Sexual violence, which encompasses acts of forced sex such as rape and unwanted sexual behaviors, can occur through physical force, verbal threats, intimidation, and/or verbal coercion (Institute of Medicine, 2011). Data on sexual assault in the United States reveal that between 25% and 50% of women will report an experience of attempted or completed forced sex during their lifetime (Jozkowski & Sanders, 2012; Tjaden & Thoennes, 2006). Female college students, aged 18-25 years, are significantly more likely to experience sexual violence, compared with women of any other age group and compared with same women of the same age who are not on college campuses (Breiding et al., 2014; Fisher, Cullen, & Turner, 2000; U.S. Department of Justice, 2008). Forced sexual contact has been linked to a variety of poorer health outcomes among women, including anxiety, lower self-rating of health status, emotional problems, depression, and decreased work productivity (Campbell, 2002; Jozkowski & Sanders, 2012).
Background
Sexual consent is most often defined as freely given verbal or nonverbal willingness to engage in sexual activity (Humphreys & Brousseau, 2010; Humphreys & Herold, 2007). Lack of clear consent is a preceding and defining component of forced sex and other forms of sexual violence (Gross, Winslett, Roberts, & Gohm, 2006). Researchers have begun to understand the behaviors, beliefs, and attitudes surrounding sexual consent and their influence on forced and unwanted sexual encounters. College men and women have reported similar definitions of consent, but women were more likely to communicate with their partners using verbal cues, whereas men utilized more nonverbal actions and cues during the consent process. In addition, the level of intimacy dictated how consent was communicated, with participants reporting more direct, verbal approaches to consent involving vaginal intercourse and more nonverbal actions for sexual activities such as kissing, touching, and oral sex (Jozkowski, Peterson, Sanders, Dennis, & Reece, 2014).
Other researchers continue to highlight possible sex differences in the communication of consent. In a qualitative study involving 44 male and female participants between the ages of 18 and 25 years, more than half of the male participants reported behavior that supported established gender norms for sexual behavior. These included viewing women as the gatekeepers of sex, having multiple partners, and the possibility of needing to use coercive behavior to obtain sex. This contrasted with female participants desiring relationships and monogamy (Masters, Casey, Wells, & Morrison, 2013). In another qualitative study, female participants depicted a process of consent for sexual activity that was often implied by the situation and not explicitly verbal or clear (Fantasia, 2011). In this study, participants described wanting sexual consent that was informed and involved both partners, despite describing situations in which consent did not occur. Similarly, a different researcher reviewed the literature on sexual consent and found that participants of both genders used nonverbal communication that was not always clear or direct more often than verbal communication (Beres, 2010).
Finally, college students have been reported to use a combination of verbal and nonverbal communication during the sexual consent process, which may involve traditional sex scripts that reflect cultural norms about sexual activity (Jozkowski & Peterson, 2013). In a study of 185 male and female college students, researchers documented that men used behaviors that could be described as aggressive such as verbally demanding sex and telling female partners that they were going to have sex instead of asking for consent. Female participants reported that they often looked for nonverbal cues from male partners such as the initiation of noncoital behaviors (kissing, touching) as indication that sex might occur.
Researchers have also explored the relationship between alcohol consumption and decisions to engage in sexual activity. Alcohol use among college students has been linked to sexual risk-taking behaviors, including a decrease in safer sexual practices and nonuse of contraception, and perpetration of nonconsensual sexual activity from intoxicated male partners (Abbey & McAuslan, 2004; Goldstein, Barnett, Pedlow, & Murphy, 2007). Alcohol use, a risk factor for sexual victimization, may also be a factor in whether women acknowledge and report the sexual assault because of fear of victim blaming and inability of remembering details (Cleere & Lynn, 2013; Gidycz et al., 2007; Testa & Livingston, 2009).
The impact of alcohol use before consent for sexual activity has also been explored. Among 828 male and female college students, researchers documented that less than half of the female participants reported they would have participated in the sexual encounter if they were sober (Labrie, Hummer, Ghaidarov, Lac, & Kenney, 2014). Other researchers have reported similar results. In a qualitative study of college students' hookup experiences (defined as sexual encounters without the expectations of a relationship), participants reported sexual coercion, especially in the context of alcohol consumption. Approximately 80% of the college students interviewed described consuming alcohol either before or during the hookup, and female participants described verbal and physical coercion by male partners when they were too intoxicated to resist (Downing-Matibag & Geisinger, 2009).
