Yes, (Most) Men Know What Rape Is: A Mixed-Methods Investigation Into College Men’s Definitions of Rape

Abstract

Sexual violence, including rape, is a pervasive problem on college campuses in the United States. Although men perpetrate the majority of sexual violence, men’s attitudes, experiences, and perspectives are not typically included in research on rape and sexual violence. We addressed this empirical gap through our mixed-methods analysis of 365 college-aged men’s definitions of the term “rape.” Our analysis via consensual qualitative research revealed that men’s definitions fit into nine primary domains: lack of consent, taken advantage of, sex, sexual activity, unwanted, gender/sex-specific, harm to victim, relationship, and emotional response, as well as a miscellaneous domain. Further, using chi-square tests of independence, we compared responses from men with and without histories of sexual violence perpetration. Findings showed that the definitions generated by men with a history of perpetration were less likely to include nonpenetrative sexual violence and were more likely to use gender/sex-specific language. We conclude that most young men have a generally accurate understanding of rape, though perpetrators’ understandings may be somewhat narrower and more limited than those without a history of perpetration. We end with recommendations for refocusing sexual education curricula to better aid in the prevention of sexual violence perpetration. Specifically, given that (most) men know what rape is, educators should emphasize the cultural and situational factors that make rape more likely so all people can reduce the risk of sexual violence and take proactive precautions to prevent it.

Sexual violence (SV) occurs on a spectrum, with rape representing the most extreme end. At the time of this writing, the U.S. Department of Justice defines rape as “penetration, no matter how slight, of the vagina or anus with any body part or object, or oral penetration by a sex organ of another person, without the consent of the victim” (United States Department of Justice, 2012).¹ Although still fairly narrow, this updated definition was broadened in 2012 to recognize a wider range of sexual activities beyond penile–vaginal intercourse that constitute rape. This expanded definition also acknowledges the potential for any person to be victimized by rape, rather than incorrectly implying that men are exclusively the perpetrators. Despite clear definitions set out by lawmaking bodies, researchers and laypeople alike tend to provide different definitions of the term (Haugen et al., 2018; Muehlenhard et al., 1992).

Although people of any gender can be victimized by rape (e.g., Aosved et al., 2011; Griswold et al., 2020), men, and college-aged men in particular, comprise the majority of rape perpetrators (Barone et al., 2007; Flood, 2006; Koss, 1993; Ybarra & Mitchell, 2013). Indeed, in a recent systematic review of 78 independent studies, the average prevalence rates of SV perpetration and rape in a combined sample of 25,524 college-aged men were 29% and 6%, respectively (Anderson et al., 2019; see also Muehlenhard et al., 2017). Indeed, sexually aggressive strategies are common among college-aged men (Warkentin & Gidycz, 2007), and research suggests that men of this age group underreport their use of sexual assault strategies (Krebs et al., 2007; Strang & Peterson, 2017), with some studies suggesting this underreporting is intentional (Strang & Peterson, 2020). To comprehensively prevent SV, men’s attitudes, experiences, and knowledge must be empirically examined; yet, men are often left out of research and conversations about SV (Anderson et al., 2019). Rather than examining gender differences in college students’ definitions of the term (see Haugen et al., 2018), this study exclusively focuses on college-aged men’s definitions of rape to develop a better understanding of these men’s perceptions of rape. Further, we explore potential differences in these definitions between men with a history of sexual violence perpetration and those without.

Sexual Scripting Theory and Rape Myths

One proposed reason why sexual violence perpetration is common among college-aged, heterosexual men is the sociocultural normativity of these behaviors. Sexual scripting theory (Beres, 2014; Gagnon & Simon, 1973; Simon & Gagnon, 2003) is a psychological framework that situates sexual attitudes and behaviors in a social context. The theory proposes that, much like the “script” of a play or movie, “sexual scripts” give individuals directions for how they should (and should not) act in specific sexual situations. These scripts are written and reinforced by a variety of sources, including, but not limited to, mainstream media (Kim et al., 2007), magazines (Carpenter, 1998), pornography (Bridges et al., 2016), and music videos (Rodgers & Hust, 2018). Sexual scripts vary on the cultural, interpersonal, and intrapsychic levels (Gagnon & Simon, 1973), and college campuses serve as a powerful cultural context for the understanding and enactment of sexual scripts among young adults (Berntson et al., 2014). Although sexual scripts are multifaceted (Morrison et al., 2015) and differ based on factors such as age, culture, and sexual orientation (e.g., Bowleg et al., 2015; Eaton et al., 2016), sexual scripts and gender norms are inherently intertwined, as traditional gender roles dictate the options that are perceived as normative and appropriate to individuals in specific sexual contexts (Eaton & Rose, 2011). For example, in heterosexual sexual relationships, traditional gender-sexual norms specify that men are to be the initiators and pursuers of sex, and women are to be the gatekeepers and/or passive recipients of men’s sexual advances (Byers, 1995; Laner & Ventrone, 2000). These scripts serve as a powerful cultural force, and aspects of heteronormative sexual scripts have been identified even in nonheterosexual relationships (Hoppe, 2011; Pham, 2016).

