Power, desire, and pleasure in sexual fantasies.

Most people engage in sexual fantasy at least occasionally (Hsu et al., 1994; Kinsey, Pomeroy, & Martin, 1948; Kinsey, Pomeroy, Martin, & Gebhart, 1953; Sue, 1979) and such fantasies are generally recognized as part of a healthy sexuality (Hariton & Singer, 1974). Although other aspects of sexuality have received more attention from researchers, many studies concerning sexual fantasy have nevertheless been conducted (for a review see Leitenberg & Henning, 1995). However, this work has tended to be descriptive in nature, reporting on the age of onset and incidence or frequency of sexual fantasies, summarizing fantasy content, and presenting information on gender differences (and similarities) in these areas. Many interesting questions about sexual fantasy have not yet been considered.

In the research reported here, we investigated several aspects of sexual fantasy that are important both theoretically and practically. We considered sexual fantasies of power--both dominance and submission. Although the relationship between dominant or aggressive sexual fantasies and aggressive sexual acts is not well understood, the presence of a causal link between the two would have important implications for the prevention of rape and sexual abuse. Thus, a fuller understanding of dominance and submission in both men's and women's fantasies is a worthwhile research goal. We also explored the role of pleasure and desire in sexual fantasies. These two concepts seem fundamental to sexuality, yet have not been adequately examined in the context of sexual fantasy. Finally, we investigated the relationship between sexual fantasies and attitudes relevant to sexual aggression.

DOMINANCE AND SUBMISSION IN SEXUAL FANTASY

There are numerous reasons to suspect that merging power with sexuality might be dangerous. Specific behavioral fusions of dominance and sex include rape, sexual harassment, and sexual abuse--serious offenses that often result in long-lasting physical, emotional, and/or psychological damage to victims (Browne & Finkelhor, 1986; Gutek & Koss, 1993; Resick, 1993). Cognitive connections between power and sex may also be problematic in that they are associated with aggressive behavior (Mussweiler & F6rster, 2000; Zurbriggen, 2000) as well as with harmful attitudes and beliefs (Bargh, Raymond, Pryor, & Strack, 1995; Pryor & Stoller, 1994). A recent meta-analysis also found that adherence to a "hypermasculine" gender role (one that emphasizes dominance and aggressiveness) is associated with sexual aggression perpetration (Mumen, Wright, & Kaluzny, 2002).

In spite of these findings, it is not clear whether fusing dominance with sexual arousal in the context of fantasy would have similar negative effects. On the one hand, clinical evidence suggests that fantasies play a role in the etiology of criminal sexual aggression. Sex offenders frequently fantasize about their crimes, both before and after committing them (Leitenberg & Henning, 1995; MacCulloch, Snowden, Wood, & Mills, 1983; Prentky et al., 1989), and modifying fantasies is often part of the treatment plan for sexual offenders (Johnston, Ward, & Hudson, 1997). On the other hand, fantasies of force are also relatively frequent among community controls as well as among men convicted of only nonsexual crimes (Langevin, Lang, & Curnoe, 1998; Rokach, Nuthrown, & Nexhipi, 1988). For example, Crepault and Couture (1980) reported that 33% of men who had never been convicted of a sexual crime fantasized at least sometimes about raping a woman.

One interpretation of these findings is that aggressive fantasies do not necessarily lead to aggressive behavior because only a subset of men who have aggressive fantasies have committed sexual crimes. Another explanation is that aggressive fantasies are closely linked to aggressive behavior; however, only some men are arrested, charged, and convicted of sexual assault. The few studies that included participants from community or student populations and that obtained self-reported measures of sexually aggressive behavior suggest that there is a connection between aggressive fantasies and aggressive behavior. Greendlinger and Byrne (1987) found that men who reported fantasies of force were more likely to report past use of coercion in sexual relationships; they also indicated a higher likelihood of raping a woman if they knew they wouldn't be caught. Smeaton and Byrne (1987) also found that men with coercive sexual fantasies reported a higher likelihood of raping a woman if they knew they wouldn't be caught. On the other hand, they found no correlation between male participants' coercive fantasies and their reported likelihood of pushing the confederate in the experiment "farther than she says she wants to go sexually." The evidence suggests, then, that fantasies of force might be problematic; however, because of the paucity of research on nonclinical samples, this conclusion remains tentative.

Another problem with the literature on force in fantasies is that it has tended to focus only on men and only on fantasies of dominance. Yet fantasies of submission are common among women. Pelletier and Herold (1988) found that 51% of their female sample reported fantasies of being forced to submit sexually, and Knafo and Jaffe (1984) noted that the fantasy reported most frequently during intercourse for women was "I imagine that I am being overpowered or forced to surrender." This fusion of submission and sex does not, however, appear to carry the same risks as does a fusion of dominance and sex. Women who report fantasies of submission have more positive attitudes about sex (Strassberg & Lockerd, 1998) and are less sexually guilty and more open to a variety of sexual experiences (Pelletier & Herold, 1988; Strassberg & Lockerd, 1998). Moreover, although sexual fantasies of submission may be more common among survivors of childhood sexual abuse (Gold, 1991), sexual victimization as an adult is apparently not predictive of fantasies of submission (Gold, Balzano, & Stamey, 1991). Submissive fantasies in women may therefore be one aspect of a relatively open, positive, guilt-free sexuality. Again, however, because of the small number of studies, this conclusion is tentative. In addition, we know almost nothing about fantasies of dominance in women or fantasies of submission in men. Therefore, one of the main goals of our study was to gain a better understanding of fantasies of both dominance and submission in a community sample that included men and women as participants.

PLEASURE AND DESIRE IN SEXUAL FANTASY

It seems almost tautological to state that sexual desire and sexual pleasure are of central importance to sexuality. To be sure, there are other motivations for engaging in sexual behavior besides the pursuit of physical pleasure, and neither desire nor pleasure are necessary for sexual acts to take place. However, the desire for pleasurable physical sensation is clearly a centrally important sexual motive. This is especially true when one considers sexual fantasies, which are private mental events whose sole purpose would seem to be to induce pleasurable feelings of sexual desire and arousal. It is surprising, then, that virtually no research on sexual fantasies has focused on desire and pleasure.

It is especially interesting to think about the role that desire and pleasure might play in the sexual fantasies of women, given that these have often been difficult feelings for women to express in reality. Vance (1989) has written extensively on the ways that desire and pleasure are problematic for women. She notes that an expression of sexual desire has two consequences for women. First, it invites danger--the danger of rape, of pregnancy, and of harassment, as men tend to interpret any expression of female desire as an invitation for a sexual encounter. Second, women are traditionally the "gate-keepers" of sexuality and are responsible for the control of sex. So if a woman expresses desire, she is indicating that she is giving up that responsibility. In doing so, she violates the traditional feminine role.

Expressions of sexual pleasure are just as difficult for women, according to Vance (1989). Patriarchal cultures are hostile to sexuality in general, and especially to women's sexuality. Again, there is a conflict involving danger; if a woman gives in to sexual pleasure, she invites the danger of violence and punishment. Pleasure and safety are in opposition for women, which forces them to choose one over the other. The realm of fantasy may be a private and safe sphere in which women can experience desire and pleasure free from danger. An empirical assessment of the presence and role of pleasure and desire in women's (and men's) sexual fantasies would help to shed light on the speculations of Vance; our second goal in this research was to conduct such an assessment.

RELATIONSHIP BETWEEN SEXUAL FANTASIES AND ATTITUDES

One of the central questions regarding fantasy concerns its relationship to attitudes and behaviors. Is fantasy distinct from reality, with what one imagines completely unrelated to what one actually does? Or are there connections between the frequency or content of fantasies and real-world behaviors and relevant attitudes? Previous research suggests that such connections do exist. As discussed above, men who fantasize about sexual aggression may also be more likely to engage in sexual aggression. In addition, frequency of sexual fantasy has been correlated with more frequent and more varied experiences of sexuality (Pelletier & Herold, 1988), less sex guilt (Moreault & Follingstad, 1978), and more liberal attitudes about women's place in society (Brown & Hart, 1977).

