"By the Semi-Mystical Appearance of a Condom": How Young Women and Men Communicate Sexual Consent in Heterosexual Situations.

In the fall of 1990, students and administrators at Antioch College in Yellow Springs, Ohio, joined forces to develop a mutual sexual consent policy. It required, among other things, that all Antioch students obtain consent from their partners prior to engaging in any sexual contact and before proceeding to the next level of sexual intimacy, unless the sexual activity was mutually initiated. Consent was defined as "the act of willingly and verbally agreeing to engage in specific sexual contact or conduct" (Antioch College, 1990, p. 1).

When the national press brought this policy to the attention of the general public in the fall of 1993, it created an international controversy. The policy was discussed and critiqued on the front pages of newspapers ranging from the New York Times to the Bangkok Post, on every major U.S. television network, and even on the comedy show Saturday Night Live (Guskin, 1994). In a response to the public's reaction, Alan Guskin (1994), President of Antioch College, discussed the reasoning behind the mutual consent policy. The goal of the policy was to enable women and men to communicate freely about their sexual wishes in an open and honest manner. Ideally, this would reduce the incidence of sexual assault on campus. Reflecting on the international press coverage this policy received, Guskin (1994) stated,

   I believe it's not just sex that has created the reaction, but the Antioch
   requirement that students talk about sex! Talking about it with someone
   whom you desire; getting consent before having sex; having to think about
   sexual acts that you are about to do; communicating with a partner about
   your interests. (p. 2)

Sexual consent is an important issue. Consent is a central issue in defining rape both in research and in legal cases (rape is often defined as sex without consent; see Burt & Albin, 1981; Estrich, 1987; Muehlenhard, Powch, Phelps, & Giusti, 1992; Sanday, 1996). It has even been suggested that gender-based misunderstandings regarding consent can lead to rape (Abbey, 1991; see also Crawford, 1995). Unfortunately, there has been little research to help clarify how people perceive and communicate sexual consent.

Defining Sexual Consent

Sexual consent is defined differently by different people: Furthermore, the term is often used without being defined (Muehlenhard et al., 1992). There are several conceptual issues that should be considered.

One such issue is whether consent should be defined as a mental act--that is, the decision that one agrees to engage in sex--or a physical act--that is, the verbal or nonverbal expression of agreement (Muehlenhard, 1995/1996; Muehlenhard et al., 1992). Defining consent as a purely mental act is problematic. If consent is defined as a mental act, then one person can never truly know if another has consented. This allows for the possibility of misunderstandings, or claims of misunderstandings, that can lead to sexual assault (Abbey, 1982, 1987; Bart & O'Brien, 1985; Warshaw, 1994).

Defining consent as a physical act is also problematic. If consent is defined as the expression of agreement, this expression can be verbal or nonverbal. It has been argued that consent should be verbally expressed and that relying on nonverbal expressions of consent is unacceptable (Antioch College, 1990; Sanday, 1996). Nonverbal signals can be ambiguous, which creates the potential for miscommunication. Reliance on nonverbal signals allows one person to selectively interpret another's signals or to claim miscommunication as an excuse for rape. However, restricting consent to verbal expressions is problematic, for most sexual scripts do not involve explicitly giving verbal consent (Muehlenhard, 1995/1996; Muehlenhard et al., 1992). As Muehlenhard et al. (1992) observed, "if definitions of sexual assault were to include all sexual behavior that occurs without explicit verbal consent, they would include much behavior in which those involved consider themselves to be willing participants" (p. 31). Definitions of consent need to allow for the possibility that people use a variety of verbal and nonverbal behaviors to communicate consent.

Consent as a mental act and consent as a physical act do not always correspond with each other. For example, research indicates that some people engage in token resistance to sex, indicating no to sexual behavior while meaning yes. It is unclear how frequently this occurs. Some studies (e.g., Muehlenhard & Hollabaugh, 1988; O'Sullivan & Allgeier, 1994) suggest that a substantial minority of women and men have engaged in such behavior. A more recent study (Muehlenhard & Rodgers, 1998), however, suggests that many respondents who reported engaging in token resistance had misunderstood the questionnaire and, in fact, had not engaged in this behavior, although some had. Rape can result when a man misinterprets a woman's no to mean yes (e.g., see Bart & O'Brien, 1985, p. 10; Warshaw, 1984, p. 91). However, the idea that rape is frequently a result of men misunderstanding women's refusals has been challenged as an excuse allowing sexually aggressive men to avoid responsibility for their behavior (e.g., Crawford 1995). Nevertheless, the above studies suggest that feelings of willingness and expressions of willingness are distinct and should both be included in defining consent.

Some research and legal definitions of rape are based on the idea that nonconsent should be assumed until someone actively consents, whether verbally or nonverbally. Other definitions of rape, however, are based on the idea that consent should be assumed until someone refuses or resists--that being passive should be interpreted as consent (see Estrich, 1987; Muehlenhard et al., 1992, for a discussion). The latter conceptualization is problematic because there are many reasons why someone might not resist an unwanted sexual initiation (e.g., they might be confused, embarrassed, or afraid). If making no response is treated as consent, coercive sex could be viewed as consensual.

Finally, when conceptualizing consent, it is important to consider the context in which consent occurs. A person may be unable to give or withhold consent freely as a result of alcohol or drugs, threat of harm, economic coercion, or compulsory heterosexuality (Muehlenhard, 1995/1996; Muehlenhard & Schrag, 1991; Tong, 1984).

For the purposes of this paper, we defined consent to engage in sexual activity as the freely given verbal or nonverbal communication of a feeling of willingness. This definition conceptualizes consent as both a mental act (the feeling of willingness) and a physical act (the communication of this willingness) and takes context into account by requiring that consent be expressed freely.

Empirical Research

The vast majority of research on sexual communication has focused almost exclusively on sexual initiations and refusals (e.g., Byers & Heinlen, 1989; McCormick, 1979; O'Sullivan & Byers, 1992, 1993). This research is often framed using the theory of the "traditional sexual script," in which men are responsible for initiating and pursuing sexual encounters whereas women are responsible for restricting the level of sexual activity, serving as gate keepers  in sexual encounters (Gagnon & Simon, 1973; Peplau, Rubin, & Hill, 1977; Perper & Weis, 1987; Safilios-Rothschild, 1977; Sprecher & McKinney, 1993). Noticeably absent in the traditional sexual script is the concept of sexual consent.

We identified three empirical studies--one published and two unpublished--that specifically addressed sexual consent. In addition, there are other lines of research that, although not specifically addressing consent, provide data that inform the discussion of sexual consent.

Research on sexual consent. Byers (1980) provided participants with lists of behaviors thought to show agreement or refusal to sexual intercourse. They were asked to rank the three most important forms of communication a woman can use to show sexual consent. Byers found that the highest percentage of participants ranked "fondles male's genitals" as the most important way a woman could communicate consent to sexual intercourse (31% of women and 31% of men). A slightly smaller percentage of participants ranked "clear verbal consent" as the most important way for women to communicate consent to sexual intercourse (26% of women and 28% of men). The author did not assess whether gender differences were statistically significant, and this study examined only perceptions of women's consent, not perceptions of men's consent. Nevertheless, these findings suggest that both nonverbal and verbal signals are important in signaling consent.

Hall (1995) examined consent signals used by heterosexual female and male college students for a variety of sexual behaviors, ranging from kissing to anal intercourse. Of the participants who said that they had been in a situation in which they and their partner had engaged in wanted sexual activity, participants reported indicating it was okay to continue most frequently for penile-vaginal intercourse (78.8%) and anal intercourse (73.4%) and least frequently for the male's having an orgasm (35.1%). There were no significant gender differences in women's and men's reports of indicating consent for various behaviors. These findings suggest that in examining consent, it is important to specify what kind of sexual behavior is in question.

Hall (1995) also asked participants how they indicated yes to their partner for their most recent wanted sexual experience. Most participants reported that they indicated yes both verbally and nonverbally (60.9%), although some reported that they indicated yes only nonverbally (28.2%) or only verbally (11.0%). When participants were asked to report how they nonverbally indicated yes, 75.0% of women and 72.0% of men reported that they "kissed," 70.8% of women and 69.5% of men "got closer," 62.0% of women and 69.5% of men "intimately touched," and 35.4% of women and 33.9% of men "smiled." There were no significant gender differences on any of these variables. However, significantly more women than men indicated that they "hugged or caressed" (72.4% and 63.6%, respectively) and "did not move away" (58.9% and 47.5%) to indicate consent. Participants were not asked how they indicated yes verbally, so it is not clear how participants interpreted this item when responding to the questionnaire. Hall's study suggests that there are few gender differences in the use of consent signals, that nonverbal consent signals are more popular than verbal consent signals, that people often use a combination of consent signals, and that people consent at varying rates for different behaviors.

Burrow (1997) examined women's and men's perceptions of women's consent and nonconsent to sexual intercourse, similar to Byers (1980). In this study, participants read vignettes in which a man initiates sexual intercourse with a woman while they are on a date. Women were asked to rate which of 12 verbal and nonverbal behaviors would be likely to indicate their consent if they were the woman in that vignette. Men were asked to rate which of the behaviors would most likely signal the woman's consent. Parallel questions were asked in which the same 12 items stated in the negative (e.g., "She would/would not touch Rick in intimate places") were used to assess which behaviors represent nonconsent.

