The Importance of Substance-Related Sexual Victimization

Abstract

Data on risk perception, sexual victimization, and substance use were obtained via surveys from 496 female college students to determine what factors influence risk perception using a written vignette in which participants make a hypothetical decision to leave a potentially risky situation. Experiences of substance-related (SR) victimization, rather than forcible victimization, were associated with significantly delayed risk perception. SR victimization victims reported feeling uncomfortable significantly later and leaving the scenario significantly later than non-victims. SR victimization victims also had significantly higher scores on heavy episodic drinking (HED), marijuana use, alcohol-related tolerance, and blackouts. Both substance use (HED and marijuana use) and alcohol-related problems (tolerance and blackouts) mediated the link between SR victimization and risk perception in the form of behavioral leave response. In contrast, only HED and tolerance mediated the link between SR victimization and risk recognition. Findings suggest the importance of differentiating types of victimization in predicting risk perception and of addressing substance use in sexual victimization risk reduction interventions.

Keywords

sexual victimization risk perception substance use college women

Sexual victimization is a widespread problem on college campuses, with approximately 1 out of 5 women reporting attempted or completed rape (Krebs, Lindquist, Warner, Fisher, & Martin, 2009a; Walsh, DiLillo, & Messman-Moore, 2012). A growing body of research has examined women’s risk factors for rape and sexual assault, most notably past victimization (Humphrey & White, 2000), victim alcohol use (Krebs, Lindquist, Warner, Fisher, & Martin, 2009b; Mohler-Kuo, Dowdall, Koss, & Wechsler, 2004; Ullman, 2003), and impaired risk perception (Loiselle & Fuqua, 2007). Sexually victimized women typically display more impaired risk perception skills and behavior (Breitenbecher, 2008), and impaired risk perception predicts future sexual victimization (Messman-Moore & Brown, 2006). Among college women, the experience of substance-related (SR) victimization is far more common than forcible victimization (Krebs et al., 2009a), and women’s alcohol use is associated with increased sexual assault risk, particularly for alcohol-related assaults (Mohler-Kuo et al., 2004). Not surprisingly, college women’s alcohol use is linked both to increased risk for sexual victimization as well as impaired risk perception (Gidycz et al., 2007), yet few studies have examined whether women’s risk perception is linked to specific types of sexual victimization and alcohol use. This next step seems warranted, given recent studies suggesting a stronger relationship between women’s substance use and SR victimization (Krebs et al., 2009b; Mohler-Kuo et al., 2004), and calls into question whether previous links between risk perception and victimization extend to all types of sexual victimization or instead are linked specifically to SR experiences.

Risk Perception

Risk perception related to sexual victimization is comprised of several widely researched constructs that generally fit into two categories: risk recognition and behavioral response (Breitenbecher, 1999; Messman-Moore & Brown, 2006; Pumphrey-Gordon & Gross, 2007). Risk recognition can involve the respondent’s ability to identify personal risk in a risky situation, often measured by proxy variables such as “feeling uncomfortable” in a potentially risky hypothetical scenario (Breitenbecher, 1999; Messman-Moore & Brown, 2006). Risk recognition may also be assessed from an observer perspective, by indicating when an actor has engaged in an act of sexual assault (i.e., “gone too far”; Marx & Gross, 1995). Risk perception also involves a behavioral response, or the actions the respondent takes once potential risk has been recognized. The ability to recognize risk is important, but the behavioral response of leaving is crucial to reducing victimization risk (Messman-Moore & Brown, 2006; Naugle, 2000).

Previous research on risk perception reflects diverse methodology that typically utilizes hypothetical vignettes, including those in a written format (Messman-Moore & Brown, 2006; Walsh et al., 2012; Yeater & Viken, 2010), an audio format (Marx & Gross, 1995; Soler-Baillo, Marx, & Sloan, 2005), or a video format (Breitenbecher, 1999; Naugle, 2000), with varied instructions including suggestions to “pretend that you are participating in each activity as it is described” to assess personal risk (Messman-Moore & Brown, 2006, p. 171), or asking participants to observe interactions between actors and indicate if sexual aggression had occurred (the observer perspective). For example, Marx, Calhoun, Wilson, and Meyerson (2001) instructed women to press a button to indicate “if and when they thought the man has gone too far” (p. 27). Risk perception has also been assessed in analog studies in which the participant consumes alcohol (e.g., Loiselle & Fuqua, 2007; Pumphrey-Gordon & Gross, 2007; Testa, Livingston, & Collins, 2000). In addition to these paradigms, some studies include a “risk estimate rating” where women estimate the likelihood they will experience sexual victimization in the future (Gidycz et al., 2007). While risk estimates are valuable contributions to risk perception research, such variables do not appear to measure risk recognition and behavioral response. Thus, there is variability in how risk perception is operationalized in terms of the perspective of the participant (e.g., participant vs. observer) as well as the methodology involved in assessment, which warrants caution in comparing results across studies.

