participants reported how sexually alluring, desirable, exciting, skilled, and confident they believed the target to be as a sexual partner. The five items were averaged to form the desirable sexual partner index (α = .93). The extent to which participants viewed the target as a selfish sexual partner was measured by averaging participants’ ratings of the target on the following traits: selfish, self-involved, self-focused, and self-centered sexual partner (α = .90). Participants also rated the extent to which they viewed the target as a controlling, dominant, stubborn, selfish, and cold romantic partner. These items were combined to create the controlling romantic partner index (α = .82). Further analyses revealed that the selfish sexual partner and controlling romantic partner indices were strongly correlated, r(232) = .79, p < .001. We therefore combined these two indices into a selfish partner general index (α = .91) accounting for the perception that targets put their own needs above those of their partners. Participants also rated the target as an overall romantic partner, by responding to the question, “Overall, what kind of girlfriend/boyfriend do you think this person would be?” on a scale of 1 (very bad) to 7 (very good). Participants rated the target first on the romantic partner variables and then on the sexual partner variables.

Perceptions of Targets’ Sexual Practices

Following the ratings of the target, participants indicated how frequently they believed the target to use safe sex practices. Participants responded to the following items on a scale of 0 (0 % of the time) to 10 (100 % of the time): “How often do you think this person uses safe sex during sexual encounters?”, “How often do you think this person uses contraception during sexual encounters?”, and “How often do you think this person has unprotected sex?” The final question was reverse scored and the three items were averaged together to create a safe sex index (α = .86). Finally, participants indicated how many people they believed the target to have had sex with in their lifetime. Five participants reported a number of sexual partners that was greater than three SDs above the mean (M = 4.38, SD = 3.18). In order to limit the effect of these outliers on the data while still accounting for participants’ extreme responses, we replaced their estimates with a number that was exactly 3 SDs above the mean (Field, 2013). This strategy produced the same results as removing the outliers entirely but does not result in the loss of any data.

Results

Table 1 shows the correlations between the variables of interest. To examine the effect of participant gender, target gender, and target sexual role on participants’ perceptions of the target as a potential sexual or romantic partner, a 2 × 2 × 3 ANOVA was conducted for each dependent variable. Table 2 reports the means and F values for the main effect of targets’ sexual role on all Study 1 variables. A significant main effect of target sexual role on the agentic sexual partner variable indicated that the experimental manipulation was successful: Participants rated highly agentic targets as significantly more sexually agentic than controls, who were rated as significantly more sexually agentic than targets who were low in agency (see Table 2).

Table 1 Correlations for Study 1 dependent variables by target gender
Full size table
Table 2 Main effect of target sexual role on participants’ ratings of the target as a romantic and sexual partner in Study 1
Full size table

Perceptions of Targets as Sexual and Romantic Partners

Examining the desirable sexual partner index, we found main effects of target gender and target sexual role (see Table 2 for F value). Because both main effects were superseded by a significant target gender × target sexual role interaction, F(2, 223) = 4.82, p = .009, we only report the findings for the interaction. Participants rated the female target as more desirable than the male target in the agentic condition (M F  = 5.62, SD F  = 0.97; M M  = 4.70, SD M  = 1.04), t(81) = 4.14, p < .001, d = 0.94, and the low agency condition (M F  = 3.59, SD F  = 1.20; M M  = 3.10, SD M  = 0.93), t(75) = 2.04, p = .045, d = 0.46. Participants did not rate the targets differently in the control condition (t < 1). There was also a significant participant gender × target sexual role interaction, F(2, 223) = 4.15, p = .017, in which women rated the agentic target as more desirable (M = 5.42, SD = 0.89) than did men (M = 4.87, SD = 1.25), t(81) = 2.32, p = .023, d = 0.51.

We also found a significant main effect of target sexual role for the selfish partner index. Targets in the high agency condition were viewed as significantly more selfish sexual and romantic partners than were targets in the control condition, who in turn were rated as significantly more selfish than targets in the low agency condition (see Table 2).

For the overall romantic partner question, there was a main effect of target sexual role, whereby highly agentic targets were viewed as worse romantic partners than controls, with low-agency targets not differing from either group. In addition, there was a main effect of gender in which women (M = 5.15, SD = 0.80) rated targets in general as better romantic partners than did men (M = 4.80, SD = 1.03), F(2, 223) = 9.13, p = .003, d = 0.38.