Although much is known about the prevalence of sexual assault, nonconsensual sexual activity, and alcohol use among college students, less is known about how consent for sexual activity is communicated and whether a past history of forced sex is associated with sexual consent communication. Therefore, the primary aim of this study was to explore the association between histories of forced sex and sexual consent awareness, attitudes, beliefs, and behaviors. Given the association between alcohol use and nonconsensual sexual activity documented by previous researchers, a secondary aim examined overall alcohol use among the women in this sample and explored the relationship between alcohol use and a history of forced sex among this sample of college women.
Theoretical Perspectives
This research was conducted using constructs derived from the Theory of Planned Behavior (TPB), which guided the selection of study measures. The TPB is a widely used model that was designed to predict and explain human behavior in specific contexts, in particular, those behaviors over which individuals may have incomplete volitional control (Ajzen, 1991, 2002). As sexual consent involves another person, negotiating and communicating consent may not be under complete control of the individual woman. The TPB helps to determine salient attitudes, beliefs, and controls that predict behavioral intentions and may contribute to the communication of consent. This theoretical framework has been utilized effectively in previous studies with adolescent and young adult populations (Jemmott, Jemmott, Fong, & Morales, 2010; Jemmott, Jemmott, & Villarruel, 2002).
METHODS
Design
After institutional review board approval from the participating university, a cross-sectional electronic survey (via Survey Monkey) was sent to approximately 5900 female undergraduate and graduate students enrolled at a large public university in northeastern United States. The data were collected during the fall semester of 2012. With permission from the institutional review board and Office of Institutional Research and Reporting, student email addresses were obtained, and an introductory email was sent to all female students who were currently enrolled. The email introduced the study and overall purpose and provided a link to the informed consent page of the survey. If students met the inclusion criteria and electronically acknowledged they were eligible, they were allowed to proceed with the survey. Participants were informed that all responses were anonymous, no identifying information was being collected, and any question(s) could be skipped. Completion of the survey implied consent. Students were given the option of entering a contest to win one of two Apple iPod Touch devices that were given away as remuneration for time and participation. The link to the prize survey was separate from the study survey and could not be used to identify individual responses.
Sample
The sample consisted of female college students. Inclusion criteria were (a) currently or previously sexually active, (b) aged 18-25 years, (c) enrolled either full or part time at the university, and (d) able to read and understand English. As this study explored variables related to sexual consent and sexual behaviors, women who had never been sexually active were excluded. The rationale for including students through the age of 25 years is the higher rates of unwanted and nonconsensual sexual activity among this group (Workowski, Berman, & Centers for Disease Control and Prevention, 2010; Flack et al., 2007; Gross et al., 2006). Sample size calculations were estimated using Demidenko's approach for logistic regression techniques (Demidenko, 2007). Power analysis revealed that a minimum of approximately 354 participants was required to achieve a power of 0.80 given an alpha of 0.05 and an estimated odds ratio effect size of 2.
Measures
Forced sex, the outcome variable, was measured by a dichotomous (yes/no) response to the question, "Have you ever been forced to have sex when you didn't want to?". This question was adapted from the Abuse Assessment Screen, a screening tool commonly used in clinical practice (Soeken, McFarlane, Parker, & Lominak, 1998).
Demographics
Demographic characteristics included age, race, gender, ethnicity, living situation, and year in school. Data on race and ethnicity were collected using established terminology (National Institutes of Health, 2001). Additional information was gathered on sexual and reproductive history through standardized questions utilized by the Centers for Disease Control and Prevention (Gavin et al., 2009; Groves, Mosher, Lepkowski, & Kirgis, 2009). Sample questions included age at first intercourse, preferred sexual partner (men, women, or both), relationship status, and type of sexual activity (oral, vaginal, anal) the participant engages in. All participants were also asked if they used alcohol before sexual activity.
