TIM LAMBERT HAS MORE on the seemingly interminable John Lott coding error question.
I'm inclined to agree with Mark Kleiman that the "con" is always in econometrics, and I'm not sufficiently knowledgeable to opine on the statistical questions. But perhaps Lott will respond on his site.
UPDATE: Lambert has more on Lott today. I am, as I've said before, not competent to assess the accuracy of Lott's work, though at this point I'd be quite reluctant to rely on it. He's wrong, however, to suggest that Lott's accuracy bears on the constitutional right to bear arms, though it may bear on the policy question of whether concealed weapons reduce the crime rate or not.
ANOTHER UPDATE: Kleiman emails to clarify a fuzzy memory of mine regarding something he had said earlier: "That why it's so hard to take the 'con' out of econometrics: there are always sample definition, coding, and model-specification decisions to be made, and more often than not they're big enough to matter. That's why judging econometric work, except for the occasional finding of an intra-ocular correlation (i.e., one that hits you right between the eyes), depends so much on knowledge of the character of the person doing it."
I thought I had remembered this post, but I wasn't able to find it until he sent me the link. Sorry.