18 May 2006
Harvard School of Public Health doctoral candidate Janet Rosenbaum has been in the news lately, following the publication of her study of virginity pledges in the American Journal of Public Health, as well as her recent IQSS seminar. (Full disclosure: Janet is a friend of mine. I'll address her as Ms. Rosenbaum for this entry.) Since it's certainly a hot topic, it's no surprise how much attention her findings have received; first, the big news agencies picked it up, then the blogosphere took their shift - mainly over the "controversy" resulting from the study. (See pandagon.net for an example.)
But I think the more relevant part of the whole debate is the point Ms. Rosenbaum was trying to make about surveys and self-reporting: we use these data to make broad, sweeping conclusions on social phenomena, and while they are the best we have, they aren't up to the best standard we could achieve.