Given our interest in the role of power in sexual fantasy, we especially wanted to consider a possible relationship between sexual fantasies and attitudes related to sexual aggression perpetration: for example, endorsement of victim-blaming rape myths, a belief that relations between men and women are hostile and adversarial, and conservative beliefs about women's place in society. A third major aim of our study, then, was to explore whether the content of sexual fantasies was related to these specific attitudes, in both women and men.

THE PRESENT STUDY

Our goal was to look at the content of sexual fantasies in a community sample of men and women in early and middle adulthood. The vast majority of research on sexual fantasies has used a checklist method to assess the content of people's fantasies. This method, while expedient, has several drawbacks. Most importantly, it constrains participant responses to the preselected list of fantasy themes chosen by the researcher. In addition, it yields little detailed information about the fantasies. Because of these disadvantages, we chose to collect open-ended accounts of sexual fantasies. We then developed a coding system that allowed us to look in a detailed way at dominance, submission, pleasure, and desire.

We were also interested in the correlation of fantasy themes with attitudes relevant to sexual aggression. We therefore asked participants to complete self-report measures concerning rape myth acceptance, the belief that relationships between men and women are adversarial, and attitudes about women's place in society.

METHOD

Participants

Participants were 85 men and 77 women between the ages of 21 and 45, recruited from a midsized Midwestern city via notices advertising pay for participation in a study on romantic relationships. These individuals were part of a larger pool of 102 men and 92 women, but because of probable differences among bisexual, homosexual, and heterosexual participants and because of the small number of nonheterosexual participants, we analyzed only data from heterosexual participants. Thus, 2 (1%) participants who did not indicate their sexual orientation and 22 (11%) participants who identified as primarily bisexual or homosexual were excluded. We also excluded 8 (5%) additional participants because they did not write any sexual fantasies.

Participants provided demographic information about their age, ethnic background, education, and marital status. The mean age for men (29.5 years) did not differ significantly from the mean age for women (29.9 years), t(160) = -.37, p = .71. Most participants were White: the frequencies for men were 64 (75.3%) White, 3 (3.5%) African American, 7 (8.2%) Asian, 2 (2.4%) Latino, 4 (4.7%) mixed heritage, and 5 (5.9%) of other ethnic-racial background or uncodable; for women the frequencies were 55 (71.4%) White, 4 (5.2%) African American, 3 (3.9%) Asian, 5 (6.5%) Latina, 5 (6.5%) mixed heritage, and 5 (6.5%) of other ethnic-racial background or uncodable. The sample was highly educated. All participants had at least a high school education, and 124 (76.5%) had at least a bachelor's degree. About one third of the participants were fulltime students: 33 (38.8%) of the men and 27 (35.1%) of the women. Eighteen of the men (21.2%) and 19 of the women (24.7%) were currently married.

Procedure

The data reported here were obtained in conjunction with a larger study on power and sexuality (Zurbriggen, 2000). After an initial phone screening, participants were mailed packets containing open-ended sexual fantasy questions, attitude scales, and other measures not reported here. Participants completed these questionnaires in the privacy of their own homes.

Sexual Fantasy Measure

Participants were asked to write two of their favorite or most frequent sexual fantasies. Most participants (151) wrote two fantasies; 11 additional participants wrote only one fantasy. The instructions were as follows:

   Some daydreams are sexual in nature. In fact, surveys have shown
   that 95% of men and women say that they have had sexual daydreams
   or fantasies in one context or another. As with non-sexual
   daydreams, sexual fantasies can be about lots of different things,
   and having a fantasy doesn't necessarily mean that you would like
   it to come tree. It's just a story that you make up in your head to
   help you feel sensual and aroused.

   Think of a sexual fantasy that you may have recently had, and
   write about it in the space below. You can write about the first one
   that comes to mind, or, if you think of several, write about one
   that is common for you, or that is one of your favorites. Again, it
   is very important that you be as detailed as possible. What is the
   scene? What is your role? What are the series of events? Who is
   doing what to whom? What are you thinking and feeling? Please
   be as concrete and detailed as you can, while letting the fantasy
   flow along freely. Remember, you are writing anonymously, so
   feel free to write anything you like.

We developed a coding system to score fantasies for 11 different themes: power (dominance and submission), desire (experienced by the self or by another person in the fantasy), sexual pleasure (experienced by the self or by another person), emotional-romantic, explicit-sexual, interaction with one partner, interaction with multiple partners, and frame. In this article we focus on the themes of power, desire, and pleasure, but we include data concerning other fantasy themes for completeness.

We coded all themes as either present (1) or absent (0) in each fantasy. These codes were summed to produce the total score, which could range from 0 to 2, because participants wrote two fantasies. All fantasies were scored independently by the first and second authors, neither of whom had access to additional information about the participants while they were coding. We resolved disagreements through discussion. Interrater reliability was acceptable (kappa = .78, p < .001; kappa ranged from .59-.91 for individual themes). The observed proportion of overall agreement was .91 (range = .80.-96); proportion of agreement for positive ratings (theme present) was .91 (range = .82-.98); and proportion of agreement for negative rating (theme absent) was .87 (range = .77-.97).

Power: Dominance and Submission

We scored for dominance and submission whenever there was an indication of one person in the fantasy controlling another. We scored for dominance when the self exerted power over another person in the fantasy and for submission when another person in the fantasy exerted power over the self. We coded six subcategories of dominance and submission: verbal control (violent and nonviolent), physical control (violent and nonviolent), sadomasochism, and general (e.g., power differential by age or role, or general statements about power such as "I like the way it makes me feel--dominant and in control"). We scored a fantasy for power if one or more of the subcategory themes was present. Further information about subcategory scoring is presented in the Appendix.

Desire

We coded for desire when there was an indication of sexual arousal, showing that the sexual activities were wanted. Descriptions of sexual arousal, freely-chosen masturbation, or actions or words that encouraged the partner to continue a sexual activity indicated desire. Two examples are "As we're kissing, he backs me up against the door and presses into me touching me all over. I am overwhelmed and surprised but so turned on by the spontaneity of it all"; and "I'd be in the perfect outfit from Victoria's Secret. My husband would be so sexually aroused that he'd tear it off me." Desire was coded as "self" or "other" depending on which character in the story indicated his or her desire. Thus, the first example was scored as desire-self, whereas the second example was desire-other.

Sexual Pleasure

We coded for sexual pleasure when descriptions indicated feelings of sexual satisfaction or orgasm. We also included verbalizations or sounds indicating sexual pleasure in this category. We did not score sexual arousal alone as pleasure; some indication that the activities portrayed provided sexual gratification was necessary. Sexual pleasure was scored as "self" or "other" depending on which character experienced pleasure. An example of sexual pleasure-other is "I would lick her breasts and she would be in total ecstasy"; an example of sexual pleasure-self is, "He goes down on me and brings me to climax as my body shudders uncontrollably."

Emotional-Romantic

Emotional-romantic referred to affection, positive emotions, or romantic story lines. We coded a story as romantic if it involved references to the setting (such as a beach or a moonlit sky), a build-up preceding a sexual encounter (such as a romantic dinner or the existence of a committed relationship), or descriptions of the partner in terms of his or her emotional response and personal, not physical, characteristics. An example of this category is "My fiance takes me to a wonderful, intimate Italian restaurant and I have a glass of red wine and some calamari. We have a wonderful meal with a delicious dessert like tiramisu. Later we go back to our waterfront hotel and make love."

