According to the results of a survey released last month by the Association of American Universities, roughly one in four women will experience sexual assault during their time in college and one in five women at UT.
That’s deeply troubling. But what’s almost as troubling is that, because of the survey’s limitations, we still know next to nothing about what sexual assault looks like at UT — meaning that we also know next to nothing about how to prevent it.
The survey did include some useful details. For instance, it’s helpful for law enforcement to know that sexual assaults are more likely to occur off-campus. But for the UT System’s four-year study to provide more specific results, it needs to revise the AAU’s methodology.
Despite its large sample size, the AAU survey still doesn’t tell us how many students are actually victims of sexual assault. As President Greg Fenves wrote in an email to the student body following the survey’s publication, even one assault is intolerable. But any self-reported study, especially one with an abysmal 13 percent response rate, will be distorted by sampling bias, perhaps even beyond the point of reliability.
There are two simple approaches through which UT could address that problem. The first is to select a random sample of students, then collect enough demographic information to draw helpful conclusions from the data, including breakdowns based on race, age, classification and Greek affiliation. Alternatively, the University could make the survey mandatory by tying its completion to a student’s registration status, as it does with the “Know Your Line” safety module that freshmen must complete.
The AAU survey gave us the most comprehensive evidence yet that sexual assault is a major problem on college campuses. The UT System will do its students a disservice if it merely parrots that result, without shedding light on any potential solutions.
Shenhar is a Plan II, economics and government junior from Westport, Connecticut. He is an Associate Editor. Follow him on Twitter @jshenhar.