Burrow (1997) found gender differences in perceptions of sexual consent: Male participants rated verbal behaviors as significantly more likely to represent the woman's consent in the vignette than female participants rated the behaviors as representative of their own consent in the vignette. There were no gender differences in ratings of behaviors representing nonconsent. These results are complicated by some methodological concerns. First, female and male participants received different instructions in completing this questionnaire: Women were asked to imagine themselves in the scenario when rating behaviors, whereas men were asked to rate the behaviors as an outside observer, thereby confounding gender and response set. Second, it is unclear if all the women were able to imagine themselves as the woman in the scenario. Third, many of the questions regarding nonconsent included double or triple negatives, which may have been confusing for respondents (e.g., "She definitely would not 'not touch Rick in very intimate places' to show nonconsent"). Thus, Burrow's study suggests that there may be gender differences in interpreting whether a woman's behavior indicates consent, but many questions remain.

Research on related topics. Research on related topics can also inform the discussion of sexual consent. In a line of research designed to examine sexual initiations and refusals, Byers and colleagues examined "positive responses" to sexual initiations. Such responses include one aspect of our definition of consent--the expression of willingness--but do not address the feeling of willingness or the context in which the individual is free to give or withhold consent.

In Byers and Heinlen's (1989) study, female and male participants recorded all sexual interactions with their partners over the course of a week. Two trained raters then coded these descriptions. Of the participants who described an initiation in which the partner responded positively, 41.8% of the positive responses were nonverbal, 31.3% were verbal, and 26.9% were both verbal and nonverbal. Nonverbal responses included initiating the sexual activity and continuing the sexual interaction. Verbal responses included requesting clarification, saying yes, or making an invitation. Neither information regarding gender differences in these responses nor operational definitions of the responses were provided.

In a study by O'Sullivan and Byers (1992), female and male participants recorded aspects of their sexual interactions in dating situations over a two-week period, including responses to sexual initiations. Nonverbal responses were more common than verbal responses: 90.9% of the women and 82.9% of the men reported making a positive nonverbal response to an initiation. In contrast, 56.3% of the women and 49.9% of the men reported making a positive verbal response to an initiation. It is unclear whether there were significant gender differences in the use of these positive responses. Consistent with the studies specifically on consent, both nonverbal and verbal responses were used, with nonverbal responses predominating.

Another related area of research involves gender differences in perceptions of sexual interest and intent. This line of research relates to one aspect of our definition of consent--perceptions of someone's sexual interest or desire--but interest and desire are not the same as willingness (see Muehlenhard & Rodgers, 1998); furthermore, neither the expression of willingness nor the context are typically addressed in these studies.

Numerous studies using a variety of methodologies have found that, on average, men view persons and interactions more sexually than do women. Measures used to assess sexual perceptions include ratings of sexiness (e.g., Johnson, Stockdale, & Saal, 1991; Saal, Johnson, & Weber, 1989), seductiveness and promiscuity (e.g., Abbey, 1982; Abbey & Melby, 1986; Abbey, Cozarelli, McLaughlin, & Hamish, 1987), perceptions of willingness to engage in sex (e.g., Botswick & DeLucia, 1992; Muehlenhard, 1988), and perceptions of a woman's sexual interest in a man (e.g., Shotland & Craig, 1988). Given the overwhelming evidence that gender affects perceptions of sexual intent, it is likely that gender affects perceptions of sexual consent as well. The tendency of men to view interactions more sexually than women may lead men to interpret signals as more representative of consent than do women. If the theory of the traditional sexual script is accurate and men initiate sexual intercourse more often than women do, then men would interpret women's consent signals more often than women would interpret men's consent signals. Thus, men's perceptions or misperceptions of consent signals become more important than women's perceptions or misperceptions of consent signals.

A final topic useful to this discussion involves a schema for conceptualizing consent signals. In a review of the sexual initiation and refusal literature, McCormick (1987) categorized initiation and refusal signals as either verbal or nonverbal, and as either direct or indirect. The verbal/nonverbal categorization reflects a commonly understood distinction between communication approaches, one often used in sexual signal research (e.g., Byers & Heinlen, 1989; Byers & Lewis, 1988; Metts, Cupach, & Imahori, 1992; Perper & Weis, 1987). The direct/indirect categorization of sexual signals is based on research on more general influence strategies (Falbo, 1977; Falbo & Peplau, 1980). According to McCormick (1987), direct sexual influence signals are strategies in which influencees are aware of how power is being used on them, and indirect sexual influence signals are strategies that depend on the influencees' ignorance of how power is being used to influence them (McCormick, 1979, 1987). McCormick's definitions of direct and indirect include the use of power to influence another, inconsistent with the definition of consent used in this study. Moreover, a person cannot convey her or his willingness to have sexual intercourse without the other person's awareness, as the definition of indirect implies.

We therefore used McCormick's categories of verbal and nonverbal in our consent category system, but we reconceptualized the distinction between direct and indirect. Specifically, for our purposes we defined direct consent signals as signals that are straightforward and unambiguous and indirect consent signals as signals that are ambiguous.

The Present Study

The existing literature on consent provides us with important information about consent, but many unanswered questions remain. For example, we do not know whether there are significant gender differences in how consent is conveyed and inferred. We do not know what kinds of signals are perceived as most indicative of sexual consent or if consent is situationally specific, such that the meaning of a signal varies depending on the nature of the relationship, the partner's initiation strategy, and so forth. Data suggest that nonverbal consent signals are used more often than verbal consent signals, but our information about these consent signals--especially about verbal signals--is limited. Finally, the information we have regarding sexual consent provides us with lists of alternately vague catchall terms (e.g., "verbal consent") without explanation of what those terms mean and specific behaviors (e.g., "smiling"), with no organizational schema to aid our understanding of consent.

The present study was an exploratory investigation of how young women and men convey and infer sexual consent. Female and male participants were asked to imagine themselves in scenarios in which sexual intercourse was initiated either verbally or nonverbally by either themselves or their date, along with a list of possible responses someone could make to such initiations. First, participants were asked to indicate whether they could imagine themselves in each scenario. This information was important because it would decrease the validity of the results to analyze the responses of participants who could not imagine themselves in the situations. Given the traditional sexual script, it could be that many women are unable to imagine themselves initiating sex and that many men are unable to imagine their dates initiating sex. Also, many people may be uncomfortable talking about sex (consider the controversy regarding the Antioch College verbal consent policy and the predominance of nonverbal over verbal responses found by Byers and Heinlen, 1989; Hall, 1995; and O'Sullivan and Byers, 1992); thus, some individuals may be unable to imagine themselves or their dates making verbal sexual initiations.

Next, participants were asked to rate how indicative each response would be of their own sexual consent as a participant in a hypothetical scenario (hypothetical self-consent ratings) and how indicative each response would be of their hypothetical date's sexual consent (hypothetical date-consent ratings). They were also asked to rate how frequently they use each response to signal their sexual consent in actual situations (actual self-consent ratings). We addressed the following research questions:

   1. Does the ability to imagine oneself in the scenarios vary as a function
   of participant gender and type of initiation (verbal or nonverbal) in
   self-initiation and date-initiation scenarios?

   2. What consent signals do women and men report using the most? The least?
   That is, which actual self-consent ratings were highest and lowest?

   3. Are there gender differences in how often women and men report using
   consent signals in actual situations, similar to the differences found by
   Hall (1995)? In other words, are there gender differences in actual
   self-consent ratings?

   4. Do people actually use the signals they rate as indicative of their
   consent in the scenarios? That is, do actual self-consent ratings correlate
   with hypothetical self-consent ratings?

   5. Hall's (1995) study suggested that there were differences in how consent
   was conveyed based on the type of sexual behavior participants were
   consenting to. Similarly, does the type of initiation (verbal or nonverbal)
   affect ratings of how indicative the signals are of consent in the
   scenarios, providing evidence that consent is situationally specific? That
   is, does the type of initiation in the scenario affect hypothetical
   self-consent and hypothetical date-consent ratings?

   6. Numerous studies have found that men tend to view themselves, other men,
   and women in a more sexualized fashion than do women (e.g., Abbey, 1982;
   Muehlenhard, 1988). Does participant gender affect participants' ratings of
   how indicative signals are of their own consent and their date's consent in
   the scenarios? In other words, are there gender differences in hypothetical
   self-consent and hypothetical date-consent ratings?

   7. Are there discrepancies between women's and men's ratings that would
   suggest gender-related miscommunication, lending support to the idea that
   gender-based miscommunication can lead to rape? That is, do women's
   hypothetical date-consent ratings differ significantly from men's
   hypothetical self-consent ratings, and do men's hypothetical date-consent
   ratings differ significantly from women's hypothetical self-consent
   ratings?

METHOD

Pilot Study

Purpose. The purpose of the pilot study was to gather information regarding how women and men communicate sexual consent both verbally and nonverbally. The responses to the pilot questionnaire were used to create items for the questionnaire used in the second phase of the study.

Participants. The pilot participants were 39 female and 28 male introductory psychology students from the University of Kansas. The demographics of the participants were as follows: 88.1% European American, 4.5% Asian American, 3.0% African American, 1.5% Hispanic American, and 3.0% international students. Their mean age was 19. Their participation was solicited through posted sign-up sheets that did not mention the topic of the study. Each respondent received credit toward a course research requirement.

Questionnaire. Participants were asked whether they had ever had sexual intercourse. If they had not, they were asked to answer the questionnaire the way they thought they would if they had experienced sexual intercourse; thus, everyone answered the questions regardless of their sexual experience, protecting their privacy during data-gathering sessions. Participants who had never had sexual intercourse were eliminated from the analyses, leaving a total of 22 women and 22 men.