Despite these methodological differences, multiple studies have identified impaired risk perception as a correlate and predictor of sexual victimization (Breitenbecher, 2008; Messman-Moore & Brown, 2006; Soler-Baillo et al., 2005; Testa et al., 2000). Unfortunately, the vast majority of studies use retrospective designs which limit temporal inferences regarding prediction. However, one 8-month prospective study found that not only is prior sexual victimization associated with impaired risk perception, delayed risk recognition and behavioral response (i.e., staying later) predicted prospective sexual victimization and rape (Messman-Moore & Brown, 2006). Thus, sexual victimization is a likely outcome as well as a consistent predictor of impaired risk perception. However, more research is needed on additional factors that may influence risk perception difficulties among women with and without a history of sexual victimization.

Alcohol Use and Risk Perception

Alcohol is relevant to risk perception research because alcohol use impairs the ability to focus on threatening factors in a risky situation (Steele & Josephs, 1990), and victim alcohol consumption precedes the vast majority of college women’s sexual victimization experiences (Mohler-Kuo et al., 2004). Despite these theoretical links, alcohol and risk perception research is limited given that researchers cannot actually observe the interactions between alcohol use and reactions in real-life risky sexual situations. However, analog designs in which participants are administered alcohol facilitate the examination of alcohol’s impact on risk perception. In one such analog study (Testa et al., 2000), women who consumed alcohol perceived more benefits and fewer negative consequences of entering a hypothetical risky situation and reported a higher likelihood of participating in risky sexual behaviors than women who had not consumed alcohol. Loiselle and Fuqua (2007) also found that women who consumed alcohol reported less resistance during a hypothetical sexual assault, and utilizing the Marx and Gross (1995) audio vignette, had significantly longer response latencies (indicating when the man should stop) than women in the placebo group (who were deceived to believe they had received alcohol when they had not). Although the analog design allows researchers to directly examine alcohol consumption and risk perception judgments, (in most states) research participants must be at least 21 years old to consume alcohol. This is a significant limitation given that greater risk for sexual victimization is associated with younger age, with risk for college sexual victimization decreasing significantly after freshman year (Humphrey & White, 2000; Krebs et al., 2009a). Preliminary findings suggest a need for additional research that examines alcohol consumption, risk perception, and sexual victimization.

The Relevance of Type of Sexual Victimization

Emerging research suggests it may becrucial to distinguish sexual victimization that occurs through one of two methods: (a) a perpetrator’s threats or use of physical force, or (b) the victim’s inability to consent or resist due to intoxication or incapacitation. For college women, the experiences of SR victimization are far more common than forcible victimization. Some studies estimate that between 8.5% and 13.1% of college women have experienced SR victimization (Krebs et al., 2009a; Lawyer, Resnick, Bakanic, Burkett, & Kilpatrick, 2010) compared with 1.9% to 3.4% experiencing forcible victimization (Krebs et al., 2009a; Mohler-Kuo et al., 2004), and in one study of college women, alcohol or drugs were a factor in all sexual victimization experiences (Lawyer et al., 2010). In addition, frequency of women’s alcohol, marijuana, and other drug use is more strongly associated with SR victimization than forcible victimization (Krebs et al., 2009b; Mohler-Kuo et al., 2004). Thus, different types of sexual victimization exist and some types (e.g., SR) are more prevalent among college women, suggesting a need for future studies to consider different forms of sexual victimization. Moreover, given the significant link between alcohol consumption and impaired risk perception (Loiselle & Fuqua, 2007), additional studies are needed to determine to what extent different types of victimization are associated with impaired risk perception.

Research Aims and Hypotheses

The main aim of the current study is to extend current sexual victimization research by examining the relationship between two different types of victimization and risk perception. Given the strong relationship between heavy alcohol use and risk perception, as well as between heavy alcohol use and SR victimization (but not forcible victimization), it is hypothesized that SR victimization will be uniquely associated with impaired risk perception, and that forcible victimization will have a weaker or non-significant association. A secondary aim of the current study is to determine whether substance use and related variables (i.e., tolerance, blackouts) mediate the relationship between victimization and risk perception.

