Icosian Reflections

…a tendency to systematize and a keen sense

that we live in a broken world.

On the AAU Survey and the Crimson

I've got an op-ed in the Harvard Crimson today, expressing my concern that an important narrative is missing from the discussions of the AAU sexual assault climate survey. Excerpt:

When male survivors are invisible, they face stigma against seeking help. Though male and female survivors of sexual assault seek out institutional resources at roughly the same (low) rates, male survivors are 60 percent more likely than female survivors to speak to no one—not even a friend—after an assault. (31.2% versus 19.3% for assault by force; 38.1% versus 23.3% for assault by incapacitation.) And so male students make up more than a quarter of silent survivors, in large part because we so rarely acknowledge that they exist at all. (...)

Those numbers, by the way, come from tables 3.1a,c and 3.5a,b in the full report. Below, I've got few thoughts that didn't make it into the published version.

disclosure: I am, at least on paper, still a Crimson editor on the Design Board and Data Science Team. My published-articles count is now...one.


I've heard (and more or less believed) "one in three women" for longer than I can remember. And so the fact that the report basically confirms that number doesn't mean all that much to me. I suppose if you doubted that number but believe it now, then the report is big news, but otherwise, this isn't a new crisis because we already knew that it was a crisis, and hopefully, were already acting accordingly. Waiting until the evidence is undeniable before updating your beliefs and actions is just bad strategy.

If you've been calibrating your actions to an unknown quantity that suddenly becomes known to you, conservation of expectation means that you should end up adjusting upwards as often as you end up adjusting downwards, and sometimes, the extra information doesn't change your plans at all. If extra information will most likely cause you to update in a predictable direction, why haven't you already updated?

Still, I think that the survey results are hugely important, -- just not for the reasons that everyone is mostly talking about. It's the surprising results we should be updating on, not the concerning ones.

For an example that didn't have a place in the Crimson piece:

  • Something like 15 percent of on-campus sexual assaults of female students, and 12 percent of on-campus rapes of female students, were in 'a space used by a single-sex student [organization], other than a fraternity or sorority house'[1] [3.1a; 3.5a,b]. Fewer than 4 percent of assaults of male students take place in such locations [3.1c; 3.5c], mainly by incapacitation, rather than force.
  • The combined membership of the final clubs which host parties is not larger than 4% [wiki].

This is actually news to me, and I also think that the Crimson is badly messing up in not reporting on it.


My original submission went in almost two weeks ago, on September 25 (less than a week after the survey results were released). For the whole first week, I was (rather selfishly) worried that someone at the Crimson would sit down with the report, read it, write a long-form piece on its most surprising statistics, and take the wind out of my sails. After all, the title isn't quite so catchy if it's "six hundred and fifty-nine words".

But nope; this is the Crimson, after all. Thirteen days, four articles (3,4), and 2600 words later, we're now at seventy-one words. (Students Will Host Identity-Based Discussions on Sexual Assault repeated the sentence "For LGBAQN undergraduate males, that rate of experiencing nonconsensual penetration and sexual touching was 10.9 percent in that time, compared to 2.7 percent among their straight male peers.")

22% of sexual assault survivors are male, and male survivors get less than 1% of the Crimson's column inches. And that's just sad.


I'm grateful to Christina, Connor, Cyndia, and Kara for their edits and input; the early drafts I showed to them were definitely not worthy of print, even in the Crimson. Incidentally, if you'd like to be on my go-to list for when I feel like I need editing help, just drop me a line. I guarantee you that the usual suspects would appreciate a break.