This year has not been a peaceful one for Harvard’s unrecognized final clubs. After months of intense administrative pressure to reform membership policies, a University-wide task force report in March took the clubs to task for “a culture often inimical to Harvard’s mission.” Citing data from a University-wide sexual misconduct survey conducted last year in tandem with the Association of American Universities, the report argued that single-gender final clubs significantly contribute to campus sexual assault.
“The qualitative and quantitative information before us is deeply troubling and requires a strong response from Harvard,” the task force wrote.
On Wednesday, though, a professional statistical analyst, commissioned by the 225-year old Porcellian Club, sharply criticized the Harvard sexual assault prevention task force’s interpretation of survey data and recommendations for action on final clubs.
The analysis, conducted by statistical firm Welch Consulting employee Jora B. Stixrud and sent to Dean of the College Rakesh Khurana Wednesday, focused on two figures central to the task force’s recommendations: 47 and “at least 15.”
Forty-seven represents the percent of female College seniors surveyed by the College “participating in the Final Clubs” who reported experiencing “nonconsensual sexual contact” during their undergraduate years, while more than 15 represents the percent of women reporting “nonconsensual penetration involving physical force” who said the incident occurred at a space used by a single-gender social organization that was not a fraternity or a sorority.
In a letter to Khurana accompanying Stixrud’s 14-page analysis, newly-instated graduate president of the Porcellian, David T. Lawrence ’67 wrote that the club commissioned the report because of what he called Harvard’s unwillingness to respond to “expressed concern about the statistical analyses” of the sexual assault prevention task force.
Lawrence wrote that he took issue with “repeated public statements by Harvard officials incorrectly implying that the 2015 AAU survey established statistically a direct connection between sexual assault and final clubs.”
“Harvard’s persistent use of misleading talking points has distracted the public and the student body from the undisputed fact established by the Final Report that the vast majority of sexual assaults occur on Harvard property under Harvard control,” Lawrence argued.
According to the AAU survey, 87 percent of on-campus “nonconsensual penetration involving physical force” among female students occurred in Harvard dormitories, which task force chair and former University Provost Steven E. Hyman wrote was “[c]onsistent with a residential campus such as Harvard College” in a letter published alongside the survey results in September.
In a recent letter to the editor of the Wall Street Journal, University Spokesperson Jeff Neal emphasized the importance of the 47 percent statistic, adding that “While the data shows there is no single cause of sexual assault, it is clear that the social dynamics fostered by single-gender social organizations warrant our sustained attention.”
Lawrence wrote that the Porcellian believes in the “central message” of the Task Force report “that sexual assault at Harvard is a very serious problem requiring immediate attention and effective action.” The Porcellian “has an important part to play in reducing sexual assault,” he added.
Concluding, Lawrence wrote that he would share the analysis and letter—both of which were obtained by The Crimson—with graduate leaders of all other final clubs. Another graduate member of the Porcellian said that neither Welch nor Stixrud had any previous affiliation with the club, and that Welch was paid a regular hourly rate.
Khurana, through College Spokesperson Rachael Dane, declined to comment for this story. Hyman also declined to comment.
Neal welcomed any scrutiny to the University’s sexual misconduct survey and the task force’s prevention recommendations.
“The final report of Harvard’s Task Force on the Prevention of Sexual Assault and the sexual conduct survey results speak for themselves,” Neal wrote in an email. “ Both are posted online and we welcome any interested party to review them in detail.”
‘CORRELATION DOES NOT IMPLY CAUSATION’
Welch’s analysis, titled “The AAU Sexual Assault Survey Data Cannot Substantiate Claims Regarding Harvard Final Clubs,” had three main criticisms of the 47 percent statistic.
First, Stixrud, who holds a Ph.D. in economics from the University of Chicago, argued that the task force report conflates correlation and causation, calling the distinction between the two “one of the first lessons taught in most introductory statistics courses.”
Stixrud called any claims of causality between the final clubs and sexual assault prevalence “particularly weak” in relation to the 47 percent statistic, because the survey “does not ask about the timing of the sexual contact vis-a-vis the timing of involvement in the Clubs.”
“A woman who received an unwanted kiss in her dorm room while a freshman would be counted as part of this figure if she later joined an all-female Final Club in her sophomore year,” Stixrud wrote.
Second, the analysis posited that even the wide disparity between the average rate of nonconsensual sexual contact among senior College women in different student groups and the 47 percent rate reported by women who “participated in clubs,” could be the result of imprecision in weighting the raw data to account for non-respondents.
Stixrud argued that women participating in various activities could further skew activity-specific statistics, as any one of a woman’s affiliated activities could hypothetically contribute to the risk of sexual assault.