Method

Participants

The current study involved an initial sample of 546 college women recruited from an Introduction to Psychology course at a mid-sized Midwestern university. Of these women, 47 were excluded because they reported impossible responses on the Risk Perception Survey (RPS; Messman-Moore & Brown, 2006, indicating they would “leave” the vignette before indicating they “felt uncomfortable”) and an additional 3 women did not provide complete information on the Sexual Experiences Survey (SES; Koss, Gidycz, & Wisniewski, 1987), resulting in a usable sample of 496 undergraduate women. Participant ages ranged from 17 to 23 years (M = 18.71; SD = 0.85); most were first-year students (65.9%) or sophomores (27.4%). The sample was predominantly Caucasian (92.7%), with other racial groups represented as follows: African American (2.0%), Asian/Asian American (1.8%), Biracial (0.8%), Hispanic/Latina (0.6%), Native American (0.2%), and other (0.8%). Five (1.0%) participants did not provide their race. Participants tended to come from middle- to upper-class backgrounds (over 60% reported an annual family income of US$75,000 or higher). Racial and socioeconomic characteristics reflected the larger university population.

Measures

Risk perception

The RPS (Messman-Moore & Brown, 2006) is a written vignette consisting of 25 chronological statements describing an episode of sexual assault in a social situation and progresses with increasingly greater risk for sexual victimization, culminating in rape by an acquaintance (see Appendix). The vignette contains clear (e.g., sexual comments, verbal persuasion) and ambiguous (e.g., alcohol consumption) risk factors for sexual victimization. The RPS was used to assess two aspects of risk perception: risk recognition, the item number when the participant felt uncomfortable (i.e., discomfort score), and behavioral response, the item number when the participant would leave the situation (i.e., leave score). Participants were instructed to immerse themselves in the hypothetical scenario, and indicate when (if ever) they would feel uncomfortable and when (if ever) they would leave the scenario. Higher scores indicate greater risk perception impairment, which corresponds to greater risk for sexual victimization.

Sexual victimization

Sexual victimization was assessed with the SES, a widely used self-report measure (Koss et al., 1987). This measure was modified to assess specific sexual experiences in greater detail (e.g., additional questions were added regarding oral–genital contact). Questions regarding alcohol-facilitated victimization were added according to suggestions by Muehlenhard, Powch, Phelps, and Giusti (1992). Prevalence since age 14 (yes/no) and number of incidents were assessed. SR victimization was defined as experiences of attempted or completed oral, anal, or vaginal penetration when unable to give consent or resist due to alcohol or drugs. Forcible victimization was defined as experiences of attempted or completed oral, anal, or vaginal penetration because of threats or physical force.

Alcohol and drug use

A series of questions were used to examine the frequency and patterns of recent alcohol and marijuana use. These questions were based on National Institute on Alcohol Abuse and Alcoholism definitions of a standard drink and were consistent with national surveys on substance use (Wechsler & Nelson, 2008). Patterns of substance use were assessed with questions examining heavy episodic drinking (HED), “How often do you consume 4 or more drinks in one setting,” and marijuana use, “In the past month, how often have you used marijuana,” with responses based on a 7-point scale where 0 is not at all and 7 is more than once a day. Tolerance was assessed with the question, “How many drinks does it take for you to ‘feel the effects’” and was assessed with a 7 point scale where 0 is I do not drink and 7 is more than 6 drinks. Finally, frequency of blacking out, “How often have you blacked out from drinking alcohol,” was assessed with an open-ended written (numerical) response.

Procedure

The Committee of Human Subjects in Research (i.e., institutional review board) approved all procedures for the project. Participants were recruited using an online tool (experimentrix.com) as part of the Introduction to Psychology student participant research pool. Participants completed the study in the psychology building in groups of approximately 5 to 40 women. All surveys were assigned an identification number to ensure anonymity. Packets were counterbalanced to ensure that surveys on victimization, substance use, and risk perception were not always delivered in the same order (i.e., victimization was sometimes assessed prior to risk perception, and vice versa). Debriefing provided participants with information on the purpose of the study and provided campus and community counseling resources.