Explicit-Sexual

Explicit-sexual referred to detailed imagery describing explicit sexual acts or anatomical detail. Anatomical detail included detailed descriptions of specific body parts, either one's own or a partner's. We scored the mention of generally accepted sexualized body parts (nipples, penis), sexual fluids (semen or vaginal secretions), or a nonsexual body part that was described in further detail as explicit-sexual. Examples of this category include "My lover is a beautiful woman with almost silvery blond hair, long and wavy, green eyes, and a straight distinctive nose, and sensuously shaped lips"; and "My wife is rubbing my legs and penis. The other woman is rubbing my chest. I am rubbing her breasts and stomach."

Interaction With One Partner and Interaction With Multiple Partners

When a participant wrote a fantasy involving interactions between him or herself and only one participating partner, we scored it as interaction with one partner. We scored for interaction with multiple partners when a participant wrote of interactions between him or herself and two or more partners.

Frame

We coded for frame when the words "I fantasize" or "I imagine" or "I dream" occurred in the fantasy. The category of frame indicated that the author qualified what was written, specifically stating that it was imagined and not real. An example of a frame is "I fantasize about sleeping with a co-worker."

Attitude Measures

Rape Myth Acceptance

The 19-item Rape Myth Acceptance Scale (RMA; Burt, 1980) measures stereotypical and false beliefs about rape. An example item is "Any healthy woman can successfully resist a rapist if she really wants to." Coefficient alpha was .80 for men and .80 for women.

Attitudes Toward Women

We used a 15-item version of the Attitudes Toward Women Scale (AWS; Spence, Helmreich, & Stapp, 1973) to assess traditional beliefs about the place of women in society. A sample item is "Women should worry less about their rights and more about becoming good wives and mothers." Higher scores imply more liberal, egalitarian attitudes about women's rights. Coefficient alpha for this scale was .78 for men and .69 for women.

Adversarial Sexual Beliefs

The 9-item Adversarial Sexual Beliefs Scale (ASB; Burt, 1980) assesses the degree to which participants believe the two sexes are hostile toward one another. An example item is "In a dating relationship a woman is largely out to take advantage of a man." Coefficient alpha for this scale was .71 for men and .74 for women.

Social Desirability

We used 18 items from the 20-item impression management subscale of the Balanced Inventory of Desirable Responding (BIDR; Paulhus, 1991) as a measure of social desirability. One item ("I have done things that I don't tell other people about") was excluded because it did not correlate with the rest of the scale. A second item ("I never read sexy books or magazines") was dropped because the content overlapped with the central questions of the study. Coefficient alpha for this scale was .80 for men and .80 for women.

RESULTS

Because expressing a desire to dominate or submit to one's partner during sex is not socially accepted, we expected that participants who were concerned with maintaining a positive image would be less likely to report fantasies of power. As expected, social desirability was negatively associated with dominance in sexual fantasies, r = -.22, N = 162, p = .005. Submission was not correlated with social desirability, r = -.05, N = 162, p = .57. Accordingly, we conducted all correlational analyses reported here both with and without social desirability partialled out. In all cases, there were only small changes in the magnitude of the correlation coefficients and substantive conclusions were not affected. For simplicity, we report zero-order rather than partialled correlation coefficients for all analyses.

Attitudes

The means and standard deviations of attitude variables for men and women are displayed in Table 1. Men were more likely than women to endorse stereotypical and false beliefs about rape, t(160) = 4.85, p < .001. They also scored higher on the ASB, expressing the belief that the sexes are hostile toward one another, t(160) = 3.06, p = .003. Finally, men's lower scores on the AWS indicate that they held more traditional beliefs about the place of women in society, t(160) = -4.49, p < .001.

As has been found previously (Burt, 1980; Osland, Fitch, & Willis, 1996; Ryan & Kanjorski, 1998), we found that RMA, ASB, and AWS were highly intercorrelated for both men and women (see Table 2). Not surprisingly, people who believed rape myths were also likely to believe that relations between the sexes are adversarial and to hold conservative beliefs about women's place in society.

Sexual Fantasies

Gender Differences in Frequencies of Fantasy Themes

The means and standard deviations of fantasy themes for men and women are displayed in Table 1. Men's sexual fantasies were more sexually explicit than women's, t(160) = 4.11, p < .001, and women's sexual fantasies were more emotional and romantic than men's, t(160) =-2.69, p =.008. In addition, men's fantasies involved more interactions with multiple partners than did women's, t(160) = 1.97, p = .05.

There were also gender differences in fantasized dominance and submission. Men's fantasies included more portrayals of the self as dominant and in power than did women's, t(160) = 2.45, p = .02. Women's fantasies did not include significantly more portrayals of submission than did men's, t(160) = -1.62, p = .11. However, in a repeated measures ANOVA, the interaction between gender and type of power fantasy (dominance vs. submissive) was reliable: F(1,160) = 7.75, p = .006. Although men were equally likely to fantasize about dominance and submission, women were more likely to fantasize about submission.

Other interesting gender differences involved sexual desire and sexual pleasure. Men's fantasies mentioned a partner's sexual desire more frequently than did women's fantasies, t(160) = 3.09, p = .002. Although men's fantasies were equally likely to include desire-self and desire-other, paired t(84) = -.58, p = .57, women's fantasies were marginally more likely to include desire-self than desire-other, paired t(76) = 1.69, p = .10. Men also described their partners as experiencing sexual pleasure more frequently than did women, t(160) = 3.24, p = .001. Although men's fantasies were marginally more likely to include sexual pleasure-other than sexual pleasure-self, t(84) = -1.74, p = .09, women's fantasies were significantly more likely to include sexual pleasure-self than sexual pleasure-other, paired t(76) = 2.04, p = .04. In a repeated measures ANOVA, this interaction between gender and sexual pleasure (self vs. other) was reliable, F(1, 160) = 7.14, p = .008. In other words, men focused on theft own and on their partners' desire and pleasure, but women tended to focus on their own desire and pleasure rather than those of their partners.

Intercorrelations of Fantasy Themes

We computed Pearson correlation coefficients for all 11 fantasy themes, separately for men and women; these are reported in Table 3. We were most interested in correlations involving dominance, submission, pleasure, and desire; these findings are discussed below. Correlations involving other themes were conducted for exploratory purposes; for completeness, Table 3 includes the results of those analyses.

Dominance. For men, dominance was positively correlated with explicit sexual imagery (r = .31, p = .004), marginally correlated with sexual pleasure-self (r = .19, p = .08), and marginally negatively correlated with emotional-romantic imagery (r = -.20, p = .07). For example, one man wrote the following:

   Another frequent sexual fantasy is having anal sex with a woman.
   I did this a lot in reality with my ex-girlfriend and really enjoyed
   it. She let me pull her hair and asked me to be rough with her. I
   like the way it makes me feel; dominant and in control. Usually
   we're in a bed and both totally naked. Sometimes the woman is
   on "all fours" and sometimes I grab her breasts and squeeze them.
   Then I usually push her flat on her stomach and continue being
   rough until I climax deep inside of her.

As in the men's fantasies, dominance in women's fantasies was positively correlated with explicit sexual imagery (r = .29, p = .009), but in contrast to the men's fantasies, dominance in women's fantasies was positively correlated with sexual pleasure-other (r = .31, p = .006), rather than being correlated with sexual pleasure-self. In other words, women who fantasized about being sexually dominant tended to describe their partners' sexual pleasure rather than their own. For example, one woman's fantasy was as follows:

   I would love to drive to the factory where my husband is an
   engineer, walk to his office, pull his clothing off, push him onto
   the top of his desk and make love to him. I know he would be
   surprised and pleased. A variation on this theme is that I would go
   to his office, hide under his desk and when he returns perform oral
   sex with him at his desk. Again he would be surprised and
   pleased.