Participants were asked a series of open-ended questions regarding how they and their most recent sexual partner communicated consent for sexual intercourse. For example, participants were asked the following question:

   Have you ever been in a situation in which you definitely wanted to have
   sexual intercourse with someone, and they wanted to have sexual intercourse
   with you? If so: (a) How, if at all, did your partner communicate his or
   her consent verbally? (b) How, if at all, did your partner communicate his
   or her sexual consent nonverbally?

They were also asked how they communicated consent to their partner verbally and nonverbally. The intent was to identify the ways in which people communicate their sexual consent.

Procedure. The pilot questionnaires were administered in single-sex groups by a female researcher (the first author). After signing consent forms, participants completed the questionnaires anonymously, seated in alternate seats to protect their privacy. When they were finished, they were debriefed and given information about the purpose of the study and the phone numbers of the researchers in case they had questions.

Results. Participants' responses to the pilot questions regarding how they and their partners communicate consent to sexual intercourse were compiled into a list of behaviors. These lists were used to construct the questionnaire used in the second part of the study.

Main Study Participants

The participants for the second phase of the study were 214 female and 210 male introductory psychology students at the University of Kansas. Participation was solicited in the same manner described above, and respondents received research credit for participating. The demographics of the participants were as follows: 84.7% European American, 3.1% African American, 3.1% Asian American, 1.7% Hispanic American, 1.4% biracial, 0.7% Native American/Alaskan, 3.1% international students, and 2.4% other. Their mean age was 19. A total of 46 participants were dropped from the analyses for the following reasons: 29 failed accuracy checks (to be described later), and 9 had incomplete questionnaires. International students (n = 13) were dropped to maintain the focus on cultural norms in the United States. Because the focus was consent in heterosexual situations, the one participant who reported being homosexual was dropped. Participants older than 26 (n = 3) were excluded to allow for an examination of traditional-aged college students. The sum of those in the exclusionary categories is greater than 46 because several participants were in more than one category. This left a final sample of 378 participants (188 women and 190 men).

Measures

Cover sheet. Participants were asked to complete a set of demographic questions as well as questions regarding their sexual experience history. These included questions about their sexual orientation, whether they had ever engaged in sexual intercourse, and the number of partners with whom they had engaged in sexual intercourse. For the purposes of this study, sexual intercourse was defined exclusively as penile-vaginal intercourse. Participants were asked to use this definition and were reminded of it throughout the questionnaire to ensure that they were all thinking of the same kind of sexual behavior when answering the questions.

Consent questionnaire. There were parallel versions of this questionnaire: one depicting verbal heterosexual initiations and one depicting nonverbal heterosexual initiations. Each participant was randomly assigned to complete one of the two versions. Women and men received gender-appropriate forms. Each participant was asked to read and imagine being in two scenarios: two verbal initiation scenarios in the verbal version or two nonverbal initiation scenarios in the nonverbal version.

To obtain hypothetical date-consent ratings, participants were asked to imagine themselves making a sexual advance. Each participant read the following scenario with one of the two final sentences:

   Self-Initiation Scenario

   You are very attracted to your date and would like to have sexual
   intercourse with her/him. You have been out several times, but the two of
   you have not had sexual intercourse (penile-vaginal intercourse) together
   before. The two of you are finally alone in a private place. [Verbal
   version: ]You start to kiss her/him, and you decide to make a sexual
   advance by asking her/him directly, "Will you have sex with me?" [Nonverbal
   version:] You make a sexual advance by sitting close to her/him, kissing
   her/him, and then starting to undress her/him.

After reading the scenario, participants were asked to indicate whether they could imagine themselves in the scenario. If so, they were asked to rate whether each of 34 behaviors (see Table 1) would indicate that their date was giving sexual consent, using a scale ranging from 0 (does not show his/her consent to sexual intercourse) to 6 (definitely shows his/her consent to sexual intercourse). Most of the items on this list came from the pilot study, although a few (e.g., "She/He says `no'") were added so that not all the behaviors would be positive.

Table 1. Subscales and Factor Loadings Derived From Factor Analysis of Actual Self-Consent Ratings, Hypothetical Self-Consent Ratings, and Hypothetical Date-Consent Ratings.

Signal                                                   Loadings
                                                           for
                                                          Actual
                                                       Self-Consent

Direct verbal signals (.76; .78; .49)(a)
  You say, "I want to have sex with you"; She/He          .846
    says, "I want to have sex with you"
  You say, "I want you"; She/He says, "I want you"        .445(b)
  You say, "Yes"; She/He says, "Yes"                      .522
  You say, "I would like to sleep with you";              .774
    She/He says, "I would like to sleep with you"
  You say, "I consent to sexual intercourse";             .531
    She/He says, "I consent to sexual intercourse"

Direct nonverbal signal
  You don't say anything--you just start having
     intercourse with her/him;
    She/He doesn't say anything--she/he just starts       .682
     having intercourse with you

Indirect verbal signals (.69; .74; .83)(a)
  You ask if she/he has a condom;                         .886
    She/He asks if you have a condom
  You talk about the importance of using birth
     control if you do have sex;
    She/He talks about the importance of using            .418(e)
     birth control if you do have sex
  You suggest she/he get a condom out;                    .881
     She/He suggests you should get a condom out
  You talk about your positive feelings
     about having sex with her/him;
    She/He talks about her/his positive feelings          .057(e)
     about having sex with you

Indirect nonverbal signals (.91; .94; .94)(a)
  You rub, fondle, and touch her/him sexually;            .771
    She/He rubs, fondles, and touches you sexually
  You put your hands down her/his pants;                  .708
    She/He puts her/his hands down your pants
  You help her/him undress you;                           .749
    She/He helps you undress her/him
  You undress her/him; She/He undresses you               .815
  You hug and caress her/him;                             .543(g)
    She/He hugs and caresses you
  You touch and kiss her/him in return;                   .695
    She/He touches and kisses you in return
  You kiss her/him in return;                             .655
    She/He kisses you in return
  You get physically closer to her/him;                   .517(g)
    She/He gets physically closer to you
  You start having dry sex with her/him
     (humping with clothes on);
    She/He starts having dry sex with you                 .411
    (humping with clothes on)
  You smile; She/He smiles                                .564

Intoxication signals (.92; .94; .95)(a)
  You say, "I'm really drunk";                            .881
    She/He says, "I'm really drunk"
  You say, "I'm feeling a little drunk";                  .884
    She/He says, "I'm feeling a little drunk"

Direct refusal signal
  You say, "No"; She/He says, "No"                        .609

No response signals (.88; .89; .91)(a)
  You do not resist her/his sexual advances;              .869
    She/He does not resist your sexual advances
  You don't stop her/him from kissing you and
    touching you sexually;
    She/He doesn't stop you from kissing her/him          .830
    and touching him/her sexually
  You do not say no; She/He does not say no               .849
  You let her/him take your clothes off;                  .575
    She/He lets you take her/his clothes off

Signal                                                   Loadings
                                                            for
                                                       Hypothetical
                                                       Self-Consent

Direct verbal signals (.76; .78; .49)(a)
  You say, "I want to have sex with you"; She/He           .838
    says, "I want to have sex with you"
  You say, "I want you"; She/He says, "I want you"         .657
  You say, "Yes"; She/He says, "Yes"                       .729
  You say, "I would like to sleep with you";               .641
    She/He says, "I would like to sleep with you"
  You say, "I consent to sexual intercourse";              .617
    She/He says, "I consent to sexual intercourse"

Direct nonverbal signal
  You don't say anything--you just start having
     intercourse with her/him;
    She/He doesn't say anything--she/he just starts        .442
     having intercourse with you

Indirect verbal signals (.69; .74; .83)(a)
  You ask if she/he has a condom;                          .694
    She/He asks if you have a condom
  You talk about the importance of using birth
     control if you do have sex;
    She/He talks about the importance of using             .704
     birth control if you do have sex
  You suggest she/he get a condom out;                     .569
     She/He suggests you should get a condom out
  You talk about your positive feelings
     about having sex with her/him;
    She/He talks about her/his positive feelings           .604
     about having sex with you

Indirect nonverbal signals (.91; .94; .94)(a)
  You rub, fondle, and touch her/him sexually;             .865
    She/He rubs, fondles, and touches you sexually
  You put your hands down her/his pants;                   .865
    She/He puts her/his hands down your pants
  You help her/him undress you;                            .783
    She/He helps you undress her/him
  You undress her/him; She/He undresses you                .804
  You hug and caress her/him;                              .650
    She/He hugs and caresses you
  You touch and kiss her/him in return;                    .725
    She/He touches and kisses you in return
  You kiss her/him in return;                              .613
    She/He kisses you in return
  You get physically closer to her/him;                    .638
    She/He gets physically closer to you
  You start having dry sex with her/him
     (humping with clothes on);
    She/He starts having dry sex with you                  .694
    (humping with clothes on)
  You smile; She/He smiles                                 .507(h)

Intoxication signals (.92; .94; .95)(a)
  You say, "I'm really drunk";                             .868
    She/He says, "I'm really drunk"
  You say, "I'm feeling a little drunk";                   .877
    She/He says, "I'm feeling a little drunk"

Direct refusal signal
  You say, "No"; She/He says, "No"                        -.488(i)

No response signals (.88; .89; .91)(a)
  You do not resist her/his sexual advances;               .806
    She/He does not resist your sexual advances
  You don't stop her/him from kissing you and
    touching you sexually;
    She/He doesn't stop you from kissing her/him           .488(j)
    and touching him/her sexually
  You do not say no; She/He does not say no                .803
  You let her/him take your clothes off;                   .450(j)
    She/He lets you take her/his clothes off

Signal                                                   Loadings
                                                           for
                                                       Hypothetical
                                                       Date-Consent