Sexual Consent
The Sexual Consent Scale-Revised (SCS-R) is a 39-item, self-report questionnaire with five subscales designed to assess attitudes and behaviors about the negotiation of sexual consent (Humphreys & Brousseau, 2010; Humphreys & Herold, 2007). Developed using constructs from the TPB, the SCS-R uses a 7-point Likert-type scale ranging from 1 (strongly disagree) to 7 (strongly agree) to examine the theoretical constructs of sexual consent. Reliability for the SCS-R is 0.89, and internal consistency for each subscale ranges from 0.72 to 0.85. Subscales measure positive attitudes toward establishing consent ([alpha] = 0.83), lack of perceived behavioral control ([alpha] = 0.85), sexual consent norms ([alpha] = 0.72), indirect behavioral approaches to consent ([alpha] = 0.78), and awareness of consent ([alpha] = 0.72). For this sample, internal consistencies of the subscales were as follows: positive attitudes toward establishing consent ([alpha] = 0.82), lack of perceived behavioral control ([alpha] = 0.91), sexual consent norms ([alpha] = 0.78), indirect behavioral approaches to consent ([alpha] = 0.55), and awareness of consent ([alpha] = 0.75).
Alcohol Use
The Alcohol Use Disorders Identification Test-Consumption Items (AUDIT-C) is a three-item scale used to measure hazardous drinking (Dawson, Grant, Stinson, & Zhou, 2005). The AUDIT-C is a modified version of the 10-question AUDIT. The items ask about frequency of drinking, quantity consumed at a typical occasion, and frequency of heavy episodic drinking. Each item has five answer choices and is scored from 0 to 4. The AUDIT-C is scored on a scale from 0 to 12. Higher scores indicate greater levels of drinking and the greater likelihood that the person's drinking is affecting their safety. Specifically for women, a score of 0 on the AUDIT-C reflects no alcohol use, and a score of >3 reflects hazardous drinking in women. The sensitivity and specificity for hazardous drinking among women are 0.66 and 0.94, respectively (Bradley et al., 2003).
Analysis
All statistical analyses were conducted using SPSS 21 with a p < 0.05 as the criterion for statistical significance. Analyses included descriptive statistics, including tabulation of categorical variables and calculation of means, standard deviations (SDs), ranges of continuous variables, and scores for individual scales. Differences in sexual consent scores were assessed via logistic regression models based on a history of forced sex. Potential confounders of age, race, ethnicity, year in school, and living situation were controlled for via block entered, main-effect covariate adjustments in the final models.
RESULTS
Sample Description
The final sample, with complete data, consisted of 925 female students. The mean age of the participants was 20.6 years. The participants were predominately White (75.2%), non-Hispanic (89.6%), and fairly equally distributed among year in school with 18.7% freshmen, 22.9% sophomores, 24.5% juniors, and 22% seniors. A smaller percentage (11.2%) of participants reported being graduate students. Of those who did not live at home with parents, most resided in the residence halls (33.2%) or in an apartment with roommates (17.2%). The average age of first intercourse was 16.7 years, which is consistent with national data for U.S. adolescents (Guttmacher Institute, 2013). The participants overwhelmingly identified as heterosexual; 94.2% of women reported their preferred sexual partner as male. Relationship status varied, and participants reported multiple types of relationships. Although most participants (66.4%) reported being in a monogamous relationship, approximately 33% also reported being involved in casual hookups with known and unknown partners. Forced sex was reported by 22% of the participants (see Table 1 for complete sample characteristics).
Sexual Consent
Participants reported positive overall attitudes toward communicating sexual consent with their sex partners (mean = 5.75) and had similar positive scores for sexual consent awareness (mean = 4.92). Participants reported using indirect behavioral approaches to communicate sexual consent (see Table 2 for mean score, range, and SD on all sexual consent subscales).
Alcohol Use
Greater than half (59%) of all participants reported alcohol use before sexual activity. Past year use of alcohol was examined, and approximately 24% of all participants reported drinking two or three times per week during the past year. In addition, when asked "how often they consumed six or more alcoholic drinks in a row," 23% of the participants reported drinking this amount monthly.