Sexual scripts can normalize violent sexual behavior toward women and can make it difficult for women to actively and firmly say no to unwanted sex (Frith, 2013; Frith & Kitzinger, 2001). Sexual scripts are particularly dangerous when they are laden with rape myths, or false beliefs that place blame on the victims and absolve the perpetrators of SV (Aronowitz et al., 2012; Edwards et al., 2011; Ryan, 2011). Indeed, endorsement of rape myths (e.g., “women enjoy rape,” “rape is an accident,” “women lie about being raped,” and “women ask to be raped”; Edwards et al., 2011; McMahon & Farmer, 2011) is not uncommon among college-aged men (Martinez et al., 2018; O’Connor et al., 2018; Suarez & Gadalla, 2010). College-aged men’s support for rape myths has been linked to various adverse psychological and behavioral outcomes, including hostile attitudes toward women (Chapleau et al., 2007), problematic drinking behaviors (Hayes et al., 2016), adversarial sexual beliefs and acceptance of interpersonal violence (Lonsway & Fitzgerald, 1995), and various forms of prejudice toward marginalized social groups (Aosved & Long, 2006).

Further, and important for the current study, rape myths are common in perpetrators’ justification for their sexually violent behaviors (e.g., Edwards et al., 2011). Given the legitimizing power of these myths, debate exists regarding whether men who perpetrate SV recognize their behaviors as violent (Anderson et al., 2019; Koss, 1998; Snyder & Higgins, 1988), and some have suggested that this discrepancy may reflect a misunderstanding of the definition of rape (Edwards et al., 2014). For example, in a qualitative study assessing convicted sex offenders’ motivations for their sexually violent behavior (Beech et al., 2006), rather than perceiving their behaviors as internally and personally motivated, perpetrators attributed their sexually violent behaviors to five implicit theories: a belief that men are naturally hostile, a sense that women are dangerous and evil, a perception of women as sexual objects to be violated at will, a perceived uncontrollable male sex drive that cannot be reasonably contained, and entitlement to whatever they would like. Further, in a recent content analysis of 68 anonymous narratives from sexual assault perpetrators taken from the social media website Reddit, a majority of responses used rape-normalizing sexual scripts to explain their behaviors (Hipp et al., 2017). Much like the convicted sex offenders, some posts suggested that the perpetrators coerced their partners on the basis of gender-stereotyped sexual scripts (e.g., “I had this impression that girls didn’t want to be seen as sluts or whores so they would pretend that they didn’t want to have sex when really they did”) or denied their partners’ resistance when sexual activity was initiated. Others engaged in victim blame and hostility, attributed their behaviors to hormones, or described their victims as objects or less than human to justify their behaviors. Taken together, this research suggests that the sexual scripts of perpetrators may be warped by rape myths; yet, it remains unclear whether rape myths are fundamental components of college-aged men’s definitions of rape, particularly those who have no prior history of SV perpetration.

Current Study

These findings indicate a need to examine college-aged men’s basic understandings of rape to examine, in an open-ended manner, to what degree they are influenced by sexual scripts and rape myths. In the present study, we provided college-aged men—with and without past histories of SV perpetration—the opportunity to freely provide definitions of the word “rape” to explore the extent to which rape myths and sexual scripts are embedded in these men’s basic definitions of the term. We chose a qualitative paradigm to allow for inductive analysis of the data, given the paucity of research on this topic, followed by a quantitative approach, which allowed us to make meaningful distinctions between men with and without a history of SV perpetration.

Method

Participants and Procedure

In total, 488 college men recruited from the Psychology undergraduate participant pool at the University of North Dakota participated in the study. Data were coded to saturation, meaning that we did not anticipate that we could extract new domains in future results (Williams & Hill, 2012). The final sample consisted of responses from 365 male students (Mage = 19.88, SD = 4.56). Most participants were Caucasian or White (n = 305; 83.6%), with others identifying as African American (n = 29; 7.9%), Asian American (n = 12; 3.3%), Native American (n = 7; 1.9%), or that their racial group was not listed in the choices provided (n = 26; 7.2%). Participants primarily identified as heterosexual or straight (n = 319; 86.2%), gay (n = 20; 5.5%), bisexual (n = 19; 5.2%), asexual (n = 2; .5%), demisexual (n = 1; .3%), heteroflexible (n = 1; .3%), or did not provide a response (n = 3; .8%). The demographic composition of this sample was consistent with the university as a whole (https://und.edu/analytics-and-planning/data-and-reports/2019.html). A majority of the sample was sexually active (n = 218; 59.7%). Two hundred and one participants (55.1%) identified as single, with others indicating that they were single but dating (n = 58; 15.9%), in a committed relationship (n = 99; 27 .1%), engaged or married (n = 4; 1.10%), or did not provide a response (n = 3; .8%). Based on scores acquired from combining scores on two measures of sexual perpetration history (see below), 140 participants (38.4%) were included in the perpetrator category.