Direct verbal signals (.76; .78; .49)(a)
  You say, "I want to have sex with you"; She/He          .776
    says, "I want to have sex with you"
  You say, "I want you"; She/He says, "I want you"        .387(c)
  You say, "Yes"; She/He says, "Yes"                      .165(c)
  You say, "I would like to sleep with you";              .758
    She/He says, "I would like to sleep with you"
  You say, "I consent to sexual intercourse";             .055(d)
    She/He says, "I consent to sexual intercourse"

Direct nonverbal signal
  You don't say anything--you just start having
     intercourse with her/him;
    She/He doesn't say anything--she/he just starts       .707
     having intercourse with you

Indirect verbal signals (.69; .74; .83)(a)
  You ask if she/he has a condom;                         .817
    She/He asks if you have a condom
  You talk about the importance of using birth
     control if you do have sex;
    She/He talks about the importance of using            .741
     birth control if you do have sex
  You suggest she/he get a condom out;                    .828
     She/He suggests you should get a condom out
  You talk about your positive feelings
     about having sex with her/him;
    She/He talks about her/his positive feelings          .573
     about having sex with you

Indirect nonverbal signals (.91; .94; .94)(a)
  You rub, fondle, and touch her/him sexually;            .802(f)
    She/He rubs, fondles, and touches you sexually
  You put your hands down her/his pants;                  .800(f)
    She/He puts her/his hands down your pants
  You help her/him undress you;                           .737(f)
    She/He helps you undress her/him
  You undress her/him; She/He undresses you               .704(f)
  You hug and caress her/him;                             .803(f)
    She/He hugs and caresses you
  You touch and kiss her/him in return;                   .717(f)
    She/He touches and kisses you in return
  You kiss her/him in return;                             .801(f)
    She/He kisses you in return
  You get physically closer to her/him;                   .766(f)
    She/He gets physically closer to you
  You start having dry sex with her/him
     (humping with clothes on);
    She/He starts having dry sex with you                 .695(f)
    (humping with clothes on)
  You smile; She/He smiles                                .681(f)

Intoxication signals (.92; .94; .95)(a)
  You say, "I'm really drunk";                            .876
    She/He says, "I'm really drunk"
  You say, "I'm feeling a little drunk";                  .868
    She/He says, "I'm feeling a little drunk"

Direct refusal signal
  You say, "No"; She/He says, "No"                        .657

No response signals (.88; .89; .91)(a)
  You do not resist her/his sexual advances;              .721(f)
    She/He does not resist your sexual advances
  You don't stop her/him from kissing you and
    touching you sexually;
    She/He doesn't stop you from kissing her/him          .779(f)
    and touching him/her sexually
  You do not say no; She/He does not say no               .679(f)
  You let her/him take your clothes off;                  .688(f)
    She/He lets you take her/his clothes off

Note. Factor loadings are presented only for the factor to which each item was ultimately assigned, For each item, the actual self-consent signal and hypothetical self-consent signal given in response to the date's initiation are presented to the left of the semicolon; the hypothetical date-consent signal given in response to the participants' initiation is presented to the right of the semicolon.

(a) These numbers represent the coefficient alphas for actual self-consent ratings (n = 285), hypothetical self-consent ratings (n = 352 or 353), and hypothetical date-consent ratings (n = 272), respectively.

(b) This Direct Verbal item loaded more highly with the Direct Refusal item (-.618).

(c) These two Direct Verbal items loaded more highly with the Indirect Nonverbal and No Response items (.423 and .464, respectively) and with the Direct Nonverbal item (.430 and .442, respectively),

(d) This Direct Verbal item loaded more highly on its own factor (.688).

(e) These two Indirect Verbal items loaded more highly on their own factor (.535 and .733, respectively),

(f) The Indirect Nonverbal items and No Response items loaded on the same factor.

(g) These two Indirect Nonverbal items loaded more highly with the No Response items (.564 and .634, respectively),

(h) This Indirect Nonverbal item loaded more highly with the Direct Nonverbal item (-.532).

(i) This Direct Refusal item loaded negatively with the Direct Verbal items.

(j) These two No Response items loaded more highly with the Indirect Nonverbal items (.642 and .699, respectively).

To obtain hypothetical self-consent ratings, participants were asked to imagine that their date had made either a verbal or nonverbal sexual advance. The order of the self-initiation and date-initiation scenarios was counterbalanced across participants. Each participant read the following scenario with one of the two final sentences:

   Date-Initiation Scenario

   You and your date have been out several times, but the two of you have not
   had sexual intercourse (penile-vaginal intercourse) together before. The
   two of you are finally alone in a private place. [Verbal version:] He/She
   starts to kiss you and then asks you directly, "Will you have sex with me?"
   [Nonverbal version:] He/She sits close to you, kisses you, and starts to
   undress you.

After reading the scenario, they were asked to indicate whether they were able to imagine themselves in the scenario. If so, they were asked to rate whether each of 34 behaviors, parallel to those in the previous section, would show that they were giving sexual consent in response to the described sexual advance on a scale ranging from 0 (does not show your consent to sexual intercourse) to 6 (definitely shows your consent to sexual intercourse).

For both self- and date-initiation scenarios, participants unable to imagine themselves in the scenario were asked to respond as outside observers; these participants were dropped from analyses related to the scenario they could not imagine (n = 106 for self-initiation scenarios and n = 24 for date-initiation scenarios). Accuracy checks were included to ensure that participants were reading each item rather than responding randomly. Participants were asked to respond to an item with a specific number (e.g., "he yawns--answer this question with an eight"). The first part of these questions (e.g., "he yawns") was included so the items would blend in with the other items; the second part instructed participants how to respond if they were reading carefully. Anyone failing an accuracy check was eliminated from the study (n = 29). One of the 34 consent behaviors was eliminated from analysis because of inconsistent wording across forms, leaving a total of 33 behaviors.

To obtain actual self-consent ratings, all participants were asked how frequently they engaged in each of the 34 behaviors in order to indicate sexual consent (as discussed, only 33 were analyzed). They were asked to rate each item on a scale ranging from 0 (never do this to show consent) to 6 (always do this to show consent). They were reminded that we were interested in how often they did the behavior to indicate sexual consent, not just how often they did each behavior. Participants who had never had penile-vaginal sexual intercourse were asked to imagine how they might consent; their responses were eliminated from analyses of this section.

Procedure

The questionnaires were administered in single-sex groups by the first author. Participants were seated in alternate desks to protect their privacy. After they read and signed consent forms, they completed the questionnaires anonymously. Afterwards, the researcher distributed and read aloud a debriefing script about the purpose of the study. They were also provided with the researchers' phone numbers in case they had questions.

RESULTS

Descriptive Data

Of the 378 participants in the final sample, 364 (96.3%) identified themselves as heterosexual, and 14 (3.7%) identified themselves as bisexual. Questions regarding their previous sexual experience revealed that 93, or 24.6%; 21.8% of the women and 27.4% of the men, [chi square] (1, N = 378) = 1.56, p = .210, of the participants had never engaged in penile-vaginal intercourse and 285 (75.4%) of the participants had engaged in penile-vaginal intercourse with a mean number of 5.16 partners. The modal number of partners was one.

Participants had indicated whether they were able to imagine themselves in the scenarios depicted. To address Research Question 1, whether ability to imagine oneself in these scenarios varied as a function of participant gender and type of initiation depicted (verbal or nonverbal), two logistic regressions were conducted. One was performed with gender of participant (female or male) and type of initiation (verbal or nonverbal) as the predictors and the ability to imagine their date initiating sexual intercourse as the outcome. A test of the full model, with both predictors against a constant-only model, was not statistically significant, [chi square] (2, N = 377) = 3.25, p = .197, indicating that neither gender of participant nor type of initiation affected the ability to imagine one's date initiating sexual intercourse (see Table 2).

Table 2. Percentages of Participants Who Could Imagine Themselves in Each Scenario as a Function of Participant Gender, Type of Initiation, and Initiator

                            Gender of participants
Type of initiation             Women           Men

                            Date-initiation scenarios

Verbal                      90.59% (85)    93.26% (89)
Nonverbal                   93.20% (103)   97.00% (100)
Across type of initiation   92.02% (188)   95.24% (189)

                            Self-initiation scenarios

Verbal                      48.24% (85)    73.03% (89)
Nonverbal                   66.99% (103)   96.04% (101)
Across type of initiation   58.51% (188)   85.26% (190)

Type of initiation          All participants

                            Date-initiation scenarios

Verbal                      91.95% (174)
Nonverbal                   95.07% (203)
Across type of initiation   93.63% (377)

                            Self-initiation scenarios

Verbal                      60.92% (174)
Nonverbal                   81.37% (204)
Across type of initiation   71.96% (378)

Note. Numbers in parentheses are ns indicating the total number of participants on which the percentages are based.

A second logistic regression was performed with gender of participant (female or male) and type of initiation (verbal or nonverbal) as the predictors and the ability to imagine oneself initiating sexual intercourse as the outcome. A test of the full model with both predictors against a constant-only model was statistically significant, [chi aquare] (2, N = 378) = 57.12, p [is less than] .001. Both gender and type of initiation significantly predicted participants' ability to imagine themselves initiating sexual intercourse. Significantly more men than women could imagine themselves initiating sexual intercourse, [chi square] (1, N = 378) = 33.48, p [is less than] .001. Participants were more likely to be able to imagine themselves initiating sexual intercourse nonverbally than verbally, [chi square] (1, N = 378) = 21.34, p [is less than] .001 (see Table 2).