Results

Descriptive statistics and bivariate correlations of demographic and study variables were computed to examine potential covariates (Table 1). Data were examined for skew and kurtosis and several participants (n = 8) had significant outlier scores (greater than three standard deviations above the mean) on the number of blackouts variable. To conserve their data on the remaining substance use variables, these participants were assigned a value one unit above the next highest score on the blackout variable (i.e., truncation). This significantly reduced the skewness (12.88 to 2.44) and kurtosis (216.68 to 7.03) within acceptable ranges of normality ( skewness < 3 and kurtosis < 10; see Kline, 1998).

Table 1.

Correlations for Sexual Victimization, Risk Perception, and Substance Use Variables.

	1	2	3	4	5	6	7	8
1. SR victimization	—
2. Forcible vic	.21**	—
3. Discomfort	.11*	−.02
4. Leave	.12**	.06	.68**
5. HED	.32**	.10*	.24**	.26**
6. Marijuana use	.10*	.03	.15**	.17**	.34**
7. Tolerance	.23**	.06	.28**	.31**	.61**	.21**
8. Black outs	.18**	.09	.16**	.21**	.43**	.30**	.36**
M (%)	20.2%	8.5%	9.91	14.03	1.97	0.37	2.17	1.60
SD	—	—	4.32	4.52	1.59	0.90	1.48	2.52

Note. SR victimization = substance-related sexual victimization; Forcible vic = forcible sexual victimization; Discomfort = discomfort score (risk recognition); Leave = leave score (behavioral response); HED = heavy episodic drinking.

p < .05. **p ≤ .01.

Sexual Victimization Prevalence

Based on responses to the modified SES, 16.1% (n = 80) of the sample reported experiences of SR victimization only, 4.4% (n = 22) reported forcible victimization only, 4.0% (n = 20) reported both SR victimization and forcible victimization, and 75.4% (n = 374) reported no experiences of sexual victimization. In this sample and consistent with previous research (Krebs et al., 2009b), SR victimization was more common than forcible victimization. In addition, roughly half of the women who experienced forcible victimization also reported SR victimization (20 out of 42 forcible victimization victims), but only one fifth of the women who experienced SR victimization also reported a forcible victimization (20 out of 100 SR victimization victims). Given the design of the SES, we could not determine whether both forms of coercion took place in the same episode of victimization.

Alcohol and Other Substance Use

In the past month, 55.5% of the sample reported consuming alcohol one or more times a week. In addition, 40.9% of the sample reported consuming four or more drinks in one sitting (i.e., HED) at least once a week, 18.1% reported having to consume four or more drinks before feeling the effects of the alcohol, and 50.6% reported blacking out at least once. Marijuana use in the past month was reported by 18.9% of the sample, with 11.7% reporting using two or more times. Other drug use was infrequent and all participants who reported other drug use also reported using marijuana, therefore analyses involving drug use included marijuana use only.

Sexual Victimization, SR Variables, and Risk Perception

As hypothesized (see Table 1), SR victimization was significantly positively correlated with HED, marijuana use, tolerance, and blackouts, as well as impaired risk perception (both discomfort and leave scores). Findings suggest that women with a history of SR victimization engage in higher levels of substance use as well as report greater negative alcohol-related outcomes. In contrast, forcible victimization was not associated with risk perception, and was generally not associated with substance use variables, with the exception of higher levels of HED. To determine whether SR victimization victims “left late” (i.e., after going to the acquaintance’s apartment; a point in the vignette in which escape would be difficult), a chi-square analysis was conducted. Almost 41% of the total sample reported leaving late; 50.0% of women who experienced SR victimization “left” the scenario after going to the acquaintance’s apartment compared with 38.6% women who had not experienced SR victimization, χ²(1, N = 496) = 4.26, p < .05. To examine whether repeated victimization experiences (i.e., revictimization) account for the relationship between victimization and risk perception, two additional variables were computed. For both SR victimization and forcible victimization victims, the number of incidents of attempted or completed acts was summed, and women with multiple victimization experiences were compared with women with a single victimization experience on risk perception scores with t tests. For both SR victimization and forcible victimization, there were no differences between the two victimized groups (ps > .20). Given that forcible victimization was not correlated with risk perception (even when revictimization was examined), and given that there were no differences between women with one or more SR victimization experiences, additional analyses focused on SR victimization (collapsing revictimized and singly victimized women into one group). ANOVAs were conducted to compare SR victimization victims and non-victims on substance use variables including HED, tolerance, blackouts, and marijuana use. Women who experienced SR victimization had higher scores on all substance use variables (see Table 2).

