Abstract
Using a community sample (n = 296), we investigated the associations between sexual behavior norm beliefs, acceptance of partner rape, judgments that non-consensual partner sex is “wrong not rape,” and decisions if non-consensual partner sex should be charged as rape. Sexual behavior norm beliefs were associated both directly and indirectly with latter components in the model related to acceptance of non-consensual partner sex judgments and charging rape judgments. In addition, participant gender moderated the model, such that many of the associations between the variables were stronger for males than for females. The results have implications for understanding how individuals label rape between intimate partners.
Dammit, when you get married, you kind of expect you’re going to get a little sex.
Intimate partner rape is a wide-scale problem that affects many women throughout the world and across cultures (Russell, 1990). There is debate surrounding the prevalence of intimate partner rape, with some national surveys reporting that intimate partner rape occurs more frequently than other forms of rape (i.e., acquaintance and stranger), accounting for 29-53% of all rapes (Basile, Chen, Black, & Saltzman, 2007; Black et al., 2011; Kilpatrick, Edmunds, & Seymour, 1992; Tjaden & Thoennes, 2006). However, the National Crime Victimization Survey (NCVS), which has been argued to be more representative than the national surveys mentioned above (Rand & Rennison, 2005), found that victims were more commonly raped by acquaintances (45%) than by intimate partners (33%; Bachman, 2000). Weighing in on the debate, Spivak (2011) concluded that the proportion of intimate partner and acquaintance rapes are nearly the same (26% vs. 28%, respectively). Differences in the prevalence rates of intimate partner rape can vary depending on the method of the data collection (e.g., sample differences, recruitment methods), semantics (e.g., using the word “rape” or not), and differential definitions of rape and “intimate partner” (e.g., Logan, Walker, & Cole, 2015; Rand & Rennison, 2005; Spivak, 2011).
In addition to methodological complications, intimate partner rape can be underreported if victims do not report or label their experience as rape. Although many rape victims do not seek help (George, Winfield, & Blazer, 1992; Kilpatrick et al., 1992), women who experience intimate partner rape may be even less inclined to seek help and report the incident. For example, cultural myths that family issues are “private matters” (e.g., Logan et al., 2015) or fear of an abusive partner’s retaliation, given that intimate partner rape often occurs during the ongoing context of domestic abuse (e.g., Frieze, 1983; Hanneke & Shields, 1985; Logan & Cole, 2011; Logan et al., 2015), may deter women from seeking help in intimate partner abuse situations. Society has generally invalidated intimate partner rape by not considering it “real rape.” For example, many people hold social norm beliefs related to how partners should behave in a relationship (e.g., “wifely” duty; Basile, 1999; Bennice & Resick, 2003; Lazar, 2010; Logan et al., 2015), or that intimate partner rape is not a serious crime that should be reported (Bergen, 1996; Frieze, 1983; Hanneke & Shields, 1985). Thus, when investigating how individuals conceptualize intimate partner rape as a crime, we must consider various social and cultural norms related to intimate relationships.
Researchers posit that gendered social norm beliefs about relationship behavior trickle down the line to eventually affect our decision to label and/or prosecute intimate partner rape (Bridges, 1991). However, the literature lacks research that attempts to bridge the connections between social norm beliefs, acceptance of rape, and decisions to label and prosecute rape. Previous research has typically focused on either how certain social norm beliefs, such as typical expectations for sexual intercourse in a relationship, are associated with victim blaming (Bridges, 1991; Burt, 1980; Mayerson & Taylor, 1987), or how justifications for rape/victim blaming are associated with labeling rape (Emmers-Sommer & Allen, 1999) and legal decision making regarding intimate partner rape (e.g., Auster & Leone, 2001; Kirkwood & Cecil, 2001). Thus, in an attempt to investigate the complexity of how social norm beliefs eventually and indirectly affect our decision to label and prosecute intimate partner rape, the present study tested a model that incorporates social norm beliefs about sex in relationships, acceptance of rape, labeling rape, and decisions to prosecute rape. Using a structural equation model (SEM) approach allowed us to investigate the relationship between each of these factors simultaneously rather than in isolation as seen in previous research. Integrating the known relationships between social norms, victim blaming, and legal decision making has implications for initiatives that work toward validating intimate partner rape, both culturally and legally.
Invalidating Intimate Partner Rape
Intimate partner rape is unlike non-intimate partner rape, in that victims are often in an ongoing relationship with their abuser and, as a result, have previously had consensual sex with the partner. In addition, intimate partner rape is unique as victims may continue to have consensual sex with their abuser after the rape incident and can have difficulty identifying a non-consensual sexual encounter as rape (Basile, 1999; Bergen, 1996; Finkelhor & Yllö, 1985; Koss, 1993; Logan et al., 2015). Further complicating the dynamic of intimate partner rape, the process of labeling intimate partner rape as a crime is influenced by social and cultural beliefs about “real rape.” Despite reports that 80% of the general U.S. population believe that husbands use force to have sex with their wives (Basile, 1999), marital rape is not often thought of as “real rape” (Bennice & Resick, 2003). For example, 26% of Kirkwood and Cecil’s (2001) undergraduate participants reported that they did not consider it rape when a husband engages in forced vaginal intercourse with his wife.
