Introduction

Sexual assault is a prevalent societal problem that can result in a variety of damaging physical, psychological, social, and behavioral health consequences (Centers for Disease Control and Prevention, 2007). Research on sexual assault among ethnic minority groups, especially Asian Americans, is minimal. Asian Americans are reportedly the fastest growing ethnic minority group in the U.S. (U.S. Census Bureau, 2004). In the most recent bulletin from the National Crime Victimization Survey, although all other types of violent crime rates decreased, sexual assault rates increased 25% in 2007 from 2005 (Rand, 2008). These reports demonstrate the importance of continued work in sexual assault, especially among the rising population of Asian Americans.

Because 95% of sexual perpetrators are male (Bureau of Justice Statistics, 1995), theories and models of sexual aggression have underscored the importance of studying males’ attitudes toward rape. Researchers have consistently found that Asian American college students hold more rape-supportive attitudes than their non-Asian counterparts (Kennedy & Gorzalka, 2002; Lee, Pomeroy, Yoo, & Rheinboldt, 2005; Mori, Bernat, Glenn, Selle, & Zarate, 1995). To date, such studies have not identified factors contributing to this phenomenon. Because research has shown that attitudes are malleable and that rape-supportive attitudes are predictive of men’s perpetration of sexually aggressive behavior (Malamuth, Linz, Heavey, Barnes, & Acker, 1995), rape-supportive attitudes have been targeted for primary prevention of male sexual aggression (Brecklin & Forde, 2001). By examining Asian American cultural constructs, the current study aimed to advance the understanding of this tendency of Asian American college men to hold more rape-supportive attitudes than their non-Asian counterparts.

Misogyny

It has been posited that Asian American men experience a different cultural socialization process than White Americans in relation to sexual aggression (Hall & Barongan, 1997; Hall, Teten, DeGarmo, Sue, & Stephens, 2005). Although Asian American college men are a heterogeneous group made up of multiple ethnicities, they share several unifying cultural values (Kim, Li, & Ng, 2005), including patriarchal values (Ho, 1990; Kim & Ward, 2007; Mills & Granoff, 1992; Okazaki, 2002). These shared values can result in a cultural socialization process that is more supportive of traditional sex roles than mainstream American culture. Inextricably a part of patriarchal values is misogyny (Boler, 2005), a general hostility, mistrust, and devaluing of women. Misogyny has not only been linked to rape (Shotland, 1985), but misogynistic beliefs are predictive of sexually aggressive behavior and positively correlated with rape-supportive attitudes. Furthermore, Malamuth et al.’s (1995) confluence model, a comprehensive explanatory model of men’s sexual aggression, includes misogynistic beliefs as a component of explaining sexual aggression. Research has not specifically investigated misogynistic beliefs among various ethnic groups. Doing so is important because they may vary as a function of cultural socialization, particularly given the different cultural influences that have been found to be related to sexually aggressive behavior among Asian American men (Hall et al., 2005).

Cultural Constructs

Although misogyny has not been investigated directly, other cultural socialization constructs related to rape-supportive attitudes have begun to be examined among Asian American college men. Two studies examined the concept of acculturation and its relationship with rape attitudes, and both suggest the importance of further investigation of this relationship. Acculturation refers to an Asian American adopting or participating in more “Western” mainstream beliefs and behaviors. Thus, the higher the acculturation, the more affiliated one is with “Western” identity, and the more one is culturally socialized to mainstream American culture. Related to rape attitudes and acculturation, Kennedy and Gorzalka (2002) reported a decrease in tolerance of rape myths with increased residence in Canada for college students of Asian origin. Kennedy and Gorzalka attributed this effect to an exposure to Western values, implying a possible acculturation effect, but they did not directly measure acculturation. Similarly, Mori et al. (1995) found that among Asian American college students, those who were categorized as high on acculturation held more prosocial attitudes toward rape victims and endorsed rape myths less than those who were categorized as low on acculturation. However, because median splits were performed on acculturation level in Mori et al.’s study, the robustness of this finding is unclear (for a discussion about the drawbacks of dichotomizing a continuous variable, see MacCallum, Zhang, Preacher, & Rucker, 2002). Nonetheless, despite these studies’ shortcomings, the available evidence suggests that the more acculturated one is, the less one would hold rape-supportive attitudes.