There was one additional gender difference in the relationship between dominance and other fantasy themes. For men, dominance was positively correlated with interaction with one partner (r = .21, p = .05). In other words, when men fantasized about their own dominance, they tended to do so in the context of a dyadic interaction. For women, however, dominance was marginally positively correlated with interaction with multiple partners (r = .19, p = .10). So when women fantasized about their dominance, they tended to do so when fantasizing about many partners. The following examples highlight this contrast.

Male:

   This concerns a woman who is swimming with me in a bikini. I
   look appreciatively at her body and she half-blushes and
   half-beckons me. I swim up to her and kiss her and fondle her. We
   then climb out of the water and then I notice that her wet clothes
   are transparent. I begin fondling her and she requests me to stop.
   But her request only enhances my desire for fondling her more
   vigorously. This continues till we reach the car by which time I am
   so aroused that I strip her and myself and push her onto the
   backseat of the car and make passionate love to her.

Female:

   Two guys paying all their attention to me, waiting on me hand and
   foot, doing whatever I wanted them to do to me. I would make
   them take off my clothes, smother me in lotion, then a full body
   massage. Making sure my whole body is tingling they would take
   their turns entering me. I wouldn't be satisfied until both of them
   have cum and then I would do the same. We would all sleep
   together in my bed, naked, listening to Miles Davis as the summer
   heat made us sweat.

Submission. Submission in men was not significantly correlated with any other fantasy themes. However, submission in women was correlated with explicit sexual imagery (r = .23, p = .04) and marginally correlated with interaction with multiple partners (r = .21, p = .06). Women's fantasies of submission (as well as dominance as discussed above) tended to occur in the context of threesomes or group sex. One woman wrote the following:

   I often fantasize about being a submissive to a couple. In the
   fantasy I live with the couple or am hired out to the couple to do
   whatever they would like me to do. Often there is bondage and
   most often I am serving them orally. While I act as though I am
   humiliated and resistant, I am really enjoying it.

Sexual desire and sexual pleasure. The pattern of intercorrelations among the two sexual pleasure variables (self and other) and the two sexual desire variables (self and other) differed by gender. For women, desire-self and desire-other were correlated (r = .38, p = .001), as were sexual pleasure-self and sexual pleasure-other (r = .48, p < .001); however, three of the other four intercorrelations were below .15 (and not statistically reliable). For men, in contrast, all six of the intercorrelations among sexual pleasure (self and other) and desire (self and other) were greater than .20, and all were statistically reliable. For men, the average of these six correlations was .43, whereas for women the average correlation was .24. Thus, it appears that the constructs of sexual pleasure and desire were more closely linked for men than for women. A one-factor structure might be adequate to explain the intercorrelations among these four variables in men. For women, however, a two-factor structure (sexual pleasure and desire) may be required.

To test the hypothesis of a gender difference in factor structure, we performed a series of confirmatory factor analyses. In the first analysis, we tested the two-factor stacked model depicted in Figure 1. In this model, two latent factors are hypothesized. The sexual pleasure factor is comprised of two indicators: pleasure-self and pleasure-other. The desire factor is also comprised of two indicators: desire-self and desire-other. The two factors were allowed to correlate, and all parameters were estimated separately across gender. This model provided an excellent fit to the data: [chi square](2) = 4.36, p =. 11; NFI = .96; CFI = .98; IFI = .98; GFI = .99; standardized RMR = .04. The estimated correlation between the two factors was high (r = .78) for the men, but low (r = .28) for the women.

[FIGURE 1 OMITTED]

We conducted a second analysis in which the correlation between the sexual pleasure and desire factors was constrained to be equal across gender. The significance test of the change in [chi square] between the two models indicated that fit was reliably worse in the second model ([chi square] = 4.05, df = 1, p < .05). We therefore concluded that the correlation between the factors was higher in men than it was in women. The size of this correlation (r = .78) for the men suggested that a one-factor model might be appropriate for them. Accordingly, we ran a separate CFA analysis in which all four indicators (sexual pleasure-self, sexual pleasure-other, desire-self, and desire-other) were modeled as being caused by a single latent factor. For men, this model provided an excellent fit to the data: [chi square] (2) = 6.79, p = .03; NFI = .92; CFI = .94; IFI = .94; GFI = .96; standardized RMR = .05. For women, however, the one-factor model was rejected because of inadequate fit: [chi square] (2) = 10.90, p = .004; NFI = .71; CFI = .72; IFI = .75; GFI = .94; standardized RMR = .11.

From these analyses, we concluded that a one-factor structure that comprises sexual desire and sexual pleasure for both self and other was present in the sexual fantasies of the men. For women, however, a two-factor structure was present. Whereas for men desire and pleasure were in some sense synonymous, for women the two were only moderately correlated. At least in their fantasies, then, women seem to have a more nuanced or differentiated sexuality, in which desire and pleasure need not co-occur.

Other correlations. For both women and men, moderate correlations between explicit-sexual and both desire and sexual pleasure were present. In part, however, these correlations reflect an overlap in scoring criteria (i.e., several of the criteria for scoring pleasure and desire necessitated the use of explicit language). Most other correlations were not statistically reliable. For both men and women, though, frame was negatively correlated with at least one theme of pleasure or desire, suggesting that a need to distance oneself from one's fantasies might interfere with some of the more positive aspects of sexuality.

Relationships Between Sexual Fantasies and Attitudes

One of the most important questions about sexual fantasy concerns its connections with real-world behavior. This question is particularly important when one considers fantasies of dominance, including raping or otherwise overpowering someone. Because attitudes such as rape myth acceptance have consistently been found to correlate with measures of aggressive sexual behavior (Dean & Malamuth, 1997; Osland et al., 1996; Spence, Losoff, & Robbins, 1991; Truman, Tokar, & Fischer, 1996), investigating the relationship between fantasies of dominance and relevant attitudes may help us to understand the links, if any, between these fantasies and aggressive behavior.

Correlations between dominance and submission in fantasies and sexual attitudes are presented in Table 2. Women's fantasies of dominance and submission were uncorrelated with the sexual attitudes measured. For men, however, fantasies of dominance were positively correlated with rape myth acceptance (r = .22, N = 85, p = .05). This correlation with an attitudinal measure that reflects acceptance of stereotypical and false beliefs about rape (including victim blaming) suggests that men's fantasies of dominance may be problematic, to the other hand, men's fantasies of submission were marginally negatively correlated with rape myth acceptance (r = -.20, N = 85, p = .07). Thus, not all fantasies of force were related to rape myth acceptance; it was specifically men's fantasies of their own dominance that were related to this attitude scale.

The overall pattern of correlations between fantasies of dominance and the three attitudinal variables is also instructive. These correlations are graphed in Figure 2. Although only the correlation (for men) between RMA and dominance reached the standard level of statistical significance, it appears that the relationship between fantasies of dominance and these important attitudinal variables was different for men than for women. We used Fisher's z transformation to test for gender differences in the magnitude of these correlations. We found a significant difference for RMA (z = 2.18, p = .01) and a marginally significant difference for AWS (z = -1.40, p = .08). The difference for ASB was in the same direction, but not statistically significant (z = -1.03, p = .15). Combining across these three contrasts yields a composite significance level that is highly reliable (p = .0002). Overall, then, the data suggest that dominance in fantasies was related to harmful attitudes about women and about sexuality for men but not for women, lf anything, dominance in women's fantasies may be related to more positive attitudes (i.e., less acceptance of rape myths and more egalitarian attitudes about women's place in society).

