Wednesday, November 27, 2013

Is it one-in-four, one-in-125, or one-in-1,876?

We are loathe to attack rape statistics because we don't want to trivialize a serious social pathology. But time after time after time, we read articles about rape where the authors see fit to inflate the stats on rape, and then go out of their way to trivialize the victimization of the wrongly accused. They just can't help themselves -- they simply have to take a shot at the sort of advocacy we do here by declaring that "only" a small percentage of rape claims are false. They typically add -- even though they have no way of knowing whether it's true -- that the percentage is the same as every other crime.

Here's a newsflash: you don't have to lie about numbers to underscore the seriousness of rape. You hurt your own cause by saying things that are ridiculous, and that are hurtful to the community of the wrongly accused. We don't think rape should be trivialized, and we don't think false rape claims should be trivialized.

We received a fascinating email from a long-time reader, who took issue with rape stats in a recent widely read article. He furnished his own analysis, which, try as we might, we haven't been able to refute. We paraphrase it below. Stay with this one until the end.

The latest bit of rape hysteria is courtesy of Slate. It appears under this none-too-subtle, top-of-the-homepage headline box: RAPE IS MUCH, MUCH MORE PREVALENT THAN WE THOUGHT. THE U.S. HAS UNDERCOUNTED IT FOR YEARS.

We'll save you the trouble of reading and cut right to the important part: the numbers are much higher than we thought, the sky is falling. So much, in fact, that there are 1.27 Million sexual assaults a year. Or roughly 6.5 times as many as we previously thought.

Never mind that the new number is almost 15 times the number reported in real-live crime statistics. We suppose that just proves it must be true. And never mind that these numbers come from surveys where every rape allegation is uncritically accepted and untested against competing claims of innocence. The irony is that for those rape claims that are actually reported to police, every serious study ever conducted shows that the majority of claims can't be classified one way or the other as rape or non-rape. It's only when the claims are unreported that they are uncritically and automatically accepted as rape, so long as they meet whatever definition is assigned to them in the survey -- and surveys are notorious for engorging the definition of rape beyond legal recognition.

Fine. Let's go with the 1.27 Million per year figure.

If you take a population average for the last three years (2010-2012), you get 311,610,092. US population is 51% female (actually 50.9%, but I round off). Which gives you an average female population of 158,921,147. Now here's the link to the FBI's Crime Stats. The average number of reported rapes for those same three years is 84,715. (Which, again, is 15 times fewer than the number quoted in the Slate piece). Divide the population by the number of reported rapes, and you get... wait for it....

...1-in-1,876.

That's right. 1-in-1,876. Almost exactly the number reported in Chad Hermann's landmark article which took a three-year average for three universities.

But, again, back to the original point.  Even if we ignore these real, live, actual crime reporting stats and accept that 15x as many figure (which would mean that 93.33% of all rapes, not 90%, go "unreported"), and accept that those numbers (taken from surveys and interviews) are indeed correct, that still leaves us at 1-in-125. Which is a damned far cry from 1-in-4.

All of which proves, yet again, that people who spout these numbers simply can't -- or, more accurately, won't -- do simple math. The one-in-four is not in the same universe as the number you get if you use THEIR OWN STATS.

And these are the same types of people who just can't resist trivializing the victimization of the wrongly accused by suggesting there are too few of them to worry about.