Also related to acculturation is the construct of ethnic identity. Defined by Tajfel (1981), ethnic identity is one’s self-concept derived from a combination of one’s knowledge of membership in a social group plus the value and emotional significance attached to that membership. Having conducted the sole study examining ethnic identity and sexual aggression, Hall et al. (2005) found that among Asian American college men, ethnic identity served as a protective factor against perpetrating sexually aggressive behavior. This finding would suggest that stronger ethnic identification may be associated with holding attitudes less supportive of rape, yet ethnic identity in relation to rape-supportive attitudes remains unexplored.

It is important to note that by comparing Asian Americans to White Americans, we are not implying that White values are the standard of normality. The current study investigated an overlooked group (Asian Americans) and compared them to the group that has been traditionally researched, White Americans. Our purpose was to emphasize differences in cultural socialization that may affect rape-supportive attitudes and to advance our understanding of these differences by incorporating culturally relevant constructs.

Present Study

The current study investigated misogyny, acculturation, and ethnic identity—constructs reflective of shared cultural socialization experiences among Asian American men. Specifically, we sought to replicate findings that Asian American men have more rape-supportive attitudes compared to White men and investigate the relationship between rape-supportive attitudes and misogyny, acculturation, and ethnic identity. Rape-supportive attitudes were measured in the context of an acquaintance rape vignette, where greater rape-supportive attitudes were indicated by blaming the victim in the vignette more and the perpetrator less, perceiving the victim’s refusal to be less credible, and defining the vignette scenario as a “rape” less. Three hypotheses were tested: (1) Asian American men would make more rape-supportive judgments of the vignette than White men; (2) misogyny would mediate the relationship between ethnicity and rape vignette judgments—namely, higher misogynistic beliefs will be associated with ethnicity and will account for the variance in the relationship between ethnicity and rape-supportive judgments; and (3) among Asian American men, acculturation would account for significant variance in each rape vignette judgment, above and beyond variance accounted for by generational status, such that higher acculturation would be associated with lower rape-supportive vignette judgments, and ethnic identity would account for significant variance in each rape vignette judgment, such that higher ethnic identity will be associated with lower rape-supportive vignette judgments.

Method

Participants

Participants were male undergraduates (N = 377) from three large public universities in the U.S. Participants self-identified as either Asian/Asian American (n = 155) or White/Caucasian (n = 222). The mean age of the Asian Americans was 20.57 years (SD = 4.00) and 20.44 years (SD = 6.36) for the Whites. Approximately, 57% of Whites and 47% of Asian Americans had consensual sexual intercourse with at least one female. Mean ages of first consensual sexual intercourse with a female were similar for Asian Americans (M = 17.41 years, SD = 1.89) and Whites (M = 17.45 years, SD = 1.48). Approximately 48% (n = 74) of Asian Americans and 3% (n = 7) of Whites identified as “first generation,” 27% (n = 42) of Asian Americans and 6% (n = 13) of Whites as “second generation,” 22.6% (n = 35) of Asian Americans and 88% (n = 194) of Whites as “third generation or higher,” and 1% (n = 3) of Whites endorsed “unknown/unsure.” No Asian Americans endorsed “unknown/unsure” for generational status. Asian American participants consisted of Chinese (n = 53, 33.5% of this sample), Japanese (n = 29, 18.7%), Korean (n = 20, 12.9%), Filipino (n = 11, 7.1%), Vietnamese (n = 9, 5.2%), or mixed-Asian (e.g., Korean and Chinese; n = 32, 20.6%). Mixed ethnicity participants, i.e., those who were of Asian and non-Asian descent (e.g., White and Filipino), were excluded from analyses.