Factor Analyses

Participants' ratings were factor analyzed to guide the creation of subscales to be used in subsequent analyses. The meanings of the 33 behaviors had been rated on a 7-point Likert scale ranging from 0 (does not show consent to sexual intercourse/never do this to show consent) to 6 (definitely shows consent to sexual intercourse/always do this to show consent). Participants rated these behaviors three times: once to rate how often they use various signals to show their consent in actual situations (actual self-consent ratings), once to rate how indicative each behavior would be of their own consent in scenarios (hypothetical self-consent ratings), and once to rate how indicative each behavior would be of their date's consent in scenarios (hypothetical date-consent ratings). To increase the validity of the results, the actual self-consent ratings of participants who had not had penile-vaginal intercourse were eliminated, as were the hypothetical self- or date-consent ratings of participants who could not imagine themselves in these scenarios.

For the factor analyses, female and male participants' responses were combined. A principal-components factor analysis with a varimax rotation was conducted on each type of rating, resulting in three factor patterns. For the hypothetical self-consent factor analysis, six factors emerged. For the hypothetical date-consent factor analysis, seven factors emerged. For the actual self-consent factor analysis, eight factors emerged. The factor loadings were examined for similarities across the three analyses. Conceptual considerations were used in combination with the loadings to guide the development of subscales. Six items were dropped to facilitate conceptual clarity. We created seven subscales (see Table 1). Four of the subscales paralleled the categories used by McCormick (1987): direct verbal signals (e.g., "She/He says, `Yes'"), direct nonverbal signals ("She/He doesn't say anything--she/he just starts having intercourse with you"), indirect verbal signals (e.g., "She/He asks if you have a condom"), and indirect nonverbal signals (e.g., "She/He rubs, fondles, and touches you sexually"). The fifth subscale represented statements about intoxication (e.g., "She/He says `I'm really drunk'"), the sixth represented a direct refusal ("She/He says, `No'"), and the seventh represented making no response (e.g., "She/He does not resist your sexual advances"). Two of these seven subscales (the direct nonverbal subscale and the direct refusal subscale) consisted of only one item each. Although this is less than ideal, we decided to retain them because they represented important concepts for expressing sexual consent or nonconsent. The items on these subscales showed satisfactory reliability, with a median Cronbach's alpha of .89 and a range of .69 to .95, with one exception: For hypothetical date-consent ratings of direct verbal signals, Cronbach's alpha was only .49 (see Table 1).

Calculating subscale scores. Once the subscales were derived, subscale scores were calculated for each participant. Participants' subscale scores were determined by calculating the mean of their responses to the items on the subscale. In this way, subscale scores retained the same range of potential values as the original ratings (see Table 1 for a list of the items on each subscale). Subscale scores for each participant were computed separately for the hypothetical self-consent, hypothetical date-consent, and actual self-consent ratings. Subscale scores were not calculated for the hypothetical self- or date-consent ratings of participants who were unable to imagine themselves in those scenarios or for the actual self-consent ratings of participants who had not had penile-vaginal intercourse. Subscale scores were used as dependent variables in subsequent analyses.

Self-Reported Use of Consent Signals

To answer Research Question 2, addressing which consent signals women and men use most frequently, the actual self-consent ratings of participants who had experience with penile-vaginal intercourse were analyzed separately for women and men. Both women and men reported that they most often showed their consent to sexual intercourse by making no response. For each gender, 21 within-subject t tests were used to compare each of the seven subscales with every other subscale; using the Bonferroni technique, alpha was set at .05/21=.002. Women's no response subscale scores were significantly higher than any other subscale scores except for the indirect verbal subscale; men's no response subscale scores were significantly higher than all their other subscale scores (see Table 3). Both women and men reported almost never using statements about intoxication or direct refusals to show their consent.

Table 3. Women's and Men's Self-Reported Use of Consent Signals

                          Gender of participants
Factor                    Women(a)           Men(b)

Direct verbal        3.21 (1.56)(c)      3.19 (1.49)(c)
Direct nonverbal     3.06 (2.21)(cd)     3.21 (2.21)(c)
Indirect verbal      3.65 (1.39)(abd)    2.85 (1.52)(c)
Indirect nonverbal   3.29 (1.44)(bc)     3.96 (1.25)(b)
Intoxication         0.26 (0.85)(e)      0.65 (1.19)(d)
Direct refusal       0.60 (1.57)(e)      0.40 (1.01)(d)
No response          3.85 (1.63)(a)      4.53 (1.61)(a)

                                          Adjusted
Factor                   F       p     [[Eta].sup.2]

Direct verbal          0.01   .9249       .000
Direct nonverbal       0.33   .5653       .000
Indirect verbal       21.57   .0001       .068
Indirect nonverbal    17.35   .0001       .055
Intoxication          10.29   .0015       .032
Direct refusal         1.69   .1947       .002
No response           12.48   .0005       .039

Note. Scale ranged from 0 (never do this to show consent) to 6 (always do this to show consent). Table entries are means with standard deviations shown in parentheses. Participants who had never engaged in sexual intercourse were dropped from this analysis. Within each column, means with the same subscript do not differ significantly from each other at p < .002, calculated using the Bonferroni technique. Fs refer to tests of gender differences. Adjusted [[Eta].sup.2] s refer to estimated effect sizes of gender differences.

(a) n = 146.

(b) n = 136.

Gender differences in the use of consent signals. To address Research Question 3, whether there are gender differences in how often women and men report using consent signals in actual situations, a between-subjects multivariate analysis of variance (MANOVA) was conducted with gender of the participant as the independent variable and actual self-consent ratings as the dependent variables. Only participants who had experienced penile-vaginal intercourse were included in these analyses. There was a significant gender difference, F(7, 274) = 10.38, p [is less than] .001. Follow-up univariate analyses of variance (ANOVAs) showed that men reported using indirect nonverbal signals, statements about intoxication, and no response to signal consent more frequently than did women (see Table 3). Women reported using indirect verbal signals to show consent more frequently than did men. To estimate effect size, adjusted eta squared was calculated for each effect (Judd & McClelland, 1989). Whenever the formula for adjusted eta squared (1 - [1 - [n.sup.2]] [[df.sub.total]/[df.sub.error]]) produced a negative number, it was set to 0. Using Cohen's (1977) guidelines, construing .02 as a small effect,. 13 as a medium effect, and .26 as a large effect, all effect sizes for gender were in the small range, suggesting that although there were significant gender differences in reports of how frequently different signals were used, the gender differences were small. There were no gender differences in any of the direct signals: direct verbal signals, direct nonverbal signals, or a direct refusal.

Self-reported behaviors and hypothetical self-consent ratings. To address Research Question 4--whether people actually use the signals they rate as hypothetically indicating their own consent to sexual intercourse--we calculated Pearson's correlations between participants' actual self-consent ratings and hypothetical self-consent ratings. Only the responses of participants who had experienced penile-vaginal intercourse and could imagine themselves responding to a partner's sexual initiation were included. Participants' self-reported use of signals correlated positively and significantly with their ratings of how indicative these behaviors were of their own sexual consent, although the magnitude of these correlations ranged from moderate to low: for direct verbal signals, r = .18, p = .004; for direct nonverbal signals, r = .22, p [is less than] .001; for indirect verbal signals, r = .23, p [is less than] .001; for indirect nonverbal signals, r = .59, p [is less than] .001; for statements about intoxication, r = .43, p [is less than] .001; for a direct refusal, r =. 18, p = .003; and for no response, r = .49, p [is less than] .001.

Women's and Men's Ratings of the Meaning of Consent Signals in Hypothetical Scenarios

Type of initiation and gender differences. To answer Research Questions 5 and 6, whether the type of initiation (verbal or nonverbal) and participant gender affected ratings of how indicative signals were of consent in the hypothetical scenarios, two 2 x 2 between-subjects MANOVAs were performed on participants' subscale ratings of hypothetical date-consent signals and hypothetical self-consent signals. For the first MANOVA, the independent variables were the gender of the participants (female or male) and the type of initiation (verbal or nonverbal), and the dependent variables were the seven hypothetical self-consent subscale scores. Participants who could not imagine their partner initiating sexual intercourse were eliminated from this analysis. For the second MANOVA, the independent variables were participant gender and type of initiation, and the dependent variables were the seven hypothetical date-consent subscale scores. Participants who could not imagine themselves initiating sexual intercourse were eliminated from this analysis. Participants were included whether or not they had experienced penile-vaginal intercourse because even inexperienced persons might find themselves in situations where they need to convey or infer consent or lack of consent to sex. There were no significant interactions between gender and type of initiation in either MANOVA, so these variables will be discussed separately.

Type of initiation. The two MANOVAs revealed that type of initiation (verbal or nonverbal) significantly affected hypothetical self-consent ratings, F(7, 343) = 6.62, p [is less than] .001, and hypothetical date-consent ratings, F(7, 262) = 4.84, p [is less than] .001. Follow-up ANOVAs showed that participants rated indirect nonverbal signals as more indicative of their consent and of their dates' consent in response to verbal initiations than in response to nonverbal initiations (see Table 4). However, they indicated that indirect verbal signals would be more indicative of their consent and of their dates' consent in response to nonverbal initiations than in response to verbal initiations. In both cases the effect sizes were small, suggesting that the type of initiation only had a small effect on the ratings of these signals. Participants also rated a direct refusal as less indicative of their own consent in response to verbal initiations than in response to nonverbal initiations, although these ratings were low and the effect size was very small; there was a similar but nonsignificant trend for hypothetical date-consent ratings. The type of initiation had no significant impact on the ratings of hypothetical direct verbal, direct nonverbal, intoxication, or no response signals in the scenarios.