Ethics approval was granted from the University of North Dakota, a large university in the Midwestern United States. Participants were recruited through the undergraduate psychology participant pool, and participants were eligible if they identified as both men and students. The entire study took place online. Eligible participants were directed to a Qualtrics survey where they provided their informed consent. As part of a larger study, participants responded to several questionnaires related to sexual violence, which were presented in a counterbalanced order. Participants were debriefed via survey on their distress and perceived benefits of participation, and were compensated with one course credit toward the completion of a psychology course. No monetary compensation was provided.

Materials

Aside from the main prompt-in question (i.e., “There are several activities that are considered sexual assault, one of which is rape. How would you define rape?”), two measures were administered to each participant, which, when combined, identified if they had a history of SV perpetration. Two different measures were employed and combined because past literature indicates that some men underidentify certain types of sexual perpetration (Krebs et al., 2007; Strang & Peterson, 2017, 2020).

Sexual Experiences

The Sexual Experiences Survey—Short Form Perpetration (SES-SFP; Koss et al., 2007) contains 35 items describing seven sexual behaviors (e.g., vaginal sex, oral sex, unwanted touching) and five potential tactics for initiating each of these behaviors (verbal criticism, verbal pressure, alcohol incapacitation, threats of sexual force, and physical force). Participants read the following prompt before completing items, “These are questions about things you have done to someone else since the age of 14.” Participants responded “yes” or “no” to indicate whether they had ever engaged in the behavior. Past research has established the convergent validity of the SES-SFP in samples of college-aged men (Anderson et al., 2017; Johnson et al., 2017).

Conflict Tactics

The Revised Conflict Tactics Scale-2 (CTS-2; Straus et al., 1996) consists of two parts, perpetration and victimization. Each section contains 39 items that measure how often people have used or been subjected to physical and psychological strategies to negotiate conflicts with their partners since the age of 14. The measure consists of five subscales: Negotiation, Psychological Aggression, Physical Assault, Sexual Coercion, and Injury. For the purposes of this study, only the Sexual Coercion subscale was measured (e.g., “used force to make partner have sex,” “used threats to make partner have anal sex”). The CTS-2 and its subscales have demonstrated good validity in samples of college-aged students (Anderson et al., 2017; Straus et al., 1996).

Analysis

Qualitative data were analyzed using Consensual Qualitative Research-Modified (CQR-M; Spangler et al., 2012), an analytic method for analyzing short and straightforward qualitative data. The modified version of CQR was employed to account for the large amount of data provided and the brevity of each individual data piece. CQR-M is a rigorous, replicable, inductive methodology whereby independent raters establish a coding scheme by consensus (Hill et al., 2005). In CQR-M, independent raters engage in triangulation through ongoing discussion to find shared meaning in responses (Spangler et al., 2012).

By its very nature, CQR is “constructivist, with a hint of postpositivism” (Hill et al., 2005, p. 197). Researchers working within a CQR framework recognize knowledge as socially situated, arising out of learned and understood meaning, rather than rote memorization. Therefore, we did not anticipate that participants would respond with textbook definitions of terms or complete and objective answers. Rather, we assessed consistent domains within men’s conceptual and schematic definitions of rape to analyze the types of information participants regarded as sufficient and necessary for defining the term. We maintained reflexivity throughout the process (Hill, 2012), searching for instances where our thinking may be influenced by our own experiences, consulting with other members of the research team to increase transparency, and arriving at consensual conclusions.

It is important to note that the social locations of the research team can influence the ways that data are collected, analyzed, and presented (Yeh & Inman, 2007); our analysis is situated within the context of our identities as researchers. The coding team consisted of four undergraduate research assistants, who varied in gender, race, and history of victimization. Coders were supervised and moderated by a Ph.D. student in psychology who identifies as a radical feminist, Jewish, cisgender woman. This was the first coding experience for the four undergraduate researchers, although the supervisor of the team had expertise in both SV research and CQR-M methodology. The entire project was overseen by a White, cisgender professor of psychology, who had extensive knowledge of the research surrounding SV perpetration. To reduce the likelihood of within-team bias and increase trustworthiness (Williams & Hill, 2012), the data and coding scheme were sent to an outside researcher (Ph.D. candidate, feminist, White, cisgender woman, no sexual assault experience), who recommended clarifications and minor revisions to the coding.