Table 2.

Substance Use Differences by History of Substance-Related (SR) Sexual Victimization.

	Total	SR Victimization	No SR Victimization
	N = 496	n = 100	n = 396	F
HED	1.97 (1.59)	2.96 (1.24)	1.71 (1.57)	54.25**
Tolerance	2.17 (1.48)	2.84 (1.28)	2.01 (1.48)	26.34**
Black outs	1.60 (2.52)	2.50 (2.58)	1.37 (2.46)	16.20**
Marijuana use	0.37 (0.90)	0.55 (1.05)	0.33 (0.85)	5.01*

Note. HED = heavy episodic drinking.

p < .05. **p ≤ .01.

Mediation Models

A parallel multiple mediation model in PROCESS for SPSS (Hayes, 2013; Model 4) was used to examine if substance use variables mediated the relationship between SR victimization and risk perception. Two aspects of heavy alcohol use (HED and tolerance) were highly correlated (see Table 1) so analyses were split into two models to reduce multicollinearity. The first set of analyses examined variables related to frequency of substance use: HED and marijuana use. The second set of analyses examined variables indicative of problematic drinking: tolerance and blackouts. Finally, as recommended by Hayes (2013), bootstrapping techniques were used to compute a robust set of bias-corrected confidence intervals (CI) around the sampling distribution of substance use variables. For the following regression models, unstandardized regression coefficients for individual paths are presented in Figure 1. Model information is included in the text.

Figure 1.

Mediation models for substance-related sexual victimization and risk recognition.

Risk recognition

First, a linear regression provided evidence that SR was directly related to risk recognition (Path C in Baron & Kenny’s [1986] terminology). In a model where SR victimization was then entered as a predictor of HED and marijuana use (Path A₁ & A₂, respectively), SR victimization was a significant predictor of HED, F(1, 491) = 54.18, p < .001, R² = .02, and marijuana use, F(1, 491) = 5.01, p < .05, R² = .01. HED and marijuana use were then added as mediators in a model examining the effects of SR victimization on risk recognition. HED related to risk recognition above and beyond SR victimization (Path B₁) but marijuana use was not significant (Path B₂) in the full mediation model, F(3, 489) = 11.42, p < .001, R² = .07. With HED and marijuana use in the model, the relation between SR victimization and risk recognition (Path C’) was no longer significant. HED use therefore accounted for the relationship between SR victimization and risk recognition. Bootstrapping techniques were used to test the indirect effects of SR victimization on risk recognition through HED. This yielded a bias-corrected CI around the indirect effect of HED (0.70, SE = 0.18) that did not contain zero (95% CI = [0.38, 1.11]) providing further evidence that HED mediated the relation between SR victimization and risk recognition.

A similar model was used to test tolerance and blackouts (two variables related to problematic drinking) as mediators of the relationship between SR victimization and risk recognition. In a model where SR victimization was entered as a predictor of tolerance and blackouts (Path A₁ & A₂, respectively; see Figure 1), SR victimization was a significant predictor of both tolerance, F(1, 472) = 25.45, p < .001, R² = .05, and blackouts, F(1, 472) = 16.35, p < .001, R² = .03. Tolerance and blackouts were then added as mediators in a model examining the effects of SR victimization on risk recognition. Tolerance related to risk recognition above and beyond SR victimization (Path B₁) but the blackouts variable was not significant (Path B₂) in the full mediation model, F(3,470) = 12.88, p < .001, F(3, 470) = 12.88, p < .001, R² = .08. Bootstrapping techniques were used to test the indirect effects of SR victimization on risk recognition through tolerance. This yielded a bias-corrected CI around the indirect effect of tolerance (0.56, SE = 0.15) that did not contain zero (95% CI = [0.31, 0.92]), providing further evidence that tolerance mediated the relation between SR victimization and risk recognition.

