Saturday, September 24, 2016

An Unacceptable Question

Reproduced below are the introductory paragraph and one of the bullet items from Harvard's "Guide to Unacceptable Interview Questions." (See this FAS web page for top level link.)

Why would information that may not be asked of a faculty candidate disqualify a student from team captaincies, Rhodes nominations, and leadership of student organizations?

That's not a rhetorical question. Why is asking about a candidate's club memberships prohibited? Perhaps it is Harvard policy not to discriminate against candidates on the basis of organizations they join--in which case our motion does nothing more than afford students the same protection. Or Harvard is bound by some Massachusetts or federal law which codifies the same principle. If anyone knows the rationale for the prohibition, please do share. In any case, this does seem to bear on the hypothetical I posed to the Faculty Council, about how the faculty might feel if the policy applied to them.
It is essential for all members of a search committee to be aware of these guidelines and follow them in both spirit and letter. Avoid any direct or indirect questions that touch on material that may not be asked. This information about an applicant should never be discussed with regard to his or her candidacy for a position. 
Inquiry into an applicant’s membership in non-professional organizations (e.g., clubs, lodges, etc.)  


  1. Asking about club membership is prohibited because it enables employers to discriminate without asking directly about any prohibited topics. It's a workaround. Let's say candidate X is a member of an club that excludes Jews. By finding out whether X is a member of that club, the employer has found out that he's not Jewish without directly asking. There's an EEOC guideline about this specifically:

  2. Perhaps the inconsistency can be explained by the following logic:

    Employers have a long history of finding surreptitious ways to illegally discriminate against those in protected classes, and so rather than trust them to determine which information to use and not to use in making an employment decision, the best policy is to keep it out of their hands entirely. (This would seem to protect well-meaning, non-discriminatory employers too, because when it comes time to defend a lawsuit, it's much easier to show that they didn't discriminate because they didn't have the information in the first place.)

    In the college context, however, the logic might (I say might because I'm speculating on what the thought process would be) run as follows: as a policy matter, students in attendance should be protected from being the targets of discrimination and various forms or violence (including sexual violence). To foster the right environment, then, it makes sense to exclude those who engage in those kinds of behaviors.

    There is a measurement problem here: Since the tendency of individuals to engage in those kinds of behaviors can only be DIRECTLY measured after the fact (at which point it is too late), one might try to do this pre-emptively by using some proxy such as membership in a club that endorses the ills sought to be avoided (the link would be that if one holds membership in a clubs that excludes Jews, by voluntarily being a member, one endorses the views espoused by the club).

    The difference between the "good" use - trying to save students from harm, and the "bad" use - trying to keep otherwise qualified people out of a job for race/sex-based reasons that have nothing to do with their ability to do the job - critically depends on the ability of those applying the policy to make the link between the proxy and the behavior sought to be avoided.

    I don't think there is a reliable way to do this.