It's terrifically shoddy statistical work. You'd get dinged for this in a college class. But it's in a book written by a celebrated economist and a leading journalist. Moreover, the topic isn't whether people prefer chocolate or vanilla, but whether people should drive drunk. It is shoddy statistical work, in other words, that allows people to conclude that respected authorities believe it is safer for them to drive home drunk than walk home drunk. It's shoddy statistical work that could literally kill somebody. That makes it more than bad statistics. It makes it irresponsible.
But hey, it makes for a fun and unexpected opener.
Levitt teaches at my school and I can't say as I've ever heard anything good about him. His reputation is that he's full of himself, a bit of a jackass, and--much, much worse--a bad teacher. You'd be hard-pressed to find a big-name prof at this school, especially in econ, who isn't a bit full of himself and a bit of a jackass, but that's what we expect from big-name profs here. It's all part of the schtick. It works for them because they're legitimately great scholars and effective, enjoyable teachers. Copping the attitude and not delivering the pedagogy is really just not okay.
It's pretty typical economist fare to go for the cheap thrill or the controversy rather than the staid-but-valuable contribution to the field, and I'm not saying Levitt is obliged to aim for a Nobel rather than NYT-blog fame--but your cool stories have to at least make sense if it's going to be a good kind of pop culture fame. Ezra's take was satisfying to read as a longtime anti-fan of Levitt; God knows every time I've ever taken the time to read his blog entries I've finished them angry, sardonic, and unimpressed.