Did You Actually Read the AAU Sexual Assault Survey?

Students and administrators have been rather abuzz lately with the release of the results from a sexual assault survey from the Association of American Universities’ (AAU) that covered 27 schools, including Cornell.
 
President Elizabeth Garrett initially sent every student an email regarding the study’s findings and the numerous highly effective and meritorious actions the University is taking to reduce the prevalence of sexual assault on campus. That week, the Cornell Sun devoted a section of its front page Tuesday and Wednesday to related coverage and President Garrett and Vice President Ryan Lombardi discussed sexual assault at the Student Assembly meeting.
 
But how many “concerned” Cornellians have actually read the survey, or at least browsed some of the more important sections? By important, I am referring not so much to the numbers, but to the definitions, because if the definitions behind what the survey is intending to measure are faulty, nonsensical, or don’t make sense, then the numbers themselves don’t really mean anything.
 
For example, did you know the AAU considers unwanted kissing a crime (sexual battery)?
For example, did you know that persistently asking someone out on a date or for dinner is sexual harassment?
For example, did you know that telling “offensive” jokes is sexual harassment?
For example, did you know that if your partner has ever made a decision for you, you are a victim of intimate partner violence?
For example, did you know that the words “rape” and “assault” were never used in the survey questions—only vague terms like “sexual touching”?
 
If you answered mostly or only “no” to these questions, don’t feel bad. Campus administrators have no vested interest in elucidating these facts and challenging these types of studies for their vague or ridiculous definitions and methodologies because that would mean (1) coming to terms with the true cases of sexual assault on campus (over-drinking, hookup culture, lack of law enforcement, and reluctance to incur bad PR from publicized expulsions) and (2) giving up on their legal power grabs (e.g. trying to eliminate due process rights for the accused).
 
A good analysis of the study’s numerous shortcomings can be read here: “The latest big sexual assault survey is (like others) more hype than science”.
 
One commonly cited flaw is the low response rate of 19%. While such a response rate does not preclude statistical conclusions, last year former Vice President Susan Murphy ’73 PhD ’94 said a response rate of 40-50% would be needed for an accurate picture of Cornell. Either way, a more substantive problem is the response bias inherent in a voluntary survey: those who are victims or think they have been victims of sexual assault are naturally more likely to respond to a voluntary survey. In fact, the AAU survey authors themselves admit this response bias, writing, “Two of these three analyses provide evidence that non-responders tended to be less likely to report victimization. This implies that the survey estimates related to victimization and selected attitude items may be biased upwards.”
 
But to belabor the point, it is the efficacious definitions that are most egregious. When the survey basically defines sexual harassment as any social activity imaginable, it is no wonder 50% of Cornell respondents reported being victims of sexual harassment since coming to Cornell. As another example, consider the finding that 88% of Cornell respondents have been victims of intimate partner violence, which only makes sense if students think they’re victims of intimate partner violence because their partners once told them to wear a coat because it’s cold outside.
 
Even then, of those who said they were victims of nonconsensual sexual touching due to physical force, 83.7 percent felt the incident was not serious enough to report.