The AUDIT-C identified women with hazardous drinking patterns. The mean score for the AUDIT-C was 4.07 for the participants and reflected responses to all three questions. Seventy-two percent of female students had an AUDIT-C score of >3, which is considered hazardous drinking for women.
Primary Aim
This logistic regression model explored the association between theoretical constructs related to sexual consent (measured by consent subscales) and a history of forced sex (see Table 3). Awareness of sexual consent and type of behavioral approach to communicate consent were associated with a history of forced sex. Specifically, women with greater awareness of sexual consent were significantly more likely to have a history of forced sex (OR = 1.553, p < 0.000, 95% CI [1.324, 1.820]). In addition, women who utilized more nonverbal, indirect behavioral approaches to communicate sexual consent were significantly less likely to have a history of forced sex (OR = 0.683, p = 0.002, 95% CI [0.537, 0.869]). There were no statistically significant differences found between associations of attitudes, subjective norms, and perceived behavioral control regarding sexual consent and a history of forced sex (see Table 4 for correlation matrix).
Secondary Aim
This logistic regression model examined the relationship between alcohol use and misuse and the history of forced sex. Alcohol use before sex and hazardous drinking were not significantly associated with forced sex (OR = 1.516, p = 0.105, 95% CI [0.916, 2.509]).
DISCUSSION
This is one of the first studies to examine the theoretical constructs of sexual consent awareness, attitudes, beliefs, and behaviors in a large sample of college women who reported a history of forced sex. Examination of these sexual consent constructs is the beginning step in planning and designing effective programs to decrease nonconsensual sexual activity on college campuses.
Two sexual consent variables, constructs from the TPB in the SCS-R measure, were associated with forced sex. Women who had higher scores on the sexual consent awareness subscale were significantly more likely to report a history of forced sex compared with women without a history (p < 0.001). A past experience of forced sex potentially influenced participant's current consent awareness; greater consent awareness could be a result of a previous forced sex experience. An example of questions that measured this construct included "I have discussed sexual consent issues with a friend" and "I have heard sexual consent issues being discussed by other students on campus." Although there have been no studies examining awareness of consent postassault, it is likely that students who had a history of forced sex may have discussed the incident with friends or peers, even if they did not report the assault to law enforcement or seek physical or mental health care (Fisher, Daigle, Cullen, & Turner, 2003). In addition, it is possible that the experience of forced sex made subsequent sexual consent an important issue for them.
Use of indirect behavioral approaches to communicate sexual consent was a statistically significant sexual consent construct (p = 0.002). Women who used more nonverbal communication cues during the consent process were significantly less likely to report a history of forced sex. Sample questions that measured these behavioral approaches included "I don't have to ask or give my partner sexual consent because I have a lot of trust in my partner to 'do the right thing'" and "Typically I communicate sexual consent to my partner using nonverbal signals and body language." Although it is impossible to determine causality related to forced sex, nonverbal communication patterns have the potential to be misinterpreted compared with verbal communication that may be more clear and direct. Therefore, women who have not experienced forced sex may be more comfortable with nonverbal cues and communication, whereas women with a history of forced sex may use communication strategies that are more direct and clear. Prior victimization may influence communication patterns and place a higher value on direct, verbal communication for women who have experienced a past episode of forced sex.
Similar to existing research on college students and alcohol use (White & Hingson, 2014), the participants in this study reported high rates of alcohol use. Alcohol use has been associated with sexual assault and nonconsensual sexual activity (Abbey & McAuslan, 2004; Downing-Matibag & Geisinger, 2009; Goldstein et al., 2007). Previous researchers have suggested that high rates of alcohol consumption among college students may lead to decreased sexual decision making, difficulty refusing sexual activity, and a poorly communicated refusal that could be misinterpreted by a sexual partner (Masters et al., 2013). In this analysis, alcohol use before sex and hazardous drinking (measured by the AUDIT-C) were not significantly associated with forced sex. Although these results are not consistent with previous findings, factors such as timing of alcohol consumption, exact amount of alcohol used, and partner characteristics were not measured in this study. Conclusions related to alcohol findings are limited. Further exploration of these associations is an important area for future research.
