Abstract
Almost all research on sexual assault victimization among undergraduate university students pertains to incidents that occur on domestic college and university campuses. The purpose of the present study was to investigate the prevalence of sexual assault victimization and related factors among undergraduates in the context of study-abroad programs. Two hundred eight female students (52% response rate) from a small university in the northeastern United States who had recently studied abroad responded to an online survey containing measures of sexual assault, posttraumatic stress responses (PSR), and alcohol consumption. Almost 19% of the respondents indicated one or more types of sexual assault victimization. Approximately 17% reported non-consensual sexual touching, 7% attempted rape, 4% rape, with 9% reporting attempted rape or rape. As in domestic studies, victimization in this sample was related positively to alcohol consumption and PSR. Use of force was the most frequently reported perpetrator tactic. In sum, the high rates of sexual assault victimization reported by this sample during study abroad replicate previous findings. This context requires further attention from sexual assault researchers, especially given the increasing numbers of university students engaging in study abroad, and from campus support personnel who may be unaware of the likelihood of assault in this context.
Almost all research on sexual assault among undergraduate university students pertains to incidents that occur on domestic college and university campuses. Reviews of this literature (Fisher, Daigle, & Cullen, 2010) indicate that approximately one in four to five female students is victimized sexually by the end of their university years. This prevalence rate has remained stable since the groundbreaking work of Koss, Gidycz, and Wisniewski (1987) in the 1980s, characterizing campuses in both the United States and Canada (DeKeseredy & Schwartz, 1998). The most common risk factor—and perpetrator tactic—for sexual assault victimization is victim incapacitation by means of alcohol consumption, a factor that has been of increasing concern with the rise of binge-drinking among U.S. students (Abbey, 2002; Wechsler et al., 2002). Psychological consequences, including symptoms of posttraumatic stress disorder (PTSD), are also commonly associated with sexual assault victimization (Zinzow et al., 2011), underscoring further the fact that campus sexual assault is a significant public health issue. While much is known about the prevalence of campus sexual assault and some of its risk factors and consequences, surprisingly little is known about the specific contexts within which assaults take place.
One context that has received almost no attention in this regard is study abroad. According to the Institute of International Education (2011), more than 270,000 U.S. college students were engaged in some type of study-abroad program in 2009-2010, and these numbers have increased steadily since 2007-2008. Almost half of the students on the campus where the present study was conducted study abroad, most in either 3-week or semester-long programs. The purpose of the study reported in this article was to assess sexual assault among university students during study abroad using standard methods for assessing sexual assault, alcohol consumption, and posttraumatic stress responses (PSR).
Previous Research on Sexual Assault During Study Abroad
Although there is some research on health issues related to study abroad (Hartjes, Baumann, & Henriques, 2009), including alcohol consumption (Pedersen, Larimer, & Lee, 2010; Wielkiewicz & Turkowski, 2010), we could find only two reports of research focusing on sexual assault of students while abroad. In the first of these, Hummer, Pedersen, Mirza, and LaBrie (2010) report on “sexual alcohol-related consequences” during study abroad in a sample of 152 students from a west coast university. Sexual consequences addressed in this study ranged from “regretted” sexual situations to “pressured or forced” sex while intoxicated or following alcohol consumption. Twenty-three percent of women and 32% of men reported regrettable sexual encounters caused by alcohol consumption while abroad, and 4% of women and 8% of men indicated unwanted sexual experiences due to drinking. Three percent of women and 8% of men reported being “pressured or forced” to have sex because of intoxication, and the same percentages of women and men reported having “pressured or forced” someone else to have sex because of intoxication. The questions used to elicit this information seem to have been designed for this study, and while they demonstrated good internal reliability, the absence of standardized measures of sexual assault make it difficult to compare these findings with those of domestic studies.
The second study of sexual assault during study abroad was conducted by Kimble, Flack, and Burbridge (2013). Among their sample of 218 junior and senior women who had studied abroad recently from a small northeastern college, 38% reported one or more incidents of victimization on the revised version of the Sexual Experiences Survey (RSES; Koss et al., 2007). Most of the women reporting victimization indicated non-consensual sexual touching (27.5% of the sample), with smaller proportions reporting attempted or completed rape (6.0% and 4.6%, respectively). Kimble and colleagues indicated that these victimization rates were significantly higher than domestic rates on the home campus, and they speculated that the increased rates might have been caused by increased access to alcohol, weaker networks of friends, and lack of familiarity with the culture and surroundings. The vast majority of perpetrators (almost 87%) across all types of sexual assault in this study were non-student local residents. Rates of assault varied to some extent by country and continent, but not by language fluency (in non-English-speaking countries). Use of the RSES by these investigators allows for a comparison of their prevalence rates with those found in domestic studies, indicating a relatively high prevalence of sexual assault in the study-abroad experiences of this cohort.
