Comprehension of Sexual Consent as a Key Factor in the Perpetration of Sexual Aggression Among College Men

Full Terms & Conditions of access and use can be found athttp://www.tandfonline.com/action/journalInformation?journalCode=wamt20

Download by: [Peter Warren] Date: 29 September 2015, At: 06:46

Journal of Aggression, Maltreatment & Trauma

ISSN: 1092-6771 (Print) 1545-083X (Online) Journal homepage: http://www.tandfonline.com/loi/wamt20

Comprehension of Sexual Consent as a Key Factorin the Perpetration of Sexual Aggression AmongCollege Men

Peter Warren, Suzanne Swan & Christopher T. Allen

To cite this article: Peter Warren, Suzanne Swan & Christopher T. Allen (2015) Comprehensionof Sexual Consent as a Key Factor in the Perpetration of Sexual Aggression AmongCollege Men, Journal of Aggression, Maltreatment & Trauma, 24:8, 897-913, DOI:10.1080/10926771.2015.1070232

To link to this article: http://dx.doi.org/10.1080/10926771.2015.1070232

Accepted online: 15 Jul 2015.

Submit your article to this journal

Article views: 11

View related articles

View Crossmark data

Comprehension of Sexual Consent as a KeyFactor in the Perpetration of Sexual Aggression

Among College Men

PETER WARREN and SUZANNE SWANDepartment of Psychology, University of South Carolina, Columbia, South Carolina, USA

CHRISTOPHER T. ALLENUniversity of Massachusetts, Lowell, Lowell, Massachusetts, USA

The comprehension of sexual consent is a crucial factor inhealthy sexual relationships. This study examined the connec-tion between the understanding of sexual consent and perpe-tration of sexual aggression. We surveyed 217 heterosexualmale college students (M age = 20.9 years) using measures ofsexual aggression, comprehension of sexual consent, rape mythacceptance, conformity to masculine norms, peer support ofabuse, and attachment to abusive peers. We tested modelsexamining factors related to comprehension of consent andthe extent to which comprehension of consent was related toperpetration of sexual aggression. Rape myth acceptance, peersupport of abuse, and conformity to masculine norms werefound to predict comprehension of consent, which mediatedthe relationship between the social and cognitive variablesand sexual aggression.

KEYWORDS adult, college, intimate partner violence, males,relationships, sexual abuse, sexual aggression, sexual assault,sexual consent

Sexual assault, defined as any kind of nonconsensual sexual activity, is apervasive problem for women during their college years. The Sexual

Received 8 April 2014; revised 22 April 2015; accepted 23 April 2015.Address correspondence to Peter Warren, Department of Psychology, University of South

Carolina, 1512 Pendleton St., Columbia, SC 29205. E-mail: [email protected]

Journal of Aggression, Maltreatment & Trauma, 24:897–913, 2015Copyright © Taylor & Francis Group, LLCISSN: 1092-6771 print/1545-083X onlineDOI: 10.1080/10926771.2015.1070232

897

Dow

nloa

ded

by [

Pete

r W

arre

n] a

t 06:

46 2

9 Se

ptem

ber

2015

Victimization of College Women Study (Fisher, Cullen, & Turner, 2000)reported a prevalence rate of 27.7 attempted or completed rapes per 1,000female students during the single academic year studied. When this annualrate is projected over the current average of 5 years that students attendcollege, as many as one fifth to one quarter of college women could beexpected to experience at least one attempted or completed sexual assault.Unfortunately, more recent studies (e.g., Fisher, Daigle, & Cullen, 2010;Kilpatrick, Resnick, Ruggiero, Conoscenti, & McCauley, 2007) have found noevidence that these rates are declining.

The perpetration of sexually aggressive acts has been associated withfactors that imply a lack of concern for others’ boundaries and rights (e.g.,wanting power, dominance, and control; Schatzel-Murphy, Harris, Knight, &Milburn, 2009) and flawed interpretation of cues regarding those boundariesand rights (Abbey, Clinton-Sherrod, McAuslan, Zawacki, & Buck, 2003; Farris,Treat, & Viken, 2010). The goal of this study was to create a model foridentifying factors that are related to the perpetration of sexual aggression,starting from a core internal cognitive perspective of the perpetrator andworking backward to uncover potential factors that might predict the devel-opment of those aggression-predictive cognitions.

As mentioned earlier, one of the key elements that allows perpetration ofsexual aggression to occur is disregard for the rights and boundaries ofanother, either through the belief that the victim’s rights are not as importantas the perpetrator’s, or a faulty understanding of those rights (e.g., men are“entitled” to sex with a woman in certain situations). In contrast, sexualconsent involves people freely communicating a willingness to engage insexual activity (Hickman & Muehlenhard, 1999). For a person to grant consentfor a sexual act, that person must meet the local age of consent, have thephysical and mental capacity to provide that consent, and all parties involvedmust agree to participation (Rape, Abuse, and Incest National Network, 2009).For a person to commit a sexually aggressive act, the perpetrator must makethe decision to disregard not only another’s wishes and boundaries, but alsothe basic human right to consent to sexual activity. Men who perpetratesexual assault tend to be more likely to misperceive a woman as beingreceptive to sexual overtures than those men who have not perpetratedsexual assault (Abbey & McAuslan, 2004; Ward, 2000).

