This is a tour de force of reporting on empirical research in an area that could hardly be of greater social importance. Drum deserves 100x more attention for this than he got for breaking the Romney 47% video in September.

Concidentally via http://election.princeton.edu/2012/12/22/scientific-american... from a couple weeks ago, here's a link to Reyes' paper on the lead/crime theory:

http://www3.amherst.edu/~jwreyes/papers/LeadCrimeBEJEAP.pdf

And here's a striking critique of the Freakonomics abortion/crime theory, which I suppose most people here have heard of:

http://www.economist.com/node/5246700?story_id=5246700

Someone inspected Donohue and Levitt's code and found a bug that meant they hadn't controlled for what they claimed they had. "Fixing that error reduces the effect of abortion on arrests by about half, using the original data, and two-thirds using updated numbers."