Research summarized in this section establishes that sexual assault takes place during at least some study-abroad programs. The results found by Hummer and colleagues are consistent with previous research demonstrating a relationship between assault and alcohol consumption. The findings reported by Kimble and colleagues indicate that prevalence rates of assault may be higher during study abroad as compared with domestic rates and may also differ from most domestic cases with respect to victims’ relationships to perpetrators. The study reported below was designed both to replicate the main findings of previous research and to assess PSR to sexual assault incidents during study abroad.
Purpose
In the present study, we surveyed a cohort of female undergraduates who had recently studied abroad from a small university in the northeastern United States. In addition to measuring sexual assault, this study included independent, standard measures of alcohol consumption and PSR. We hypothesized the following:
Method
Sample
The sample in this study consisted of 208 undergraduate women from a small university in the northeastern United States. All participants were between the ages of 20 and 25 at the time of data collection. Most (87.5%) were Caucasian, with the remainder Asian/Asian American (7.2%) and Hispanic (2.4%). Most (96.2%) indicated a heterosexual orientation, with the remainder bisexual (2.4%) or homosexual (1.0%). Less than half (43.3%) considered themselves “religious.” More than half (62.5%) were in their third year at the university, and the rest were in their fourth year. Most (62.5%) were members of sororities, and a minority (13.0%) were members of varsity sports teams.
A majority of the sample (69.7%) reported studying abroad for a single semester, 21.2% for a full academic year, and 9.1% during a short-term summer program. Most (71.2%) studied abroad in Europe, 9.6% in Australia or New Zealand, 8.2% in Asia, and 3.8% each in Africa or Central America. Approximately one third (33.2%) participated in study-abroad programs organized by their own university, with the remainder participating in programs directed by other academic institutions or study-abroad organizations. Most (67.3%) reported that a majority of their student peers during study abroad came from colleges and universities other than their own.
Measures
Revised sexual experiences survey (RSES)
The short-form victim version of the RSES (Koss et al., 2007) was used to assess sexual assault. The Sexual Experiences Survey (SES) is the most frequently used self-report measure in research on sexual assault victimization (Kolivas & Gross, 2007). Anthony and Cook (2012) recommend use of the most recent revision of the SES (RSES; Koss et al., 2007) to provide a consistent, consent-based, gender-neutral measure for comparison purposes across studies. In contrast to the original SES, the RSES uses explicitly consent-based language in all items, and thus positive responses are consistent with most legal and campus-based definitions of sexual assault. The short form of the RSES contains questions about seven types of sexual assault (non-consensual non-invasive sexual contact/touching, attempted and completed non-consensual oral, vaginal, and anal assault). Each item is endorsed in relation to one of five sets of perpetrator tactics, arranged in order of increasing severity: verbal pressure from the perpetrator, expression of anger from the perpetrator, victim incapacitation due to intoxication, threat of physical harm from the perpetrator, and use of physical force by the perpetrator. The standard form of the RSES also requires respondents to indicate the frequencies of assaults by perpetrator tactic and within one or both of two temporal reference periods: past 12 months, since age 14. Reference period indicators were replaced in the present study by three categories of perpetrator: “a fellow [name of home university] student,” “a non-[name of home university] student,” and “a non-student, local resident.” The internal consistency (Cronbach’s alpha) of the RSES in the present sample was very high (.99).
Posttraumatic stress disorder checklist (PCL-C)
The civilian version of the Posttraumatic Stress Disorder–Checklist (PCL-C; Weathers, Litz, Huska, & Keane, 1994) was used to assess PSR. The PCL-C is a 17-item self-report screening instrument commonly used to screen for the Diagnostic and Statistical Manual of Mental Disorders (4th ed.; DSM-IV; American Psychiatric Association, 1994) symptoms of PTSD (Ruggiero, Del Ben, Scotti, & Rabalais, 2003). This measure was adapted in the present study to specify that each symptom rating was to be made “in relation to the worst experience you indicated having on the previous [RSES] scale.” Scores on the PCL-C in the present study reflected PSR, rather than symptoms, because determination of the latter requires clinical evaluation. Sum scores on the PCL-C range from 17 to 85. Internal consistency (Cronbach’s alpha) in the present sample was very high (.97).