Measures

Misogynistic Beliefs

Similar to how Hall et al. (2005) operationalized Malamuth et al.’s (1995) hostile masculinity construct of the confluence model of sexual aggression, a composite score was created from three scales posing questions on perceptions of women to assess misogynistic beliefs (Lonsway & Fitzgerald, 1995). The Adversarial Heterosexual Beliefs (AHB) scale, a component of Hall et al.’s (2005) hostile masculinity composite, measured beliefs about the adversarial nature of male–female relationships at an individual and societal level. Items included, “In all societies, it is inevitable that one sex is dominant,” and “It’s impossible for men and women to truly understand each other.” There was good internal consistency with this sample for each ethnic group (Cronbach’s α White Americans (WA) = .85; α Asian Americans (AA) = .83). Also a component of Hall et al.’s (2005) hostile masculinity composite, the Revised Hostility Towards Women (HTW) scale measured perceptions of women and included items such as “I think that most women would lie just to get ahead,” and “Sometimes women bother me by just being around.” The HTW scale with this sample was also found to have good internal consistency (Cronbach’s α WA = .87; α AA = .81). Lastly, because we examined rape attitudes specifically as a part of misogyny, we included the Rape Myths Acceptance Scale (RMAS), which assessed the attitudes and generally false beliefs about rape that function to deny and justify male sexual aggression. Sample items included, “When women talk and act sexy, they are inviting rape,” “If a woman doesn’t physically fight back, you can’t really say that it was rape.” The RMAS showed good internal consistency with this sample (Cronbach’s α WA = .92; α AA = .91). Participants rated each item on a 7-point Likert scale. Z-scores of the three scales were calculated (accounting for all reverse-coded items) and summed to create a composite score representing the variable of “misogynistic beliefs.” The composite score range was minimum score of −5.52 and a maximum 8.18.

Acquaintance Rape Vignette

The vignette, originally developed for a previous study (George & Martinez, 2002), presented an acquaintance rape. The vignette described a female character who clearly refused sexual advances made by a male acquaintance (a neighbor), which ultimately led to forced vaginal penetration without consent.Footnote 1 See Appendix for full vignette. After reading the acquaintance rape vignette, participants rated how much they blamed the perpetrator in the vignette (Perpetrator Blame), how much they blamed the victim (Victim Blame), how credible they perceived the victim’s refusal (Refusal Credibility), and the extent to which they defined the incident as rape (Rape Defined). With the exception of Rape Defined (which was a single item), each judgment was assessed with multiple items. All items were rated on a 10-point continuous scale, where 1 represented rape-supportive and 10 represented not rape-supportive.

Perpetrator Blame and Victim Blame were adapted from measures used by a vignette study conducted by George and Martinez (2002), which incorporated the different dimensions of cause, responsibility, and blame (Calhoun & Townsley, 1991). To assess Perpetrator Blame, participants were asked five questions including, “How much was it the man’s fault that he engaged in sexual activity with the woman?” Reliability in the current study was acceptable for Asian Americans (Cronbach’s α = .84) and low for White Americans (Cronbach’s α = .35) for the variable of Perpetrator Blame. Participants were also asked how much they blamed the victim for the events that unfolded in the vignette with six items that were combined to create Victim Blame (Cronbach’s α WA = .86; α AA = .81). A sample question is “In your opinion, to what extent did the woman cause what happened sexually in the story you read?”

Participants additionally evaluated how credible the victim’s refusal (Refusal Credibility) was by answering six questions such as, “How much did the woman really want the man to have sex with her?” and “When the woman said no, how much did she mean it?” (Fors, 1993). Internal consistency was acceptable with this sample (Cronbach’s α WA = .71; α AA = .83). Last, the degree to which participants defined the vignette as rape (Rape Defined) was indexed by a single item, used in previous research (e.g., George & Martinez, 2002), asking, “To what extent would you describe the behavior of the man toward the woman as rape?” (Bridges, 1991).

Data transformations were conducted to address skewness and kurtosis in accordance with guidelines provided by Tabachnick and Fidell (2007). Refusal Credibility was square rooted and both Perpetrator Blame and Rape Defined were log transformed to reduce the non-normality of the distributions. Victim Blame did not require data transformation.