[FIGURE 2 OMITTED]

Although our main interest was in examining the link between sexual fantasies of power and attitudes related to sexual violence, for exploratory purposes we also computed correlations between these attitude measures and the other nine fantasy themes. Most correlations were not statistically significant. However, for men there was a marginally significant negative correlation between desire-other and RMA (r = -.18, p = .09). Men who fantasized about the desire of their partner were less likely to endorse rape myths.

For women, three fantasy themes were significantly correlated with RMA: emotional/romantic (r = .24, p = .03), interaction with one partner (r = .23, p = .04), and interaction with multiple partners (r = -.31, p = .007). Interaction with multiple partners was also correlated with AWS, r = .27, p = .02. In other words, women with more stereotypically feminine fantasies had less liberal attitudes about women and expressed more belief in victim-blaming rape myths.

DISCUSSION

We had several goals in conducting this study. We wanted to investigate dominance and submission in men's and women's fantasies in a more detailed fashion. We were especially interested in whether there were indications that such fantasies might be dangerous or problematic, and whether this was true for women as well as for men. We also wanted to explore the place of desire and pleasure in sexual fantasies, particularly given that feminist scholars (e.g., Vance, 1989) have argued that it can be dangerous for women to express desire and seek pleasure in their real lives. In addition, the coding system that we developed allowed us to look at other aspects of sexual fantasy, including explicitness and emotional or romantic content.

Replication of Previous Findings

As predicted, we replicated previous findings concerning gender differences in attitudes and fantasy themes. We found that men were more accepting of rape myths, were more likely to believe that relations between men and women are adversarial, and were more conservative in their attitudes about women's place in society. In addition, men's fantasies were more sexually explicit and were more likely to include multiple sexual partners; women's, on the other hand, were more emotional-romantic and more likely to include a single sexual partner. There was also a gender difference in fantasies of dominance and submission, with men fantasizing more about dominance and women fantasizing more about submission.

We also predicted a particular pattern of intercorrelations among the attitude variables, for both men and women. As expected, a belief in rape myths was positively correlated with the belief that relations between men and women are adversarial, and both of these variables were negatively correlated with egalitarian beliefs about women's place in society.

Gender Differences in Desire and Pleasure

Intriguing gender differences concerning desire and pleasure in sexual fantasies emerged from our analyses. In their fantasies, men mentioned the sexual desire and sexual pleasure of their partners as well as themselves; however, women tended to mention only their own desire and pleasure and not the desire and pleasure of their partners. lf Vance (1989) is correct that it is dangerous for women to express desire and pleasure in the real world, it would make sense that these would be important themes within the safe sphere of fantasy. In addition, male desire and pleasure are often privileged in real-world heterosexual interactions, such that women are encouraged to focus on their partner's desire and pleasure rather than on their own (Byers, 1996). It makes sense, then, that in the realm of fantasy women might choose to emphasize their own needs rather than those of their male partners.

Desire and pleasure pose different problems for heterosexual men. In the real world, it may be difficult for a man to know when his female partner experiences desire and pleasure. Compared to the more obvious physical displays associated with male arousal and orgasm, there may be no external indication of desire or pleasure for a woman (or the man may worry that the external indicators are not veridical, that the woman is "faking it"). In addition (as discussed above) a woman may be reluctant to communicate the full extent of her desire or pleasure because of the risks of being misunderstood or harshly judged. Thus, part of the challenge of sexual relations for many men is in correctly interpreting their partners' changing levels of desire and pleasure. Performance-focused sexual scripts that charge the man with the responsibility for orchestrating pleasure and orgasm for both partners exacerbate this challenge. To be a good lover, he must incite desire and arousal, and then bring his partner to a state of pleasure and orgasm; however, he may be limited in the information available to him as to whether he is succeeding at these tasks. Is it any wonder, then, that the women that inhabit men's fantasies give clear and unambiguous indications of their sexual desire, arousal, pleasure, and eventual orgasm? In the world of fantasy, men find reassurance that their partners desire them and that they are good lovers.

The other major gender difference related to desire and pleasure concerned the interrelationships among these two themes. For men, desire and pleasure were strongly correlated. When men wrote about desire (of the self or of the partner) they also wrote about pleasure. For women, however, desire and pleasure comprised two distinct factors. One could be present without the other. Although fantasy is clearly a space where imagination can take us beyond our past experiences, we suspect that, in this case, the differing models of pleasure and desire (revealed in fantasy) might spring from differences in lived experiences (of heterosexual intercourse). One indication of sexual pleasure (both in the real world and in our coding system) is experiencing orgasm. Although men certainly have no guarantee of having an orgasm every time they feel sexual desire, it may be that, once a sexual encounter actually begins, the man's orgasm is a fairly reliable outcome. For women, in contrast, heterosexual intercourse is less likely to result in an orgasm (Haavio-Mannila & Kontula, 1997; Morokoff, 2000). Thus, women have more real-world experiences of feeling desire without the (eventual) intense pleasure of orgasm, and this is perhaps reflected in their fantasies, with desire sometimes being mentioned without a concurrent mention of sexual pleasure.

Fantasies of Dominance: Problematic for Men but Neutral or Beneficial for Women

There were a variety of indications in our data that fantasies of dominance might be problematic for men. Most importantly, these fantasies were correlated with scores on the RMA measure. Men who fantasized about dominance were more likely to believe that women provoke rape by their dress or behavior, that women lie about sexual assault, and that some women deserve to be raped. Given the consistent finding that belief in rape myths is associated with perpetration of sexual aggression (Dean & Malamuth, 1997; Osland et al., 1996; Spence et al., 1991; Truman et al., 1996), this result suggests that even for "normal" men (i.e., men with no known criminal history or clinical diagnoses), fantasies of dominance might lead to sexually aggressive behavior. Of course, because our study was correlational, we cannot rule out the possibility that committing aggressive acts or holding victim-blaming attitudes about sexual aggression leads one to fantasize about dominance. We also cannot rule out "third variable" explanations (e.g., that previous victimization causes both aggressive behavior and fantasies of dominance). Further research is needed to explore these causal hypotheses.

In contrast to these findings for men, women's fantasies of dominance did not correlate with problematic attitudes about gender and rape. If anything, fantasy dominance for women was associated with less victim-blaming of sexual assault survivors and more liberal and egalitarian attitudes about women. It may be that women who hold more progressive attitudes concerning sexuality and gender roles feel more freedom to experiment in their fantasies with activities that contradict traditional gender roles. Conversely, the experience of fantasizing about breaking away from the submissive role prescribed by the traditional sexual script may lead to more progressive attitudes.

Our analysis of the fantasy themes that co-occurred with the theme of dominance also suggested that men's dominance fantasies may be qualitatively different from women's fantasies of dominance. For men, dominance fantasies tended to be sexually explicit, to not mention romance or emotion, to involve just one partner, and to include references to the man's own pleasure. For women, dominance fantasies also tended to be sexually explicit; however, they were likely to involve multiple partners and to focus on the partners' sexual pleasure. Men's dominance fantasies thus tended to be more narcissistic and objectifying; women's, on the other hand, were more likely to specifically incorporate concern for the partner. Whereas empathy and concern could conceivably inhibit aggressive acts, objectifying the other person would likely remove inhibitions against acting aggressively. Thus, the nature of men's fantasies of dominance might put them at greater risk of committing aggressive acts rather than merely fantasizing about them. Further research on this question would be valuable.

Fantasies of Submission: Relatively Benign

In contrast to fantasies of dominance, fantasies of submission appeared relatively benign. If anything, such fantasies might be protective against rape myth acceptance in men, and do not appear to contribute to a victim mentality in women.