Given that comprehension of sexual consent represents a set of beliefsregarding boundaries, rights, and respect for others, it becomes clear thatunderstanding when another person has given sexual consent could bepredicted by a number of social and cognitive factors related to violence.Adams-Curtis and Forbes (2004), in a review of literature focusing onsexual coercion of college women, found that male perpetrators tend tobe characterized by a set of related variables: rape-supportive beliefs ofsexual behavior, extreme or dysfunctional views of masculine norms andgender roles, and peer groups that are encouraging of the preceding

898 P. Warren et al.

Dow

nloa

ded

by [

Pete

r W

arre

n] a

t 06:

46 2

9 Se

ptem

ber

2015

attitudes. What has not been examined in previous research is the role ofcomprehension of sexual consent as a cognitive factor that contributes tothe perpetration of sexual aggression. Specifically, we proposed thatcomprehension of sexual consent would mediate the relationship betweencognitive, attitudinal, and social factors related to violence and theperpetration of sexual aggression. The next sections discuss each of thesefactors.

FACTORS RELATED TO COMPREHENSION OF SEXUAL CONSENT

Cognitive Predictors of Comprehension of Sexual Consent

Violation of others’ rights typically involves at least some form of internaljustification on the part of the perpetrator. Rape myths, or the acceptanceof beliefs that justify sexual violence against women, are related to awillingness to commit sexual assault and sexual coercion (Bohner, Jarvis,Eyssel, & Siebler, 2005). Rape myths blame women for provoking asexual assault (e.g., by the types of clothing they wear, by theirdemeanor, by being alone, by drinking, by being out at night, etc.;Payne, Lonsway, & Fitzgerald, 1999). Additionally, men who are morelikely to misinterpret a woman’s sexual interest tend to find sexualcoercion more acceptable than those who have a better understandingof sexual consent (Abbey, Parkhill, Jacques-Tiura, & Saenz, 2009). Rapemyths function to reduce an individual’s cognitive dissonance at commit-ting or endorsing acts that are morally unacceptable by blaming thevictim and are indicative of a more general motive to “rationalize theway things are so that existing social, economic, and politicalarrangements tend to be perceived as fair and legitimate” (Jost &Hunyday, 2005, p. 206).

Another powerful predictor of sexual assault is the endorsement ofextreme beliefs of masculinity and traditional gender roles (Greene &Davis, 2011). Although so-called traditional views of masculinity can involvepositive traits, such as independence and resilience (Kilmartin, 2009), themasculine gender role is often related to beliefs of power and dominanceover women and the superiority of the male gender. Furthermore, researchhas indicated that negative beliefs about masculinity are related to actualabusive behaviors (McDermott & Lopez, 2013). Malamuth’s (1998) conflu-ence theory of sexual assault states that such beliefs and attitudes areassociated with sexual assault perpetration among college males, especiallythose who tend to engage in indiscriminate sexual behavior. Additionally,traditional masculinity is frequently characterized by the belief that violenceis an acceptable strategy to achieve a desired goal (Tilley & Brackley, 2005).Taken together, the factors that make up this set of cognitive heuristicsregarding male behavior have been shown to be related to the perpetration

Comprehension of Consent and Sexual Aggression 899

Dow

nloa

ded

by [

Pete

r W

arre

n] a

t 06:

46 2

9 Se

ptem

ber

2015

of violence and sexual aggression in a relationship context (Adams-Curtis &Forbes, 2004; Allen, Swan, & Raghavan, 2009). Specifically, given the focuson dominance, power, and superiority over women, traditional male genderroles might also be associated with lack of regard for the rights, boundaries,and sexual consent of others.

Social and Normative Predictors of Sexual Consent

In addition to the previously mentioned cognitive factors, the impact of socialinfluence and social norms of peer groups are important factors in attemptingto understand sexual aggression (Fabiano, Perkins, Berkowitz, Linkenbach, &Stark, 2003). Both exposure to violence and environments that are encoura-ging of abuse have consistently been found to be related to the perpetrationof violence (Malamuth, 1998). For men, peer groups exert powerful influenceover attitudes toward women and beliefs regarding appropriate behavior.There is also some evidence to suggest that a man’s peer group acceptanceof sexual aggression is predictive of his own perpetration of sexual aggression(Rosen, Kaminski, Parmley, Knudson, & Fancher, 2003). These peer groupsare important to the internalization of social norms regarding masculinity, aswell as associated beliefs regarding appropriate behavior in relationships(Casey & Beadnell, 2010). As such, attachment to and association with thosepeers can strongly influence an individual man’s ideas about appropriatesexual behavior in relationships.

THIS STUDY

This study examined the relationship between comprehension of sexualconsent and the perpetration of sexual aggression. Furthermore, we alsoinvestigated the predictive relationships among comprehension of sexualconsent and related cognitive and social factors, examining comprehensionof sexual consent as a mediator of the relationship between social andcognitive factors and sexual aggression. To accomplish these aims, we usedpath analysis to examine the links between comprehension of consent andthe perpetration of sexual aggression. Rape myth acceptance, peer support ofabuse, attachment to abusive peer groups, and conformity to masculine normswere examined as predictors of comprehension of sexual consent. In theproposed model, we hypothesized that comprehension of sexual consentwould mediate the relationship between person-level cognitive factors (rapemyth acceptance and conformity to masculine norms) as well as social peer-level influences (peer support of abuse and attachment to abusive peers) onthe perpetration of sexual aggression.

900 P. Warren et al.

Dow

nloa

ded

by [

Pete

r W

arre

n] a

t 06:

46 2

9 Se

ptem

ber

2015

METHOD

Participants and Study Procedures

A convenience sample of 217 self-identified heterosexual males was recruitedfrom available undergraduate psychology classes at a large university in thesoutheastern United States. Participants ranged in age from 18 to 46 years old(M = 21.07 years, SD = 3.3 years). Sexual orientation was assessed as part of ademographics questionnaire. Because the focus of the study was male-on-female sexual aggression, only heterosexual males were retained for thesample.