Even if individuals acknowledge that intimate partner rape occurs, many studies have shown that victim blaming increases and the perceived level of victim harm decreases when the relationship between the victim and offender becomes more intimate (e.g., Jeffords & Dull, 1982; Kirkwood & Cecil, 2001; Monson, Byrd, & Langhinrichsen-Rohling, 1996; Monson, Langhinrichsen-Rohling, & Binderup, 2000; Shotland & Goodstein, 1992). For example, Auster and Leone (2001) surveyed undergraduate students and found that significantly fewer males, relative to females, agreed that marital and stranger rape should be treated as similar crimes (males = 44.4%, females = 63.9%), and that husbands who rape their wives should be prosecuted (males = 39.4%, females = 61.1%). Likewise, 27% of Kirkwood and Cecil’s (2001) participants believed that the relationship between the victim and offender should be taken into account when sentencing rapists. Importantly, male participants were significantly more likely than female participants to endorse this belief.
Social and Cultural Influences on Labeling Intimate Partner Rape
The invalidation of intimate partner rape as “real rape” is particularly problematic, given that intimate partner rape often occurs in the context of coercive and psychologically abusive relationships, which do not fit the schema of “real rape” or the legal definition of rape in many states (Raphael & Logan, 2009; Schulhofer, 1998). When limited or no physical force is used in an intimate partner rape situation, the context under which the assault occurs and the intentions of both the victim and perpetrator become vital factors in labeling the situation as rape (Logan et al., 2015). Gendered social and cultural beliefs about rape and sexual behavior in relationships are likely to affect how victims do or do not view their experience as rape, as individuals get information from culture and societal norms that help them form schemas of what is normal (Gavey, 2005). Studies have shown that victims are less likely to label an experience as rape when the perpetrator is an intimate partner (Hammond & Calhoun, 2007; Kahn, Jackson, Kully, Badger, & Halvorsen, 2003; Layman, Gidycz, & Lynn, 1996; Littleton, Axsom, & Grills-Taquechel, 2009; Littleton, Breitkopf, & Berenson, 2008). For example, only 30-45% of victims of partner rape (forced sex) labeled their experience as rape compared with 55-72% of victims of non-partner rape (Kahn et al., 2003; Layman et al., 1996; Littleton et al., 2008).
Cultural Myths
Cultural myths about sexual behavior in a relationship feed beliefs that sex is a “private” matter and a “wifely duty” (Basile, 1999; Bennice & Resick, 2003; Logan et al., 2007; Russell, 1990). Such cultural myths have roots in old beliefs that a wife is a husband’s private property (Bennice & Resick, 2003; Green, 1988; Small & Tetreault, 1990), and that marriage implies continuous consent to sexual activity throughout the relationship (i.e., precedence of consent; Lazar, 2010; Logan et al., 2015). Consequently, these beliefs play a part in why both the general public and women in abusive relationships do not perceive intimate partner rape as “real” rape (Basile, 1999; Bennice & Resick, 2003; Logan et al., 2015; Raphael & Logan, 2009; Russell, 1990). In addition, intimate partners may be reluctant to label their experiences as rape to avoid labeling their significant others as rapists. Peterson and Muehlenhard’s (2007) study found evidence of this reasoning as participants felt uncomfortable that they would be labeling their boyfriends as rapists if they came forward about the rape or even admitted to themselves that the experience was rape.
Legal and Social Norms
Rape culture is built on social norms that contribute to our notion of understanding what constitutes “real rape” and “real victims” (Raphael, 2013; Raphael & Logan, 2009). Schulhofer (1998) argued that the law is an important contributor to our understanding of rape in society and that individuals likely use this information in their schemas of “real rape.” States with narrow definitions of rape (i.e., physically forced intercourse) and that do not include elements regarding coercion or threats send the message that non-consensual sex that does not meet the state’s narrow definition of rape is not rape (Schulhofer, 1998). Relatedly, legal and social norm beliefs that “real rape” is violent and that “real victims” fight back affect how individuals form schemas of rape (Raphael & Logan, 2009). Such beliefs can result in the failure of both victims and members of the justice system to label an incident, particularly involving coercion, as rape.
Gendered social norms also dictate what is normal in a relationship, and substantial evidence in the literature suggests that gender-role beliefs are associated with rape-supportive beliefs (e.g., Bridges, 1991; Burt, 1980; Mayerson & Taylor, 1987). For example, aggressiveness or persistence to obtain sex from a woman is not only considered normal but in fact is romanticized in our society (e.g., Basile, 1999; Bennice & Resick, 2003; Muehlenhard, Andrews, & Beal, 1996; Raphael, 2013; Schulhofer, 1998). Because males are socialized to pursue sex (Bennice & Resick, 2003; Bridges, 1991; Finkelhor & Yllö, 1988; Raphael, 2013), there is likely a culturally salient expectation for males to seek sex in a relationship and for females to provide sex or “give in” to sex. Furthermore, research suggests that males are more likely than females to endorse gender-role expectations, and that these expectations can lead to judgments that rape is an extension of normal male–female sexual interactions (Bridges, 1991).
