Most available research on consent given by women for sexual intercourse has examined coercive/aggressive situations such as rape (Muehlenhard & Linton, 1987). These studies have focused on nonconsent (which usually has to be established for a judgment that a rape occurred) rather than on consent. How consent is given a typical dating situation has been studied very little.
According to Muehlenhard, Powch, Phelps, and Giusti (1992), there are two ways in which consent and nonconsent can be defined: (a) nonconsent is assumed unless explicit consent is given, and (b) consent is assumed unless explicit nonconsent is given. The Antioch College Sexual Offense Policy adopted in 1990 assumes the first definition of nonconsent (Hall, 1998). This policy states that consent for sexual behavior must be (a) verbal, (b) mutual, and (c) reiterated for every new level of sexual behavior. In other words, nonconsent is assumed unless explicit verbal consent is given. Interestingly, this policy is not based on research. Therefore, the intentions of the policy may be good, but the practical application may be difficult if the policy requires behavior which differs greatly from the way consent is usually given. If this policy accurately describes typical consent patterns, then men and women give explicit verbal consent for every sexual behavior. Research on consent, however, demonstrates that couples do not typically follow the pattern required by the Antioch policy (Byers, 1980; Hall, 1998; Hickman, 1996).
Muehlenhard (1996) states that consent can be thought of as a
mental and/or verbal act. If consent is a mental act (an internal
decision) then a partner may try to infer consent from the other person's
nonverbal behavior. This can easily lead to miscommunication.
Consent can also be a verbal act (explicit statement) which is less likely
to lead to miscommunication. Research suggests that most sexual consent
is not given in a clear manner (Muehlenhard, 1996). The purpose of
the present study was to identify what types of behavior heterosexual men
and women believe women would in fact use to show clear consent and nonconsent
for sexual intercourse.
Verbal and Nonverbal Consent for Sexual Behavior
Only a few studies have empirically examined issues of consent in heterosexual dating couples (Byers, 1980; Hall, 1998; Hickman, 1996). Consent and nonconsent for intercourse have been found to occur both verbally and nonverbally. The way in which people give consent is important because miscommunication of one's intentions can lead to unwanted sex or rape (Byers, 1980; Muehlenhard, 1988).
Byers (1980) examined perceptions of how women communicate sexual consent and nonconsent for penile/vaginal intercourse. She suggested that miscommunication of consent and nonconsent could be associated with sexual aggression. The participants (106 males and 87 females) were mailed questionnaires in which they were asked to rank order (out of a series of behaviors) the three most important behaviors they thought a woman would use to clearly communicate consent to intercourse, and the three most important to communicate nonconsent. Byers found that both male and female participants indicated that fondling the male's genitals, a nonverbal response, was chosen by the largest percentage of participants for communicating that a woman consents to intercourse. The next largest percentage chose clear verbal consent. Byers also noted that saying "No" was chosen by the largest percentage of both females (48%) and males (44%) as the most important method for communicating nonconsent to intercourse. It is interesting to note that a nonverbal behavior was perceived as being the most important method for a woman to communicate consent for intercourse, while a verbal behavior was perceived as the most important method for a woman to communicate nonconsent. Byers also found that males and females are very similar in their perceptions regarding how women communicate consent and nonconsent to intercourse.
Hall (1998) examined how consent is given for several types of sexual behavior by heterosexual men and women in their own actual dating situations. He was interested in expanding the literature on sexual consent in positive sexual interactions. Hall defined positive sex as a situation in which both people say "yes" and mean "yes" to engaging in sexual behavior. Participants were given a questionnaire which asked them to indicate if they had ever been in a situation in which they and their partner consented to sexual intercourse or other intimate sexual behavior. If participants responded "yes," the next part of the questionnaire asked them to describe their most recent sexual experience in which they consented and how consent was given (verbal/nonverbal). Hall used a table of sexual behaviors (e.g., kissing, hugging, etc.) on which participants listed in order the behaviors that occurred. The participants were then asked to indicate for every behavior that occurred whether consent was given and, if so, whether it was given verbally or nonverbally.
