My Faults My Own

One's ponens is another's tollens.

Sex: Statistics and Student Opinions

This week, MIT released Survey Results: 2014 Community Attitudes on Sexual Assault, making them one of the first schools to release such broad survey data on sex crimes. These are the results of a survey emailed to MIT undergraduate and graduate students last April, which had a response rate of 35% from 10,831--3,844 total responses.

I wish to be clear: Without good reason to believe otherwise, I'm taking these statistics as probably representative of MIT's peer institutions as well, and in no case do I mean to critique MIT specifically by citing them. If anything, the school deserves praise for its dedication to transparency by publishing such detailed statistics.

Now, MIT is clear that the document they've published should be taken as initial, not final, results:

"This document is a summary of the most pertinent results corresponding to questions asked in the survey; it is intended to be an initial summary of survey results. Throughout the upcoming academic year we will work with the community to use the survey data to answer additional important questions. New findings will be posted to, where the full text of the survey questions and other related information can also be found."

They are also upfront about acknowledging the inevitability of response bias:

"Response bias is expected in virtually any voluntary survey, particularly one focused on a narrow topic. While we invited all enrolled graduate and undergraduate students to take this survey, and more than one-third responded, it is not possible to know if students self-selected in or out of the survey in a way that would bias our results. For example, it is difficult to determine whether students who have experienced sexual assault were more or less likely to respond to the survey. This does not make the findings from the survey any less accurate; it simply means that the rates based on those who responded to the survey cannot be extrapolated to the MIT student population as a whole, and cannot be validly compared to results from other surveys."

Keeping all that in mind, though, there are interesting things we do find in the report. In no particular order (with all emphasis mine):

  • 6.5% (208) of respondents "indicated one or more unwanted sexual behaviors involving force, physical threat, or incapacitation."
  • 15% (122) of female undergrads and 4% (25) of male undergrads report being sexually harassed.
  • either* 2% (13) or 5% (32) of male undergraduate respondents and either* 10% (81) or 17% (133) of female undergraduate respondents report being sexually assaulted. (The lattermost seems to be literally the only statistic any media site sees fit to report on, predictably. See, look--even I felt compelled to bold it.)
  • 5% (38) of female undergrads and 1% (5) of male undergrads report being raped.
  • 6% (44) of male undergraduate respondents and 12% (100) of female undergraduate respondents report "Unwanted Sexual Behavior—sexual touching or kissing, oral sex or sexual penetration (attempted or completed)—not involving force, physical threat, or incapacitation and did not indicate they had been sexually assaulted or raped"
  • "When survey participants were asked about their behavior towards others, 1.9% of the respondents said they had acted in a way that would be considered unwanted sexual behavior, and another 2.2% indicated they were unsure if they had behaved in this way. More than one in five undergraduate respondents indicated knowing a perpetrator."
  • Between 10% and 15% of sexual assaults are perpetrated by females against males. (This is a composite statistic I multiplied out, not one written up in the report.)

* Interestingly, when respondents were asked whether they had been sexually assaulted, 2% and 10% of male and female undergraduate respondents, respectively, reported that they had, but when asked in separate questions whether they had experienced "sexual touching or kissing", "attempted oral sex", "oral sex", "attempted sexual penetration", or "sexual penetration", 5% and 17% reported that they had experienced at least one. HuffPo quotes MIT Chancellor Cynthia Barnhart:

"That indicates to us there is confusion around what sexual assault is, and that's why it's imperative, I think, that we open up this dialogue and increase education about what constitutes sexual assault and consent."

I'm not certain if any of this is news. 17% of female students being sexually assaulted is (1) terrifying and (2) not the 1-in-3 statistic that occasionally gets bandied about. It's worth noting that, by asking current students about events occurring thus far in their college career, the survey is sampling students who have completed an average of 2.5 years of college (SD 1.58), so the numbers for "in all of college" are probably (sublinearly) higher.

But really, what I read in these statistics is a great deal of confusion. When 12% (165) of undergraduate respondents report experiencing an "unwanted sexual behavior" which meets the criteria of sexual assault, but only half of them indicate that they'd been sexually assaulted...something's wrong. The signs only get more troubling when students are asked to rate their agreement with particular statements (from "strongly disagree" to "strongly agree"):

  • 67% of undergraduate respondents agreed that "Rape and sexual assault can happen unintentionally, especially if alcohol is involved."
  • 23% of undergraduate respondents agreed that "When someone is raped or sexually assaulted, it's often because the way they said 'no' was unclear or there was some miscommunication." (note: 'often')
  • 33% of undergraduate respondents agreed that "Sexual assault and rape happen because men can get carried away in sexual situations once they've started." (don't forget: between 10% and 15% of sexual assaults are perpetrated by females against males)

In case you missed it: two out of three undergraduate respondents believe "rape and sexual assault can happen accidentally", and almost a quarter believe that rape and sexual assault are "often" the result of miscommunication. I'm...going to have a lot to say about this tomorrow.

Fortunately, there are a few encouraging statistics from this category:

  • Only 8% of respondents agree that "An incident can only be sexual assault or rape if the person says 'no.'"
  • Only 7% of respondents agree that "Many women who claim they were raped agreed to have sex and then regretted it afterwards."
  • Only 14% of respondents agree that "Rape and sexual assault happen because people put themselves in bad situations." (1 in 7 isn't bad, considering how often it seems we hear this meme)
  • 97% of respondents agree that "It is important to get consent before sexual activity."

Anyway, the full report is here, with more details at I've got more to say about what I think these results mean, but I think I'll leave them for part 2, and keep this mostly just a summary of the MIT report. Once again, good on MIT for collecting and publicly publishing the results of this (I think) unprecedented survey.