Initially, the project supervisor generated a list of domains, or broad categories of repeated ideas extracted from the research data, relevant to the project and created a codebook with operational definitions of the domains, which was provided to the research assistants. The initial domains included lack of consent, sexual activity, sex, taken advantage of, unwanted, harm to victim, gender/sex-specific, relationship, emotional response, multiple people, crime, reasons for rape, age of victim, fear, and shared responsibility. Based on low frequency, multiple people, crime, reasons for rape, age of victim, fear, and shared responsibility were then grouped together into a miscellaneous domain.

The coders read the responses and voted on the codes by consensus after independently reviewing the data and taking notes. In some cases, additional, noteworthy variant domains and subdomains (e.g., reasons for rape—dominance, reasons for rape—pleasure) were extracted. We allowed for any individual response to fit into several domains, given that they varied in length and complexity.

Results

Overall, the analysis revealed nine primary domains (lack of consent, taken advantage of, sex, sexual activity, unwanted, gender/sex-specific, harm to victim, relationship, emotional response) and a miscellaneous domain for notable variant domains (see Table 1 for frequencies and exemplary quotes). We conducted exploratory analyses to assess differences between individuals who reported lifetime experiences of SV perpetration and those who did not (see Table 2). Given the spirit of qualitative research is hypothesis-generating and exploratory, we did not make specific predictions about the domains men with perpetration histories might provide.

Table 1.

Chart of Domains, Subdomains, and Examples

Domain	Subdomain	Number of occurrences, prevalence	Example (additional labels)
Lack of consent		233, 63.84% Typical	“Forcibly having sex or sexual acts with someone without there [sic] proper permission” (Sex, sexual activity, taken advantage of—force)
	Ignoring verbal commands	36, 9.86% Variant	“Forcing sexual activity on someone who has verbally said no” (sexual activity, taken advantage of—force)
	Ignoring nonverbal cues	13, 3.56% Variant	“The victim does not give consent, is fighting you and you continue to make moves, is screaming NO or STOP” (Lack of consent—ignoring verbal cues)
Taken advantage of		216, 59.18% Variant	“Unwanted/unwilling sexual acts; any sexual activity against somebody’s will” (Unwanted, taken advantage of)
	Force	154, 42.19% Variant	“Forcing a sexual act on someone who does not give consent; keeping someone bound/held down with intent of committing a sexual act” (Taken advantage of—force, sexual activity)
	Manipulation	8, 2.16% Variant	“Any way you’d take advantage of a fellow human weather [sic] it be through pressure, influence, or disregard”
	Drugged	7, 1.89% Variant	“Using some kind of force or method (drugging for example) to prevent the victim from preventing the perp from penetrating the victim vaginally, anally or orally” (Taken advantage of—force, sex)
	Incapacitated	30, 8.22% Variant	“Rape is taking advantage of someone without their consent. Such as a girl that is too drunk to walk or function. If the girl or guy is blacked out drunk and you partake in sexual activity with them then that is rape” (Lack of consent, gender/sex-specific, sexual activity)
Sex		176, 48.22% Variant	“Any form of penetration with an instrument, object, with a penis”
Sexual activity		154, 42.19% Variant	“Sexually touching another person without their consent” (Lack of consent)
Unwanted		98, 26.85% Variant	“Having a form of sex with someone without them wanting you to” (Unwanted, sex)
Harm to victim		23, 6.30% Variant	“Forcing yourself upon/inside an individual, abusing” (Sex, taken advantage of—force)
	Emotional harm	7, 1.92% Variant	“Rape is when someone takes advantage of another individual sexually, robbing the person of their self-esteem and taking away something that was not theirs to take” (Taken advantage of, sexual activity, emotional response, emotional harm)
	Physical harm	12, 3.29% Variant	“Physically injuring and forcing a female who has said no to participate in sexual activity” (Gender/sex-specific, taken advantage of—force, lack of consent—ignoring verbal cues, sexual activity)
Gender/Sex-specific		20, 5.48% Variant	“A man forcing a women to engage in any sexual activity such as sex or even contact with either party’s genitals” (sexual activity, sex, taken advantage of—force)
Relationship		9, 2.47% Variant	“Rape is when on [sic] partner forces their partner to have sex with them” (Sex, taken advantage of—force)
Emotional response		6, 1.64% Variant	“One of the most disgusting and degrading things you can do to a female” (Gender/sex-specific)
Multiple people		5, 1.37% Variant	“Rape is when someone or when a group of people force someone to have sexual activity without their consent” (Taken advantage of—force, lack of consent, sexual activity)
Crime		5, 1.37% Variant	“Rape; a crime that is committed to a person that does not give consent to sexual activities but is forced into them” (Lack of consent, sexual activity, taken advantage of—force)
Reasons for rape	Dominance	2, .55% Variant	“Forcing women or men to follow your command” (Taken advantage of—force)
	Pleasure	2, .55%	“To me, rape is one person gaining sexual satisfaction from another person without that person’s consent or knowledge of the sexual act taking place” (Sexual activity, lack of consent)
Age of victim		2, .55% Variant	“Unlawful sexual activity and usually sexual intercourse carried out forcibly or under threat of injury against the will usually of a female or with a person who is beneath a certain age or incapable of valid consent—compare sexual assault, statutory rape” (Crime, sexual activity, sex, taken advantage of—force, gender/sex-specific, lack of consent, harm to victim—physical, harm to victim—emotional)
Fear		1, .27% Variant	“When a party doesn’t agree to by own accord [sic] or doesn’t feel safe with the partner, yet insertion occurs” (lack of consent, sex, relationship, fear)
Shared responsibility		1, .27% Variant	“I would define rape as forced sexual intercourse with someone who does not want to have it. If someone is passed out and is unconscious it is morally wrong. If the victim is intoxicated they should be held responsible for their actions as well as aware of their actions be held partially responsible for the accident” (Unwanted, taken advantage of—force, sex, incapacitated, victim)

