Mark Blumenthal, the famed Mystery Pollster, has an excellent post on the problems with Steven Freedman’s much-cited paper “The Exit Poll Discrepancy” (PDF). The short version is this: Freedman doesn’t really know what he’s talking about – he has a PhD in organization studies from MIT’s Sloan School of Management and is not an expert in polling. And, like many real or wanna-be public intellectuals, he’s gone outside his field of expertise and basically made a fool of himself (this is Richard Posner’s thesis in Public Intellectuals: A Study of Decline). The exit polls were inaccurate, but the errors were most likely driven by relatively boring and mundane factors that Freedman wrongly dismisses as insufficient. The problem, of course, is that the combination of Freedman’s academic imprimatur and his insinuations that Bush’s victory might be illegitimate have further inflamed the tinfoil hat brigades, and the corrections will go largely unnoticed. Where is peer review when you need it?