Perceptions of Targets’ Sexual Practices

The main effect of target sexual role on frequency of targets’ safe sex practices was significant. Participants believed that highly agentic targets used safe sex practices and contraception less frequently than less agentic targets or controls. In addition to a target sexual role main effect for the number of sexual partners participants believed targets to have, there was a significant target gender × target sexual role interaction, F(2, 216) = 5.08, p = .007. Participants believed highly agentic targets and controls had significantly more sexual partners than target who were low in agency. Gender played an important role within the high agency condition where participants believed the female target had more sexual partners (M = 5.84, SD = 2.89) than the male target (M = 4.23, SD = 1.80), t(80) = 3.02, p = .003, d = 0.60. This finding suggests there is a unique perception of sexually agentic women; they are believed to have had more sexual partners than similarly agentic men.

Discussion

As expected, sexually agentic targets were seen as significantly more desirable sexual partners, but also more selfish sexual and romantic partners compared to targets who were low on agency. Overall, participants believed high-agency targets, compared to controls and low-agency targets, to be riskier sexual partners, with more previous partners and less frequent safe sex practices. Interestingly, we did not find that female targets were rated as more selfish than male targets when they were sexually agentic. Although the overall pattern of results suggests that people rate sexually agentic men and women fairly similarly, there were a few significant interactions with target gender. Specifically, people believed the agentic female target to be more desirable than the low agency targets or the similarly agentic male target. Further, they believed the highly agentic female target to have the highest number of previous sexual partners. It is important to note that the indices on which we found differences for the ratings of the sexually agentic male and female targets were not unambiguously negative and, in fact, were even positive in the case of sexual desirability. In Study 2, we considered men’s and women’s own beliefs about these agency-linked perceptions to further explore potential reasons for gender differences in sexually agentic behavior.

Study 2

Though men and women may experience similar—though not identical—penalties and benefits in person perception for sexually agentic behavior, they may differ in the importance and value placed on these domains. In other words, women may care more about being perceived as safer sexual partners than men do while men care about being perceived as skilled lovers more so than women do. In fact, because agency goes against prescribed gender roles for women (Prentice & Carranza, 2002; Rudman et al., 2012), women may believe it is more negative, compared to men, to be viewed as an agentic sexual partner. Similarly, we expected that women would rate being perceived as a selfish partner and having many previous sexual partners more negatively, compared to men, because selfishness and sexual promiscuity are generally proscribed for women (Milhausen & Herold, 1999; Rudman et al., 2012).

Method

Participants

In order to directly compare Study 2 results to the previous sample, we dropped any participants who did not identify as heterosexual (n = 13) from the analyses, resulting in a total N = 124 (67 men, 57 women). The mean age of the sample was 19.02 years (SD = 1.36). The racial distribution of the sample was as follows: 43 (35.0 %) White, 39 (31.7 %) East Asian, 21 (17.1 %) South Asian, 9 (7.3 %) Latino/Hispanic, 5 (4.1 %) Black/African American, and 6 (4.9 %) participants who listed multiracial or other ethnicities. A little less than half of the participants (43.5 %) reported having had sex in the past and 37.9 % reported being in a relationship at the time of data collection. Participants completed the measures online in exchange for course credit for a psychology class.

Procedure and Measures

Participants completed a short survey (approximately 10 min long) in which they indicated how positive or negative it would be to be viewed as an agentic and desirable sexual partner, as well as a selfish partner overall, using the same traits as in Study 1. For all of these items, the scale ranged from 1 (very negative) to 7 (very positive). They also indicated how important it was to them to be viewed in those ways by an opposite sex partner on a scale of 1 (very important to me NOT to be viewed this way) to 7 (very important to me to be viewed this way). An example valence, or rating, item would be, “How negative or positive would it be if people thought you were a dominant sexual partner?” An importance question for the same item would be, “How important is it to you that potential partners view you as a dominant sexual partner?” For each index, the valence and importance scales were strongly correlated. Therefore, we combined these two scales into one single index, with higher scores indicating that participants rated the items more positively and thought it was more important to be viewed in that way.

Sexual and Romantic Partner Traits

For agentic sexual partner (assertive, dominant, and aggressive), valence and importance ratings were correlated at r(121) = .72, p < .001. The Cronbach alpha for the total scale was .86. For desirable sexual partner (exciting, alluring, desirable, confident, and skilled; α = .94), valence and importance ratings were correlated at r(121) = .78, p < .001. Valence and importance ratings for the selfish partner items (self-centered, focused on self, self-involved, and selfish sexual partner; cold, controlling, dominant, stubborn, and selfish romantic partner) were strongly correlated, r(122) = .79, p < .001, and were combined to create one selfish partner index, α = .91.

