Apparently the likelihood of a woman being raped on campus is closer to 1 in 100 than to 1 in 5. Would you call that an epidemic?

Greg Piper at The College Fix has the story:

Shoddy Statistics Behind Bill On ‘Campus-Rape Epidemic,’ Former U.S. Official Says

The Campus Accountability and Safety Act, introduced by a bipartisan group of senators last week, relies on statistics from a poorly designed survey, according to a former chief economist of the U.S. Department of Labor.

Writing in MarketWatch, Diana Furchtgott-Roth says the Centers for Disease Control survey that said 18 percent of women have been raped and White House report that said one in five women have suffered the same in college are contradicted by other reports. That’s probably because of flawed methodology:

The survey starts out with 201,881 phone numbers, but whittles it down to only 8% of the original sample. Fifty-three percent of the phone numbers are of “unknown eligibility,” hence unused, without stating the reason.

Further, it is the surveyors, rather than the respondents, who determine whether the respondent has been raped. It is unclear why the CDC did not just ask whether the respondent has been raped. In the CDC’s own words, “the survey is designed to assess violence in a way that is consistent across states.”

Then, some questions are ambiguous. For example, the survey asks, “When you were drunk, high, drugged, or passed out and unable to consent, how many people have had vaginal sex with you?” To a typical respondent, it is not clear whether the condition of “unable to consent” applies to “drunk, high, drugged,” or if “unable to consent” is a separate condition. Respondents who had consensual sex while drunk but were able to consent (not exactly an uncommon act on campuses these days) may answer affirmatively, leading the surveyor to wrongfully count this as an instance of rape.

Using Bureau of Justice Statistics figures and a University of Michigan professor’s compiled data sets, Furchtgott-Roth comes up with 1.3 rapes per 1,000 persons for the population at large and 3-5 percent of college women.