Alcohol use and disorders identification test (AUDIT)
The AUDIT (Babor, Higgins-Biddle, Saunders, & Monteiro, 2001) is a standard 10-item self-report measure of alcohol consumption, and problems related to alcohol consumption. Scores on the AUDIT range from 0 to 40. Internal consistency (Cronbach’s alpha) within the present sample was moderately high (.83).
Procedures
All of the procedures used in this study were approved by the university’s institutional research review board prior to data collection. Email invitations to participate in a study titled “Unwanted Sex During Study Abroad” were sent to 400 female students randomly chosen from those in the third- and fourth-year classes but limited to those who had recently participated in study-abroad programs and who were, at the time of data collection, currently residing at the university. A positive response to the email invitation took the participant to a web-based survey, the first page of which contained a consent form. The purpose of the project given in the consent form was “to obtain a better understanding of the kinds of sexual experiences that college students have while studying abroad.” Those who gave their consent were then presented with the demographics questionnaire, followed by the RSES, PCL-C, and AUDIT. The last page of the survey contained a debriefing statement, which included information about how to register for a cash lottery prize (five prizes of US$50 each), information about the purpose of the survey, and contact information for the psychological counseling service at the university (in the event of untoward reactions to the survey questions). Two hundred eight out of the 400 students invited completed the survey, resulting in an overall response rate of 52%. Missing data were treated by excluding cases pairwise in all analyses.
Results
Sexual Assault Victimization
The overall prevalence rate for women reporting experiences of any type of sexual assault victimization (non-consensual sexual touching, attempted oral, vaginal, or anal rape, and completed oral, vaginal, or anal rape) during their study-abroad program was 18.8%. These 39 respondents reported a total of 98 incidents of sexual assault. Thirty-five respondents (16.8% of the sample) reported one or more incidents of non-consensual sexual touching. Fourteen respondents (6.7% of the sample) reported one or more incidents of attempted rape (collapsed across anal, oral, and vaginal types). Eight respondents (3.8% of the sample) reported one or more incidents of completed rape (collapsed across anal, oral, and vaginal types). The prevalence rate of attempted or completed rape collapsed across all six categories (attempted or completed anal, oral, or vaginal types) was 8.
Across all seven types of sexual assault, the primary perpetrator tactics reported on the RSES were verbal pressure from the perpetrator (2.4% of the sample, 12.8% of victims), expression of anger by the perpetrator (2.4% of the sample, 12.8% of victims), victim incapacitation due to intoxication (6.3% of the sample, 33.3% of victims), threat of physical harm by the perpetrator (5.3% of the sample, 28.2% of victims), and physical force used by the perpetrator (11.5% of the sample, 61.5% of victims; total percentages of victims reporting different perpetrator tactics and identities exceeded 100 because categories of sexual assault were not mutually exclusive within respondents). Likewise, across all seven types of assault, the perpetrators identified by victims in follow-up questions on the RSES were: student from the same university as the victim (11.1% of the sample, 59.0% of victims), student from a different university than the victim (8.7% of the sample, 46.2% of victims), and non-student resident of the host country (7.7% of the sample, 41.0% of victims; total percentages of victims reporting different types of sexual assault exceeded 100 because categories of assault were not mutually exclusive within respondents). The finding that most perpetrators were fellow university students is inconsistent with our expectation, based on the findings of Kimble et al. (2013), that most perpetrators would be non-student local residents.
Victimization, Alcohol Consumption, and PSR
The mean total score on the alcohol consumption measure (AUDIT) for the sample was 7.91 (SD = 4.43), which falls just below the level defined by Babor et al. (2001) as “indicators of hazardous and harmful alcohol use, as well as possible alcohol dependence” (p. 19). As expected, the Pearson correlation between total AUDIT scores and frequency of any sexual assault victimization was positive and moderately strong, r = .38, p = .000 (n = 163 for those completing both measures).
The mean total score on the measure of PSR (based on PCL-C) for those completing this measure (n = 39) was 25.05 (SD = 14.21), which is comparable with the mean PCL-C score obtained by Ruggiero et al. (2003; M = 29.4, SD = 12.9) in a study of female and male college students most of whom had reported exposure to one or more extremely stressful events. As expected, the correlation between frequency of any sexual assault victimization and PCL-C sum scores was also positive and substantial in magnitude, r = .63, p = .000 (n = 39), suggesting that number of assaults was associated with a greater degree of posttraumatic stress.