Behavioral response

First, a linear regression provided evidence that SR was directly related to behavioral response (Path C; see Figure 2). SR victimization was then entered as a predictor of HED and marijuana use (Path A₁ & A₂, respectively) and SR victimization was a significant predictor of both HED, F(1, 491) = 54.18, p < .001, R² = .10, and marijuana use, F(1, 491) = 5.01, p < .05, R² = .01. HED and marijuana use were then added as mediators in a model examining the effects of SR victimization on behavioral response. HED and marijuana use were related to behavioral response above and beyond SR victimization (Paths B₁ & B₂) in the full mediation model, F(3, 489) = 13.85, p < .001, R² = .08. With HED and marijuana use in the model, the relation between SR victimization and behavioral response (Path C’) was no longer significant. HED and marijuana use therefore accounted for the relationship between SR victimization and behavioral response. Bootstrapping techniques were used to test the indirect effects of SR victimization on behavioral response through HED. This yielded a bias-corrected CI around the indirect effect of HED (0.77, SE = 0.20) that did not contain zero (95% CI = [0.41, 1.21]) and a bias-corrected CI around the indirect effect of marijuana use (0.11, SE = 0.08) that did not contain zero (95% CI = [0.01, 0.32]) providing further evidence that HED and marijuana use mediate the relation between SR victimization and behavioral response.

Figure 2.

Mediation models for substance-related sexual victimization and behavioral response.

Finally, tolerance and blackouts were examined as mediators between SR victimization and behavioral response. SR victimization was entered as a predictor of tolerance and blackouts (Paths A₁ & A₂, respectively; see Figure 2) and SR victimization was a significant predictor of both tolerance, F(1, 472) = 25.45, p < .001, R² = .05, and blackouts F(1, 472) = 16.35, p < .001, R² = .03. Tolerance and blackouts were then added as mediators in a model examining the effects of SR victimization on behavioral response. Tolerance and blackouts related to behavioral response above and beyond SR victimization (Paths B₁ & B₂) in the full mediation model, F(3,470) = 19.38, p < .001, R² = .11. With tolerance and blackouts in the model, the relation between SR victimization and risk response (Path C’) was no longer significant. Tolerance and blackouts therefore accounted for the relationship between SR victimization and behavioral response. Bootstrapping techniques were used to test the indirect effects of SR victimization on risk response through tolerance and blackouts. This yielded a bias-corrected CI around the indirect effect of tolerance (0.68, SE = 0.17) that did not contain zero (95% CI = [0.39, 1.05]) and a bias-corrected CI around the indirect effect of blackouts (0.21, SE = 0.11) that did not contain zero (95% CI = [0.04, 0.48]) providing further evidence that tolerance and blackouts mediate the relation between SR victimization and behavioral response.

Discussion

This study aimed to explore risk perception deficits related to specific types of sexual victimization. As expected, impaired risk perception was linked to previous experiences of SR victimization, rather than forcible victimization. Specifically, SR victimization victims recognized risk and indicated they would leave the scenario significantly later than women who had not experienced SR victimization. Although there were not quantitatively large differences in risk recognition between groups, differences in behavioral response to the risk (i.e., leaving the scenario) appear to be meaningful. For instance, half of the SR victimization victims indicated they would go to the man’s apartment, isolating themselves with the perpetrator. In essence, SR victimization victims were more likely to decide to leave the risky scenario at a point in which “escape” may in reality be much more difficult. Women who reported forcible victimization had similar risk perception scores to women who did not report any sexual victimization, suggesting the experience of SR victimization may have a unique relationship with risk perception.

Risk perception impairments have been linked to both sexual victimization history (Breitenbecher, 1999; Loiselle & Fuqua, 2007) and prospective sexual victimization (Messman-Moore & Brown, 2006), but more information is needed about the predictors of risk perception related to specific types of victimization. Previous studies have used the same risk perception measure (RPS) without examining different types of sexual victimization, and found that women with a history of victimization had more impaired risk perception (Breitenbecher, 2008; Messman-Moore & Brown, 2006). Based on the present findings, it seems possible that experiences of SR victimization may have heavily influenced the link between victimization and risk perception impairments found in earlier studies, although findings could be specific to the current vignette under investigation. Broad victimization definitions (e.g., collapsing all types of victimization into one group) may dampen or even negate potential effects driven by unique sexual victimization characteristics. The present findings reinforce the importance of examining different types of victimization, particularly SR experiences.