Hall found that participants gave permission for some sexual behaviors, but not for every behavior in a positive sexual interaction. Permission giving was most associated with initial sexual behaviors, oral sex, and intercourse. Hall also found that most permission giving for sexual behaviors other than intercourse (e.g., kissing, touching) was nonverbal. About 80% of male and female participants reported giving permission for penile/vaginal intercourse by either verbal or nonverbal means. Participants who gave permission for intercourse indicated giving verbal permission about half of the time and nonverbal about half the time. From this data, it is apparent that roughly 50% of the time males and females are indicating by some nonverbal behavior (e.g., smile, get closer, etc.) that they want to engage in penile/vaginal intercourse. Since nonverbal behavior may be more easily misinterpreted than verbal behavior, a belief in this type of consent may lead to unwanted sexual activity.
Hickman (1996) examined how women and men give and perceive sexual consent in heterosexual situations. Her study was an exploratory investigation and she was interested in looking at many factors including (a) what consent behaviors men and women report using the most/least and (b) whether each gender reports using different or similar consent behaviors. The participants (214 females and 210 males) were given two scenarios in which they were asked to imagine themselves being one of the people involved. The scenarios were constructed so that the participant or her/his date initiated sex verbally or nonverbally. After each scenario, participants were asked to rate 34 verbal and nonverbal behaviors to indicate how each would show their consent and their dates' consent to sexual intercourse. In addition, participants reported how frequently they actually used each of the 34 behaviors to indicate consent in real situations.
Most relevant to the current study, Hickman found that there were
small differences in the kinds of behaviors men and women reported themselves
using to indicate consent. She defined direct verbal behaviors (e.g.,
"I want to have sex with you") and direct nonverbal behaviors (e.g., just
starts having intercourse) as being unambiguous. In addition, she
defined indirect verbal behaviors (e.g., "Do you want to have sex") and
indirect nonverbal behaviors (e.g., touches and kisses) as being ambiguous.
For indicating consent, she found that men and women equally reported using
direct verbal and direct nonverbal behaviors. Men reported using
indirect nonverbal behaviors more than women to indicate consent.
In contrast, women reported using indirect verbal behaviors more than men
to indicate consent. Hickman stated that the actual numerical differences
between men and women's self-reported use of consent behaviors were quite
Purpose and Hypotheses
Byers (1980) was the only study found in our literature review
that examined men and women's judgments of how women communicate both consent
and nonconsent. In contrast, Hall (1998) and Hickman (1996) examined
the behaviors of men and women in consent only situations. For this
study, we focused on perceptions of both consent and nonconsent given by
women in heterosexual encounters, and developed a multi-item scale to increase
the reliability of the measurement of behaviors used to indicate consent/nonconsent.
Based on the literature reviewed, we hypothesized that (a) participants
would perceive verbal responses as being more likely for indicating nonconsent
than consent, (b) participants would perceive nonverbal responses as being
more likely for indicating consent than nonconsent, and (c) male and female
responses would not differ significantly in the consent versus nonconsent
situations. Predicting no gender differences is consistent with Byers'
(1980) findings that males and females reported similar behaviors for a
woman to communicate consent and nonconsent, and with Hall's (1995) findings
that males and females reported communicating consent similarly.
In addition, although Hickman (1996) found gender differences in the self-reported
use of consent behaviors, these were numerically small.
Participants were 385 students (118 male, 267 female; M age = 23.0) enrolled in undergraduate psychology courses at a community college and a private university in northern California. The demographics of the participants were as follows: 50.4% European American, 16.1% Latin American, 9.1% South East Asian, 6.8% Pacific Island American, 6.5% Asian American, 5.2% Other American, 2.9% African American, 1.6% Native American, and 1.6% Non American. Participation was voluntary and participants received no credit for being in the study.
The dating questionnaire consisted of a brief demographic section and one of four questionnaires with a vignette describing a date of a heterosexual couple.