One hundred fifty-three (72.9%) participants identified themselves asWhite, 33 (15.7%) identified as African American, and 24 (11.4%) were ofanother racial background, mixed, or did not report. Parental levels of educa-tion (for both mother and father) ranged from grade school to doctorate, withthe modal level reported as having completed a 4-year college degree.Regarding participants’ academic year, the sample consisted of 25 (11.9%)freshmen, 50 (23.8%) sophomores, 64 (30.5%) juniors, 63 (30.0%) seniors, and8 (3.8%) students who did not report. Forty-five (21.4%) participants statedthey belonged to a fraternity, whereas 159 (75.7%) did not belong to afraternity, and 6 (2.9%) did not report. The mean grade point average for allparticipants was 3.2 (minimum = 2.0, maximum = 4.0, SD = .49). Sixty-five(31.0%) participants reported being single, 45 (21.4%) were dating, 74 (35.2%)were in a relationship, 8 (3.8%) were in a cohabitating relationship, 9 (4.3%)were engaged, 5 (2.4%) were married, and 4 (1.9%) did not report.

Surveys were administered in either a classroom setting using a pen-and-paper form or as an online survey, and students were offered a financialincentive or course credit for participation. No statistical differences werefound between online and classroom-administered groups on any variablesof interest. Informed consent was obtained from all participants, and this studywas approved by the university’s institutional review board.

Measure of Sexual Experiences

SEXUAL EXPERIENCES SCALE

The Sexual Experiences Scale (SES–Short Form; Koss et al., 2007) is aninstrument used to assess sexual aggression and victimization; the victimiza-tion data were not examined in this study. This 30-item self-report measureuses items that ask the respondent to rate the frequency with which he hasperpetrated sexual aggression as well as the means by which the violationtook place (e.g., verbal pressure, showing displeasure or getting angry,through intoxication, threat of physical harm, or by actual use of force). Anexample of an item indicating perpetration of sexual aggression is “I had oralsex with someone or had someone perform oral sex on me without their

Comprehension of Consent and Sexual Aggression 901

Dow

nloa

ded

by [

Pete

r W

arre

n] a

t 06:

46 2

9 Se

ptem

ber

2015

consent by . . .,” with the respondent identifying the means of aggression froma list of verbally, physically, and substance-abuse-related coercive acts.

Although the SES–Short Form asks respondents if they have had eachexperience in the last 12 months, we were interested in focusing on experi-ences within the past semester, so we modified the time frame such thatrespondents indicated if an experience had occurred within the past 4months. This time frame allowed for us to assess the most recent events inthe respondents’ lives. Previous studies using the SES have found low baserates when the use of sexually aggressive behaviors is categorized (DeGue,DiLilo, & Scalora, 2010). To address this, we calculated a dichotomous vari-able indicating whether or not a given respondent had committed any act ofsexual aggression in the past 4 months.

Measures of Cognitive Predictors

COMPREHENSION OF SEXUAL CONSENT/COERCION SCALE (CCC)

This 10-item instrument, created by Gibson and Humphrey (1993), wasdesigned to measure a respondent’s perception that sexually coercive oraggressive behaviors are acceptable methods to obtain consent for sexualactivity. Respondents were instructed to read two vignettes describing situa-tions in which consent is not given and sexual coercion or assault occurs.They then rated their agreement on a 5-point Likert scale ranging from1 (strongly agree) to 5 (strongly disagree) with a series of five statementsregarding the level of consent described in each vignette.

An abbreviated sample vignette from the CCC is as follows:

Mai and Kirk are new students at a small college. . . . Tonight, they hadtheir first date. After coming back to Kirk’s apartment, he mixes somedrinks using his brother’s stock of alcohol. They have each had severaldrinks. Mai is clinging to Kirk, laughing and giggling . . . Suddenly, shekisses him on the lips. Kirk is surprised but kisses her back. Mai walksover to a sofa and he joins her. Kirk pushes her down on the couch andbegins pulling his clothes off. Mai pushes against him as if attempting topush him away. She says nothing. Kirk gets undressed while kissing Mai.He begins to undress her. Mai freezes for a moment: She is afraid. Thenshe tries to put her clothes back on, but Kirk grabs them out of her handand throws them across the room. He pushes her back on the sofa. Mailooks at the clothes and up at Kirk; for a moment she is unsure what to do.She is afraid of what Kirk will do if she tries to get the clothes. She tries toget up from the couch but has trouble getting her balance. She falls backonto the couch. Kirk begins to kiss Mai and to have sex with her.

An example of an item measuring understanding of sexual consent fromthe CCC is:

902 P. Warren et al.

Dow

nloa

ded

by [

Pete

r W

arre

n] a

t 06:

46 2

9 Se

ptem

ber

2015

Kirk mixed a number of really strong drinks for the two of them. Hethought that this might make them more relaxed and might lead tosomething happening. It was okay for Kirk to have sex with Mai afterdoing this.

Notably, the CCC does not include the word consent. Rather, it depictsscenarios in which sexual activity occurs, even though consent is not given.Questions are then asked about aspects of the scenario to ascertain if respon-dents think that the behaviors are acceptable. Higher scores indicate that respon-dents believe that the coercive behaviors used in the vignettes to obtainnonconsensual sex are not acceptable. This measure offers examples that areboth realistic and relevant to the college context, and it has been recommendedfor use in college interventions assessing understanding of sexual consent (Davis,2000). The CCC has also been previously used with a college population (Potts,2009). Cronbach’s α for the CCC in this study was .73 (M = 4.14, SD = 0.65).