Safe Sex Practices

To determine how participants would feel if others viewed them as risky sexual partners, participants reported how positive or negative it would be, as well as how important it was to them to be viewed as never using safe sex practices and using safe sex practice infrequently. Again, the valence and importance ratings were strongly correlated, r(121) = .73, p < .001. The Cronbach alpha for the final 4-item safe sex practices index was .82.

Number of Sexual Partners

Finally, participants indicated how positive or negative it would be if potential partners viewed them as having more sexual partners than they actually have and having more sexual partners than other people their age. They also indicated how important it was to them to be viewed as having more sexual partner than they actually have or than other people their age. Valence and importance ratings were strongly correlated, r(121) = .64, p < .001, and combined into a total number of sexual partners index, α = .88.

Results

To examine gender differences in participants’ ratings of the different sexual and romantic partner traits, we conducted a multivariate analysis of variance (MANOVA) with the agentic sexual partner, desirable sexual partner, selfish partner, safe sex, and number of sexual partners indices as the dependent variables and gender as the independent variable. We decided to use a MANOVA because we expected the gender differences for each of these variables to be in the same direction (with women rating each index less positively than men).

As expected, the main effect of gender was significant, F(5, 118) = 7.34, p < .001. As a follow-up, we examined the effect of gender at the univariate level for each of the dependent variables (see Table 3). Men and women rated being viewed as an agentic and desirable sexual partner similarly. Furthermore, the importance and positivity of being seen as an agentic partner was correlated with the importance and positivity of being seen as desirable for both men and women, r(122) = .46, p < .001. Compared to men, the women reported significantly lower scores on the selfish partner, safe sex, and number of sexual partner indices. By far the largest effect (d = 1.07) was the difference between men’s and women’s rating of the sexual partners index. Women believed it was much more negative than men to be viewed as having more sexual partners than others their age and it was much more important to them not to be viewed that way by potential partners.

Table 3 Univariate level tests of gender differences in ratings of Study 2 variables
Full size table

General Discussion

Overall, we did not find that sexually agentic women were rated much more negatively than sexually agentic men. Although sexually agentic targets were viewed as more desirable, riskier, and more selfish partners, this finding was true for both male and female targets. However, it appears that even when people’s perceptions of sexually agentic men and women were fairly similar, targets’ evaluations of these perceptions were not necessarily the same. Participants viewed sexually agentic men and women as similarly selfish partners, both sexually and romantically, yet women believed it was much more negative than men to be viewed as a selfish partner. The same pattern held for perceptions of safe sex practices. Furthermore, women may be penalized for their sexual agency by appearing especially sexually experienced (i.e., having more previous sexual partners than others their age). Compared to men, women believed it was much more negative to be viewed as having many sexual partners and it was more important to them not to be viewed this way (a large effect).

Women may view this perception of sexual experience as more negative because experienced women are more likely to be stigmatized starting at a young age. Girls who show an interest in sex, or act assertively with regards to sex, are often labeled as “sluts” or “whores” (likely in an attempt to keep them in their place, as a desire for female sexual submission is related to both sexism and support for societal inequities) (Eder, Evans, & Parker, 1995; Rosenthal, Levy, & Earnshaw, 2012; Rudman et al., 2013). Research using the National Longitudinal Study of Adolescent Health found that a greater number of sexual partners correlated positively with male peer acceptance and negatively with female peer acceptance (Kreager & Staff, 2009), likely because male sexual experience is related to prestige while female sexual experience is not (Jonason & Fisher, 2009). Thus, beginning at an early age, women learn that sexual experience leads to potential negative outcomes not only in relationships with romantic partners, but also in their peer relationships. Regardless of whether women actually experience negative romantic or sexual outcomes due to their number of previous sexual partners, if they believe a sexual double standard exists and they fear repercussions, women are likely to at least report having had fewer sexual partners. Furthermore, because sexual agency is linked to perceptions of sexual experience, women may be less likely to communicate their desires and initiate sex in an attempt to appear less experienced than they actually are. As outlined above, a lack of sexual agency for women can negatively impact their own sexual satisfaction (Kiefer & Sanchez, 2007; Sanchez et al., 2006) as well as their partner’s satisfaction (Sanchez et al., 2012b).