Acculturation

The Suinn–Lew Asian Self-identity Acculturation Scale (SL-ASIA; Suinn, Rickard-Figueroa, Lew, & Vigil, 1987) is a 25-item scale that assessed acculturation for the Asian American participants. Participants were asked to rate on a scale of 1–5 how much they behave in accordance to and identify with “Asian” culture, “American” culture, or both (1 = highest affiliation with solely Asian culture, 3 = affiliation with both Asian and American culture equally, and 5 = highest affiliation with solely American/non-Asian culture). This scale showed good internal consistency with this sample (Cronbach’s α = .91).

Generational Status

Responding to a single item, each participant indicated his generational status by choosing one of the following: first generation (born outside the U.S.), second generation (at least one of his parents was born outside the U.S.), third generation (where at least one of his grandparents was born outside the U.S.), fourth generation, fifth generation, or unknown.

Ethnic Identity

The Multigroup Ethnic Identity Measure (MEIM; Phinney, 1992) assessed ethnic identity with 12 items. Asian American participants were asked to rate on a scale of 1 (strongly disagree) to 4 (strong agree) to what extent they agreed with items involving awareness, interest, involvement, and emotional significance of one’s ethnic identity (Cronbach’s α = .91).

Procedure

Data were collected as part of a larger multi-site study (see Hall et al., 2005). Participants’ informed consent was collected prior to data collection, in compliance with each site’s Internal Review Boards. Included in a larger packet of questionnaires were measures on cultural constructs and attitudes toward women. After responding to these measures, participants read a vignette portraying an acquaintance rape, the most common type of rape on college campuses (Abbey, Ross, McDuffie, & McAuslan, 1996; Fisher, Cullen, & Turner, 2000; Koss, Gidycz, & Wisniewski, 1987; Sampson, 2002). After reading the vignette, participants made rape judgments by rating the events that unfolded and the characters involved.

Results

Hypothesis 1: Ethnic Differences in Rape Vignette Judgments

Mean ratings for the four rape vignette judgments for both ethnic groups are shown in Table 1. As a whole, participants tended to endorse attitudes that indicated relatively high Perpetrator Blame (M = 9.46, SD = 1.03), low Victim Blame (M = 2.42, SD = 1.38), high victim’s Refusal Credibility (M = 9.18, SD = 1.26), and strong beliefs that the incident was a rape (M = 9.77, SD = .96). Independent samples t-tests revealed significant differences by ethnicity for each rating: Perpetrator Blame, t(178.55) = 6.03, Victim Blame, t(248.92) = 5.42, Refusal Credibility, t(211.78) = 6.19, and Rape Defined, t(159.16) = 4.04, all ps < .001. For all four rape vignette judgments, Asian Americans’ responses were more supportive of rape and rated the vignette with more rape-supportiveness than White American participants (see Table 1). Cohen’s d (1988, 1992) measured the magnitude of the effect of ethnicity, which resulted in a “medium” effect size for Rape Defined and “large” effects for the remaining outcomes.

Table 1 Means, SDs, and effect sizes of rape judgments
Full size table

Hypothesis 2: Misogyny as a Mediator

Misogynistic beliefs were evaluated as a mediator of the relationship between being of Asian descent (participants’ ethnic background was dummy coded as Asian American = 1 and White American = 0) and each of the four ratings of the rape vignette. Therefore, four analyses were conducted—i.e., one for each of the dependent variables. In all cases, the predictor (X) was ethnicity and the mediator (M) was misogynistic beliefs. The four analyses differed only in the dependent variable (Y), one of the four rape judgments. Each mediation model had one of four rape judgment ratings as its Y: Perpetrator Blame, Victim Blame, Refusal Credibility, or Rape Defined. According to Baron and Kenny’s (1986) recommendations, mediation is indicated when (1) there is a significant relationship between X and Y, (2) there is a significant relationship between X and M, (3) there is a significant relationship between M and Y, controlling for X, and (4) the effect of X on Y is either no longer significant or is substantially reduced, according to the Sobel (1982) test, when controlling for M.