Note. The final counts were tabulated and labeled as general (nearly all), typical (more than 50%), or variant (less than 50%) on the basis of frequency (Hill et al., 2005). Domains with a small number of responses (≤5) were grouped into a miscellaneous category (Hill et al., 2005). Although Hill et al. (2005) suggest that, in samples larger than 15, only domains with one response should be included in the miscellaneous section, due to the large sample size in this study, we chose to expand this to five or fewer responses. Most responses were categorized into two or more domains, as noted by the additional domains listed under each quote.

Table 2.

Results of Chi-Square Tests of Independence

	Total n = 365	History of perpetration n = 140	No history of perpetration n = 225	χ 2	p
Lack of consent	233	83, 59.29%	150, 66.67%	2.04	.15
Sexual activity	154	47, 33.57%	107, 47.56%	6.92	<.01
Taken advantage of	216	79, 56.43%	137, 50.89%	.71	.40
Sex	176	74, 52.86%	102, 45.33%	1.96	.16
Unwanted	98	44, 31.43%	54, 24.00%	1.41	.12
Harm to victim	23	13, 9.29%	10, 4.44%	3.43	.06
Gender/Sex-specific	20	13, 9.29%	7, 3.11%	6.35	.01

Consensual Qualitative Analysis

Lack of Consent

Lack of consent was defined as “at least one person in the encounter does not or cannot agree or give permission for the action.” Most responses were explicit that the behavior was “unauthorized,” or that consent was not granted by at least one member of the encounter. Responses could be categorized as ignoring verbal commands, ignoring nonverbal cues, and incapacitated. Ignoring verbal commands was defined as “perpetrator of the act disregards spoken demands to stop.” Responses in this subdomain included the victim stating that they did not want to engage in the encounter, such as by saying no,” “stop,” or that they felt uncomfortable. Ignoring nonverbal cues was defined as “perpetrator of the act disregards physical demands to stop made before and/or during the act.” Some examples of nonverbal cues included “signs,” “struggle,” “laying hands out,” or “fighting you.” Several responses indicated that sexual activity with someone who could not provide consent to sexual activity due to incapacitation constituted rape. Examples of incapacitation included being “passed out,” “black out drunk,” “not in the (state of) mind,” “under the influence,” “messed up,” or “intoxicated.”

Taken Advantage Of

A majority also indicated that rape involves taking advantage of another person, which was defined as “employing a tactic to make someone engage in sexual activity who is unwilling.” Most responses in this subdomain could be further categorized into force, manipulation, or drugged. Responses were coded into the force category if they indicated that rape included “exerting physical strength onto someone” (e.g., “keeping someone bound/held down” and “forcing [oneself] onto another individual”). Manipulation was defined as “mentally/verbally coercing the victim into a sexual act,” and responses included “pressure,” “influence,” “disregard,” and emotional or verbal manipulation. When participants mentioned the use of “substances to alter the victim’s state of consciousness,” responses were classified into the drugged subcategory. Responses were included in this category if the drugs or “substances” were used to disarm the victim.

Sex

Many explicitly noted that “the act of [penile] penetration (i.e., oral, vaginal, anal, intercourse)” must occur for the encounter to constitute as rape, a domain we labeled as sex. Specifically, responses in this category referred to “sex,” “intercourse,” or “penetration.”

Sexual Activity

Others indicated that rape included “any physical or sexual act done to another person (that does not include penetration; i.e., groping, kissing, touching genitals).” Examples of “sexual activity” included “sexual act(s)” or “actions,” “sexual advances,” “sexual relations,” “sexual touching,” or “sexual anything.”