Research investigating expectations for sex in hypothetical date rape vignettes has found that undergraduate male participants’ ratings of vignette characters’ expectations to have sex following a date are positively associated with victim blaming and/or rape-myth acceptance (Basow & Minieri, 2011; Claire Morr & Mongeau, 2004; Muehlenhard, Friedman, & Thomas, 1985). Although the abovementioned research investigated sexual expectations in the context of date rape, the impact of expectations for sexual behavior should also affect how individuals perceive intimate partner rape. Sexual expectations or sexual behavior norms in intimate partner relationships (e.g., a male can pressure a female partner to have sex if they have not had sex in a while) may lead to the legitimizing of female partners giving in to unwanted or coerced sex with a partner. For example, Basile (1999) found evidence for non-physical coerced sex, or rape by acquiescence, in a sample of victimized women and found that the women sometimes gave in to having sex: “It [sex] was something that I didn’t want to do, and it was something that I didn’t feel I could not do,” one participant explained (p. 1048). Another one of Basile’s participants discussed that she was raised to believe that sex was a duty that a married woman was supposed to perform. Such expectations for women to “give in” can result in the difficulty of labeling more “ambiguous rape” experiences that involve non-physical coercion and can result in coerced sex or even forced sex being seen as normative behavior (Littleton, Axsom, & Yoder, 2006).
The Present Study
We know that the process of labeling a situation as rape in intimate relationships is influenced by social and cultural norms. However, to our knowledge, there has yet to be a published study linking social norms about sex in relationships to the acceptance of non-consensual sexual behavior in relationships, and further to how such beliefs affect decisions to label and prosecute intimate partner rape. Thus, the present study tested a statistical model that investigated both the direct and indirect effects of normative beliefs about sexual behavior within a relationship on judgments of physical and non-physical intimate partner rape as a crime.
Specifically, we tested the following hypotheses based on the relevant literature. Hypothesis 1 (H1): Based on the research related to effects of sexual expectations in perceptions of rape (Basow & Minieri, 2011; Claire Morr & Mongeau, 2004; Muehlenhard et al., 1985), we hypothesized that participants’ partner sexual behavior norms (e.g., expectations for sexual behavior and acceptability to demand sex when haven’t had sex for a week to a month) would directly predict their perceived acceptability of non-consensual partner sex (e.g., acceptability to verbally demand sex, coerce sex, and physically force sex). Hypothesis 2 (H2): Based on the idea that normative beliefs supporting male pursuit of sex and female acquiescence to sex can lead to adherence to rape myths (Basile, 1999; Bridges, 1991; Littleton et al., 2006), we hypothesized that both participants’ partner sexual behavior norms and non-consensual partner sex judgments would also directly predict participants’ judgments about whether behaviors (i.e., physically forced sex and coerced sex) are “wrong but not rape.” Hypothesis 3 (H3): Based on the research suggesting that intimate partner rape is viewed as less of a real crime than non-intimate partner rape (e.g., Jeffords & Dull, 1982; Kirkwood & Cecil, 2001; Monson et al., 1996; Monson et al., 2000; Shotland & Goodstein, 1992), we hypothesized that “wrong but not rape” judgments would be negatively associated with participants’ charging partner rape judgments, which include decisions regarding whether to charge non-consensual partner sex (both physically forced and coerced) as rape. Hypothesis 4 (H4): Based on the research that suggests social norms affect how both physically forced and coerced rape is viewed as a crime (e.g., Basile, 1999; Bennice & Resick, 2003; Emmers-Sommer & Allen, 1999; Raphael, 2013; Raphael & Logan, 2009), we hypothesized that earlier measures in the model (e.g., partner sexual behavior norms) would also indirectly predict the later variables (e.g., charging partner rape judgments). Finally, Hypothesis 5 (H5): Based on gender differences in perceptions of partner rape (e.g., Auster & Leone, 2001; Jeffords & Dull, 1982; Kirkwood & Cecil, 2001; Monson et al., 1996, Monson et al., 2000; Shotland & Goodstein, 1992) and male socialization to pursue sex (Bennice & Resick, 2003; Bridges, 1991; Finkelhor & Yllö, 1988; Raphael, 2013), we hypothesized that participant gender would moderate the model, such that the associations between variables would exist for both males and females but would be stronger for males.
Method
Participants
We recruited 296 (50% females) community members online via Mechanical Turk, a participant recruitment service hosted by Amazon.com (www.mturk.com; Buhrmester, Kwang, & Gosling, 2011). All participants were at least 18 years old and were U.S. citizens. Participants’ age ranged from 18-70 years, with an average age of 33.4 years (SD = 11.8). The vast majority of participants identified as non-Hispanic Caucasian (74%), with 8.4% of participants identified as African American, 7.1% as multiracial, 5.1% as Hispanic, and the remaining 5% identified as Asian, Middle-Eastern, or “Other.” With regard to participants’ self-reported marital status, 36% were married, 31.5% were not in a romantic relationship, 20.2% were in a romantic relationship and were living with that partner, and 12.3% were in a romantic relationship but were not living with that partner.
Measures
The online survey consisted of questions in the following five domains: (a) participant demographic information, (b) partner sexual behavior norms, (c) acceptability of non-consensual partner sex, (d) “wrong but not rape” judgments, and (e) charging partner rape judgments. Participants were randomly assigned to one of two conditions: a married condition, which referred to the female partner as a wife and the male partner as a husband, and a cohabiting condition, which referred to the female as a live-in girlfriend and the male as a live-in boyfriend. In each question involving the term coercion, participants were given the following dictionary definition of coercion: “force to act or think in a certain way by use of pressure, threats, or intimidation; compel.”