Each of the four rape vignette judgments was regressed on the predictor, ethnicity, and was significant for each model, fulfilling Criterion 1 (see Fig. 1 for values). In order to evaluate Criterion 2, the mediator (misogynistic beliefs) was regressed on ethnicity and found to be significant (β = .18, p = .001). Criterion 3 was examined by regressing each rape vignette judgment on the mediator (misogynistic beliefs), while controlling for the predictor (ethnicity). Finally, we assessed the reduction in strength of this relationship for Criterion 4. In all four models, the relationship between the predictor of ethnicity and each dependent variable remained significant when including the mediator of misogynistic beliefs, indicating that there was not full mediation. However, using the Sobel test, a significant reduction was found in all but one of the rape vignette judgments (Rape Defined, Sobel test = −2.28, ns), indicating that partial mediation was found for the remaining three judgments: Perpetrator Blame (Sobel test = 2.70, p = .007), Victim Blame (Sobel test = 3.26, p = .001), and Refusal Credibility (Sobel test = 2.89, p = .004).

Fig. 1
figure 1

Path diagrams of regression analyses depicting the role of misogynistic beliefs as a mediator of the relationship between ethnicity and four rape vignette judgments. Note. Numbers in parentheses represent values incorporating the mediator. Bolded numbers in parentheses represent significantly reduced beta weights when controlling for the mediator, according to the Sobel test. Analyses were run with transformed variables. Directions of relationships noted reflect the directions of the relationships among the raw dependent variables. *p < .01, **p < .001, ***p < .0001

Full size image

Hypothesis 3: Generational Status, Acculturation, and Ethnic Identity

A series of hierarchical multiple-regression analyses was used to determine if acculturation was associated with rape vignette judgments, above and beyond the effects of generational status, and if ethnic identity was associated with rape vignette judgments above and beyond the effects of acculturation and generational status. To assess for multicollinearity, tolerance and VIF values were examined. Following the standard guidelines that VIF values greater than 10 (Bowerman & O’Connell, 1990; Myers, 1990) and tolerance values under .1 are causes for concern (Fields, 2005), multicollinearity was not a concern in the current regression models, with the largest VIF value as 2.47 and the lowest tolerance value as .41.

As shown in Table 2, generational status accounted for significant variance in each rape-vignette-judgment, and adding acculturation to the model resulted in generational status becoming nonsignificant and acculturation accounting for significant variance in each rape-vignette-judgment. Adding ethnic identity to each model resulted in a small but significant increase in variance explained for one of the four rape vignette judgments, Perpetrator Blame, and near significance for Victim Blame (p = .05). Consistent with our hypothesis, as generational status and acculturation increased, rape-supportive vignette judgments decreased. Inconsistent with our hypothesis, as ethnic identity increased, rape-supportive vignette judgments increased.

Table 2 Hierarchical regression analyses predicting rape vignette judgments among Asian American men
Full size table

Discussion

The present study provided evidence for the importance of including cultural constructs when investigating judgments of an acquaintance rape and rape-supportive attitudes among Asian American college males. Although both White and Asian American men generally rated the acquaintance rape vignette with unsupportive attitudes toward rape, this was less true for Asian Americans who judged the vignette with significantly more rape-supportiveness than their White counterparts. Misogynistic beliefs partially mediated the relationship between ethnicity and rape vignette judgments of blaming the perpetrator, blaming the victim, and judging how credible the victim’s refusal of the perpetrator, but not defining the vignette as a rape. As predicted, higher generational status was significantly associated with less rape-supportive vignette judgments but when adding acculturation, this relationship became non-significant, and only acculturation was significantly associated with less rape-supportive vignette judgments. Ethnic identity was associated with two of the four rape vignette judgments, Perpetrator Blame and Victim Blame, above and beyond the effects of acculturation. Inconsistent with our hypothesis, higher ethnic identity was associated with greater rape-supportive blame judgments.