Unwanted

Some mentioned that rape is unwanted, which was defined as “at least one person in the encounter lacks desire for the action.” Although similar to lack of consent, we differentiated between the two categories in our analysis because sex can technically be unwanted but consensual or nonconsensual but “wanted” (Impett & Peplau, 2002; Peterson & Muehlenhard, 2007). Responses in this domain sometimes noted that the other person “did not want it” or that it was “unwanted.”

Harm to Victim

Other responses were categorized as harm to victim. Answers fit into this category if they indicated that there was any form of harm to the victim as a result of the incident. Some responses were broad and general, suggesting that rape involved any form of harm, such as “abuse.” Others specified physical versus emotional harm. Some indicated that rape involves physical harm, or “bodily injury to the victim.” Physical harm responses included the use of “brutality” or “violence,” causing “injury,” or “hurting” the other person. Some participants observed that rape involved emotional harm, which was defined as mental or psychological trauma to the victim. Examples of emotional harm included “robbing the person of their self-esteem” and “humiliat[ion].” Notably, with only one exception, every response that was characterized as emotional harm also mentioned physical harm.

Gender/Sex-Specific

Several participants provided responses that were categorized as gender/sex-specific if their definition mentioned that the perpetrator or victim is male or female (or had male or female sexual parts). Gender/sex-specific responses unilaterally described men, “males”, or people with penises as the perpetrators and girls, women, “females”, or people with vaginas were described as the victims.

Relationship

A few responses were categorized into the relationship domain, particularly if the response implied that “the sexual act happens with a partner or significant other.” Relationship responses included phrases such as “partners,” and “member[s] of sexual relationship[s].”

Emotional Response

Some provided emotional responses to the question. Responses were categorized into this domain when the answer suggested that “an intense feeling was evoked from the respondent.” Responses with strong emotions attached sometimes referred to rape as “childish,” “immature,” “horrible,” and “a sick, disgusting act.”

Miscellaneous

Some responses were endorsed infrequently, though their inclusion was noteworthy. For example, some responses were categorized as multiple people, if they “mentioned more than one person committing or receiving the act.” Answers in this domain referred to “groups” or “one or more people.” Responses were categorized as crime if they used legal terms in their definition, describing rape as being against the law. Responses were categorized into the crime subdomain when the respondent recognized the act is criminal in nature. Answers included those with phrases such as “crime” or “unlawful.” A few responses were categorized into the reasons for rape domain because they made explicit mention of “why the perpetrator commits the sexual act.” These responses were categorized as dominance (e.g., “want to feel dominant,” and “follow[ing] your command”) or pleasure (perpetrator performs the act to feel sexual satisfaction or forces the other person to “please” them). Two responses included the age of the victim. Specifically, these answers indicated that victims were under 18 years old. One response was also categorized into a fear domain because it uniquely mentioned the trepidation of the victim. Finally, one response suggested shared responsibility for rape because it indicated that both the victim and the perpetrator “should be held partially responsible for the accident.”

Exploratory Analysis: Differences Between Perpetrators and Nonperpetrators

Given that most participants provided definitions of rape that were generally consistent with, if not broader than, the legal definition, we examined potential differences between individuals who reported SV perpetration over the course of their lifetime. We employed a chi-square test of independence to identify potential differences between the two categories because the groups were categorical. Using 2 (perpetrator/nonperpetrator) by 2 (absence/presence) chi-square analysis to compare between domains with 20 or more responses² (i.e., lack of consent, taken advantage of, sex, sexual activity, unwanted, harm to victim, gender/sex-specific), we identified a statistically significant difference between perpetrators and nonperpetrators in their use of the sexual activity domain (χ² (1, N = 365) = 6.92, p = <.01), as well as the gender/sex-specific domain (χ² (1, N = 365) = 6.35, p = .01). Harm to victim trended toward significance (χ² (1, N = 365) = 3.43, p = .06), with definitions from men with a history of SV perpetration more likely to be categorized into this domain but did not cross the p < .05 threshold for statistical significance. No other tests achieved or trended toward statistical significance ( p = .15–.40; see Table 2). In other words, participants without a history of SV perpetration were more likely to recognize a broad definition of rape that included a wide range of sexual activities beyond penetration that constituted rape, and less likely to use gender/sex-specific language in their definitions, compared to men with lifetime experiences of SV perpetration. However, participants with and without a history of SV perpetration did not significantly differ on five of the seven primary categories, suggesting that even men with a history of perpetration are generally just as knowledgeable about what constitutes rape as men without such history.