Demographic Information
Participants reported their demographic information at the end of the survey. Participants indicated their gender (0 = female, 1 = male), age, race, and marital status. Due to the fact that the majority of participants identified themselves as White and there was little variability among non-White participants, race was coded such that 0 = non-White or multiracial and 1 = White. In the analyses, marital status was coded, such that 1 = not in a cohabitating relationship, 2 = in a non-married, cohabitating relationship, and 3 = in a married, cohabitating relationship.
Partner Sexual Behavior Norms
We used four partner sexual behavior norm variables based on five questions related to sexual behavior norms in intimate partner relationships. We created these measures based on studies that investigated expectations for sex in a date rape context (Basow & Minieri, 2011; Claire Morr & Mongeau, 2004; Monson et al., 2000; Muehlenhard et al., 1985), which asked participants to rate how much a female is obligated to have sex and expects to have sex with a male date/partner. However, we are not aware of any published research that has specifically used the sexual behavior norm measures as they appear in the present study.
Participants indicated on a 5-point scale (1 = definitely disagree, 2 = disagree, 3 = neither agree nor disagree, 4 = agree, 5 = definitely agree) how much they agreed or disagreed with each of the five statements. Two of the variables were related to participants’ endorsement of a male partner’s expectations for sexual intercourse: (a) a male should expect to have sexual intercourse with his female partner at least once every other day and (b) a male should expect to have sexual intercourse with his female partner at least once a week. The following two statements were highly correlated (r = .92), therefore we averaged the items: It is acceptable for a male to demand sex from his female partner if they had not had sex in 1 week, and it is acceptable for a male to demand sex from his female partner if they had not had sex in 1 month, to create a single variable labeled, (c) “acceptable for a male partner to demand sex after a short latency”: We created the acceptability to demand sex after a short latency variable as a means to establish if expecting sex leads to the justification of demanding sex from a female partner if a male has not had sex after only a day or week. The fourth and final partner sexual behavior norm variable consisted of the following statement: (d) A female should be expected to have sexual intercourse with her male partner whenever he wants.
Non-Consensual Partner Sex Judgments
Three variables reflected each participant’s general acceptance of three different forms of non-consensual partner sex. We created these measures based on Burt’s (1980) Acceptance of Interpersonal Violence Against Women scale. Participants rated how much they agreed or disagreed on a 5-point scale (1 = definitely disagree, 2 = disagree, 3 = neither agree nor disagree, 4 = agree, 5 = definitely agree) with the following three statements: (a) it is acceptable for a male to coerce his female partner to have sex with him, (b) it is acceptable for a male to verbally demand his female partner have sex with him, and (c) it is acceptable for a male to physically force his female partner to have sex with him.
“Wrong But Not Rape” Judgments
Two “wrong but not rape” variables reflected the following two statements, with which participants rated how much they agreed or disagreed on a 5-point scale (1 = definitely disagree, 2 = disagree, 3 = neither agree nor disagree, 4 = agree, 5 = definitely agree): (a) if a male coerces his female partner to have sex it is wrong but not rape and (b) if a male physically forces his female partner to have sex it is wrong but not rape. We created these two questions based on the construct of “It Wasn’t Really Rape” within the Illinois Rape Myth Acceptance Scale (Payne, Lonsway, & Fitzgerald, 1999), which speaks both to the justification of a rapist’s behavior and labeling an incident as “real rape.”
Charging Partner Rape Judgments
Finally, participants completed two charging partner rape judgment questions. We adapted this measure from Auster and Leone (2001); however, we asked participants to rate how much they agreed that a male partner should be charged with rape (vs. a dichotomous yes/no question) and included a question for physically forced and coerced rape. Participants rated on a 5-point scale (1 = definitely disagree, 2 = disagree, 3 = neither agree nor disagree, 4 = agree, 5 = definitely agree) how much they agreed with the following statements: (a) a male who coerces his female partner to have sex should be charged with rape and (b) a male who physically forces his female partner to have sex should be charged with rape.
Procedure
Participants completed an online consent form and were directed to the survey, which was administered via Surveymonkey.com. Participants read instructions informing them that they would be asked about their thoughts and expectations related to heterosexual intimate partners. We assured participants that their responses were confidential. Participants answered questions in the following single order: partner sexual behavior norms, acceptability of non-consensual partner sex, wrong but not partner rape judgments, and charging partner rape judgments. The order of the questions reflects the direction of the associations in our predicted SEM. After completing the survey, participants provided demographic information and were given electronic copies of the consent form and debriefing form.