The relationship between SR victimization and risk perception was mediated by SR variables; however, the specific links accounting for this relationship varied depending on which aspect of risk perception was examined. For the behavioral leave response, both substance use (HED and marijuana use) and alcohol-related problems (tolerance and blackouts) mediated the link, whereas for risk recognition (i.e., discomfort), only HED and tolerance (i.e., the number of drinks “it takes for you to feel the effects”) mediated the link. Findings highlight the importance of assessing different aspects of substance use and associated problems. Both alcohol and marijuana consumption may impair women’s ability to leave potentially risky situations, whereas HED and tolerance to the effects of alcohol appear relevant to both detecting risk as well as leaving potentially risky situations. SR victimization victims reported significantly higher levels of tolerance and consumed almost twice the number of drinks before they felt the effects of the alcohol, compared with women who did not report SR victimization. Higher tolerance estimates may be a reflection of more regular and consistent heavy drinking patterns. SR victimization victims reported needing at least 3 to 4 drinks to begin feeling the effects of alcohol, thus not experiencing a physiological effect until engaging in HED. Findings suggest that experiences of SR victimization are inextricably intertwined with heavy alcohol consumption and impaired risk perception. It is critical that researchers continue to examine the impact of heavy substance use and certain types of sexual victimization, namely SR victimization, on risk perception.

The RPS has been used in other studies assessing links between impaired risk perception and sexual victimization (Breitenbecher, 2008; Messman-Moore & Brown, 2006; Walsh et al., 2012). The current study’s use of the RPS allows for straightforward comparison to one of the only longitudinal studies on risk perception (Messman-Moore & Brown, 2006). In addition, Breitenbecher (2008) found that the RPS is associated with risky dating behaviors, suggesting that it may accurately assess impaired risk perception. Together with the current findings, there is growing evidence linking impaired risk perception and sexual victimization, particularly SR victimization, but future studies are needed with different stimuli to determine whether method or nature of the stimulus accounts for the present findings.

Although the findings from this study provide important information about SR victimization and risk perception, there are several limitations. Given the retrospective design, it is unknown whether risk perception is truly influenced by the examined factors, or whether risk perception would predict such behaviors. More research is needed using prospective designs to investigate how risk perception impairments influence sexual victimization over time. In addition, the RPS (Messman-Moore & Brown, 2006) is a paper-and-pencil written vignette which allowed participants to read ahead and know that the scenario culminates in victimization, potentially influencing their responses. Responses are also limited to only leaving the risky scenario in response to threat, and although this is the most assertive response, there are other strategies to handle potentially risky situations. In addition, the current study may have limited generalizability due to the inclusion of only college students, and a lack of diversity in ethnicity and socioeconomic status among college women in the current sample.

The current study is unique in its inclusion of different types of sexual victimization in relation to two components of risk perception: risk recognition and behavioral response. Future prospective studies should seek to determine if alcohol and marijuana use may indirectly increase risk for different types of sexual victimization via impaired risk perception. Targeting women’s substance use in the hopes of improving risk perception is consistent with current efforts aimed to alter such behaviors to reduce the risk for sexual victimization (Testa et al., 2000).

Footnotes

Appendix

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.

Author Biographies

Lee Eshelman, MA, is a PhD student in the Clinical Psychology program at Miami University. Her research interests include examining outcomes of sexual victimization with particular focus on the intersections between emotion regulation, PTSD, and stress reactivity.

Terri L. Messman-Moore, PhD, is the O’Toole Family Professor of Psychology and the Director of Clinical Training at Miami University. Her research examines the intersection of emotion dysregulation and interpersonal violence (particularly sexual victimization across the lifespan). Many of her studies focus on predictors and outcomes of interpersonal violence revictmization such as PTSD, dissociation, substance use, and risky sexual behavior.

Nicole Sheffer received her BA in Psychology with honors from Miami University in 2010. She received her MSW in 2014 from the University of Houston. Her professional interests include trauma, family violence, child and adolescent mental health, and child welfare.

References

Baron

R. M.

Kenny

D. A.

(1986). The moderator–mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173-1182.

Breitenbecher

K. H.

(1999). The association between the perception of threat in a dating situation and sexual victimization. Violence and Victims, 14, 135-146.

Breitenbecher

K. H.

(2008). The convergent validities of two measures of dating behaviors related to risk for sexual victimization. Journal of Interpersonal Violence, 23, 1095-1107.

Gidycz

C. A.

Loh

Lobo

Rich

Lynn

S. J.

Pashdag

(2007). Reciprocal relationships among alcohol use, risk perception, and sexual victimization: A prospective analysis. Journal of American College Health, 56, 5-14.

Hayes

A. F.

(2013). Introduction to mediation, moderation, and conditional process analysis. New York, NY: Guilford Press.

