While much of the focus surrounding sexual assault on college campuses has recently been on the frequency of victimization, the perpetrators of such behavior receive, as a whole, relatively less attention. In some ways, this makes sense. Researching the number of men responsible for perpetrating sexually violent acts against women is difficult. The accuracy of such research is crucial and depends largely on the amount of college-aged men who will willingly report perpetrating such acts. Even guaranteeing anonymity in reporting is no guarantee of total honesty. As such, there is a broad range of reported sexual perpetration rates in existing research and an overall dearth of large-scale, national epidemiological studies on this topic—in part due to, as our team discovered, the wide variety of assessment tools and administration techniques used by investigators.
To reliably translate perpetration rates into actionable policies, it is imperative to deeply examine the methods of this research. Scientifically answering questions about complex human behavior requires attention to both operationally defining terms and precisely asking research questions. How is this research being conducted? How are these men approached for research studies? In person? Online? In groups? Alone? In what ways are men being asked about their behavior? Should researchers ask directly if a respondent has ever raped anyone? Or, are researchers better off asking about this behavior without using words like “rape” or “assault”—words with heavy cultural connotation and definitions that change depending on the time and study. Field-standard measurement tools are frequently modified by investigators. How do these changes affect men’s responses, and how are other investigators supposed to interpret their results relative to the standard? What is the best way to statistically determine the rate of men sexually perpetrating? What methodological factors could affect overall reported perpetration rates?
Our research team sought to address these questions in a wide-reaching systematic review of studies examining sexual assault perpetration in college-aged men published between 2000 and 2017 (1). This resulted in 78 studies across 25,524 college men. Our goal was to examine the overall perpetration rate and how the rate of perpetration may be related to the measurement tools used in the field of sexual perpetration research. We asked the question: Does the manner in which sexual perpetration is assessed statistically impact the answer?
When statistically examined, these varied ways of asking what is supposedly the same question yielded dramatically different results. Overall, we found that almost one third of men in our sample (29.3 percent) reported perpetrating some degree of sexual violence against women, with a 6.5 percent rate of rape. However, these percentages represent averages because there were a wide variety of perpetration rates across individual studies, in large part due to the impact of varying measurement tools. In total, 16 different questionnaires were used over the 78 studies to assess perpetration, the most common being the 1982 Sexual Experiences Survey-Perpetration, and its updated iterations. Indeed, versions of the Sexual Experiences Survey (SES) accounted for 78.2 percent of the measures used. However, it was also very common for researchers to modify the SES; the SES was modified 63.8 percent of the time. Even though these modifications create difficulty in comparing results across studies, they may be useful. When we compared non-modified to modified versions, the modified measures revealed higher overall prevalence rates. We also examined the different versions of the SES. When we compared versions of the SES, the updated version (the SES – Short Form Perpetration) yielded higher rates of rape perpetration but lower rates of verbal coercion than older versions of this questionnaire. We also compared studies that used some version of the SES to non-SES using studies. Studies that did not use the SES recorded higher prevalence rates (any type of perpetration: 41.5 percent vs. 26.2 percent). We recommend future researchers use an un-modified version of the SES as these questionnaires have the best empirical support. Currently available data likely underestimates the scope of this problem.
Because the rate of perpetration significantly changes based on individual variations in measurement, this indicates a clear problem in the field. Even though researchers may desire to alter the questions in these assessments to more accurately report on outcomes, our group’s findings clearly illustrate a need to create behaviorally defined and consistent definitions of perpetration, verbal coercion, and rape that researchers can use to create innovative and empirically sound solutions to the public health crisis of sexual violence.
This project was an important step in understanding why it has been so difficult to identify the frequency of sexual perpetration and highlighted some key methodological improvements that need to be made in the literature with regard to measurement. We also identified that nearly all of the studies included in our final systematic review focused on white cisgender heterosexual/female-attracted men assaulting cisgender women. Further research should be more inclusive of individuals from marginalized and/or underrepresented groups. Forty-six percent of bisexual women have been raped in their lifetime (2). Ten percent of respondents to the 2015 U.S. Transgender Survey have been sexually assaulted over that past year alone (3). Among transgender respondents of color, Native American (65 percent), Middle Eastern (58 percent), Black (53 percent) and multiracial (59 percent) individuals were the most likely to experience sexual violence in their lifetimes (3). Investigating perpetrating sexual violence against people of color, queer, and transgender individuals (as well as those at the intersection of these identities) is radically important for the health and wellbeing of those at risk. Understanding the nature of perpetration in and against these groups is another crucial aspect of preventing future sexual violence.
The ultimate goal of scientific research is, of course, to develop accurate answers to empirical questions. But the importance of how the question is asked should not be lost in the rush for an answer.
Amanda Vitale, B.S., is a research coordinator with the Laboratory of Medical Epigenetics at the James J. Peters VA Medical Center and Nash Family Department of Neuroscience at the Icahn School of Medicine at Mount Sinai.
Alyssa Ciampaglia, M.S., is a clinical research coordinator in Maternal Fetal Medicine at the Center for Fetal Diagnosis and Treatment at the Children's Hospital of Philadelphia.
Kristin E. Silver, M.A., is a doctoral candidate at The University of Akron and currently completing her pre-doctoral psychology internship at the Durham VA Medical Center.
RaeAnn E. Anderson, Ph.D., is a clinical psychologist and assistant professor of psychology at the University of North Dakota.
(1) Anderson, R.E., Silver, K.E.S., Ciampaglia, A.M., Vitale, A.M., & Delahanty, D.L. (2019). The Frequency of Sexual Perpetration in College Men: A Systematic Review of Reported Prevalence Rates from 2000-2017. Trauma, Violence, & Abuse.
(2) Black, M.C., Basile, K.C., Breiding, M.J., Smith, S., Walters, M.L., Merrick, M., … Stevens, M. (2011). The National Intimate Partner and Sexual Violence Survey (NISVS): 2010 Summary Report. Atlanta, GA, USA.
(3) James, S.E., Herman, J.L., Rankin, S., Keisling, M., Mottet, L., & Anafi, M. (2016). The Report of the 2015 US Transgender Survey. Washington, D.C.