Results
Data Analysis Plan
First, we investigated if there were differences in mean partner sexual expectation norms, non-consensual partner sex judgments, “wrong but not rape” judgments, and charging partner rape judgments as a function of participant gender and marital status condition. Given that we found no significant differences on any of the model variables as a function of the marital status condition (i.e., the male and female in each statement were married or not), only descriptive differences as a function of participant gender are presented. Second, we used a SEM in our primary analyses to investigate the multivariate relations between gender, partner sexual expectation norms, non-consensual partner sex judgments, “wrong but not rape” judgments, and charging partner rape judgments. Using the SEM method, we were able to create a model representing how each variable related to the others, which ultimately gives us a more complete picture of the different paths from participants’ endorsed partner sexual behavior norms to their judgments of non-consensual partner sex to their decisions of charging partner rape. Third, we tested the significance of the indirect effects found in the predicted model. These indirect effects explain the associations within a model above and beyond the direct effects. Fourth, we investigated participant gender as a moderator of each individual path in the hypothesized model. Thus, we were able to determine if specific associations were stronger or weaker for one gender than the other.
Differences on Participant Gender
Correlations, means, and standard deviations are shown separately by participant gender in Table 1. We observed significant participant gender differences on the majority of the rating measures using a MANOVA. With regard to partner sexual behavior norm questions, males agreed more than females that a male partner should expect sex from a female partner both every other day, F(1, 294) = 18.93, p < .001, and once a week, F(1, 294) = 9.12, p = .003. Males versus females had higher scores for demanding sex after not having sex for a short latency (i.e., acceptable to demand sex after short latency), F(1, 294) = 22.57, p < .001. Males also agreed more than females that a female partner is expected to have sex with a male partner whenever the male partner wants, F(1, 294) = 26.37, p < .001.
Correlations, Means, and Standard Deviations (N = 360).
Note. Correlations for males are above the diagonal; correlations for female participants are below the diagonal. Means and standard deviations for males are on the right side of the table; means and standard deviations for females are on the bottom of the table.
p < .05. **p < .01. ***p < .001.
With regard to non-consensual partner rape judgments, males agreed significantly more than females that it is acceptable for a male partner to coerce his female partner to have sex, F(1, 294) = 9.90, p = .002, and it is acceptable for a male partner to verbally demand his female partner to have sex, F(1, 294) = 14.61, p < .001. However, males and females did not significantly differ on their agreement with the acceptability for a male partner to physically force his female partner to have sex.
As for “wrong but not rape” judgments, males agreed significantly more than females that a male partner coercing his female partner to have sex is “wrong but not rape,” F(1, 294) = 4.52, p = .045. However, no differences were found regarding males’ and females’ agreement that using physical force is wrong but not rape. Last, males agreed significantly less than females that a male partner should be charged with rape when he coerces his partner to have sex, F(1, 294) = 4.54, p = .034, but males and females did not differ on the agreement that a male partner should be charged with rape when he physically forces his female partner to have sex.
The Predicted SEM
We tested our hypotheses via SEM using AMOS version 20. The hypothesized model can be seen in either Figure 1 (model for females) or Figure 2 (model for males). The model proposed that participants’ sexual behavior norms would be associated with their judgments of what constitutes rape among intimate partners. Overall, the model presented in Figure 1 was a good fit for the data, χ2(72) = 160.69, p < .001; χ2 / df = 2.21; comparative fit index (CFI) = .947; standardized root mean square residual (SRMR) = .087; root mean square error of approximation (RMSEA) = .064, 90% confidence interval (CI) = [0.050, 0.078].

The hypothesized structural equation model for female participants.

The hypothesized structural equation model for male participants.
Partner Sexual Behavior Norms Predict Non-Consensual Partner Sex Judgments
Within the four sexual behavior norms, endorsement of the idea that a male partner should expect sex every week and every month was associated with endorsement of the statements that females should have sex whenever the male wants and acceptability for a male to demand sex after not having sex after a short latency. In addition, for females only, the idea that a female should have sex whenever a male wants was associated with acceptability for a male to demand sex after a short latency.
We hypothesized that these partner sexual behavior norms would also be related to participants’ perceived acceptability of non-consensual partner sex (H1). As shown in the model, endorsement of the sexual behavior norm that it is acceptable for a male to demand sex after a short latency was associated with endorsing two of the non-consensual partner sex judgments: acceptability for a male to coerce and verbally demand sex. Likewise, participants endorsing the sexual behavior norm that females should have sex whenever a male wants was associated with endorsing the non-consensual sex judgment that it is acceptable for a male to physically force sex. Thus, H1 was supported.
“Wrong But Not Rape” Judgments
We hypothesized that participants’ partner sexual behavior norms and non-consensual partner sex judgments would be related to their endorsement of statements related to labeling an incident as wrong rather than rape (H2). As shown in the model, endorsement of the partner sexual behavior norms that men should expect and demand sex was associated with agreeing that coerced sex and physically forced partner sex are wrong but not rape. In addition, higher agreement of acceptability for a male to physically force sex was related to higher agreement that physically forced sex is wrong but is not rape. To that end, H2 was supported. However, higher agreement that it is acceptable for a male to coerce sex or for a male to verbally demand sex was not associated with the idea that coerced sex is wrong but not rape.
Predicting Charging Partner Rape Judgments
Finally, we hypothesized that participants’ “wrong but not rape” judgments would be negatively related to their judgments about charging intimate partner rape (H3). As shown in the model, stronger agreement that coercion is wrong but not rape was negatively related to agreement that a male partner should be charged with rape when he coerces his female partner to have sex. Relatedly, agreement that physically forced sex is wrong but not rape was negatively associated with agreeing that a male partner who physically forces his female partner to have sex should be charged with rape. Thus, we also found support for H3.