For each rape vignette judgment, medium and large effect sizes (Cohen, 1988, 1992) were found for the significant differences between Asian American and White men. However, the differences were small in absolute terms, representing less than a one point difference on a 10-point scale. It is important to note that this ethnic difference appears to be a robust finding and is consistent with research results from the past 15 years (Kennedy & Gorzalka, 2002; Lee et al., 2005; Mills & Granoff, 1992; Mori et al., 1995). This finding further bolsters the argument for the incorporation of cultural considerations when developing theoretical models and prevention programs.

Our second hypothesis had partial support: misogynistic beliefs partially mediated the relationship between ethnicity and rape-supportive judgments. Comparable rates of sexual aggression have been found between Asian/Pacific Islander (21% in Hawaii and 32% mainland) and White (29%) college men (Hall et al., 2005). Although there is a long history indicating the effectiveness of targeting misogynistic beliefs in rape interventions with White men (Brecklin & Forde, 2001), this finding suggests that, perhaps with an appropriate culture-specific approach, targeting misogynistic beliefs may even be slightly more effective with Asian American men. These results signify a need for a more culturally nuanced understanding of misogyny. Furthermore, these findings were consistent with converging evidence that finds that incorporating cultural factors into rape prevention programs reduce rape-supportive attitudes held by Asian American males (Stephens, 2009). Because rape awareness workshops may result in more attitude change for men than women (Szymanski, Devlin, Chrisler, & Vyse, 1993) and 95% of sexual perpetrators are men (Bureau of Justice Statistics, 1995; Tjaden & Thoennes, 2006), education and intervention programs designed for men are necessary. Misogyny did not, however, mediate the relationship between ethnicity and defining the vignette as rape, which was assessed with a single item. It is possible that a multi-item measure consisted of reverse-coded items and items assessing the extent to which the vignette was a conviction-worthy crime would better detect a mediational relationship between ethnicity.

Because misogyny only partially mediated the relation between ethnicity and rape attitudes, future research must identify additional factors that contribute to ethnic differences in rape attitudes. As previously stated, Asian American men’s cultural socialization involves traditional gender roles and patriarchal values that are more highly promoted in Asian cultures (Ho, 1990; Kim & Ward, 2007; Mills & Granoff, 1992; Okazaki, 2002). Accordingly, subsequent research should incorporate direct measurements of traditional gender roles and patriarchal values and examine their effects on Asian American rape attitudes. Cross-cultural research supports this suggestion. Among Japanese college students, blaming a rape victim was correlated with minimizing rape to a greater extent than for American college students, and these differences were partially mediated by the cultural differences in endorsement of traditional gender roles (Yamawaki & Tschanz, 2005). Additionally, the Asian American cultural socialization into an interdependent orientation is an additional possible mediator contributing to this ethnic difference in rape attitudes. In an interdependent culture where group harmony is prioritized over personal goals (Hall & Barongan, 1997), contesting a perpetrator and supporting a rape victim (as opposed to ignoring the rape) may be perceived as socially deviant by violating interpersonal and group harmony. Analyzing other Asian American cultural constructs as mediators between ethnicity and rape-supportive attitudes would be valuable in explaining the variance in these attitudes.

For our final hypothesis, as evidence suggests (Kennedy & Gorzalka, 2002; Mori et al., 1995), higher generational status and acculturation of Asian American males was associated with lower rape-supportive vignette judgments. Without valuing one culture over another, given this effect of acculturation, incorporating more “Western” values could be influential in changing rape attitudes for Asian Americans, but identifying which “Western” values are instrumental for achieving this change is complicated. Evidence suggests that acculturation may be domain specific rather than an overarching status of an individual (Arends-Toth & Van de Vijver, 2004). The complex nature of the construct of acculturation makes it difficult to say which part of American culture is driving this phenomenon. Future research addressing more nuanced assessments of acculturation is necessary to begin to disentangle this relationship between higher acculturation and lower rape-supportive attitudes among Asian American men.