ILLINOIS RAPE MYTH ACCEPTANCE SCALE

Rape myth acceptance was measured using a shortened version of the IllinoisRape Myth Acceptance Scale (IRMA; Payne et al., 1999). This 20-item instru-ment was used to assess the degree to which an individual endorses items thatperpetuate rape myths and blame victims for being assaulted. Respondentswere instructed to rate their endorsement of each item, with higher scoresindicating greater degrees of rape myth acceptance. The response scaleranges from 1 (not at all agree) to 7 (very much agree). An example item ofthis measure is “When women are raped, it’s often because the way they said‘no’ was ambiguous.” For this study, we used a shortened (7-item) version inan effort to reduce the overall length of the survey. This shortened (7-item)version of the IRMA had a Cronbach’s α of .76 (M = 2.33, SD = 0.86).

CONFORMITY TO MASCULINE NORMS INVENTORY

The Conformity to Masculine Norms Inventory (CMNI; Mahalik et al., 2003) isa 94-item instrument designed to assess the degree to which an individualendorses stereotypically masculine gender role attitudes and beliefs. Respon-dents were instructed to rate their endorsement of each item on a 4-pointLikert scale ranging from 0 (strongly disagree) to 3 (strongly agree), withhigher scores indicating greater adherence to traditional masculine genderroles. An example item from the CMNI is “I often feel I need to be in charge ofthose around me.” Although this scale uses 11 subscales to measure specificaspects of masculinity, this study examined the overall scale to assess generaladherence to the masculine gender role. Cronbach’s α for the CMNI in thisstudy was .93 (M = 1.40, SD = 0.86).

Comprehension of Consent and Sexual Aggression 903

Dow

nloa

ded

by [

Pete

r W

arre

n] a

t 06:

46 2

9 Se

ptem

ber

2015

Social and Normative Predictors

PEER SUPPORT OF ABUSE SCALE

The Peer Support of Abuse Scale (PSA; DeKeseredy, 1988) is a 3-iteminstrument used to assess the degree to which an individual reports hismale peers endorse attitudes encouraging or supportive of abuse or violenceagainst women. Respondents were instructed to rate their agreement witheach item on a 7-point Likert scale ranging from 1 (strongly disagree) to7 (strongly agree), with higher scores indicating a higher degree of supportfor violence against women in a person’s peer groups. These items are:“Would your male friends agree or disagree: (1) Your dates or girlfriendsshould have sex with you when you want; (2) If a man spends money on adate, she should have sex with him in return; (3) You should respond toyour dates’ or girlfriends’ challenges to your authority by insulting them orputting them down.” Cronbach’s alpha for the 3-item version of the PSAScale in this study was .77 (M = 2.48, SD = 1.29).

ATTACHMENT TO ABUSIVE PEERS SCALE

The Attachment to Abusive Peers Scale (AAP; DeKeseredy & Kelly, 1995) is a10-item instrument, the first seven items of which were designed to measurethe frequency with which an individual’s male friends report committingabusive acts toward women. These seven items were summed such thathigher scores indicate more frequent exposure to male peers’ reports ofabusive acts. Respondents were instructed to rate their endorsement of eachitem on a 5-point Likert scale ranging from 1 (never) to 5 (almost always),with higher scores indicating more frequent reporting of acts of violence orsexual aggression toward women by the men in the respondents’ peergroups. An example item from the AAP is, “How often or seldom have yourmale friends made physically forceful attempts at sexual activity with womenthey were dating which were disagreeable and offensive enough that thewomen responded in an offended manner such as crying, fighting, screamingor pleading?” Cronbach’s alpha for the AAP in this study was .70 (M = 1.57,SD = 0.50).

Testing for Mediation

To examine if comprehension of sexual consent served as a mediator ofthe relationships between our predictor variables and sexual aggression,we used MacKinnon, Lockwood, and Hoffman’s (1998) method by whichwe constructed asymmetric confidence intervals using bias-correctedbootstrap sampling methods to accommodate the nonnormal distributionof the intervening variable effect based on the distribution of the product

904 P. Warren et al.

Dow

nloa

ded

by [

Pete

r W

arre

n] a

t 06:

46 2

9 Se

ptem

ber

2015

of random variables. Based on this method, if the confidence intervaldoes not include zero, the intervening variable effect is significant.According to the results of the power analysis method developed byFritz and MacKinnon (2007), a sample size of 148 is needed to achieve.80 power to detect the mediated effect when the effect size of the α andβ paths are between small and medium (.26). Using this criterion, thecurrent sample of 217 has adequate power to detect such an effect.

RESULTS

Descriptive and Baseline Data

It was found that 10% of the sample (n = 22) reported perpetration of sexualaggression in the last 4 months. Ten respondents reported having sexualcontact, five attempted coercion, three used coercion, six attempted rape,and four committed rape. Four participants reported lying or misleadingtheir victims, eight endorsed showing displeasure regarding criticism as ameans of coercion, five reported committing drug- or alcohol-facilitatedassault, five endorsed using threat to the victim or others, and two reportedusing physical force to commit sexual aggression against a victim. It was notedthat these figures were not exclusive because a single person might haveemployed more than one method to carry out acts of sexual aggression (Kosset al., 2007). Table 1 details the zero-order correlations among all of the lowerorder predictors used in his study; no evidence of multicollinearity or singu-larity was present. All variables of interest were examined for skew andkurtosis and found to be within acceptable limits.

Path Analyses

Path analysis was conducted using Mplus software (Muthén & Muthén, 2007).The greatest percentage of missing data for any variable in any analysis wasonly 2.8%. As Mplus uses full information maximum likelihood to estimate

TABLE 1 Zero-Order Correlations Among Predictors

Illinois Rape MythAcceptance Scale

(IRMA)

PeerSupport ofAbuse (PSA)

Attachment toAbusive Peers

(AAP)

Conformity toMasculine

Norms (CMNI)

Comprehensionof Sexual

Consent (CCC)

IRMA 1PSA .29** 1AAP .25** .33** 1CMNI .40** .41** .30** 1CCC −.35** −.38** −.21* −.38** 1

*p < .01. **p < .001.