Limitations
The results of this study must be viewed in terms of the limitations. The research was conducted at one university, and therefore, results are not generalizable to all college women or women who are not enrolled in college. In addition, the sample primarily consisted of students who self-identified as non-Hispanic and White, and results may not be representative of students of other races and ethnicities. The cross-sectional design precludes conclusions related to temporality and the timing of forced sex in relation to reported awareness, attitudes, behaviors, and beliefs regarding sexual consent as well as alcohol use. The dichotomous (yes/no) measurement of forced sex has inherent limitations. With all self-report measures, issues of social desirability, reporting bias, and selection bias may occur. The response rate was lower than expected (16%). Previous researchers have documented an approximately 30% response rate when using electronic surveys to collect violence data (Sutherland, Amar, & Laughon, 2013). However, the lower response rate could have been associated to a weather-related widespread and prolonged loss of power in the geographical area during study recruitment.
Despite these limitations, the sample was large, included nearly 1000 college women, and was representative of the university population. The selection of study variables was guided by the use of a theoretical framework and was one of the first studies to examine the relationship between awareness, attitudes, and beliefs including social norms, perceived behavioral control, and behavioral approaches associated with sexual consent and a history of forced sex.
Additional research is necessary to better understand how the theoretical constructs associated with sexual consent may vary based on the timing of an episode of forced sex or multiple occurrences of forced sex as those were not measured in this study. Given the high prevalence of nonconsensual and forced sex among women on college campuses, longitudinal research that follows women throughout the college years would be beneficial in the design of interventions to reduce sexual violence. Longitudinal designs may also give insight into potential changes in sexual consent behaviors related to the timing of forced sex, especially theoretical constructs that relate to direct and indirect behavioral approaches to sexual consent. Inclusion of male students, male-female dyads, and students in same-sex relationships in future research will assist in uncovering sex and gender differences that potentially exist in sexual consent behaviors among men and women. Finally, testing of interventions to increase sexual consent awareness and behavioral approaches, particularly consent communication techniques, should be a necessary part of comprehensive sexual violence prevention programs.
Clinical and Forensic Implications
Healthcare providers must be aware of the high rates of forced sexual activity among women. The reported rate of forced sex among study participants was consistent with national data (Jozkowski & Sanders, 2012; Tjaden & Thoennes, 2006). Providers who work with women, especially those considered to be "college aged" (18-25 years), need to be mindful of the risk of sexual violence. When reviewing sexual activity and relationship status with female patients, healthcare providers should inquire about any history of forced sex. Recommendations for sexual safety can be provided, and providers should be prepared to provide local referrals for counseling or victim services if requested by the patient.
Previous research has linked alcohol use with nonconsensual sexual activity (Abbey & McAuslan, 2004; Goldstein et al., 2007). Although that link was not found in this study, approximately 72% of the women surveyed reported episodes of hazardous drinking in the past year, and 59% reported they consumed alcohol before sexual activity. Healthcare providers should provide information on the risks of alcohol use in general and before sexual activity, including the possibility of impaired judgment and decision making. This discussion should also include the risk of nonconsensual sexual activity when unable to provide consent because of intoxication.
Forensic nurses have the expertise and training to make significant contributions to the health and safety of college women. They are in a key position to consult with colleges and universities to provide programmatic input on existing or developing sexual assault programs. In addition, they can have an invaluable role in the review of policies and procedures around sexual assault and victim services as alcohol use is widespread on college campus by both male and female students and alcohol has been linked to sexual assault (Abbey & McAuslan, 2004; Goldstein et al., 2007; White & Hingson, 2014). It is important to seek guidance on alcohol awareness messages. Raising awareness of potential risks associated with alcohol use must be structured carefully to avoid any victim blaming or shaming that could result in reluctance to seek health care or fear of reporting.
CONCLUSION
Forced sex is a significant public health issue associated with immediate and long-term health consequences for college women. This study helps to illuminate variables related to sexual consent and their potential associations to forced sex. Healthcare providers, both on and off campus, need to understand the full spectrum of sexual risks for college students including the myriad of contextual, social, individual, and partner characteristics involved in forced sexual activity. Understanding sexual consent awareness, attitudes, beliefs, and behavioral approaches to communication is an important step in intervention development for the reduction of sexual violence among college women.
References