Consent to have intercourse was varied across two levels: (a) female wants to have intercourse and consents and (b) female does not want to have intercourse and does not consent. Participant gender (male/female) was crossed with level of consent to produce four questionnaires with identical wording except for changes directly related to gender of the participant and consent variables. The vignette was adapted from the work of Hannon, Hall, Nash, Formati, and Miller (1995). Responses were adapted from Hickman (1996). The vignette followed a traditional sexual script in which the male was the initiator of sexual intercourse (Muehlenhard et al., 1992). The vignette read:
Rick and Theresa know each other from a class they take at a local university. One day after class they went to a local cafe to discuss an upcoming assignment. After a few hours of enjoyable conversation during which they admitted to being attracted to each other, they decided to meet again over the weekend and rent a movie to watch at Theresa's apartment.
During the movie they began flirting and kissing. After a while, both Theresa and Rick were sexually aroused, and both became more involved in flirting, kissing and intimately touching each other. Rick then made it clear that he wanted to have sexual intercourse (i.e., penile-vaginal intercourse).
After reading the above vignette, participants read a paragraph that asked them to assume that Theresa either wanted to have intercourse with Rick and consented or did not want to have intercourse with Rick and did not consent. Female participants were asked to put themselves in the role of Theresa and male participants were asked to indicate how they thought Theresa would respond. The female consent version read as follows:
Assume that you are Theresa and you want to let Rick know that you also want to have intercourse. Please circle a number for each of the 12 items below to indicate how likely YOU would be to give each of the following responses in order to show him that you consent to have intercourse with him.
The female nonconsent version was identical but read you do not want to have intercourse and do not consent to have intercourse in the underlined portions above. Male consent and nonconsent versions were identical to female versions except that you was replaced with she.
Participants who read consent vignettes then completed a 12-item scale describing different verbal (6 items) and nonverbal (6 items) responses to indicate consent. The response choices were developed on a continuum from less clearly to more clearly communicating consent. For example, a less clear verbal response was, "I would say, 'Maybe we should have sex.'" A more clear verbal response was, "I would say, 'I really want to have sex with you.'" The same methodology was used in developing the nonverbal responses. Verbal and nonverbal response choices were randomly ordered. Response choices began with "I would..." on the female versions and "She would..." on the male versions.
Participants who read nonconsent vignettes completed a 12-item scale describing different verbal (6 items) and nonverbal (6 items) responses to indicate nonconsent. Response choices were developed to directly parallel items on the consent scale, e.g., "I would say, 'Maybe we should have sex,'" was converted to, "I would say, 'Maybe we shouldn't have sex.'"
Each consent scale item was rated on a 7-point Likert scale ranging from (1) would definitely not do this to show consent to (7) would definitely do this to show consent. Each nonconsent item was rated on a 7-point Likert scale ranging from (1) would definitely not do this to show nonconsent to (7) would definitely do this to show nonconsent. In addition, a final item asked participants to rank order which 3 of the 12 responses Theresa would be most likely to give to show that she consents and which 3 to show that she does not consent to have intercourse with Rick.
Mean ratings for the verbal scale and for the nonverbal scale were calculated separately. Higher scores on the verbal scale indicated more endorsement of verbal items to communicate consent or nonconsent. Similarly, higher scores on the nonverbal scale indicated more endorsement of nonverbal items to communicate consent or nonconsent.
The questionnaires were distributed during a regular class meeting.
Participants received no incentive for participating, and there was no
penalty for not participating. Each participant received a copy of
an informed consent form that briefly described his/her participation in
the study. Participants were then asked to complete the questionnaire
honestly and told that there were no right or wrong answers. After
completing the questionnaire, participants were asked to deposit them in
a box at the front of the classroom. After the last questionnaire
was collected in each session, the participants were debriefed about the
purpose of the study.
Internal consistency as measured by coefficient alpha for the combined verbal and nonverbal scale was .77, for the verbal scale was .81, and for the nonverbal scale was .84. Thus internal consistency was adequate for research purposes. The verbal and nonverbal scales were analyzed separately because the design of the present study did not permit a determination as to whether the magnitude of responses to the two scales was comparable.