Comprehension of Consent and Sexual Aggression 905

Dow

nloa

ded

by [

Pete

r W

arre

n] a

t 06:

46 2

9 Se

ptem

ber

2015

missing parameters, missing data were not a concern (Young, Weckman, &Holland, 2011).

Preliminary mediation analyses revealed that attachment to abusive peershad no direct effect on the hypothesized mediator, comprehension of sexualconsent. As a relationship is a prerequisite for mediation (Baron & Kenny,1986), attachment to abusive peers was removed from the model as a med-iator but retained as a covariate due to its significant correlation with all otherpredictor variables (see Table 1). A model that is generally interpreted asproviding a good fit to the data is one that has a nonsignificant chi-squarevalue, a root mean square error of approximation (RMSEA) value ≤ .06, and acomparative fit index (CFI) and Tucker–Lewis Index (TLI) value greater thanor equal to .95 (Hu & Bentler, 1999). Fit statistics for the hypothesized pathmodel were excellent (see Figure 1), χ2(2, N = 217) = .30, p = .82,RMSEA = 0.00, TLI = 1.08, CFI = 1.00.

As shown in Figure 1, rape myth acceptance, conformity to masculinenorms, and peer support of abuse all demonstrated negative relationshipswith comprehension of sexual consent. In other words, greater acceptance ofrape myths, more conformity to masculine norms, and greater peer support ofabuse all predicted less comprehension of sexual consent (β = –.24, p ≤ .001).Comprehension of sexual consent demonstrated a negative relationship toperpetration of sexual aggression; greater comprehension of sexual consentpredicted lower levels of sexual perpetration (β = –.32, p ≤ .01).

Rape Myth

Acceptance

Conformity to

Masculine Norms

Peer Support of

Abuse

Attachment to

Abusive Peers

Perpetration of

Sexual Aggression

Comprehension

of Sexual Consent

.09

–.06

.12

–.32*

–.20**

–.24**

–.21**

.28**

.40**

.25**

.34**

.41**

.31**

FIGURE 1 Results for final path model. *Significant at p ≤ .01. **Significant at p ≤ .001.

906 P. Warren et al.

Dow

nloa

ded

by [

Pete

r W

arre

n] a

t 06:

46 2

9 Se

ptem

ber

2015

Results indicated the indirect effect of rape myth acceptance onperpetration of sexual aggression via comprehension of sexual consentto be .06. The bootstrap confidence interval for the indirect effect of rapemyth acceptance on perpetration of sexual aggression via comprehensionof sexual consent did not include zero, 95% CI [0.01, 0.12], p = .05,indicating a significant mediated effect. Similarly, the indirect effect ofconformity to masculine norms on perpetration of sexual aggressionthrough comprehension of sexual consent was found to be .07. Thebootstrap confidence interval for the indirect effect of conformity tomasculine norms on perpetration of sexual aggression via comprehensionof sexual consent also did not include zero, 95% CI [0.02, 0.11], p = .02,again indicating a significant mediated effect. Finally, the indirect effect ofpeer support of abuse on perpetration of sexual aggression via compre-hension of sexual consent was .08. The bootstrap confidence interval forthe indirect effect of peer support of abuse on perpetration of sexualaggression via comprehension of sexual consent did not include zero,95% CI [0.03, 0.13], p = .01, once again indicating a significant mediatedeffect.

DISCUSSION

Consistent with our predictions, we found that a lack of comprehension ofsexual consent was strongly related to self-reported perpetration of sexualaggression in the past 4 months. We also identified three other social andcognitive risk factors that had an indirect effect on sexual aggression: rapemyth acceptance, conformity to masculine norms, and peer support of abuse.Each of these risk factors was fully mediated by comprehension of sexualconsent. Thus, an individual’s understanding of sexual consent might beshaped and reinforced by abuse-supportive social and normative factors thatcharacterize their peer groups as well as by adherence to traditional socialnorms regarding masculinity. This finding that greater internalization of mas-culine norms is related to poorer understanding of consent suggests thatunderstanding sexual consent might be a function of a man’s belief systemregarding dominance, power, strength, and entitlement. Thus, if a man sub-scribes to masculine norms that encourage these beliefs, he might be lesslikely to develop a healthy sense of boundaries and respect for the rights ofwomen, which can in turn lead to an increased propensity for the perpetrationof sexual aggression.

Research has shown that attitudes supportive of abuse and beliefs regard-ing traditional masculinity and violence toward women are transmitted bothgenerationally (Schafer, Caetano, & Cunradi, 2004) and among peers as youngmen develop male-to-male peer group relationships (Schwartz & DeKeser-edy, 1997). In addition, it is important to note that although abusive men tend

Comprehension of Consent and Sexual Aggression 907

Dow

nloa

ded

by [

Pete

r W

arre

n] a

t 06:

46 2

9 Se

ptem

ber

2015

to belong to abusive peer groups, it might not always be that the group makesthem more abusive; instead, abusive people might tend to associate with eachother, reinforcing their existing beliefs and providing an environment that issupportive of abusive or violent acts (Raghavan, Rajah, Gentile, Collado, &Kavanagh, 2009). Taken together, these findings suggest that peer reinforce-ment, hypermasculine belief systems, and social affiliation reinforce the crea-tion of unhealthy attitudes regarding acceptable male behavior. In thisscenario, internal predispositions for violence toward women might combinewith social learning factors (Bandura, 1971) to create and perpetuate a set ofbeliefs that lend themselves to a poorer understanding of sexual consent as afunction of hostile and dysfunctional masculinity.

