Not part of the proper Freakonomics series
But, casting my eye over the Freakonomics blog, I see Levitt’s defence of his abortions ‘n’ crime piece. I have to say that there is quite a lot in it that looks very weaselly indeed. In particular this para which is Levitt’s main response to date:
3) Only when you make other changes to the specification that Foote and Goetz think are appropriate, do the results weaken further and in some cases disappear. The part of the paper that Foote and Goetz focus on is one that is incredibly demanding of the data. For those of you who are technically minded, our results survive if you include state*age interactions, year*age interactions, and state*year interactions. (We can include all these interactions because we have arrest data by state and single year of age.) Given how imperfect the abortion data are, I think most economists would be shocked that our results stand up to removing all of this variation, not that when you go even further in terms of demands on the data things get very weak.
Foote and Goetz’s main point in their paper is that Levitt & Donohue’s abortions ‘n’ crime study did not use per capita arrest data, and thus that the abortion-influenced cohorts had a lower number of arrests simply because they were smaller. This is not a technical criticism, it doesn’t have anything to do with interaction effects and it’s not “demanding on the data” (it’s demanding of good data on state-level population, which is a problem, but that’s a different issue)
I have to say that the actual substance of Foote and Goetz’s criticism is not at all well summarised by that paragraph. I always get a sinking feeling when I see someone trying to concentrate attention on tangential points.