Means and standard deviations for verbal responses are shown in Table 1. A 2 (consent/nonconsent) X 2 (gender) ANOVA showed that those in the nonconsent condition had significantly higher verbal scale scores than did those in the consent condition, F(1, 381) = 158.49, p < .001. Males had significantly higher verbal scale scores than females, F(1, 381) = 19.45, p < .001. There was a significant interaction between consent type and gender, F(1, 381) = 26.44, p < .01. A simple-effects test of the interaction showed that in the consent condition, male scores were significantly higher than female scores, F(1, 186) = 43.96, p < .001. In the nonconsent condition, there were no significant gender differences.
Means and standard deviations for nonverbal responses are shown in Table 2. A 2 (consent/nonconsent) X 2 (gender) ANOVA showed that those in the consent condition had significantly higher nonverbal scale scores than did those in the nonconsent condition, F(1, 381) = 7.72, p < .01. Females had significantly higher nonverbal scale scores than did males, F(1, 381) = 5.42, p < .05. There was no significant interaction between consent type and gender.
Table 3 contains the percent of
men and women choosing each of the top three responses as most likely to
be used to show consent and nonconsent. In the consent condition,
the top three choices for both men and women were nonverbal responses.
In the nonconsent condition, two of the top three choices for both women
and men were verbal and one was nonverbal.
Results for verbal behavior are best summarized by the significant interaction effect which is shown in Table 1. Verbal scale ratings were higher for nonconsent than for consent situations. Ratings were equally high for male and female participants in the nonconsent situation, suggesting that participants of both genders believed that women are more likely to indicate nonconsent verbally. This is consistent with Byers' (1980) finding that a verbal behavior was the top choice reported by her participants for a woman to communicate nonconsent.
Male and female verbal ratings differed significantly for the consent situation, with male ratings higher (by 1.29 points on a 7-point scale), suggesting that male participants believed women are more likely to consent verbally than did female participants. In contrast, Byers found no significant gender differences in her participants' reports of how a woman communicates consent or nonconsent. Hickman (1996) found that females reported using indirect verbal behaviors (e.g., "Do you want to have sex") more than males to indicate consent (a distinction between direct and indirect verbal behaviors was not made in the present study).
Participants reported higher ratings for nonverbal behavior in the consent condition than in the nonconsent condition. This is consistent with Byers' finding that a nonverbal behavior was the top choice reported by her participants for a woman to communicate her consent. In contrast, Hall (1998) found that women were equally likely to report using both verbal and nonverbal behaviors to give consent. Hickman also found that women equally reported using direct verbal behaviors (e.g., "I want to have sex with you") and direct nonverbal behaviors (e.g., just starts having intercourse) to indicate consent.
Female participants in the present study had significantly higher nonverbal scores than males, though the differences were small for both consent (.24 points) and nonconsent (.43 points) conditions. In contrast, Byers found no significant gender differences in her participants' reports of how a female communicates consent or nonconsent.
The selection of response items that would most likely be used
to communicate consent and nonconsent supported the verbal scale and nonverbal
scale findings. Male participants perceived women using more nonverbal
behaviors to communicate consent than nonconsent and more verbal behaviors
to communicate nonconsent than consent. Similarly, female participants
reported that they would use more nonverbal behaviors to communicate consent
and more verbal behaviors to communicate nonconsent.
Practical Implications of Findings
The findings of this study are important for college sex education
and date rape prevention programs. Students should learn that both
men and women appear to perceive women communicating consent for intercourse
nonverbally (e.g., giving intimate kisses) and nonconsent for intercourse
verbally (e.g., "I don't want to have sex"). If nonverbal responses
are typically interpreted as indicating consent, then it is critical that
nonconsent be communicated verbally and quickly so that misunderstandings
do not occur.
Given the conflicting findings between the results of the present study and previous studies, additional research is needed to clarify the relationship between verbal and nonverbal indicators of consent and nonconsent. Gender differences in perceptions of women's responses also need clarification. Methodological differences between studies may account for discrepant findings, and more valid and reliable verbal and nonverbal scales must be developed. In addition to resolving methodological issues, future studies should examine how other types of dating couples besides heterosexual college students perceive consent and nonconsent.
Return to Front Page