Whereas we found that peer support of abuse was significantly asso-ciated with lack of understanding of sexual consent, we found no suchrelationship between attachment to abusive peers and comprehension ofsexual consent. Although this might seem somewhat counterintuitive, oneexplanation might be that abuse-supportive messages conveyed by peersare much more common than peers engaging in abusive behaviors againstwomen. Indeed, our data showed a higher mean for peer support of abuseand a very low mean for peers who actually perpetrated abusive behavior.Although the vast majority of men do not engage in sexual assault (Kilmartin& Allison, 2007), many men who would not otherwise act abusively mightespouse abuse-supportive and victim-blaming beliefs when among their peergroups as a function of social norms regarding masculinity (Kimmel, 2000).

Implications

Our findings suggest at least two pathways that could lead to sexual aggres-sion. The first is a lack of understanding of sexual consent. If a potentialperpetrator does not understand what sexual consent is or why it is important,educating students about the issue of consent via sexual assault preventionprograms might be effective. The second is that the perpetrator knows that aperson did not or cannot consent (e.g., due to the influence of drugs oralcohol), but believes that having sex with that person is acceptable anyway.The perpetrator in this case might see nothing wrong with being manipulative,being coercive, using threats, or getting someone intoxicated to obtain sex.

This second avenue points to the importance of peer reinforcement ofattitudes that encourage or excuse male violence against women. Our datasuggest that peer attitudes are an important component of these beliefs.Simply teaching individual men appropriate rules and limitations regardingbehavior toward sexual partners might have some utility in the case oflegitimate skill deficits, but in most cases, this individual educational approachmight miss some key points of intervention. Rather, intervening with groupsof peers might be helpful in decreasing the propensity for sexual violence.

908 P. Warren et al.

Dow

nloa

ded

by [

Pete

r W

arre

n] a

t 06:

46 2

9 Se

ptem

ber

2015

Beliefs excusing sexual aggression are normative within some peer groups,and such change might be best achieved by a saturation approach wherein agiven peer group might reach a critical mass of attitude change by beingrepeatedly exposed to messages discouraging unhealthy attitudes towardsexual consent and encouraging of respect and more egalitarian sexual rela-tionships (McMahon & Dick, 2011).

Furthermore, masculinity represents a set of learned beliefs and beha-viors that have been central to a man’s development from an early age andreinforced by family, friends, and society. Bound up and perhaps inexplicablyintertwined in those beliefs are default rules of behavior toward women andan understanding of what is appropriate regarding sexual partners. As statedearlier, changing these rules could be accomplished through changing men’sunderstanding of acceptable behavior toward their intimate partners. How-ever, this would involve undermining many long-held and strongly reinforcedbeliefs and scripts regarding not only a man’s own behavior, but, to somemen, also the validity of the very idea of masculinity. Interventions that focuson modifying these normative beliefs of what it is to be a man while beingsensitive to the centrality of these beliefs to some men’s core ideas of theirmasculinity could prove very effective in changing the culture of traditionalmasculinity. Examples of existing interventions of this nature with evidence tosupport their effectiveness are Men Can Stop Rape (mencanstoprape.org) andChanging Carolina (Swan, Allen, & Davis, 2010). These types of interventionsdeliver messages around changing norms of violence and relationship beha-vior through group activities as well as traditional didactic education. Forthese types of programs, incorporating specific content to address issuessurrounding the societal norms regarding sexual consent could prove parti-cularly powerful in changing dysfunctional masculine norms.

Limitations

As with all socially unacceptable behavior, sexual aggression might beunderreported. The prevalence of sexual aggression in our sample was10%, which is fairly high given that respondents were reporting on theirbehaviors during the 4 months prior to completing the survey. Previousestimates of self-reported sexual assault perpetration have ranged from 1%to 3% for violent sexual assault to over 20% for alcohol- or drug-facilitatedsexual assault (Krebs, Lindquist, Warner, Fisher, & Martin, 2007). Further-more, due to the nature of these behaviors, it is likely that some men inour sample underreported the frequency with which they engaged insexually coercive or aggressive acts. However, with n = 22 participantshaving endorsed committing some form of sexual aggression ranging fromcoercion to rape, our findings are in need of replication. In addition,although there might be some conceptual difference among the behaviors

Comprehension of Consent and Sexual Aggression 909

Dow

nloa

ded

by [

Pete

r W

arre

n] a

t 06:

46 2

9 Se

ptem

ber

2015

measured in the SES, it is important to note that most of the behaviors usedin this measure meet legal definitions of sexual assault, whether it beverbal coercion, drug-facilitated sexual assault, or violent rape (Kosset al., 2007).

Although the primary aim of this study was to examine the relationshipbetween sexual assault and comprehension of sexual consent, it could be thatdifferent forms of sexual assault (e.g., verbal sexual coercion vs. physicallyforcing someone to have sex) could have differential relationships among theconstructs assessed in our models. Additionally, because we did not assessprevious history of abuse in our sample, we were unable to estimate thosepotential effects in our model. Furthermore, it is quite possible that predictorsof sexual aggression might vary by the level of severity of a given act, butgiven the low base rates for sexual aggression in our sample (n = 22; 10% ofthe sample), we were unable to empirically test this possibility. Our analyseswere conducted on a convenience sample of self-identified heterosexualmales, and, as such, results of this study might not generalize to nonheter-osexual identified males, nonstudents, or female perpetrators. It should benoted that we employed a cross-sectional design, and, thus, these findingscannot be used to imply directional causation due to lack of temporal pre-cedence among variables

CONCLUSION

Our findings suggest that an individual’s understanding of the concept ofsexual consent is related to the likelihood of that person perpetrating sexualaggression. Furthermore, comprehension of sexual consent is a multifacetedconstruct and can be influenced by one’s own cognitive beliefs regarding thejustification of sexual aggression as well as the shared beliefs of one’s peers.Interventions that work to change both internal cognitions and social normsabout sexual behavior will likely benefit from a focus on educating men aboutthe full nature of sexual consent.