Hariton (1973) has argued that women's fantasies of submission are really disguised fantasies about female power. According to this argument, most of these fantasies feature a male character who is so overwhelmed with sexual desire for the female protagonist that he is unable to stop himself from taking her, by force if necessary. There is a sense in which the woman holds the power, with the man helpless against her attractiveness and desirability. One would expect, then, that female fantasies of submission would include detailed descriptions of the man's sexual desire. Contradicting this hypothesis, we found no correlation between submissiveness and desire-other in women's fantasies. Our results therefore do not support this account of female submission fantasies.

Generalizability

Although our sample was relatively diverse along some dimensions (e.g., gender, occupation, marital status, age), it was more homogenous along others. In particular, all the results reported here are from heterosexual participants, and most of the sample was White and relatively well-educated. It is important, therefore, to consider whether our results are likely to generalize to other populations.

Most published studies of sexual fantasies have included only heterosexual participants. The few exceptions (e.g., Masters & Johnson, 1979; Price, Allensworth, & Hillman, 1985) generally found that fantasies of homosexual participants were quite similar in thematic content to the fantasies of heterosexual participants. This suggests that our findings from heterosexuals might generalize to people with other sexual orientations; however, more research must be conducted to confirm this speculation. Research using open-ended rather than checklist methods would be especially valuable.

Even less is known about the effects on sexual fantasy of ethnicity, education, or social class (Leitenberg & Henning, 1995). However, Laumann, Gagnon, Michael, and Michaels (1994) found that as education increased, so did the level of experience with performing and receiving oral sex; this was true for both men and women. Similarly, Mahay, Laumann, and Michaels (2001) reported that men and women with more education reported more interest in giving and receiving oral sex. If these differences are also seen in sexual fantasizing, our participants may have had fantasies that were more varied and less conventional than the fantasies we would have obtained from a representative sample. Even if this is true, however, it is not immediately apparent that the correlations between fantasy themes and other variables (e.g., sexual attitudes) would necessarily be different across education levels.

Concerning ethnicity, Mahay et al. (2001) found that Whites were, in general, more permissive than were African American or Hispanic participants (although there were some differences by gender). For example, Whites were less likely to say premarital sex is wrong, less likely to say religion shapes their sexual behavior, and more likely to find oral sex appealing. Again, this relative permissiveness may result in less conventional fantasies and (perhaps) different mean level scores on one or more sexual attitude variables. However, the pattern of correlations between fantasy themes and sexual attitudes would not necessarily differ by ethnicity.

One final point should be considered. Although the study was advertised as being about romantic relationships, the screening interview mentioned the sexual nature of some of the questions; therefore, research on the differences between people who do and who do not volunteer to participate in studies of sexuality is relevant as we consider the generalizability of our results. This research has generally found that people who volunteer to participate in studies about sexuality are more sex-positive (Plaud, Gaither, Hegstad, Rowan, & Devitt, 1999; Strassberg & Lowe, 1995), more unconventional and novelty-seeking (Bogaert, 1996; Wiederman, 1999), and have experience with a wider variety of sexual behaviors (Plaud et al., 1999; Wolchik, Braver & Jensen, 1985). These findings suggest that our participants probably had more frequent and more varied fantasies than would be seen in a representative sample. Once again, however, we see no obvious reason to expect that the correlations among fantasy themes and sexual attitudes would be different among our presumably less conventional and more experienced participants. Of course, further research will be needed to fully assess whether the findings we report here will generalize across individual differences such as ethnicity, education level, and sexual orientation.

Implications for Rape Prevention and Education Programs

The findings from our correlational analyses involving RMA and fantasy themes may have implications for rape prevention and education. Men who were less likely to believe victim-blaming myths about rape were more likely to mention their partners' sexual desire in their fantasies. Perhaps these men have learned to pay close attention to their partners' levels of sexual desire rather than merely assuming that desire is present. Encouraging men to think in a more detailed and concrete fashion about their partners' sexual feelings might, therefore, help to reduce the incidence of acquaintance rape. On the other hand, it is also possible that concern about the sexual desire of one's partner is just one example of a more general variable such as empathy or nurturance, and it is this more general aspect of personality that is protective against sexual aggression perpetration (Dean & Malamuth, 1997). Even if this is true, though, an intervention program that targeted sexual fantasies (e.g., by encouraging men to imagine the thoughts and feelings of their partners as well as of themselves) might lead to changes in a global measure of empathy.

As noted above, men's fantasies of submission were negatively related to RMA; men who were less likely to believe victim-blaming myths about rape were more likely to describe a fantasy in which their (female) partners took an active, powerful role. Perhaps encouraging men to see women as powerful agents of their own sexuality might help to reduce the incidence of rape. Men who fantasize about women who break from the traditional, submissive gender role and instead express their own power may be less likely to engage in acts that deny women's agency, such as rape. Of course, the precise nature of the relationship between submission fantasies and RMA cannot be determined from this study, and further research in this area is needed.

For women, belief in rape myths was associated with sexual fantasies that were more traditionally feminine, having an emotional and romantic content and including only one partner. This combination of belief and fantasy may be risky for women. Because traditionally feminine fantasies probably signal an acceptance of traditional sexual scripts that feature male agency and female passivity, women who have such fantasies may be less likely to behave assertively with men, thereby increasing their risk of sexual victimization (Byers, 1996). If sexual victimization occurs, a higher RMA score implies that the woman would be more likely to blame herself and not seek support services for herself or punishment for the perpetrator.

Conclusion

Content analysis of men's and women's sexual fantasies provided a rich picture of the roles played by power, desire, and pleasure. Pleasure and desire may serve different purposes in men's and women's fantasies, with men using fantasy to imagine female partners who are clearly aroused and easily pleased and women using fantasy as a safe space to focus on their own desire and pleasure. Although fantasies of submission were not associated with problematic attitudes for either gender, men's fantasies of dominance were linked with victim-blaming attitudes about rape. Further research will be necessary to determine the direction of causality of that link and to investigate possible links with aggressive behavior.

APPENDIX

Scoring Criteria for Subcategories of Power in Sexual Fantasies

Verbal Control

Verbal control involved spoken directions that were an attempt to control another person's actions and behaviors. Verbal control could be either violent or nonviolent. Violent verbal control was defined as using a threat of harm or an insult in verbal communication. An example of submission, verbal control-violent is "He told me to take my clothes off or else he would really hurt me," and an example of dominance, verbal control-nonviolent is "After a while I tell her to lie down and I get on top of her."

Physical Control

Physical control involved the use of physical means to control another person's actions and behaviors. Physical control could also be either violent or nonviolent; violent physical control involved actions with an intent to harm the partner. An example of submission, physical control-violent is "There is a woman. She is some kind of secretary or clerk. Some gangsters burst into her office & throw her down across a desk on her stomach while searching the room"; and an example of dominance, physical control-nonviolent is "I pick her up and take her to the couch."

Sadomasochism

Sadomasochism was scored when participants in the fantasy engaged in sadomasochistic activities, which require the consent of the partner, are generally scripted or well-planned out scenarios, and possibly involve role-play, particular fetish clothing, or additional paraphernalia (e.g., rope, handcuffs, paddles). An example of dominance, sadomasochism is "I ask him if he would mind if I tried a little bondage. He agrees immediately and I pull out the handcuffs I have bought just for the occasion." An example of submission, sadomasochism is "I will be their sex slave for all occasions."

Not Otherwise Specified

This final subcategory of both dominance and submission was used to capture aspects of power in the fantasy that were not included in the previous subcategories. Expressions of seduction in which there was a clear power differential indicated by an age difference (an older woman initiating sexual activity with a younger man), status or role difference (a supervisor seducing an employee), or knowledge difference (a more experienced partner teaching a less experienced partner about sex) were included in this subcategory. Also included were descriptions of exhibitionism or voyeurism in which those actions were intended to control the sexual response of another person. Finally, we included in this category general statements of being the powerful or the submissive partner. An example of dominance is "I like the way it makes me feel-dominant and in control." An example of submission is "He was a little aggressive. Not in a bad way but fun. He was just in control of the situation a little more than our equal status."

