The Washington Monthly’s Jesse Singal is offended by my previous post. I’m re-treading some ground here, so those who are up-to-date on the “campus-rape incidence rate” debate can probably skip this one.

He seems to doubt the notion that “there’s some giant evil lobby benefiting from false accusations of rape.” Well, not false accusations so much as false statistics. For this I’ll direct him to Heather Mac Donald’s definitive take.

He says he has “no idea what numerical system [I’m] using to determine that 70 is 1.2 percent of 24,000.” I was probably a little unclear that by “70 different students . . . each year,” I meant I was tallying rape incidents over multiple years, for the purpose of comparing the rate at this school to the supposed rate overall (“one-in-five to one-in-four over four years,” they like to say). Over a four-year college career, 70 per year adds to 280. 280/24,000=.0117.

He points to a DOJ report indicating that one in five college females will be raped over the course of her time on campus. The report is a little odd in its methodology — it starts with the notion that 3 percent of women will experience rape over the course of a school year, extrapolates that to cover the whole twelve months of a year (bringing it to 4 percent), and then multiplies by five to account for the five years in the now-typical college career (bringing it to 20 percent). And Singal doesn’t mention that by “experiences rape,” the DOJ means both completed and attempted rapes (both crimes are awful, of course, but it’s important to be clear). Going back to the 2000 study the DOJ cites, the figure for completed rape is 1.8 percent per year.

Even that figure is suspect, as Mac Donald has demonstrated:

The 2000 Department of Justice study of campus rape found that those women whom the researchers characterized as rape victims “generally did not state that their victimization resulted in physical or emotional injuries.” . . . Moreover, 65 percent of those whom the researchers called “completed rape” victims and three-quarters of “attempted rape” victims said that they did not think that their experiences were “serious enough to report” — a judgment inconceivable from a real rape victim.

Back when Mac Donald wrote this, I took a look at the National Crime Victimization Survey — which like the DOJ study includes unreported crimes, but which, to be fair, doesn’t focus specifically on college students. I found:

Per 1,000 women 12 and older, 1.4 said they’d suffered rape, attempted rape, or sexual assault in 2005. Looking at the age of the victims, those in the 16-19 group had it worst, with 3.2 2005 rape victims in every 1,000 respondents. (The latter includes men and women, so the female rate is probably around 6 per 1,000.) The 20-24 group’s rate isn’t statistically reliable, but for what it’s worth, it’s 1.1.
These numbers aren’t directly comparable to the one-in-four number, which purports to tally rape incidence over many years. But if, in a group of 1,000 women, six different ones were assaulted each year, it would take 33 years for 20 percent of them to have been victimized, and almost 42 years for a quarter of them to suffer.
Koss claims that about 25 percent of college students are raped between age 14 and graduation, and 20 percent of them suffer rape in only the four years on campus. At the very least, if both the NCVS and Koss are correct, women who don’t go to college are remarkably safer from rape than those who do.

Looking back at this, it’s actually statistically impossible for both to be right, because about 60 percent of high-school graduates go to college. Some women drop out of high school or college, of course, but no matter. Even if we take a low-end estimate of 25 percent of women being in college, and say that 20 percent of these women are raped, that’s an overall 5 percent four-year victimization rate — before we even get to rapes of women who aren’t in college. The highest number in the NCVS, 6 per 1,000 per year, amounts to a 2.4 percent four-year rate.