Additionally, generational status was no longer significantly associated with rape-supportive judgments when acculturation was included in regression models. Many studies use birthplace or generational status as an indicator of acculturation. Although these may be correlated (in the current study, r = .56, p < .001), generational status alone does not appear to tell the whole story. This was further demonstrated in each significant ΔR 2 when acculturation was added to the models, indicating that acculturation accounted for a significant amount of variance above that which was captured by generational status. Perhaps this is only true for rape attitudes. Regardless, this is particularly important to keep in mind as researchers measure acculturation.

Lastly, ethnic identity was significantly associated with only Perpetrator Blame and nearly significantly associated with Victim Blame (but not Refusal Credibility or Rape Defined) above and beyond acculturation and generational status, such that as ethnic identity increased, these rape-supportive blame judgments increased as well. Although the effect was statistically significant for Perpetrator Blame, the change in R 2 was .01 (see Table 2). Thus, it may be worthwhile to replicate these findings before drawing conclusions. However, these findings suggest that the degree to which an Asian American male valued his membership in his ethnic group had minimal or no protective impact on his rape attitudes, above and beyond acculturation. Although Hall et al. (2005) found that ethnic identity protected against sexually aggressive behavior, the current finding may be due to the difference between attitudes and behaviors, perhaps a particularly significant distinction for Asian American cultures. Behaving in a socially deviant manner (i.e., sexually aggressive) may have greater detrimental disruptive consequences than privately endorsing socially deviant attitudes in an interdependent culture that values interpersonal harmony over individual needs. Our findings suggest that, among Asian American men, ethnic identity may have a differential relationship with rape-supportive attitudes than with sexually aggressive behavior.

Study limitations must be considered. The low reliability among White Americans of the dependent variable Perpetrator Blame was universally highly blaming of the perpetrator. However, this variable was only included in the analysis of our first hypothesis, replicating previously established ethnic differences in rape-supportive attitudes. Specifically recruiting a sample of men with greater variance in rape attitudes overall would be beneficial for identifying mechanisms behind ethnic differences in rape-supportive attitudes. Although social desirability is often controlled for when examining sensitive topics such as rape, evidence also suggests that social desirability and rape myth attitudes are not necessarily correlated (Spohn, 1993). We unfortunately did not assess for social desirability and thus were unable to investigate these relationships. Further limitations of the current study are related to generalizability. Grouping all Asian ethnicities as one category of “Asian” minimized the differences between the many Asian cultures. The same could be said for other factors that were overlooked, such as mixed-ethnicity groups or generational issues. However, despite the grouping of this sample, a robust effect was found. Moreover, although our measurement of misogynistic beliefs followed a sound theoretical foundation, it did not precisely mirror that of Malamuth et al.’s (1995) confluence model, and perhaps doing so might have influenced results differently. Lastly, as cross-sectional data, inferring causality or direction of effects is inappropriate (Cole & Maxwell, 2003), and thus longitudinal studies examining these relationships are worthwhile.

Rather than directly measuring cultural constructs, sexuality researchers often use Asian race as a proxy for culture (Okazaki, 2002). Misogyny, acculturation, and ethnic identity add to our understanding of rape-supportive attitudes among Asian American college men, highlighting the importance of considering cultural factors. Our findings support that cultural socialization processes for Asian American men’s sexual attitudes differ from that of mainstream America. Research integrating cultural constructs is necessary to attend to the cultural socialization process of other marginalized groups and understanding sexuality for those who may not fit mainstream American culture. The current study takes an important first step toward elucidating the cultural context for the pattern of Asian American men’s tendency to hold more rape-supportive attitudes than their non-Asian counterparts (Kennedy & Gorzalka, 2002; Lee et al., 2005; Mills & Granoff, 1992; Mori et al., 1995). With rates of sexual assault and rape on the rise (Rand, 2008) and detrimental outcomes of sexual assault continuing to persist (CDC, 2007), findings such as these can properly inform more culturally-relevant treatment and interventions of sexual assault for an ever-diversifying country.