“[T]he exact statistical tests cannot be performed without the underlying data, which Harvard has not disclosed… All of the correlations between membership or participation in these groups and nonconsensual sexual contact are most likely statistically indistinguishable,” she argued.
“In short,” Stixrud added, “There is no statistical evidence to support the singling out of Final Clubs as opposed to other student organizations.”
Third, Stixrud argued that the term “participation,” used by the survey to determine whether a woman was involved with final clubs, was “too nebulous to be meaningful.” A woman who participated in recruitment or joined an all-female club, a woman who attended parties at an all-female club, or a woman who attended events at an all-male club could all considered participants in final clubs, according to Stixrud.
In a footnote, the task force’s report specified that “we interpret ‘participate’ as meaning primarily non-member interaction with the all-male Final Clubs through social and other related events, as well as any assaults experienced by women who have joined one of the all-women Final Clubs.”
As a comparison, Stixrud added that the number of women self-reporting “participation” in “Campus Varsity Sports” was 165 and that, by Stixrud’s count, there were “about 80 or 90” female senior varsity athletes when the survey was administered, meaning some women who were not members of a varsity team could have been counted participants.
Allison L. Miller, assistant director of Athletic Communications, said that she thinks the number of female varsity athletes in the class of 2015 was likely much higher than Stixrud’s count, though she said she could not provide an exact number by press time.
“Regardless of whether the true number of female senior varsity athletes is actually 80, 100, or 150—‘participation’ in varsity athletics is open to interpretation by the students who responded to the survey,” Stixrud wrote in an email to The Crimson.
Stixrud’s other main criticism of the survey’s finding involved the percentage of sexual assault incidents reported at single-sex organizations that are not fraternities or sororities.
In the fall, Economics professor David I. Laibson ’88, a member of the University’s sexual assault task force who helped design the AAU survey, said he “would assume that most students who identify that space are thinking about final clubs.”
Stixrud disagreed, writing that “this assumption is problematic for multiple reasons.”
More than 15 percent of women reporting “nonconsensual penetration involving physical force” on University property indicated that the incident occurred in space used by a single-sex organization that was not a fraternity or a sorority. The corresponding statistic for nonconsensual penetration by force on off-campus university property used by a single-sex organization was too small and therefore “suppressed” for confidentiality reasons.
Stixrud also wrote that the statistic “cannot possibly tell us anything about what is going on in any of the individual clubs, nor does it begin to account for differences in access to non-members provided by the various Final Clubs.” The Porcellian Club, for instance, has a strict no-guest policy.
“In sum,” Stixrud concluded. “This survey does not contain any data that would allow an analyst to support the recommendations of the Task Force that pertain to Final Clubs.”
While Hyman declined comment for this story, in a statement earlier this week, responding to another critique, he wrote that task force recommendations “were based not only on the AAU sexual conduct survey, but also on the extensive qualitative data gathered at scores of meetings with Harvard undergraduates.”
Stixrud’s analysis focused solely on the publicly-available survey data.
“I have not reviewed the qualitative reports because they have not been made publicly available, but they would not change my conclusions about the survey evidence,” Stixrud wrote in an email.
Schuyler H. Daum ’12, a recent alumna of the all-female IC club, said the task force’s report seemed accurate to her based on her undergraduate years.
“The results of the campus climate survey are not surprising—based on my experience and based on the research,” Daum, who worked at the Office for Sexual Assault Prevention and Response as an undergraduate, said. “Studies show that women who are in sororities are more likely to be sexually assaulted in college—presumably because they are more likely to end up in alcohol-rich, single-sex environments. This same dynamic is playing out at Harvard for women who participate in final clubs.”
Daum said that “the conversation should be about the final clubs creating safer spaces and aligning themselves with 21st-century values by admitting women,” rather than an argument over statistics, which Daum called “frustrating.”
Stixrud’s report comes at a time of unprecedented scrutiny for Harvard’s single-gender final clubs. In September, University President Drew G. Faust publicly criticized the clubs, and even the Harvard Corporation, the University’s highest governing body, has been involved in discussions with Khurana on how best to oversee the centuries-old organizations.
In the fall, the traditionally all-male Spee and Fox clubs accepted women into their membership. In a letter explaining their decisions to graduate members, Fox undergraduates wrote that administrators had “forced [their] hand.”
The Porcellian’s analysis also comes at a time when concern over possible administrative sanctions for undergraduates in final clubs, particularly against all-male clubs, has driven club leaders to more pointedly criticize administrators. Then-graduate president of the Porcellian and the current graduate president of the Fly Club both made statements criticizing administrative pressure of final clubs, with both accusing top administrators of some ulterior motives in scrutinizing the clubs.
Khurana is expected to submit a plan to address task force recommendations, including those on final clubs, by the end of this academic year.