Table 4. Mean Hypothetical Self-Consent Ratings and Hypothetical Date-Consent Ratings of Sexual Consent Signals in Verbal and Nonverbal Initiation Scenarios

                          Type of initiation
Factor                  Verbal        Nonverbal      F       p

                            Ratings of one's own hypothetical
                                   consent signals(a)

Direct verbal         5.28 (1.01)    5.26 (0.90)    0.01   .9297
Direct nonverbal      5.34 (1.45)    5.36 (1.44)    0.01   .9140
Indirect verbal       4.22 (1.20)    4.62 (1.00)   13.36   .0003
Indirect nonverbal    3.87 (1.27)    3.51 (1.37)    6.43   .0116
Intoxication          1.02 (1.57)    1.10 (1.56)    0.40   .5278
Direct refusal        0.12 (0.59)    0.30 (1.01)    4.08   .0442
No response           3.64 (1.48)    3.54 (1.56)    0.27   .6004

                            Ratings of dates' hypothetical
                                    consent signals(b)

Direct verbal         5.50 (0.48)    5.43 (0.54)    0.73   .3927
Direct nonverbal      5.66 (1.03)    5.61 (0.99)    0.27   .6041
Indirect verbal       4.20 (1.10)    4.53 (1.06)    4.29   .0394
Indirect nonverbal    3.91 (1.13)    3.46 (1.27)    9.44   .0023
Intoxication          1.08 (1.52)    1.10 (1.54)    0.01   .9116
Direct refusal        0.06 (0.33)    0.23 (0.91)    3.44   .0646
No response           3.47 (1.42)    3.25 (1.38)    1.77   .1839

Factor                            Adjusted
                               [[Eta].sup.2]

                     Ratings of one's own hypothetical
                             consent signals(a)

Direct verbal                      .000
Direct nonverbal                   .000
Indirect verbal                    .025
Indirect nonverbal                 .008
Intoxication                       .000
Direct refusal                     .003
No response                        .000

                     Ratings of dates' hypothetical
                             consent signals(b)

Direct verbal                      .000
Direct nonverbal                   .000
Indirect verbal                    .004
Indirect nonverbal                 .023
Intoxication                       .000
Direct refusal                     .000
No response                        .000

Note. Table entries are means with standard deviations shown in parentheses. Participants who could not imagine themselves in the situations were dropped from these analyses. Fs refer to tests of differences between types of initiations. Adjusted [[Eta].sup.2] refer to estimated effect sizes of these differences.

(a) Scale ranged from 0 (does not show your consent to sexual intercourse) to 6 (definitely shows your consent to sexual intercourse); n = 160 Verbal and 193 Nonverbal.

(b) Scale ranged from 0 (does not show her/his consent to sexual intercourse) to 6 (definitely shows her/his consent to sexual intercourse); n = 106 Verbal and 166 Nonverbal.

Participant gender. The MANOVAs revealed that gender of the participant significantly affected hypothetical self-consent ratings, F(7, 343) = 11.48, p [is less than] .001, but not hypothetical date-consent ratings, F(7, 262) = 1.85, p = .078. Follow-up ANOVAs revealed that men rated their own hypothetical direct verbal, direct nonverbal, indirect verbal, indirect nonverbal, intoxication, and no response signals as more indicative of their sexual consent than women rated their own such hypothetical behaviors (see Table 5). The effect sizes ranged from small to medium-small, suggesting that the differences between men's and women's ratings of their own hypothetical consent signals were minimal. Both women and men rated a direct refusal similarly as not indicative of sexual consent.

Table 5. Women's and Men's Mean Hypothetical Self-Consent Ratings and Hypothetical Date-Consent Ratings

                        Gender of participants
Factor                   Women           Men         F       p

                        Rating's of one's own consent signals(a)

Direct verbal         5.05 (1.13)    5.49 (0.69)   19.73   .0001
Direct nonverbal      5.11 (1.69)    5.58 (1.12)    8.85   .0031
Indirect verbal       4.10 (1.14)    4.76 (0.98)   35.44   .0001
Indirect nonverbal    3.31 (1.29)    4.03 (1.29)   27.52   .0001
Intoxication          0.55 (1.06)    1.55 (1.80)   40.48   .0001
Direct refusal        0.16 (0.73)    0.28 (0.95)    1.67   .1969
No response           3.13 (1.46)    4.02 (1.46)   34.30   .0001

                        Ratings of dates' consent signals(b)

Direct verbal         5.44 (0.53)    5.47 (0.51)    0.42   .5168
Direct nonverbal      5.59 (1.12)    5.65 (0.92)    0.11   .7352
Indirect verbal       4.30 (1.12)    4.47 (1.06)    0.87   .3510
Indirect nonverbal    3.50 (1.21)    3.72 (1.25)    1.23   .2676
Intoxication          0.77 (1.33)    1.30 (1.62)    7.12   .0081
Direct refusal        0.14 (0.66)    0.18 (0.80)    0.27   .6063
No response           3.39 (1.38)    3.30 (1.42)    0.32   .5725

Factor                     Adjusted
                        [[Eta].sup.2]

                     Rating's of one's own
                       consent signals(a)

Direct verbal                .045
Direct nonverbal             .016
Indirect verbal              .081
Indirect nonverbal           .064
Intoxication                 .096
Direct refusal               .000
No response                  .081

                        Ratings of dates'
                       consent signals(b)

Direct verbal                .000
Direct nonverbal             .000
Indirect verbal              .000
Indirect nonverbal           .000
Intoxication                 .015
Direct refusal               .000
No response                  .000

Note. Table entries are means with standard deviations shown in parentheses. Participants who could not imagine themselves in the situations were dropped from these analyses. Fs refer to tests of gender differences. Adjusted [[Eta].sup.2] s refer to estimated effect sizes of gender differences.

(a) Scale ranged from 0 (does not show your consent to sexual intercourse) to 6 (definitely shows your consent to sexual intercourse); n = 173 women and 180 men.

(b) Scale ranged from 0 (does not show her/his consent to sexual intercourse) to 6 (definitely shows her/his consent to sexual intercourse); n = 110 women and 162 men.

Assessing the potential for miscommunication. To answer Research Question 7, whether there were discrepancies between women's and men's ratings of the scenarios that would suggest gender-related miscommunication, two between-subjects MANOVAs were performed. The first compared women's hypothetical date-consent ratings and men's hypothetical self-consent ratings to assess possible gender miscommunications about sexual consent. This analysis compared how men rated their own consent signals and how women rated their male dates' signals in the scenarios to see if women and men have similar understandings of how men consent to sexual intercourse. Participants who could not imagine themselves in the pertinent scenarios were eliminated. A significant gender difference was found, suggesting that women and men have different understandings of how men signal consent to sexual intercourse, F(7, 282) = 3.96, p [is less than] .001. Follow-up ANOVAs revealed that men rated indirect verbal, indirect nonverbal, no response, and intoxication signals as more indicative of their own sexual consent than women rated these signals as indicative of their male dates' consent in scenarios (see Table 6). Effect sizes were small, suggesting that the actual gender differences were minimal. Direct verbal signals were rated as similarly indicative of a man's sexual consent by both women and men, as were direct nonverbal signals and a direct refusal.

Table 6. Women's and Men's Mean Hypothetical Date-Consent Ratings Compared With Their Mean Hypothetical Self-Consent Ratings

                      Gender of participants
Factor                   Women           Men         F       p

                       Women's mean hypothetical date-consent
                       ratings compared with men's hypothetical
                                self-consent ratings(a)

Direct verbal         5.44 (0.53)    5.49 (0.69)    0.31   .5752
Direct nonverbal      5.59 (1.12)    5.58 (1.12)    0.00   .9554
Indirect verbal       4.30 (1.12)    4.76 (0.98)   13.39   .0003
Indirect nonverbal    3.50 (1.21)    4.03 (1.29)   11.74   .0007
Intoxication          0.77 (1.33)    1.55 (1.80)   15.52   .0001
Direct refusal        0.14 (0.66)    0.28 (0.95)    1.90   .1693
No response           3.39 (1.38)    4.02 (1.46)   13.35   .0003

                       Men's mean hypothetical date-consent ratings
                         compared with women's mean hypothetical
                                 self-consent ratings(b)

Direct verbal         5.05 (1.13)    5.47 (0.51)   19.16   .0001
Direct nonverbal      5.11 (1.69)    5.65 (0.92)   13.02   .0004
Indirect verbal       4.10 (1.14)    4.47 (1.06)    9.22   .0026
Indirect nonverbal    3.31 (1.29)    3.72 (1.26)    8.96   .0030
Intoxication          0.55 (1.06)    1.30 (1.62)   25.43   .0001
Direct refusal        0.16 (0.73)    0.18 (0.80)    0.08   .7732
No response           3.13 (1.46)    3.30 (1.42)    1.18   .2777

Factor                                Adjusted
                                   [[Eta].sup.2]

                       Women's mean hypothetical date-consent
                      ratings compared with men's hypothetical
                              self-consent ratings(a)

Direct verbal                       .000
Direct nonverbal                    .000
Indirect verbal                     .041
Indirect nonverbal                  .036
Intoxication                        .048
Direct refusal                      .003
No response                         .041

                       Men's mean hypothetical date-consent ratings
                          compared with women's mean hypothetical
                                 self-consent ratings(b)

Direct verbal                       .052
Direct nonverbal                    .035
Indirect verbal                     .024
Indirect nonverbal                  .023
Intoxication                        .068
Direct refusal                      .000
No response                         .001

Note. Table entries are means with standard deviations shown in parentheses. Scale ranged from 0 (does not show his/her/your consent to sexual intercourse) to 6 (definitely shows his/her/your consent to sexual intercourse). Participants who could not imagine themselves in the situations were dropped from these analyses. Fs refer to tests of gender differences. Adjusted [[Eta].sup.2] s refer to estimated effect sizes of gender differences.