Humphrey

J. A.

White

J. W.

(2000). Women’s vulnerability to sexual assault from adolescence to young adulthood. Journal of Adolescent Health, 27, 419-424.

Kline

R. B.

(1998). Principles and practices of structural equation modeling. New York, NY: Guilford Press.

Koss

M. P.

Gidycz

C. A.

Wisniewski

(1987). The scope of rape: Incidence and prevalence of sexual aggression and victimization in a national sample of higher education students. Journal of Consulting and Clinical Psychology, 55, 162-170.

Krebs

C. P.

Lindquist

C. H.

Warner

T. D.

Fisher

B. S.

Martin

S. L.

(2009a). College women’s experiences with physically forced, alcohol-or other drug-enabled, and drug-facilitated sexual assault before and since entering college. Journal of American College Health, 57, 639-649.

10.

Krebs

C. P.

Lindquist

C. H.

Warner

T. D.

Fisher

B. S.

Martin

S. L.

(2009b). The differential risk factors of physically forced and alcohol-or other drug-enabled sexual assault among university women. Violence and Victims, 24, 302-321.

11.

Lawyer

Resnick

Bakanic

Burkett

Kilpatrick

(2010). Forcible, drug-facilitated, and incapacitated rape and sexual assault among undergraduate women. Journal of American College Health, 58, 453-460.

12.

Loiselle

Fuqua

W. R.

(2007). Alcohol’s effects on women’s risk detection in a date-rape vignette. Journal of American College Health, 55, 261-266.

13.

Marx

B. P.

Calhoun

K. S.

Wilson

A. E.

Meyerson

L. A.

(2001). Sexual revictimization prevention: An outcome evaluation. Journal of Consulting and Clinical Psychology, 69, 25-32.

14.

Marx

B. P.

Gross

A. M.

(1995). Date rape: An analysis of two contextual variables. Behavior Modification, 19, 451-463.

15.

Messman-Moore

T. L.

Brown

A. L.

(2006). Risk perception, rape, and sexual revictimization: A prospective study of college women. Psychology of Women Quarterly, 30, 159-172.

16.

Mohler-Kuo

Dowdall

G. W.

Koss

M. P.

Wechsler

(2004). Correlates of rape while intoxicated in a national sample of college women. Journal of Studies on Alcohol and Drugs, 65, 37-45.

17.

Muehlenhard

C. L.

Powch

I. G.

Phelps

J. L.

Giusti

L. M.

(1992). Definitions of rape: Scientific and political implications. Journal of Social Issues, 48(1), 23-44.

18.

Naugle

A. E.

(2000). Identifying behavioral risk factors for repeated victimization using videotaped stimulus materials. Dissertation Abstracts International, 61(2-B), 1091.

19.

Pumphrey-Gordon

J. E.

Gross

A. M.

(2007). Alcohol consumption and females’ recognition in response to date rape risk: The role of sex-related alcohol expectancies. Journal of Family Violence, 22, 475-485.

20.

Soler-Baillo

J. M.

Marx

B. P.

Sloan

D. M.

(2005). The psychophysiological correlates of risk recognition among victims and non-victims of sexual assault. Behavior Research and Therapy, 43, 169-181.

21.

Steele

C. M.

Josephs

R. A.

(1990). Alcohol myopia: Its prized and dangerous effects. The American Psychologist, 45, 921-933.

22.

Testa

Livingston

J. A.

Collins

R. L.

(2000). The role of women’s alcohol consumption in evaluation of vulnerability to sexual aggression. Experimental and Clinical Psychopharmacology, 8, 185-191.

23.

Ullman

S. E.

(2003). A critical review of field studies on the link of alcohol and adult sexual assault in women. Aggression and Violent Behavior, 8, 471-486.

24.

Walsh

DiLillo

Messman-Moore

T. L.

(2012). Lifetime sexual victimization and poor risk perception: Does emotion dysregulation account for the links? Journal of Interpersonal Violence, 27, 3054-3071.

25.

Wechsler

Nelson

T. F.

(2008). What we have learned from the Harvard School of Public Health College Alcohol Study: Focusing attention on college student alcohol consumption and the environmental conditions that promote it. Journal of Studies on Alcohol and Drugs, 69, 481-490.

26.

Yeater

E. A.

Viken

R. J.

(2010). Factors affecting women’s response choices to dating and social situations. Journal of Interpersonal Violence, 25, 1411-1428.