Today’s Washington Post announces that a White House Task Force to Protect Students from Sexual Assault  is launching a new website, NotAlone.gov, and has published a 20-page report on sexual assault at colleges and universities.  Previously, Obama’s Department of Justice and Education made headlines by issuing a settlement agreement in a sexual harassment case at the University of Montana that was widely criticized for its over-broad definition of sexual harassment, mis-statements of relevant law, and insufficient attention to First Amendment concerns.  Combined, all of these efforts suggest that deterring sexual harassment and assault are high priorities for this administration.

Indeed, nobody is in favor of sexual assault. It ought to go without saying that assault is a devastating experience for the victim. Colleges and universities should take these allegations with the utmost seriousness.

At the same time,  colleges and universities need to safeguard the rights of the accused. They also need to make sure that in protecting students from unlawful sexual harassment that they are not infringing on the First Amendment rights of other students.

It is also important to make sure that policymakers have their facts straight on this critical topic. It does nobody any good to talk about this serious and often emotionally charged issue by throwing around misleading statistics that suggest that sexual assault on campus is a worse problem that it really is. Unfortunately, the “Not Alone” report is not a model of how to engage seriously with hard data when facing contentious questions. It begins by stating the provocative claim that one in five college women have been sexually assaulted in college. The two studies that it cites in the footnote following this claim do not quite support this figure: each states that close to 20% of the women in the study experienced either an attempted or completed sexual assault. By contrast,  the percentage of surveyed women who reported experiencing completed sexual assault in both studies was about 13%. Notably, a majority of the women who reported experiencing completed sexual assault to The Campus Sexual Assault Study (60% of those classified as “physically forced victims” and 75% of the “incapacitated assault victims”) did not consider themselves to have been raped. Other studies finding similarly high numbers, such as Mary Koss’s 1987 study of campus rape for Ms. magazine, have suffered from similar methodological problems.

The report also calls for colleges to start administering “campus climate” surveys to students to determine how bad the problems of sexual assault are on campus — an interesting recommendation in light of the administration officials’ earlier positions on the use of surveys to measure student interest in sports for Title IX purposes. By way of background: in 2005, the Bush Department of Education introduced a “Model Survey” that colleges and universities could administer to assess male and female students’ respective interest in sports. If survey results indicated that fewer women than men at a particular college were interested in playing sports,  then the Department of Education would presume that disparities in that college’s athletic offerings were the result of these different levels of interest and not of bias or discrimination. Although the Model Survey might have seemed like it should have been non-controversial, it drew vehement responses: Jocelyn Samuels of the National Women’s Law Center (now Acting Assistant Attorney General at DOJ’s Civil Rights Division) questioned whether (pp. 52-54)  universities were fundamentally competent to administer such a survey accurately. For example, Samuels claimed, universities might not be able to figure out how to achieve a high enough response rate for the survey to be statistically valid, or some students might not be capable of assessing their own interest and ability to play sports. Perhaps unsurprisingly in light of all of this opposition, the Obama Department of Administration rescinded the Model Survey in 2010.  Yet the proposed campus climate survey could potentially have the same problems: some universities might administer the survey in a way that would generate low response rates, and students incapable of assessing how good they are at lacrosse might also incorrectly estimate how likely they or their friends are to intervene to stop a potential sexual assault. Strangely, however, senior Obama officials seem to take these methodological issues much less seriously in the campus climate context than they did in the sports context. Which side are Samuels et al. on?