Tests of Indirect Effects
We hypothesized that there would be numerous indirect effects in addition to the aforementioned direct effects (H4). Indirect effects allow one to observe if a variable is related to an outcome variable through (i.e., indirectly) one or more other variables. This allows us to investigate the relationship between a set variables in a more complex way rather than simply only observing direct associations within a model. To test this hypothesis, we tested the significance of indirect effects for both males and females using bootstrapping. Bootstrapping is a process by which N observations are randomly drawn, using replacement, from the original sample of N participants to create a new dataset. Analyses are conducted on the new dataset, which provides parameter estimates of interest. This process is repeated many times (often as many as 500-1,000), creating a distribution of parameter estimates (Efron & Tibshirani, 1998). Finally, a 95% CI is produced, indicating the significance of the parameter estimate. Indirect effects are computed by multiplying the path coefficient of the link between the independent variable and the mediator by the path coefficient of the link between the mediator and the dependent variable. We conducted bootstrapping with AMOS 20, and 95% CIs were based upon 500 bootstrap samples. Because there were no missing data, no cases were excluded from bootstrapping analyses. Reported CIs are for standardized indirect effects and are bias corrected.
The results of bootstrapping indicated numerous indirect effects, and all indirect effects are shown in Table 2 (for females) and Table 3 (for males). We found a number of differences in significant indirect effects between males and females. Such paths are denoted in the tables. As seen in the predicted model and in Tables 2 and 3, the initial partner sexual behavior norm variables in the model were not only directly related to the non-consensual partner judgments but were also indirectly related to participants’ judgments of what partner behavior is “wrong but not rape” and charging rape judgments. Of note, endorsement of the idea that females should have sex whenever the male wants was directly and indirectly related to non-consensual partner sex judgments, “wrong but not rape” judgments, and charging rape judgments for females but only directly related to one variable (“It is acceptable for the male partner to physically force sex”) for males.
Coefficients, Standard Errors, and Confidence Intervals of Indirect Effects for Females.
Note. All indirect effects shown are significant. Bolded paths were significant for females but not for males.
Coefficients and Standard Errors of Indirect Effects for Males.
Note. All indirect effects shown are significant. No paths were significant for males but not females.
Participant Gender as a Moderator
Last, we hypothesized that gender would moderate the associations within the SEM, such that each of the hypothesized effects, except for agreement that physically forced and coerced partner sex should be charged as rape, would be stronger for males than for females (H5). To test moderation using SEM, AMOS runs the hypothesized model separately for each group (i.e., female, male), producing unique coefficients for each path. Results indicated that not every path was significantly moderated by gender; thus, H5 was partially supported. The models are shown in Figure 1 (for females) and Figure 2 (for males); moderated paths are denoted by the letter m.
As seen in Figures 1 and 2, four paths were significantly moderated by gender. In most cases, the moderation took the hypothesized form of the associations being stronger for males. First, males who endorsed greater sexual expectations of a male partner (i.e., a male partner should expect to have sex with his female partner every other day and once a week) more strongly endorsed the idea that the female should have sex whenever the male partner wants relative to females who expected males to have greater sexual expectations. Next, males who felt that it was acceptable for the male partner to demand sex after a short latency more strongly endorsed the statement that it is acceptable for males to verbally demand sex, relative to females. Finally, males who endorsed the statement that physically forced sex is wrong but not rape yielded less agreement, relative to females, that males who physically forced sex should be charged with rape. Importantly, all of these associations were significant for females as well, but the associations were not as strong as they were for males. Only one path was moderated such that the association was stronger for females than for males. Specifically, for females, endorsement of the idea that a female should have sex whenever the male partner wants was positively associated with beliefs of acceptability for the male partner to demand sex after a short latency. This association was not significant for males.
Discussion
The present study investigated how participant gender moderated the interplay between participants’ partner sexual behavior norms, acceptance of both physical and non-physical non-consensual sex judgments in relationships, and judgments of a situation as “wrong not rape,” in determining how these associations lead to judgments about charging partner rape as a crime. Overall, participants who agreed that it is acceptable for male partners to demand sex from their female partners after a relatively short amount of time (i.e., a day/week) were more likely to (a) agree with the acceptability of men verbally demanding sex from their female partners, (b) agree with the acceptability of men coercing sex from their female partners, and (c) agree that physically forced sex, as well as coerced sex, is wrong but not rape. In turn, agreeing that coerced sex and physically forced sex are wrong but not rape was negatively related to agreeing that men who do so should be charged with rape. Finally, some of the associations in the model were stronger for male participants than for female participants, supporting the vast literature that men are more prone to victim blaming and rape-myth acceptance than are women (e.g., Auster & Leone, 2001; Kirkwood & Cecil, 2001; Lynch, Wasarhaley, Golding, & Simcic, 2013; Schutte & Hosch, 1997).