Table 1. Means and Standard Deviations for Attitudes and
Sexual Fantasy Themes

                                                Men            Women
                                             (N = 85)        (N = 77)

Variable                                     M      SD       M      SD

Attitudes
  Rape myth acceptance ***                 35.74   10.93   28.01   9.18
  Adversarial sexual beliefs **            27.48    7.83   23.75   7.68
  Attitudes toward women ***               86.84   10.61   93.77   8.83
  Social desirability                       4.56    3.59    5.34   3.73
Sexual fantasy theme
  Dominance *                               0.48    0.67    0.25   0.54
  Submission                                0.45    0.63    0.61   0.65
  Desire-self                               0.65    0.70    0.51   0.66
  Desire-other **                           0.69    0.74    0.38   0.54
  Sexual pleasure-self                      0.75    0.75    0.70   0.71
  Sexual pleasure-other **                  0.88    0.71    0.55   0.60
  Emotional-romantic **                     0.46    0.61    0.75   0.78
  Explicit-sexual ***                       1.54    0.63    1.09   0.76
  Interaction with one partner ([dagger])   1.52    0.65    1.69   0.49
  Interaction with multiple partners *      0.49    0.68    0.30   0.56
  Frame                                     1.12    0.88    0.94   0.86

Note. Scores for fantasy themes could range from 0-2. Scores for
attitude scales could range from 19-133 for rape myth acceptance, 9-63
for adversarial sexual beliefs, 15-105 for attitudes toward women
(higher scores represent more liberal attitudes about women), and 0-18
for social desirability.

([dagger]) p < .10. * p < .05. ** p <.01. *** p <.001.

Table 2. Intercorrelations of Fantasy Dominance, Fantasy Submission,
and Attitudes

Variable                            1          2                3

1. Dominance                       --         0.01            -0.13
2. Submission                    -0.12         --             -0.07
3. Rape myth acceptance           0.22 *     -0.20 ([dagger])   --
4. Adversarial sexual beliefs     0.15        0.00             0.53 ***
5. Attitudes toward women        -0.11        0.08            -0.60 ***
6. Social desirability           -0.23 *      0.05            -0.17

Variable                            4          5                6

1. Dominance                     -0.01        0.12            -0.17
2. Submission                    -0.06        0.00            -0.17
3. Rape myth acceptance           0.53 ***   -0.61 ***        -0.14
4. Adversarial sexual beliefs      --        -0.48 ***        -0.29 **
5. Attitudes toward women        -0.59 ***     --              0.22 *
6. Social desirability           -0.22 *      0.11              --

Note. For women, N = 77; for men, N = 85. Correlations for women are
shown above the diagonal; correlations for men are shown below.

([dagger]) p < .10. * p < .05. ** p <.01. *** p < .001.

Table 3. Intercorrelations of Fantasy Themes

Variable                               1               2         3

1. Dominance                          --              0.01     0.01
2. Submission                       -0.12              --     -0.06
3. Desire-self                       0.11             0.12      --
4. Desire-other                      0.13             0.14     0.45 ***
5. Sexual pleasure-self              0.19 ([dagger])  0.03     0.44 ***
6. Sexual pleasure-other             0.12            -0.07     0.25 *
7. Emotional-romantic               -0.20 ([dagger])  0.02     0.08
8. Explicit-sexual                   0.31 **         -0.11     0.22 *
9. Interaction one partner           0.22 *           0.13     0.04
10. Interaction multiple partners   -0.16             0.03     0.09
11. Frame                            0.02             0.01    -0.26 *

Variable                               4          5        6

1. Dominance                         0.13       0.06      0.31 **
2. Submission                        0.01       0.17      0.05
3. Desire-self                       0.38 ***   0.10      0.12
4. Desire-other                       --        0.09      0.25 *
5. Sexual pleasure-self              0.46 ***    --       0.48 ***
6. Sexual pleasure-other             0.43 ***   0.56 ***   --
7. Emotional-romantic                0.10       0.12      0.10
8. Explicit-sexual                   0.28 **    0.31 **   0.49 ***
9. Interaction one partner           0.11       0.05      0.11
10. Interaction multiple partners    0.02       0.15      0.07
11. Frame                           -0.26 *    -0.14     -0.19
                                                         ([dagger])

Variable                               7          8           9

1. Dominance                         -0.10       0.29 **    -0.05
2. Submission                        -0.14       0.23 *     -0.01
3. Desire-self                       -0.01       0.32 **    -0.07
4. Desire-other                      -0.03       0.27 *      0.10
5. Sexual pleasure-self               0.10       0.34 **     0.14
6. Sexual pleasure-other              0.01       0.50 ***    0.18
7. Emotional-romantic                  --       -0.16        0.45 ***
8. Explicit-sexual                   -0.13        --         0.01
9. Interaction one partner            0.24 *     0.27 *       --
10. Interaction multiple partners    -0.15       0.01       -0.67 ***
11. Frame                            -0.26 *     0.03        0.12

Variable                               10                11

1. Dominance                          0.19             -0.08
2. Submission                         0.21 ([dagger])  -0.05
3. Desire-self                        0.08             -0.08
4. Desire-other                       0.06             -0.31 **
5. Sexual pleasure-self               0.00             -0.03
6. Sexual pleasure-other              0.02             -0.19
7. Emotional-romantic                -0.40 ***          0.05
8. Explicit-sexual                    0.33 **          -0.15
9. Interaction one partner           -0.56 ***         -0.02
10. Interaction multiple partners      --              -0.01
11. Frame                            -0.12               --

Note. For women, N = 77; for men, N = 85. Correlations for women are
shown above the diagonal; correlations for men are shown below.

([dagger]) p < .10. * p <.05. ** p <.01. *** p < .001.

REFERENCES

Bargh, J. A., Raymond, P., Pryor, J. B., & Strack, F. (1995). Attractiveness of the underling: An automatic power [right arrow] sex association and its consequences for sexual harassment and aggression. Journal of Personality and Social Psychology, 68, 768-781.

Bogaert, A. F. (1996). Volunteer bias in human sexuality research: Evidence for sexuality and personality differences in males. Archives of Sexual Behavior, 25, 125-140.

Brown, J. J., & Hart, D. H. (1977). Correlates of females' sexual fantasies. Perceptual and Motor Skills, 45, 819-825.

Browne, A., & Finkelhor, D. (1986). Impact of child sexual abuse: A review of the research. Psychological Bulletin, 99, 66-77.

Burt, M. R. (1980). Cultural myths and support for rape. Journal of Personality and Social Psychology, 38, 217-230.

Byers, E. S. (1996). How well does the traditional sexual script explain sexual coercion? Review of a program of research. Journal of Psychology and Human Sexuality, 8, 7-25.

Crepault, C., & Couture, M. (1980). Men's erotic fantasies. Archives of Sexual Behavior, 9, 565-581.

Dean, K. E., & Malamuth, N. M. (1997). Characteristics of men who aggress sexually and of men who imagine aggressing: Risk and moderating variables. Journal of Personality and Social Psychology 72, 449-455.

Gold, S. R. (1991). History of child sexual abuse and adult sexual fantasies. Violence and Victims, 6, 75-82.

Gold, S. R., Balzano, B. F., & Stamey, R. (1991). Two studies of females' sexual force fantasies. Journal of Sex Education and Therapy, 17, 15-26.

Greendlinger, V., & Byrne, D. (1987). Coercive sexual fantasies of college men as predictors of self-reported likelihood to rape and overt sexual aggression. The Journal of Sex Research, 23, 1-11.

Gutek, B. A., & Koss, M. P. (1993). Changed women and changed organizations: Consequences of and coping with sexual harassment. Journal of Vocational Behavior, 42, 28-48.