(a) n = 110 women and 180 men.

(b) n = 173 women and 161 men.

Similarly, a between-subjects MANOVA was performed on men's hypothetical date-consent ratings and women's hypothetical self-consent ratings. Again, a significant gender difference was found, suggesting that women and men have different understandings of how women signal consent to sexual intercourse, F(7, 326) = 7.39, p [is less than] .001. Follow-up ANOVAs revealed that for all signals except a direct refusal and no response, women rated their own hypothetical signals as less indicative of sexual consent than men rated their hypothetical female dates' signals in the scenarios (see Table 6). Effect sizes were small, suggesting that the gender differences on these ratings were minimal. There was no significant difference between women's and men's ratings of a direct refusal, and these ratings were low, suggesting that women and men agree that a direct refusal does not indicate sexual consent. Also, there was no significant difference between men's and women's ratings of no response signal: These ratings fell in the middle of the rating scale, suggesting they are considered somewhat indicative of consent.

DISCUSSION

Sexual Consent Signals

Communicating sexual consent is far more complex than simply saying yes to a sexual initiation. The young women and men in this study reported that they use a variety of signals to indicate their consent, and that they view a variety of signals as indicative of sexual consent in heterosexual situations. These signals ranged from behaviors as vague as smiling to statements as straightforward as "I want to have sex with you." Most of the sexual consent signals in this study fell into identifiable categories of direct and indirect verbal and nonverbal consent signals.

Reported use of consent signals. Participants who had previously engaged in sexual intercourse reported that they used a wide repertoire of signals to indicate their sexual consent in actual situations: direct verbal signals, direct nonverbal signals, indirect verbal signals, and indirect nonverbal signals. They reported almost never using statements about their level of intoxication or direct refusals to signal their sexual consent; they did, however, frequently convey consent by not resisting. Self-reported use of consent signals was, at most, moderately related to ratings of how indicative the signals were of sexual consent in the scenarios. The use and interpretation of consent signals depended on other factors such as gender and the way in which sex was initiated.

Gender and Sexual Consent

Gender and the self-reported actual use of consent signals. Closer examination of the self-reported use of consent signals revealed that there were small differences between women and men in the kinds of consent signals they reported using. Women were more likely than men to use indirect verbal signals (e.g., asking if the other person has a condom), whereas men were more likely than women to use indirect nonverbal signals (e.g., touching, kissing, or caressing the other person), no response (e.g., not saying no), or statements about their level of intoxication (though the use of intoxication statements was relatively rare) to indicate their consent. These differences suggest the possibility of gender-based misunderstandings. Women and men may expect that their date would consent in the same way they would consent; if their date does or says something that they themselves would use to signal consent, they may mistakenly assume the date is signaling consent. However, the effect sizes of the differences between women's and men's self-reported use of these consent signals were small, suggesting that the actual differences are small. Additionally, both women and men reported using no response (e.g., not resisting) more frequently than other signals. They were equally likely to use direct verbal and direct nonverbal signals, and equally unlikely to use direct refusals to signal consent. There were far more similarities than differences in women's and men's self-reported use of consent signals.

Gender and perceptions of sexual consent. There were also gender differences in ratings of how indicative signals would be of one's own sexual consent in response to a date's sexual initiation in hypothetical scenarios. Men rated direct verbal signals, direct nonverbal signals, indirect verbal signals, indirect nonverbal signals, statements about intoxication, and no response as more indicative of their own sexual consent than did women. This is not surprising given that numerous studies have found that men rate women, other men, and even themselves more sexually than women do (e.g., Abbey, 1982; Johnson et al., 1991). It seems this tendency also affects perceptions of hypothetical self-consent ratings. The one exception was saying no to signal consent: Women and men rated a direct refusal--a no--as being equally unindicative of their sexual consent.

Women and men were in agreement about the meaning of their dates' sexual consent signals in hypothetical scenarios. Interestingly, although women and men have similar ideas about whether their dates' signals indicate consent, they mean different things when they themselves use these signals. That is, women interpret their male dates' signals the same way that men interpret their female dates' signals, but paradoxically, as discussed in the previous paragraph, men's signals indicate a greater level of sexual consent than do the same signals given by women. This difference sets up the potential for sexual miscommunication.

Further exploration revealed that men rated their female dates' direct verbal signals, direct nonverbal signals, indirect verbal signals, indirect nonverbal signals, and intoxication statements as more indicative of their female dates' sexual consent than women rated their own sexual consent signals. Likewise, men rated their own indirect verbal signals, indirect nonverbal signals, statements about intoxication, and no response as more indicative of their own sexual consent than women rated their male dates' signals. These findings, taken in combination with the differences between women and men in what they mean by different consent signals, suggest the potential for gender-based sexual miscommunication.

However, it is important not to exaggerate these gender differences. Although there were significant gender differences in the expected direction, effect sizes were small, and the actual ratings of signals were quite similar. Differences between women's and men's ratings of how they interpreted these signals were always less than 1 point on a 7-point scale, just as they were for ratings of how often they used these signals. Additionally, there was agreement on an important issue: Both men and women rated direct refusals as equally unindicative of sexual consent, suggesting that a direct refusal is an unambiguous signal.

Gender and sexual initiations. There were also significant gender differences in ability to imagine oneself initiating sexual intercourse in the scenarios. More men than women were able to imagine themselves initiating sexual intercourse both verbally and nonverbally. This pattern fits with the traditional sexual script, in which men initiate sexual intercourse and women are the recipients of these initiations. It also has implications for sexual consent. If men typically initiate sexual intercourse, men will be interpreting women's consent signals more often than women will be interpreting men's consent signals. Thus, women's misinterpretations of men's consent signals are less relevant to most sexual situations than men's misinterpretations of women's sexual signals.

Verbal and Nonverbal Initiations

The way in which sexual intercourse was initiated influenced what behaviors were seen as indicative of consent. Women and men rated indirect verbal signals as more indicative of their and their dates' sexual consent in response to nonverbal initiations than in response to verbal initiations. Conversely, they rated indirect nonverbal signals as more indicative of their and their dates' sexual consent in response to verbal initiations than in response to nonverbal initiations. It seems that the meanings of indirect verbal and nonverbal signals are more situationally specific than are direct verbal and nonverbal signals. The unclear nature of indirect signals may force the initiator to look to the situation for cues as to whether the respondent has actually consented. The type of initiation did not affect how participants rated direct verbal signals or direct nonverbal signals. Further research is needed to clarify the role of the type of initiation on perceptions of sexual consent and to explore the situational specificity of consent signals.

Implications

Miscommunication and acquaintance rape. Many researchers have discussed the possibility that sexual miscommunication between women and men contributes to acquaintance rape (Abbey, 1982, 1987; Bart & O'Brien, 1985; Warshaw, 1994). Indeed, anecdotal evidence suggests that gender-based misunderstandings do sometimes lead to rape (Bart & O'Brien, 1985; Warshaw, 1994). Empirical research has also found support for this hypothesis. One study found that male and female participants had different understandings of resistance messages, particularly when the messages were indirect, suggesting that miscommunication about refusals could contribute to sexual assault (Motley & Reeder, 1995). Abbey (1987), examining naturally occurring misperceptions of sexual intent, found that such misperceptions are common between women and men. Significantly more women (72%) than men (60%) reported that they had had at least one experience in which a member of the "opposite sex" misperceived their friendly behavior as sexually-interested behavior. Although most misperceptions were minor in nature and were resolved quickly, some led to forced sexual activity, ranging from kissing to sexual intercourse. However, it is unclear from these data whether the respondents' signals were misunderstood or deliberately ignored.

As a result of the miscommunication hypothesis, women have been advised to clearly communicate their sexual intentions to prevent being raped. This "prevention strategy" is problematic, for it suggests that it is women's responsibility to ensure that men understand their sexual intentions, not men's responsibility to listen to their partner or date or to get clear consent before proceeding (Crawford, 1995). It encourages victim blaming, for women are held responsible when their communication efforts "fail" and they are raped (Crawford, 1995; Warshaw, 1994). Also, this approach does not take into account women's concerns about the effect of direct messages, particularly direct refusals, on their relationship with the man involved (Motley & Reeder, 1995). Cupach and Metts (1991) suggested that both men and women prefer indirect communication because it enables them to gain sexual access and avoid explicit rejection. If the indirect communication is not responded to in a positive way, it can go unacknowledged. If it is accepted, then sexual activity can begin or continue.

Although this study found evidence that women and men do have different ideas about how indicative of sexual consent many signals are, the differences between women and men were generally quite small. Thus, it is unlikely that miscommunication about consent is a major contributing factor to acquaintance rape. It is more likely that sexually aggressive men selectively ignore or reinterpret what women say to fit what they want to hear, using miscommunication as an excuse for raping (Christopher & Frandsen, 1990; Warshaw, 1994).

One finding in the present study did signal a cause for concern, however. Both women and men reported that they most frequently signaled sexual consent by not resisting: letting their partner undress them, not stopping their partner from kissing or touching them, not saying no. Unfortunately in some cases of rape, the aggressor proceeds without resistance from the victim, perhaps because the victim is frightened, confused, or embarrassed (Estrich, 1987; Meyer, 1984; Warshaw, 1994). Thus, it could be dangerous to use not resisting as a signal of consent.