There were several key findings that warrant further discussion. One of the primary findings of the present study was that sexual behavior norms primarily had an indirect effect, rather than a direct effect, on labeling experiences as rape (vs. just wrong) and decisions to charge intimate partners with rape, through the acceptability of using force and coercing female partners to have sex. Specifically, participants who agreed that women should have sex with their male partners whenever those partners want were more likely to find it acceptable for men to use physical force on their female partners. Participants who found it acceptable for men to use physical force on their female partners were also more likely to agree that such force is wrong but not rape, and those who think such incidents are wrong but not rape are less likely to believe that men in such situations should be charged with rape. Taken together, the results not only provide evidence for and support the notion that gendered sexual norm beliefs are associated with judgments of intimate partner rape as a crime (Basile, 1999; Bennice & Resick, 2003; Raphael, 2013; Raphael & Logan, 2009), but that these beliefs are also largely indirectly associated with such overt judgments about intimate partner rape (Bridges, 1991).
A second key finding consisted of the distinct paths related to physically forced versus coerced partner sex. Although the path from sexual behavior norms to judgments that physically forced partner sex should be charged as rape was relatively simple, the path to the judgment that coerced partner sex should be charged as rape was more complex. For example, neither of the two non-consensual partner judgments pertaining to non-physical rape (i.e., acceptable to coerce sex and verbally demand sex) were associated with agreement that coerced sex is “wrong but not rape” or agreement that males who coerce female partners to have sex should be charged with rape. Instead, two sexual behavior norm variables for male participants and all three sexual behavior norm variables for female participants were indirectly related to the judgment that coerced partner sex should be charged as rape through the belief that coerced sex is wrong but not rape. It appears that it could be more difficult for individuals to acknowledge that non-physically coerced partner rape is “real rape,” which is likely because it does not fit the schema of violent rape by a stranger (Raphael & Logan, 2009).
Relatedly, the belief that females should have sex whenever their male partner wants was directly related to acceptance of physically forced sex but not coerced or verbally demanded partner sex. However, the belief that it is acceptable for a male to demand sex from his female partner after a short amount of time was associated with acceptance to coerce and demand sex but not physically force sex. Thus, the model suggests that certain sexual behavior norm beliefs are related to the acceptance of physically forced rape, while other beliefs are associated with non-physically forced partner rape. Judgments about coerced partner sex may be more directly associated with gendered social norms related to males as the pursuers of sex in a relationship (Bennice & Resick, 2003; Finkelhor & Yllö, 1988; Raphael, 2013), while the expectations for females to “give in” to sex may be associated with the acceptance of physically forced sex (Basile, 1999). Emmers-Sommer and Allen (1999) found in their model that attitudes toward women and rape-myth adherence were associated with justifications of rape by coercion; however, there is a dearth of research that compares rape (supportive) social norm beliefs that are associated with both physically forced and coerced rape. Therefore, the complex process of the impact of social norms on labeling coerced versus physically forced intimate partner sex warrants further investigation.
Another noteworthy finding in the present study was the moderation by participant gender. Specifically, there were four direct effects moderated by participant gender, three of which involved male participants more strongly endorsing sexual social behavior norms or agreeing less that physically forced partner rape should be charged as rape. These effects support the vast literature that has shown that males blame rape victims more and justify rape more than females (e.g., Auster & Leone, 2001; Kirkwood & Cecil, 2001; Lynch et al., 2013; Schutte & Hosch, 1997). In contrast, one female path involved a weak association between the idea that females should have sex whenever a male wants and acceptability for a male to demand sex after a short latency. This same path was not significant for male participants, though the rationale behind this finding is unclear. The surrounding associations between partner sexual behavior norms were much stronger for males than for females, so it is possible that, for males but not females, controlling for those associations has subsumed the unique relationship between the idea that females should have sex whenever a male wants and acceptability for a male to demand sex after a short latency. It is also possible that females more strongly identify with the obligation to have sex with a male partner. This finding demonstrates the complexities of the relationship between social norm beliefs for males versus females. Future research is needed to investigate gender differences in social norm beliefs and how these differences relate to how intimate partner rape is perceived between genders.
The results of the present study have implications for both research and practical application related to intimate partner rape. First, this study was the first published study to test a model that allowed for investigation of the associations between sexual behavior norms in an intimate relationship and judgments about physical and non-physical intimate partner rape simultaneously rather than in isolation. Typically, prior research in this area would isolate various stages in the model, such as judgments about intimate partner rape as a crime (e.g., Kirkwood & Cecil, 2001). However, using SEM allowed us to investigate various components of the process of using sexual behavior norms in a relationship to form judgments of acceptable non-consensual sex behavior and, ultimately, judgments regarding if the act is “wrong,” “rape,” and/or should be charged as a crime. Second, gendered sexual expectations to have or demand sex in relationships in relation to the formation of an “Is it rape?” judgment are underinvestigated aspects of the literature. To address this issue, the present study used expectations for sexual behavior as a proxy for sexual behavior norms in intimate partner relationships. There has been a focus on similar sexual norm expectations in date rape scenarios (e.g., Basow & Minieri, 2011; Claire Morr & Mongeau, 2004; Muehlenhard et al., 1985) but not within an intimate partner context, which involves many relationship expectancies or norms related to sex because sexual activity is a part of an intimate relationship.