REFERENCES

Abbey, A., Clinton-Sherrod, A., McAuslan, P., Zawacki, T., & Buck, P. (2003). Therelationship between the quantity of alcohol consumed and the severity ofsexual assaults committed by college men. Journal of Interpersonal Violence,18, 813–833. doi:10.1177/0886260503253301

Abbey, A., & McAuslan, P. (2004). A longitudinal examination of male collegestudents’ perpetration of sexual assault. Journal of Consulting and ClinicalPsychology, 72, 747–756. doi:10.1037/0022-006X.72.5.747

910 P. Warren et al.

Dow

nloa

ded

by [

Pete

r W

arre

n] a

t 06:

46 2

9 Se

ptem

ber

2015

Abbey, A., Parkhill, M. R., Jacques-Tiura, A. J., & Saenz, C. (2009). Alcohol’s role inmen’s use of coercion to obtain unprotected sex. Substance Use & Misuse, 44,1329–1348. doi:10.1080/10826080902961419

Adams-Curtis, L., & Forbes, G. (2004). College women’s experiences of sexual coer-cion: A review of cultural, perpetrator, victim, and situational variables. Trauma,Violence, & Abuse, 5, 91–122. doi:10.1177/1524838003262331

Allen, C., Swan, S., & Raghavan, C. (2009). Gender symmetry, sexism, and intimatepartner violence. Journal of Interpersonal Violence, 24, 1816–1834. doi:10.1177/0886260508325496

Bandura, A. (1971). Social learning theory of aggression. In J. F. Knutson (Ed.), Controlof aggression: Implications from basic research (pp. 201–250). Chicago, IL: Aldine-Atherton.

Baron, R. M., & Kenny, D. A. (1986). The moderator–mediator variable distinction insocial psychological research: Conceptual, strategic, and statistical considera-tions. Journal of Personality and Social Psychology, 51, 1173–1182.doi:10.1037/0022-3514.51.6.1173

Bohner, G., Jarvis, C., Eyssel, F., & Siebler, F. (2005). The causal impact of rape mythacceptance on men’s rape proclivity: Comparing sexually coercive and noncoer-cive men. European Journal of Social Psychology, 35, 819–828. doi:10.1002/(ISSN)1099-0992

Casey, E., & Beadnell, B. (2010). The structure of male adolescent peer networks andrisk for intimate partner violence perpetration: Findings from a national sample.Journal of Youth and Adolescence, 39, 620–633. doi:10.1007/s10964-009-9423-y

Davis, T. (2000). Programming for men to reduce sexual violence. New Directions forStudent Services, 2000, 79–89. doi:10.1002/(ISSN)1536-0695

DeGue, S., DiLillo, D., & Scalora, M. (2010). Are all perpetrators alike? Comparing riskfactors for sexual coercion and aggression. Sexual Abuse: A Journal of Researchand Treatment, 22, 402–426.

DeKeseredy, W. (1988). Woman abuse in dating relationships: The relevance of socialsupport theory. Journal of Family Violence, 3, 1–13. doi:10.1007/BF00994662

DeKeseredy, W., & Kelly, K. (1995). Sexual abuse in Canadian university and collegedating relationships: The contribution of male peer support. Journal of FamilyViolence, 10(1), 41–53. doi:10.1007/BF02110536

Fabiano, P., Perkins, H., Berkowitz, A., Linkenbach, J., & Stark, C. (2003). Engagingmen as social justice allies in ending violence against women: Evidence for asocial norms approach. Journal of American College Health, 52, 105–112.doi:10.1080/07448480309595732

Farris, C., Treat, T., & Viken, R. (2010). Alcohol alters men’s perceptual and decisionalprocessing of women’s sexual interest. Journal of Abnormal Psychology, 119,427–432. doi:10.1037/a0019343

Fisher, B., Cullen, F., & Turner, M. (2000). The sexual victimization of college women(Report No. NCJ 182369). Washington, DC: Bureau of Justice Statistics, NationalInstitute of Justice.

Fisher, B. S., Daigle, L. E., & Cullen, F. T. (2010). What distinguishes single fromrecurrent sexual victims? The role of lifestyle-routine activities and first-incidentcharacteristics. Justice Quarterly, 27, 102–129. doi:10.1080/07418820902763061

Comprehension of Consent and Sexual Aggression 911

Dow

nloa

ded

by [

Pete

r W

arre

n] a

t 06:

46 2

9 Se

ptem

ber

2015

Fritz, M. S., & MacKinnon, D. P. (2007). Required sample size to detect the mediatedeffect. Psychological Science, 18, 233–239. doi:10.1111/j.1467-9280.2007.01882.x

Gibson, D., & Humphrey, C. (1993). Educating in regards to sexual violence: Aninteractional dramatic acquaintance rape intervention (Unpublished manu-script). Minneapolis, MN: Sexual Violence Program, University of Minnesota.

Greene, P., & Davis, K. (2011). Latent profiles of risk among a community sample ofmen: Implications for sexual aggression. Journal of Interpersonal Violence, 26,1463–1477. doi:10.1177/0886260510369138

Hickman, S., & Muehlenhard, C. (1999). “By the semi-mystical appearance of acondom”: How young women and men communicate sexual consent in hetero-sexual situations. The Journal of Sex Research, 36, 258–272. doi:10.1080/00224499909551996

Hu, L., & Bentler, P. (1999). Cutoff criteria for fit indexes in covariance structureanalysis: Conventional criteria versus new alternatives, structural equation.Modeling, 6(1), 1–55.