Discussion

In this study, we first used an inductive, consensual, qualitative methodology to explore college-aged men’s conceptual definitions of “rape,” given the prevalence of rape myths and sexual scripts that may encourage men to perpetrate SV. Our findings suggest that most young men understand the basic definition of what rape is, and their definitions could be categorized into nine primary domains—lack of consent, taken advantage of, sex, sexual activity, unwanted, gender/sex-specific, harm to victim, relationship, emotional response—and a miscellaneous domain. We found that very few men’s definitions contained common rape myths (e.g., shared responsibility) or gendered sexual scripts (e.g., gender/sex-specific). Many provided broad definitions of sexual activity (e.g., groping) that constituted rape, but others suggested that penetration (i.e., sex) was necessary for rape to occur, as per the legal definition of the term. Therefore, in general, the sexual scripts of the men sampled for this study show that college-aged men are knowledgeable about the actions and behaviors that constitute rape, with some defining rape more broadly than the narrow, legal definition.

Notably, some men distinguished between nonconsensual sex/sexual activity and unwanted sex/sexual activity in their definitions. Given that lack of consent was the most common category for responses to this prompt, it stands to reason that college-aged men typically understand, in principle, that rape is usually nonconsensual. However, while the legal definition indicates that rape must be “nonconsensual,” sexual activity can be unwanted but technically consensual (i.e., sexually compliant behavior; Carter et al., 2020; Impett & Peplau, 2002), as well as wanted but nonconsensual (e.g., Peterson & Muehlenhard, 2007). The discrepancy between the number of men who indicated that rape was unwanted (n = 98) compared to nonconsensual (lack of consent; n = 233) suggests that there is cause to better develop men’s understanding of sexually compliant behaviors and the complex nature of consent. Specifically, results indicate that some men may lack awareness of the myriad pressures that people, particularly women, must navigate during sexual encounters and the consequent role of sexual compliance in sexual activity.

Further, some respondents suggested that rape occurs in the context of relationships. This domain was rather broad, as it was unclear whether participants used the term “partner” to indicate a short-term or long-term sexual partner. However, the existence of this domain suggests that some men’s foundational understanding of rape includes an acknowledgement of the likelihood of acquaintance and intimate partner rape. Additionally, while gendered definitions were variant in this study (n = 20), men who see rape as gender/sex-specific may not recognize themselves or their male peers as potential victims.

Our exploratory analyses suggest that both men with and without a history of SV perpetration can correctly identify the primary components of rape, and responses generally did not differ between groups. Specifically, groups did not differ in how often they provided responses that included lack of consent, taken advantage of, sex, unwanted, or harm to victim (though harm to victim was slightly more common among men with a history of SV perpetration). However, in some ways, men with a lifetime history of SV perpetration may perceive rape more narrowly than do men without this history; those who had committed SV in the past provided definitions of rape that were less likely to include behaviors outside of penetration that constitute rape. Although technically correct based on the 2012 definition of rape, even this new definition focuses exclusively on penetration and is influenced by a cultural acceptance of sexually violent behaviors (i.e., “rape culture”; Burt, 1980; Johnson & Johnson, 2017). This definition fails to capture myriad routine sexual violence experienced by both women (Papp & McClelland, 2020) and men (Stemple & Meyer, 2014) that can have a lasting detrimental effect on survivors.

Similarly, men with a history of SV perpetration were more likely to define rape as gender/sex-specific, with their definitions suggesting that men were the perpetrators and women the victims of rape. They were also marginally more likely to list harm as a requirement for rape to have occurred. Put another way, the rape scripts of men with a history of SV perpetration were more closely aligned with narrow, gendered scripts, requiring visible/physical harm compared to men without a history of SV perpetration. However, given the small number of men who provided gendered definitions, this finding should be interpreted with caution.

One potential explanation for this pattern of results is that perpetrators of SV use narrower definitions to normalize and justify their own sexually violent behaviors as a way to resolve the cognitive dissonance they may feel as a result of their transgression (Wegner et al., 2015). Given that we operationalized SV perpetration broadly, including both physical and psychological sexual violence, it is possible these men provided specific, narrow definitions to distance their own behaviors from “rape.” This rationale is consistent with past literature in which men with histories of SV perpetration used common cultural scripts and myths to normalize their behaviors (Beech et al., 2006; Hipp et al., 2017). When rape is narrowly defined, behaviors that do not fit within this specific definition may be regarded as less harmful and more normative.

Practice Implications

Our findings indicate that most college-aged men can correctly define rape and recognize its impact, suggesting that most young men do, in fact, know what rape is. The variant findings, especially regarding gender and relationships, point to the need for universal comprehensive, research-driven sexual education, which may reduce SV on college campuses (Banyard & Potter, 2018; Jouriles et al., 2018; Santelli et al., 2018). We echo Orchowski et al. (2020) in recommending that sexual violence interventionists consider addressing the contexts and cultures in which rape is likely and encourage men to be hypervigilant in these situations (e.g., high-risk situations with social pressure and binge alcohol use). Programs should address the topics of sociocultural pressures, masculine norms, information about the legal ramifications of sexual violence, and bystander intervention information (Orchowski et al., 2020), and educate participants on the role of sexual compliance in sexual interactions. These sexual education programs should be forthright about the gender demographics of victims and perpetrators while remaining explicitly clear that people of all genders may experience sexual violence.