The present study also has several practical implications related to policy and the prosecution of rape. The process by which victims label their experience as “real rape” is complicated and influenced by many social and cultural factors. Continuing to test models that investigate how gendered social norms are related to the justification of intimate partner rape experiences and the invalidation of rape as a real crime can provide information that may help law enforcement and victim advocates in their counseling of intimate partner rape victims. Findings from such models can also be used for initiatives to raise public awareness about the dynamics of intimate partner rape and to fight cultural myths that invalidate intimate partner rape. Finally, the present study has implications for legal scholars investigating both rape laws and rape in court. With regard to rape law research, it may be helpful to use models linking social norm expectations to beliefs about actual non-consensual acts to beliefs about those acts as crimes. For example, in the present study we found that endorsement of the idea that coercion is “wrong but not rape” was negatively associated with agreeing that coerced partner sex should be charged as rape. However, agreeing that coercion is “wrong but not rape” did not predict agreement that physically forced sex should be charged as rape. The same pattern held true for believing that physically forced sex is wrong but not rape: Endorsement of this statement was not associated with beliefs about whether to charge coerced sex. If participants believe that coercion is not the same as physically forced sex and is perhaps not even the same type of crime, this information can be used to help legal reform to expand the definition of rape in rape statutes.
Although the present study offers a unique contribution to better understanding individuals’ endorsement of social norms related to partner rape, we would like to address several limitations. First, participants were only asked about how much they agreed with statements that focused on a prototypical instance of partner abuse, in which the male was the sexual aggressor and the female was on the receiving end of the sexual pursuit. Given that females also pursue and expect sex in a relationship and can also perpetrate rape, it would be valuable for future research to also ask participants to respond to statements with non-prototypical situations (i.e., the female as the sexual aggressor). Second, and related to the first limitation, the generalizability of the results is limited to heterosexual couples and prototypical abuse situations, in which the male is the sexual aggressor. Because beliefs about roles of masculinity and femininity may be more complicated in lesbian couples, for example, with both partners female (Hassouneh & Glass, 2008), the relationships among the variables in the present model may be different or possibly moderated by the perceived masculinity of the aggressor/recipient of the aggressed behavior. Relatedly, Felson and Cundiff (2014) used data collected from the National Incident-Based Reporting System (NIBRS) and found that rates of male-to-male sexual assault were at least as high as male-to-female sexual assault. Felson and Cundiff suggested that, given the similarities in homosexual and heterosexual assaults, social beliefs or norms beyond negative attitudes toward women must be considered when investigating perpetration of sexual assault.
Third, although the present study measured demographic characteristics of the sample, other information about participants’ previous experience with intimate partner abuse or level of conservatism may also be important to incorporate or control for in the model. Fourth, the present study posed potential limitations with the use of its measures. For example, only quantitative data were used, and open-ended responses regarding beliefs about partner rape would be an excellent way to supplement a statistical model. In addition, the measures were primarily created by the authors though they were guided by previously published measures and constructs. However, the SEM methodology controls for measurement error associated with relatively novel measures. Nevertheless, future research should focus on the psychometric properties of the measures used in the present study and the reliability across populations. Finally, given that our study was cross-sectional and SEM is correlational in nature, casual inferences cannot be drawn from this study.
Future research can take several avenues to build on the current findings. First, it would be valuable to test a model of social norm beliefs and conceptualization of partner rape using both prototypical and non-prototypical abuse situations. For example, the measures used in the present study should be asked both from the point of view of a male partner as a sexual pursuant or aggressor and the female partner as the sexual pursuant or aggressor (e.g., It is acceptable for a female partner to verbally demand her male partner have sexual intercourse with her). Second, the present study did not find any significant differences in the variables in the model between conditions in which the statements described a married, cohabitating couple versus a non-married, cohabitating couple. However, it would be worthwhile for future research to focus on other relationship stages of intimate partners, such as non-cohabitating couples. Furthermore, the dynamic among partners who experience intimate partner rape can be complicated by back-and-forth separation between partners (Logan et al., 2015), and thus the relationship and dynamic among partners mentioned in the statements provided to participants should be further investigated.
Third, participant characteristics and experiences should be further incorporated into future investigations on the topic. Although the present study focused primarily on participant gender, other participant characteristics, such as race, could also be potential moderators of the model. However, given that the majority of participants in the present study were White and the racial distribution among non-White participants was very limited, we did not feel it would be appropriate to make generalizations across all non-White participants. Future research examining the impact of participant characteristics, such as race, marital status, and religiosity, should aim to recruit participants based on such characteristics, rather than measuring characteristics of the participants, so they may be properly used as moderators or quasi-independent variables. Fourth and finally, the measures used in the present study should be further developed and tested to increase the validity of the model and its variables. Relatedly, other measures should be tested in a similar model, such as beliefs about a partner’s duties in a relationship (e.g., “wifely duty”; Basile, 1999; Bennice & Resick, 2003; Lazar, 2010; Logan et al., 2015) or existing, validated scales related to gender roles or sexism, to further understand how social norm beliefs contribute to beliefs that excuse or justify partner rape.
In conclusion, the present study investigated the interplay between sexual behavior norms and beliefs about what is acceptable, wrong, or rape in intimate partner rape situations. We hope that the present study can be used as a stepping stone for future research that will investigate the complexity of gendered social norm beliefs as a means to build a framework that increases understanding of how individuals “decide” what is and is not “real” intimate partner rape. Further research in this area will hopefully allow us to better understand the mechanisms underlying the cultural and legal invalidation of intimate partner rape as a means to bring justice and validation to its victims.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