Somewhat unexpectedly, the importance and positivity of being seen as an agentic sexual partner was significantly correlated with the importance and positivity of being seen as a desirable sexual partner for women as well as men. Study 1 demonstrated that people viewed agency as especially desirable. In fact, the agentic female target was viewed as a more desirable sexual partner than the agentic male target by both men and women. Further, both men and women rated being viewed as sexually agentic positively. Thus, it appears that women may not fear being judged for initiating sexual experiences or communicating their desires. Rather, they fear the traits that are associated with sexual agency, such as perceived selfishness and promiscuity. This may help to explain why, even though women are cognizant of the desirable aspects of sexual agency, their levels of agency have yet to match their male peers (Dworkin & O’Sullivan, 2005; Sanchez et al., 2012b). It is likely a product of the mixed messages women receive, urging them to be sexually desirable (e.g., flirtatious and attractive) (Prentice & Carranza, 2002), but not too sexually experienced (Eder et al., 1995; Marks, 2008; Milhausen & Herold, 1999).

Limitations and Future Directions

We defined sexual agency as the power to initiate sexual interactions and to communicate one’s sexual desires with their partner. We emphasized these aspects of agency because they have been found to predict greater sexual satisfaction and sexual functioning for women (Sanchez et al., 2012a, b; Simms & Byers, 2013). However, this is not an all-encompassing definition of sexual agency. Sexual agency also includes the power to control contraceptive decisions and the ability to voice desire to not engage in sexual intercourse at times (sexual gatekeeping) (e.g., Impett & Peplau, 2003; Sanchez et al., 2012a), which may be viewed differently in person perception. Framed in this way, sexual agency may have been associated with greater contraceptive use, fewer sexual partners, or less selfishness in agentic targets. Considering that these outcomes (safe sex practices, number of sexual partners, selfishness) are ones that women rated the most negatively, reframing sexual agency in this way may make it more desirable for women. Future research should examine the perceptions associated with different facets of sexual agency, as well as the way people define agency in their own sexual experiences.

As a first examination of people’s perceptions of sexual agency, we limited our focus solely to sexual agency and provided participants with limited information about the targets. Yet, because sexual agency (as we defined it here) was associated with perceptions of selfishness, as well as sexual safety and experience, it is important for future research to consider the separate and additive effects of both agency and other important sexual traits and behaviors (e.g., experience). Also important for future research is examining the perceptions of sexually agentic targets within committed relationships. Research suggests that the large gender gap in sexual agency narrows somewhat within long-term relationships yet persists (Dworkin & O’Sullivan, 2005; Sanchez et al., 2012b). We would expect women in relationships to care about their partners’ perceptions of them, as well as the satisfaction they both receive from their sexual experiences. Conducting qualitative research (e.g., interviews, focus groups) with partnered men and women would provide ample information and could help to narrow the focus of surveys or experimental studies. A closer look at why some women who are with long-term partners still feel as though they cannot or should not behave with agency in their sex lives (e.g., Dworkin & O’Sullivan, 2005; Sanchez et al., 2012b) would likely be informative for women who are not in relationships as well.

An interesting area for future research to consider would be implicit attitudes toward sexually agentic, and sexually submissive, men and women. Explicitly, participants rated sexually agentic women in this study as more desirable and more experienced sexual partners compared to men, but it is possible that people have more negative attitudes toward sexually agentic women than they are likely to admit. For example, while the research on sexual permissiveness or experience produces inconsistent results with explicit designs (Crawford & Popp, 2003), asking participants to complete a similar person perception paradigm while under cognitive load results in findings that support the sexual double standard (Marks, 2008). Similarly, while people may say or believe that they are accepting of sexual agency in both men and women, if they are forced to make judgments without the possibility of thinking deeply about their responses, we might find that people perceive sexually agentic women more negatively than they report.

Conclusion

Obviously, women’s sexually agentic behavior, or lack thereof, is not due solely to their concerns about being perceived negatively. This research examined only one aspect of sexual agency. However, based on our findings, we suggest that the perceptions people have of sexually agentic men and women, and the way women view these outcomes, may influence women’s sexually agentic behavior at the beginning of a relationship. For example, women were just as likely as men to rate the sexually agentic female target as selfish and as having more previous partners than any other target. Thus, women are cognizant that sexually agentic behavior is associated with sexual experience, as well as beliefs that they are more selfish partners in general, and we found that it was very important to them not to be viewed in this way. Moving on from this initial study, it is important for future research to continue examining numerous aspects of women’s sexual and social lives when studying sexual agency. Until women no longer anticipate negative outcomes related to sexual experience or sexual agency, and until they see that agency is a common and desirable female trait, many women may be reluctant to express their desires and act with agency in their sexual lives. Unfortunately, this puts women at risk for STI/HIV infection as well as sexual problems and decreased sexual satisfaction (Farmer & Meston, 2006; Kiefer & Sanchez, 2007; Sanchez et al., 2006).