Jost, J., & Hunyady, O. (2005). Antecedents and consequences of system-justifyingideologies. Current Directions in Psychological Science, 14, 260–265.doi:10.1111/cdir.2005.14.issue-5

Kilmartin, C. (2009). The masculine self (4th ed). New York, NY: Sloan.Kilmartin, C., & Allison, J. (2007). Men’s violence against women: Theory, research,

and activism. Mahwah, NJ: Erlbaum.Kilpatrick, D. G., Resnick, H. S., Ruggiero, K. J., Conoscenti, L. M., & McCauley, J.

(2007). Drug facilitated, incapacitated and forcible rape: A national study.Charleston, SC: National Crime Victims Research & Treatment Center.

Kimmel, M. (2000). The gendered society. New York, NY: Oxford University Press.Koss, M., Abbey, A., Campbell, R., Cook, S., Norris, J., Testa, M., . . . White, J. (2007).

Revising the SES: A collaborative process to improve assessment of sexualaggression and victimization. Psychology of Women Quarterly, 31, 357–370.doi:10.1111/pwqu.2007.31.issue-4

Krebs, C., Lindquist, C., Warner, T., Fisher, B., & Martin, S. (2007). The Campus SexualAssault (CSA) study: Final report. Washington, DC: National Institute of Justice,U.S. Department of Justice.

MacKinnon, D. P., Lockwood, C. M., & Hoffman, J. M. (1998). A new method to testfor mediation. Paper presented at the Annual Meeting of the Society for Preven-tion Research, Park City, UT, June.

Mahalik, J., Locke, B., Ludlow, L., Diemer, M., Scott, R., Gottfried, M., & Freitas,G. (2003). Development of the Conformity to Masculine Norms Inventory.Psychology of Men & Masculinity, 4, 3–25. doi:10.1037/1524-9220.4.1.3

Malamuth, N. (1998). The confluence model as an organizing framework for researchon sexually aggressive men: Risk moderators, imagined aggression, and porno-graphy consumption. In R. G. Geen & E. Donnerstein (Eds.), Human aggression:Theories, research, and implications for social policy (pp. 229–245). New York,NY: Academic Press.

McDermott, R., & Lopez, F. (2013). College men’s intimate partner violence attitudes:Contributions of adult attachment and gender role stress. Journal of CounselingPsychology, 60(1), 127–136. doi:10.1037/a0030353

912 P. Warren et al.

Dow

nloa

ded

by [

Pete

r W

arre

n] a

t 06:

46 2

9 Se

ptem

ber

2015

McMahon, S., & Dick, A. (2011). “Being in a room with like-minded men”: Anexploratory study of men’s participation in a bystander intervention program toprevent intimate partner violence. The Journal of Men’s Studies, 19(1), 3–18.doi:10.3149/jms.1901.3

Muthén, L., & Muthén, B. (2007). Mplus user’s guide (5th ed.). Los Angeles, CA:Author.

Payne, D., Lonsway, K., & Fitzgerald, L. (1999). Rape myth acceptance: Exploration ofits structure and its measurement using the Illinois Rape Myth Acceptance Scale.Journal of Research in Personality, 33, 27–68. doi:10.1006/jrpe.1998.2238

Potts, D. (2009). Responding to reported rapes: A comparison of human servicesstudents’ attitudes, knowledge, and counseling skills. Doctoral dissertation.Ann Arbor, MI: Capella University.

Raghavan, C., Rajah, V., Gentile, K., Collado, L., & Kavanagh, A. (2009). Commu-nity violence, social support networks, ethnic group differences, and maleperpetration of intimate partner violence. Journal of Interpersonal Violence,24, 1615–1632. doi:10.1177/0886260509331489

Rape, Abuse, & Incest National Network. (2009). Was it rape? Retrieved from https://www.rainn.org/get-information/types-of-sexual-assault/was-it-rape

Rosen, L., Kaminski, R., Parmley, A., Knudson, K., & Fancher, P. (2003). Theeffects of peer group climate on intimate partner violence among marriedmale U.S. Army soldiers. Violence Against Women, 9, 1045–1071. doi:10.1177/1077801203255504

Schafer, J., Caetano, R., & Cunradi, C. (2004). A path model of risk factors for intimatepartner violence among couples in the United States. Journal of InterpersonalViolence, 19, 127–142. doi:10.1177/0886260503260244

Schatzel-Murphy, E., Harris, D., Knight, R., & Milburn, M. (2009). Sexual coercion inmen and women: Similar behaviors, different predictors. Archives of SexualBehavior, 38, 974–986. doi:10.1007/s10508-009-9481-y

Schwartz, M., & DeKeseredy, W. (1997). Sexual assault on the college campus: Therole of male peer support. Thousand Oaks, CA: Sage.

Swan, S., Allen, C., & Davis, K. (2010). Efficacy of changing Carolina: An interventionto prevent dating violence and sexual assault. Paper presented at the Interna-tional Family Violence Conference, Portsmouth, NH, July.

Tilley, D., & Brackley, M. (2005). Men who batter intimate partners: A groundedtheory study of the development of male violence in intimate partner relation-ships. Issues in Mental Health Nursing, 26, 281–297. doi:10.1080/01612840590915676

Ward, T. (2000). Sexual offenders’ cognitive distortions as implicit theories.Aggression and Violent Behavior, 5, 491–507. doi:10.1016/S1359-1789(98)00036-6

Young, W., Weckman, G., & Holland, W. (2011). A survey of methodologies for thetreatment of missing values within datasets: Limitations and benefits. TheoreticalIssues in Ergonomics Science, 12(1), 15–43. doi:10.1080/14639220903470205

Comprehension of Consent and Sexual Aggression 913

Dow

nloa

ded

by [

Pete

r W

arre

n] a

t 06:

46 2

9 Se

ptem

ber

2015