(Via Glenn Reynolds:) Clayton Cramer writes on the problems with Michael Bellesiles' research in Arming America; he concludes that a lack of critical analysis by historians (including poor quantitative reasoning skills) and political diversity within the discipline allowed Bellesiles' work to pass largely uncriticized. While his discussion largely centers around history, there are lessons for other disciplines — including political science.
I find political science to be a more politically diverse discipline than history (and most of the humanities and social sciences, with the exception of economics), perhaps due in part to the strong influence of economists on the quantitative part of the discipline, although the political left is predominant (the “right” of the discipline is mostly libertarian and neo-conservative; I have yet to meet a paleoconservative political scientist). However, there has been a backlash in the form of the “perestroika movement” over the past two years; for a lighthearted look, see “Some Thoughts on Perestroika on Political Science”. (For the record, I'm an empiricist who mostly does quantitative work.)