Discussion
The overall prevalence rate of any sexual assault victimization reported by participants in this study, almost 19%, falls within the range of prevalence rates reported in domestic studies of campus sexual assault using the SES (reviewed by Fisher et al., 2010). Such comparability is disconcerting, however, given the relatively brief reference period for this study in contrast with the single academic year used in many domestic studies. The more specific prevalence rate for attempted or completed rape (i.e., not including non-consensual sexual touching) of almost 9% during study abroad is also higher than rates of attempted or completed rape in domestic studies (approximately 5% during a 1-year period; Fisher et al., 2010). Thus, at least in the case of attempted or completed rape victimization, the risk appears to be greater during study abroad than on the home campus.
Findings from the present study replicate only some of the results of Kimble et al. (2013). Whereas prevalence rates for attempted and completed rape seem comparable across these two studies, for example, the rate for non-invasive sexual touching appears to be lower in the present one. In addition, proportions of perpetrators in each of the three categories also seem quite different in the two studies; while Kimble et al. report that most of their sample identified non-student residents of the host country, most in the present sample identified fellow college students. These differences may reflect varying propensities between students on these two campuses toward choosing to study abroad in locations that are popular with U.S. students (almost three fourths of the sample studied in Europe), in the company of peers from their home institution, and choosing abroad programs designed specifically for students in that university. Students on the two campuses also differ in their choices of study-abroad locations, which might entail differences in cultural expectations about non-invasive sexual touching. Differences in normative rates of alcohol consumption and sexual assault on the two campuses may also help to explain some of the differences found between these studies. Rates of sexual assault and alcohol consumption at the home university for the present cohort are relatively high (Flack et al., 2008), and studying abroad with peers from this institution may result in the importation of problematic behaviors normative on campus to the foreign context.
As expected, the amount of alcohol consumption in this sample was consistent with the ongoing, disconcertingly high rates of consumption reported in U.S. domestic campus studies (Wechsler et al., 2002). Likewise, the positive, moderate association between alcohol consumption and sexual assault victimization also mirrors findings on domestic campuses (Fisher et al., 2010). High rates of alcohol consumption appear to continue during study abroad, and thus, we should not be surprised that rates of sexual assault may be at least as high in that context as well.
The positive, substantial association between victimization and posttraumatic responses is also consistent with domestic findings (Zinzow et al., 2011) and serves to underscore the extent to which such victimization may result in significant posttraumatic stress. These assaults are likely all the more stressful because students are in locations where rape counseling and other psychological support services may be unknown to them or unavailable, or where students may be especially unlikely to use such services even when they are available, for a variety of reasons (e.g., differences in language or cultural mores, embarrassment, concern about legal consequences). Previous findings regarding high rates of reporting in anonymous surveys as compared with low rates of reporting to campus and legal authorities (Fisher et al., 2010) seem likely to be mirrored in the context of study abroad.
Although alcohol consumption in general was associated with sexual assault victimization in this study, in contrast to much of the research on domestic campus assault it was not the most frequently endorsed perpetrator tactic. Rather, the threat of physical harm and the use of physical force were both cited more than twice as frequently as intoxication (or perpetrators’ lies or anger, which were equally frequent). This finding was also surprising, and if replicated, it may indicate that fellow student perpetrators are more brazen in their behavior when outside of their domestic campus context. By the same token, perpetrators’ use of threat and physical force may make it easier for victims to identify their experiences as assault incidents, as these tactics conform more closely to common perceptions of sexual assault.
Limitations
Issues regarding the external validity of the present findings are foremost among the limitations of the present study. The need for further study of students from other types of universities and colleges is underscored by both the similarities and the differences between some of the present findings and those reported by Kimble and colleagues. The two institutions from which samples were drawn for these respective studies are both fairly small, academically “elite,” and located in the northeastern United States, although they also differ in a number of other potentially important characteristics (e.g., only one has a sorority/fraternity system). Another issue regarding the generalization of the present findings concerns the extrapolation of victimization rates from the reference periods for study abroad to those during the remainder of the undergraduate years. Given the relatively brief amount of time spent studying abroad, students may try to fit more intensive experiences into shorter amounts of time, thus perhaps resulting in increased risk. Such comparisons are also potentially problematic in that populations of students who experience sexual assault on campus may overlap only partially with those who study abroad. Socioeconomic class is likely to be an important factor in such comparisons, as students from backgrounds of greater means are better able and thus more likely to study abroad. In cases of campuses such as that from which the present sample was drawn, we should also be asking male students who study abroad about their perpetration and victimization experiences. At the very least, further research is needed simply to examine the range of assault prevalence, risk factors, and consequences, as compared with the few extant results currently available.