Haavio-Mannila, E., & Kontula, O. (1997). Correlates of increased sexual satisfaction. Archives of Sexual Behavior, 26, 399-419.

Hariton, E. B. (1973). The sexual fantasies of women. Psychology Today, 6(10), 39-44.

Hariton, E. B., & Singer, J. L. (1974). Women's fantasies during sexual intercourse: Normative and theoretical implications. Journal of Consulting and Clinical Psychology, 42, 313-322.

Hsu, B., Kling, A., Kessler, C., Knapke, K., Diefenbach, P., & Elias, J. (1994). Gender differences in sexual fantasy and behavior in a college population: A ten-year replication. Journal of Sex and Marital Therapy, 20, 103-118.

Johnston, L., Ward, T., & Hudson, S. M. (1997). Deviant sexual thoughts: Mental control and the treatment of sexual offenders. The Journal of Sex Research, 34, 121-130.

Kinsey, A. C., Pomeroy, W. B., & Martin, C. E. (1948). Sexual behavior in the human male. Philadelphia: W. B. Saunders.

Kinsey, A. C., Pomeroy, W. B., Martin, C. E., & Gebhart, P. H. (1953). Sexual behavior in the human female. Philadelphia: W. B. Saunders.

Knafo, D., & Jaffe, Y. (1984). Sexual fantasizing in males and females. Journal of Research in Personality, 18, 451-462.

Langevin, R., Lang, R. A., & Curnoe, S. (1998). The prevalence of sex offenders with deviant fantasies. Journal of Interpersonal Violence, 13, 315-327.

Laumann, E. O., Gagnon, J. H., Michael, R. T., & Michaels, S. (1994). The social organization of sexuality: Sexual practices in the United States. Chicago: University of Chicago Press.

Leitenberg, H., & Henning, K. (1995). Sexual fantasy. Psychological Bulletin, 117, 469-496.

MacCulloch, M. J., Snowden, P. R., Wood, P. J. W., & Mills, H. E. (1983). Sadistic fantasy, sadistic behaviour and offending. British Journal of Psychiatry, 143, 20-29.

Mahay, J., Laumann, E. O., & Michaels, S. (2001). Race, gender, and class in sexual scripts. In E. O. Laumann & R. T. Michael (Eds.), Sex, love, and health in America: Private choices and public policies (pp. 197-238). Chicago: University of Chicago Press.

Masters, W. H., & Johnson, V. E. (1979). Homosexuality in perspective. Boston: Little, Brown.

Moreault, D., & Follingstad, D. R. (1978). Sexual fantasies of females as a function of sex guilt and experimental response cues. Journal of Consulting and Clinical Psychology, 46, 1385-1393.

Morokoff, P. J. (2000). A cultural context for sexual assertiveness in women. In C. B. Travis & J. W. White (Eds.), Sexuality, society, and feminism (pp. 299-319). Washington, D.C.: American Psychological Association.

Murnen, S. K., Wright, C., & Kaluzny, G. (2002). If "boys will be boys" then girls will be victims? A meta-analytic review of the research that relates masculine ideology to sexual aggression. Sex Roles, 46, 359-375.

Mussweiler, T., & Forster, J. (2000). The sex [right arrow] aggression link: A perception-behavior dissociation. Journal of Personality and Social Psychology, 79, 507-520.

Osland, J. A., Fitch, M., & Willis, E. E. (1996). Likelihood to rape in college males. Sex Roles, 35, 171-183.

Paulhus, D. R. (1991). Measurement and control of response bias. In J. P. Robinson, P. R. Shaver, & L. S. Wrightsman (Eds.), Measures of personality and social psychological attitudes (pp. 17-59). San Diego, CA: Academic Press.

Pelletier, L. A., & Herold, E. S. (1988). The relationship of age, sex guilt and sexual experience with female sexual fantasies. The Journal of Sex Research, 24, 250-256.

Plaud, J. J., Gaither, G. A., Hegstad, H. J., Rowan, L., & Devitt, M. K. (1999). Volunteer bias in human psychophysiological sexual arousal research: To whom do our research results apply? The Journal of Sex Research, 36, 171-179.

Prentky, R. A., Burgess, A. W., Rokous, F., Lee, A., Hartman, C., Ressler, R., et al. (1989). The presumptive role of fantasy in serial sexual homicide. American Journal of Psychiatry, 146, 887-891.

Price, J. H., Allensworth, D. D., & Hillman, K. S. (1985). Comparison of sexual fantasies of homosexuals and heterosexuals. Psychological Reports, 57, 871-877.

Pryor, J. B., & Stoller, L. M. (1994). Sexual cognition processes in men high in the likelihood to sexually harass. Personality and Social Psychology Bulletin, 20, 163-169.

Resick, P. A. (1993). The psychological impact of rape. Journal of Interpersonal Violence, 8, 223-255.

Rokach, A., Nutbrown, V., & Nexhipi, G. (1988). Content analysis of erotic imagery: Sex offenders and non-sex offenders. International Journal of Offender Therapy and Comparative Criminology, 32, 107-122.

Ryan, K. M., & Kanjorski, J. (1998). The enjoyment of sexist humor, rape attitudes, and relationship aggression in college students. Sex Roles, 38, 743-756.

Smeaton, G., & Byrne, D. (1987). The effects of R-rated violence and erotica, individual differences, and victim characteristics on acquaintance rape proclivity. Journal of Research in Personality, 21, 171-184.

Spence, J. T., Helmreich, R., & Stapp, J. (1973). A short version of the Attitudes toward Women Scale (AWS). Bulletin of the Psychonomic Society, 2, 219-220.

Spence, J. T., Losoff, M., & Robbins, A. S. (1991). Sexually aggressive tactics in dating relationships: Personality and attitudinal correlates. Journal of Social and Clinical Psychology, 10, 289-304.

Strassberg, D. S., & Lockerd, L. K. (1998). Force in women's sexual fantasies. Archives of Sexual Behavior, 27, 403-414.

Strassberg, D. S., & Lowe, K. (1995). Volunteer bias in sexuality research. Archives of Sexual Behavior, 24, 369-382.

Sue, D. (1979). Erotic fantasies of college students during coitus. The Journal of Sex Research, 15, 299-305.

Truman, D. M., Tokar, D. M., & Fischer, A. R. (1996). Dimensions of masculinity: Relations to date rape supportive attitudes and sexual aggression in dating situations. Journal of Counseling and Development, 74, 555-562.

Vance, C. V. (Ed.). (1989). Pleasure and danger: Exploring female sexuality. London: Pandora.

Wiederman, M. W. (1999). Volunteer bias in sexuality research using college student participants. The Journal of Sex Research, 36, 59-66.

Wolchik, S. A., Braver, S. L., & Jensen, K. (1985). Volunteer bias in erotica research: Effects of intrusiveness of measure and sexual background. Archives of Sexual Behavior, 14, 93-107.

Zurbriggen, E. L. (2000). Social motives and cognitive power-sex associations: Predictors of aggressive sexual behavior. Journal of Personality and Social Psychology, 78, 559-581.

Manuscript accepted February 16, 2004

Portions of this research were presented at the 2003 annual meeting of the Society for the Scientific Study of Sexuality, San Antonio. TX. Data collection was supported by NIMH National Research Service Award F31MH11468. Analysis of data and preparation of the manuscript were supported by a grant from the Social Sciences Division at the University of California, Santa Cruz. We thank Colin Leach and Larissa Tiedens for helpful comments about an earlier version of this manuscript.

Address correspondence to Eileen L. Zurbriggen, Department of Psychology, Room 277, Social Sciences 2, University of California, Santa Cruz, Santa Cruz, CA 95064; e-mail: [e-mail].