Some theorists have also proposed that instances of SV may arise as a result of overreliance on sexual scripts and resultant miscommunication between partners (Kitzinger & Frith, 1999). However, our findings do not support the idea that sexual scripts and rape myths are embedded in men’s foundational understandings of rape, and most research suggests that miscommunication does not drive SV (Beres, 2010; Beres et al., 2014; O’Byrne et al., 2007). Therefore, rather than belaboring the point about the basic definitions of “consent,” “rape,” or “sexual violence,” we encourage sexual education programs to discuss process-based models of consent to ensure that communication about sexual desire and consent is understood clearly (Glace et al., 2020) and to discuss situations and contexts where rape is more likely to occur (e.g., in the presence of alcohol; Abbey, 2002; Lorenz & Ullman, 2016).

Given that SV continues to be a pervasive issue, mere awareness of SV is not sufficient for reducing its likelihood (see also Anderson & Whiston, 2005). More research is needed to understand what further interventions and consequences—including changes to the legal system—are needed to prevent and deter men from engaging in sexually violent behaviors. This can include the development of programs that encourage men to simultaneously support other men’s engagement in sexually healthy, consensual behaviors while challenging each other’s justifications for or denial of sexual violence. Such programs are acutely needed in subgroups or cultures where the minimization of sexual violence and the dehumanization of victims is especially pronounced (e.g., forensic populations to reduce recidivism, Greek culture on college campuses).

On a broader level, it is important that all people are empowered with the tools, skills, and resources to competently and confidently initiate, consent to, and refuse sexual activity. Our results suggest that men’s understanding about the definition of rape is insufficient for preventing sexual violence. We encourage integrating discussions of sexual ethics (e.g., Carmody, 2005) into these trainings to encourage men to accept responsibility for their sexual behaviors and act in accordance with their knowledge and personal, moral values. Further, it is not women’s responsibility to stop sexual violence from happening to them. However, research suggests that sexual assault resistance interventions can reduce women’s likelihood of victimization by empowering them to detect risks, resist attempts, and defend themselves against sexual aggression (Senn et al., 2020). Therefore, while women wait for cultural change to happen, they may wish to consider participating in some form of self-defense training, such as the Enhanced Access, Acknowledge, Act program, which has been shown to reduce women’s likelihood of experiencing rape both at 1- and 2-year follow-up (Senn et al., 2015, 2017).

Limitations and Future Directions

We acknowledge that no study is without its limitations. Importantly, given the postpositivist nature of CQR-M, it is possible that other researchers would have identified different domains or classified responses differently than this research team did. In fact, in other studies of college students’ definitions of rape (Haugen et al., 2018), the research team identified a different pattern of themes from their mixed-gender sample. It is also possible that results would be different with a more racially or culturally diverse set of participants. Although our sample was representative of the university as a whole, were this study conducted at a study with a more racially diverse student body, it is possible that the dataset may have been somewhat different, and we do not suggest that these results are generalizable outside of a North American context. More research is needed to understand how men conceptualize rape outside of North America. Further, when identifying men with a history of SV for our analyses, we used a broad definition of sexual violence perpetration, including both physical and psychological SV. However, it is possible results may vary depending on the type of SV perpetrated, and we support future work on this important topic. Researchers may wish to explore the definitions of particular subgroups of college-aged men (e.g., athletes, men who are affiliated with Greek life), and we encourage developmental research on rape myths to assess key ages for education and intervention. To test the hypothesis that men may use narrower definitions of rape to legitimize and justify their behaviours, we also support evaluating definitions over time to see if they change after SV perpetration. We encourage future researchers to consider how this line of research can be continued and expanded upon in diverse contexts.

Conclusion

Sexual education curricula need to include discussions about what rape is and is not, its function, and its impact. However, our findings suggest that men generally understand the basic definition of rape; we therefore encourage educators to channel the limited time and resources into teaching about the broad continuum of sexual violence, its relationship with substance use, situational and sociocultural motivators of rape, and the influence of gender norms, particularly in programs involving men with histories of SV perpetration. We encourage the continued examination of men’s attitudes and experiences regarding sexual violence to continue to fill this critical gap in the literature.

Public Significance Statement.

Sexual violence, including rape, is a pervasive problem on college campuses in the United States. Contrary to the myth that college-aged men do not know the definition of rape, our findings suggest that most college-aged men understand its definition and severity. However, men with histories of sexual assault perpetration provided more narrow definitions of rape compared to men without prior histories of perpetration.