Matters of internal validity pose other potential sources of limitations in this study. For example, although the SES has frequently been modified for the specific purposes of different studies, aside from reporting indices of internal consistency based on their own samples, investigators have rarely analyzed the extent to which such modifications might affect the comparability of their results with those of others. In the present context, we do not know, for example, how asking students to report about sexual assault incidents based in part on perpetrator identity (fellow student from the same U.S. university, fellow student, or non-student local resident) affects the comparability of our findings with those of previous research (although they are comparable, in this regard, to those of Kimble et al., 2013, as this aspect of the method was shared in the two studies). In a related vein, although participants were told in the email invitation and in the consent form that we were asking about their experiences during study abroad, we did not modify the general, introductory RSES instructions to indicate this restriction in reference period. Thus, it is possible that some students might have reported on incidents that took place before or after their study-abroad programs. The higher proportion of reported perpetrators from the domestic home university, as compared with the Kimble et al. findings, may be due to this limitation. Further research with RSES instructions explicitly specifying study abroad, as the reference period is required to address this issue.
Research Implications
This study is one of only a very few to date that provides evidence for sexual assault victimization among female U.S. university students during study-abroad programs. While the findings about victimization in this context may initially appear obvious, we note that research into some previously common beliefs about campus sexual assault has shown some of those beliefs to be false or oversimplified. For example, many still believe that college campuses are safe places and that most assault perpetrators are strangers (Fisher et al., 2010). These beliefs may well extend to the study-abroad context. Other common beliefs have turned out to be far more complex than initially thought. For example, the “red zone”—the highest risk period for campus sexual assault victimization—is not necessarily limited to the first few weeks of the first semester (Flack et al., 2008; Kimble, Neacsiu, Flack, & Horner, 2008). The present results, along with those of Kimble et al. (2013) and Hummer et al. (2010), provide initial evidence that study abroad entails significant risk of sexual assault and that some factors related to assault (e.g., risk from different groups of perpetrators) may differ by campus.
Given that this is only the third systematic investigation of sexual assault during study abroad and only the second to use standard measures of assault and related factors, clearly much more extensive research is required. Data from a range of college and university campuses are needed to determine overall rates of assault among U.S. students studying abroad. Likewise, rates of assault among students from other countries studying abroad in the United States are also needed. Victimization and perpetration should be examined among both female and male students. Location of study-abroad experiences and the extent to which students study abroad individually or in groups from their own campus also require further study.
Potential risk factors for assault during study abroad also require investigation, including previous histories of assault, other traumatic experiences, and the use of alcohol and other drugs. Other kinds of follow-up data should also be obtained on the potential impact of assault during study abroad, including anxiety and depression, difficulties in social interaction, and subsequent problems with academic performance.
Clinical and Policy Implications
These and other findings provide at least some empirical evidence to inform efforts at preparing students for their study-abroad experiences. Given the only partial overlap in findings between this and the Kimble et al. (2013) study, sufficient differences exist that educational efforts should be based whenever possible on locally collected data. This recommendation is consistent with recommendations from the White House Task Force to Protect Students From Sexual Assault (2014) that U.S. colleges and universities conduct campus climate surveys on sexual assault. Local data should be incorporated into the orientations that students receive while preparing for study abroad. The present findings also warrant inclusion of explicit protocols for appropriate responses by support staff and judicial personnel to reports of assaults during and after study abroad. These results should also be useful to health care professionals on home campuses, who should be aware that a substantial proportion of female students returning to campus from study abroad may have been sexually assaulted and who should ask about this possibility when assessing the mental health needs of those students. Finally, increasing reporting and acknowledgment of the extent to which campus sexual assault is a significant, ongoing public health issue (White House Task Force to Protect Students From Sexual Assault, 2014) should help support further efforts on the part of academic administrators to take serious steps toward the kinds of changes in campus cultures required to prevent these incidents, whether domestic or abroad, from happening in the first place.
Footnotes
Acknowledgements
The authors wish to thank Deirdre M. O’Connor and the anonymous reviewers of this manuscript for their critical feedback.
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.
