Campus sexual assault: what we know and what we don\ .

• Author: Beaver, William R.
• Position: Report

Sexual assault on campuses in the United States is a hot-button issue. How big is the problem? It is hard to know because many assaults go unreported. Surveys of college-age students have the potential to overcome this problem. Unfortunately, as I show in this article, the most widely cited surveys used by policy makers have flaws, and there is disagreement among them when it comes to the incidence of sexual assault on the nation's campuses. Despite the shortcomings of these surveys, the media promote the idea that campus sexual assault is a commonplace and serious problem. In response, based on this incomplete and less-than-accurate data, the federal government has implemented a series of policies that are putatively designed to lessen sexual assault but that impose heavy costs on colleges and potentially undermine due process.

Sexual assault on college campuses has captured the nation's attention. The situation has been characterized by Senator Kirsten Gillibrand (D-N.Y.) and others as an epidemic, driven by what some have called a rape culture (Toffee 2014). To make matters worse, Annie Clark, founder of the advocacy group End Campus Rape Now, maintains that colleges routinely mishandle sexual assault cases. "You hear about Amherst, and then it dies down. You hear about Yale, and it dies down. We're tired of it just popping up and everyone says it's really horrible, then nothing happens," says Clark (qtd. in Kingkade 2015). Congressman Jarod Polis (D-Colo.) stated during a congressional hearing that colleges should be able to expel any student accused of sexual assault even if that student is innocent, although he later claimed he misspoke (Burness 2015). But critics of this position maintain that such charges are largely exaggerated and blown out of proportion. Whichever the case, it is difficult to know exactly how many of these offenses actually occur. Many crimes are seriously underreported, but sexual ones even more so. The vast majority of campus sexual assaults are not reported to either law enforcement or to school officials, which obviously makes it difficult to accurately gauge how many actually occur. In theory, one way to get more accurate readings is to conduct interviews and surveys asking a broader sample of respondents about their experiences on campus.

I begin by examining the two most prominent surveys, which have done the most to shape public opinion and public policy: the Campus Sexual Assault (CSA) Study and the American Association of Universities' (AAU) Campus Climate Survey on Sexual Assault and Sexual Misconduct. Both surveys are aimed directly at college students and for that reason have gained the most attention and are deemed to be important. Then I turn to another relevant though less-prominent survey described in the report Rape and Sexual Assault Victimization among College-Age Females, 1995-2013 (Langton and Sinozich 2014), based on data from the National Crime Victimization Survey (NCVS). The NCVS is administered every year and focuses on a number of violent crimes, including rape and sexual assault, and some of its findings differ considerably from the CSA Study and AAU Survey. I discuss the results and implications of each of these surveys, with the aim of providing a clearer understanding of what we know and what we don't about campus sexual assault.

The CSA Study

The CSA Study is a web-based survey developed by RTI International, a private firm, and funded by the National institute of Justice, the research, development, and evaluation agency of the U.S. Department of Justice. The survey was administered in 2005 at two large public universities in the South and Midwest to students ages eighteen to twenty-five. In all, 5,466 women and 1,375 men took the survey, with an overall response rate of 42.5 percent, which is considered low by the survey's authors. The CSA Study uses sexual assault as an umbrella term, which includes a wide range of behaviors from rape (oral, anal, vaginal, and digital penetration) to attempted rape and forced touching of a sexual nature. The results were released in December 2007 and found that most sexual assaults occurred between individuals who knew each other and that alcohol had been consumed. Specifically, 11.1 percent of females indicated an assault had occurred while they were incapacitated by alcohol. A slightly larger number of students (13.7 percent) reported being the victim of a completed sexual assault, and 3.4 percent of female respondents report being forcibly raped. Overall, 19 percent of females had experienced a sexual assault or an attempted sexual assault since entering college. The rate for men was 3.7 percent (Krebs et al. 2007).

The major impact of the survey was not the findings per se but what was emphasized by the media. The finding most commonly cited is that almost one in five college women were the victims of sexual assault during their time on campus--a figure that does suggest a serious problem. What is often not emphasized is the wide range of behaviors included in the definition of sexual assault. There is obviously a great deal of difference between forced touching and forced penetration. The fact that 3.4 percent of females reported being forcibly raped during their time on campus was seldom emphasized. Instead, all the behaviors are lumped together. The situation was further distorted when some media outlets reported that one in five college females were the victims of rape. As James Allen Fox, a professor of criminology at Northeastern University put it, "This one in five statistic shouldn't be just taken with a grain of salt but the entire shaker" (qtd. in Contorno 2014).

The other problems with the CSA Study are methodological. First, critics pointed out that the survey was not based on a national sample but involved just two universities, which should have sent a signal to interpret the findings with some caution. The other issue is what researchers call self-selection bias, which occurs when individuals select themselves into a sample, resulting in a biased sample; it is a large problem with survey research in general. In this case, students who had experienced some form of sexual assault are more likely to complete the survey than those who had not experienced sexual assault. Although men are less inclined to participate in surveys, the fact that men are far less likely to be victimized does suggest why they were less likely to take the survey. Much the same is probably true for females. As one female student at the University of Kentucky told National Public Radio in speaking of taking a sexual assault survey, "Personally, that's not something that's affected me at this point in my life so I really wouldn't feel the need to take it" (qtd. in Smith 2015).

Despite the problems, the one-in-five figure became firmly imbedded in the nation's understanding of sexual assault on campus, along with the idea that something needed to be done. Accordingly, the Center for Public Integrity, a nonpartisan news organization, began an investigation of campus sexual assault. Its report, issued in 2010 and titled Sexual Assault on Campus: A Frustrating Search for Justice, found that most victims of sexual assault did not report it...