Directions for Future Research

Numerous participants wrote comments indicating that being in a relationship affected how they signaled consent. One woman wrote that "a smile does not mean consent in a bar to a guy I hardly know, but it does with my boyfriend." It may be that couples develop more idiosyncratic rules for interpreting each other's behavior and sexual signals as the relationship develops (Peplau et al., 1977). There may also be a sense of entitlement that encourages people to presume sexual consent in relationships (Shotland & Goodstein, 1992). Further research is needed in this area. Additionally, the current study was designed to assess how young women and men perceive and communicate consent in heterosexual situations. Further research is needed to examine how people of different ages, socioeconomic groups, ethnic groups, and sexual orientations infer and convey consent.

Conclusion

When the students and administrators at Antioch College developed mutual consent guidelines, they unintentionally created an international controversy, generating discussion about a previously neglected topic: sexual consent. At the time there was little information about how people conceptualize and communicate consent to help inform the discussion. The results of this study and related studies clearly indicate that consent is complex and can take many forms. Simply requiring that people verbally communicate consent by saying "I consent to sexual intercourse," as the Antioch College mutual consent guidelines suggested, is probably unrealistic for most people. We need to generate dialogue about the nature of consent and encourage young women and men to talk about consent, both in sexual situations and in the classroom. Such discussions would minimize the possibility of gender-based miscommunications about sexual consent and help eliminate harmful stereotypes that perpetuate rape.

REFERENCES

Abbey, A. (1982). Sex differences in attributions for friendly behavior: Do males misperceive females' friendliness? Journal of Personality and Social Psychology, 42, 830-838.

Abbey, A. (1987). Misperceptions of friendly behavior as sexual interest: A survey of naturally occurring incidents. Psychology of Women Quarterly, 11, 173-194.

Abbey, A. (1991). Misperception as an antecedent of acquaintance rape. In A. Parrot & L. Bechhofer (Eds.), Acquaintance rape: The hidden crime (pp. 96-111). New York: Wiley.

Abbey, A., Cozarelli, C., McLaughlin, K., & Harnish, R. J. (1987). The effects of clothing and dyad sex composition on perceptions of sexual intent: Do women and men evaluate these cues differently. Journal of Applied Social Psychology, 17, 108-126.

Abbey, A., & Melby, C. (1986). The effects of nonverbal cues on gender differences in perceptions of sexual intent. Sex Roles, 15, 283-298.

Antioch College. (1990). Sexual violence and safety. Yellow Springs, OH: Author.

Bart, P. B., & O'Brien, P. H. (1985). Stopping rape: Successful survival strategies. Elmsford, NY: Pergamon.

Botswick, T. D., & DeLucia, J. L. (1992). Effects of gender and specific dating behaviors on perceptions of sex willingness and date rape. Journal of Social and Clinical Psychology, 11, 14-25.

Burrow, J. J. (1997). Men's and women's perceptions of verbal and nonverbal consent for sexual intercourse. Unpublished master's thesis, University of the Pacific, Stockton, CA.

Burt, M. R., & Albin, R. S. (1981). Rape myths, rape definitions, and probability of conviction. Journal of Applied Social Psychology, 11, 212-230.

Byers, E. S. (1980). Female communication of consent and nonconsent to sexual intercourse. New Brunswick Psychology, 5, 12-18.

Byers, E. S., & Heinlen, L. (1989). Predicting initiations and refusals of sexual activities in married and cohabiting heterosexual couples. The Journal of Sex Research, 26, 210-231.

Byers, E. S., & Lewis, K. (1988). Dating couples' disagreements over the desired level of sexual intimacy. The Journal of Sex Research, 24, 15-29.

Christopher, F. S., & Frandsen, M. M. (1990). Strategies of influence in sex and dating. Journal of Social and Personal Relationships, 7, 89-105.

Cohen, J. (1977). Statistical power analysis for the behavioral sciences (Rev. ed.). New York: Academic Press.

Crawford, M. (1995). Talking difference: On gender and language. Thousand Oaks, CA: Sage.

Cupach, W. R., & Metts, S. (1991). Sexuality and communication in close relationships. In K. McKinney & S. Sprecher (Eds.), Sexuality in close relationships (pp. 93-110). Hillsdale, NJ: Lawrence Erlbaum Associates.

Estrich, S. (1987). Real rape. Cambridge, MA: Harvard.

Falbo, T. (1977). Multidimensional scaling of power strategies. Journal of Personality and Social Psychology, 35, 537-547.

Falbo, T., & Peplau, L. A. (1980). Power strategies in intimate relationships. Journal of Personality and Social Psychology, 38, 618-628.

Gagnon, J. H., & Simon, W. (1973). Sexual conduct: The social sources of human sexuality. Chicago: Aldine.

Guskin, A. E. (1994). The Antioch response: Sex, you just don't talk about it. Yellow Springs, OH: Antioch College.

Hall, D. S. (1995). Consent for sexual behavior in a college student population. Unpublished doctoral dissertation, The Institute for Advanced Study of Human Sexuality, San Francisco.

Johnson, C. B., Stockdale, M. S., & Saal, F. E. (1991). Persistence of men's misperceptions of friendly cues across a variety of interpersonal encounters. Psychology of Women Quarterly, 15, 463-475.

Judd, C. M., & McClelland, G. H. (1989). Data analysis: A model-comparison approach. New York: Harcourt, Brace, Jovanovich.

McCormick, N. B. (1979). Come-ons and put-offs: Unmarried students' strategies for having and avoiding sexual intercourse. Psychology of Women Quarterly, 4, 194-211.

McCormick, N. B. (1987). Sexual scripts: Social and therapeutic implications. Sexual and Marital Therapy, 2, 3-27.

Metts, S., Cupach, W. R., & Imahori, T. T. (1992). Perceptions of sexual compliance-resisting messages in three types of cross-sex relationships. Western Journal of Communication, 56, 1-17.

Meyer, T. J. (1984, Dec. 5). "Date rape": A serious campus problem that few talk about. The Chronicle of Higher Education, 29 (15), pp. 1, 12.

Motley, M. T., & Reeder, H. M. (1995). Unwanted escalation of sexual intimacy: Male and female perceptions of connotations and relational consequences of resistance messages. Communication Monographs, 62, 355-382.

Muehlenhard, C. L. (1988). Misinterpreted dating behaviors and the risk of date rape. Journal of Social and Clinical Psychology, 6, 20-37.

Muehlenhard, C. L. (1995/1996). The complexities of sexual consent. SIECUS Report, 24 (2), 4-7.

Muehlenhard, C. L., & Hollabaugh, L. C. (1988). Do women sometimes say no when they mean yes? The prevalence and correlates of women's token resistance to sex. Journal of Personality and Social Psychology, 54, 872-879.

Muehlenhard, C. L., Powch, I. G., Phelps, J. L., & Giusti, L. M. (1992). Definitions of rape: Scientific and political implications. Journal of Social Issues, 48 (1), 23-44.

Muehlenhard, C. L., & Rodgers, C. S. (1998). Token resistance to sex: New perspectives on an old stereotype. Psychology of Women Quarterly, 22, 443-463.

Muehlenhard, C. L., & Schrag, J. L. (1991). Nonviolent sexual coercion. In A. Parrot & L. Bechhofer (Eds.), Acquaintance rape: The hidden crime (pp. 115-128). New York: Wiley.

O'Sullivan, L. F., & Allgeier, E. R. (1994). Disassembling a stereotype: Gender differences in the use of token resistance. Journal of Applied Social Psychology, 24, 1035-1055.

O'Sullivan, L. F., & Byers, E. S. (1992). College students' incorporation of initiator and restrictor roles in sexual dating interactions. The Journal of Sex Research, 29, 435-446.

O'Sullivan, L. F., & Byers, E. S. (1993). Eroding stereotypes: College women's attempts to influence reluctant male sexual partners. The Journal of Sex Research, 30, 270-282.

Peplau, L. A., Rubin, Z., & Hill, C. T. (1977). Sexual intimacy in dating relationships. Journal of Social Issues, 33 (2), 86-109.

Perper, T., & Weis, D. L. (1987). Proceptive and rejective strategies of U.S. and Canadian college women. The Journal of Sex Research, 23, 455-480.

Saal, F. E., Johnson, C. B., & Weber, N. (1989). Friendly or sexy? It may depend on whom you ask. Psychology of Women Quarterly, 13, 263-276.

Safilios-Rothschild, C. (1977). Love, sex, and sex roles. Englewood Cliffs, NJ: Prentice-Hall.

Sanday, P. R. (1996). A woman scorned: Acquaintance rape on trial. New York: Doubleday.

Shotland, R. L., & Craig, J. M. (1988). Can men and women differentiate between friendly and sexually interested behavior? Social Psychology Quarterly, 51, 66-73.

Shotland, R. L., & Goodstein, L. (1992). Sexual precedence reduces the perceived legitimacy of sexual refusal: An examination of attributions concerning date rape and consensual sex. Personality and Social Psychology Bulletin, 19, 756-764.

Sprecher, S., & McKinney, K. (1993). Sexuality. Newbury Park, CA: Sage.

Tong, R, (1984). Women, sex, and the law. Totowa, NJ: Rowman & Allanheld.

Warshaw, R. (1994). I never called it rape (2nd ed.). New York: Harper & Row.

Manuscript accepted December 9, 1998

This paper is based on Susan E. Hickman's doctoral dissertation, conducted under the supervision of Charlene L. Muehlenhard and defended in October 1996. We wish to thank Kate Prenovost for statistical consultation. The title is taken from a participant's response to the pilot questionnaire regarding how he knew his partner was consenting to sexual intercourse.

Address correspondence to Charlene L. Muehlenhard, Department of Psychology, 426 Fraser Hall, University of Kansas, Lawrence, Kansas 66045-2160; e-mail: [e-mail].