Steven Levitt, Freakonomics
I'm part way through Levitt's new book "Freakonomics." It is a fun read, and I recommend it to all, particularly those, such as my friend Chris Russo, that think economics is "just supply and demand, dude." Still, I find the book frustratingly pithy. I know it is intended for a general, non-economist, audience, but it would be a better book were it to probe deaper into the nuts and bolts Levitt's research. In particular, I want to know how Levitt chooses a research topic, and how he develops his extraordinarily clever research methodologies. I also want a discussion that explains to the lay reader why Levitt's methods, from an applied economics perspective, are so clever.
On a positive note, the book offers a nice overview of Levitt's work, which is really fascinating stuff. See this post and this post for some tidbits on his work that are discussed in the book.
I think Dad, in particular, will enjoy the following bit, which deals with a fundamental difficulty in emperical research, that is, getting good data. The renowned sociologist William Julius Wilson assigned a graduate student to administer a multiple-choice survey to residents of Chicago’s poorest neighborhoods. The first question was “How do you feel about being black and poor?” and the five possible answers were “very bad,” “bad,” “neither bad nor good,” “somewhat good,” and “good”. After a day of trying to get answers to this question, the graduate student realized a sixth answer was needed: “fuck you.” Nice.
On a positive note, the book offers a nice overview of Levitt's work, which is really fascinating stuff. See this post and this post for some tidbits on his work that are discussed in the book.
I think Dad, in particular, will enjoy the following bit, which deals with a fundamental difficulty in emperical research, that is, getting good data. The renowned sociologist William Julius Wilson assigned a graduate student to administer a multiple-choice survey to residents of Chicago’s poorest neighborhoods. The first question was “How do you feel about being black and poor?” and the five possible answers were “very bad,” “bad,” “neither bad nor good,” “somewhat good,” and “good”. After a day of trying to get answers to this question, the graduate student realized a sixth answer was needed: “fuck you.” Nice.
posted by david burnstein at 11:22 AM
2 Comments:
My article "Pre-emptive Executions: Economist Steven Levitt contends that abortion reduces crime rates; the numbers tell a different story" in the May 9th issue of The American Conservative factually debunks the theory promoted last week in the Wall Street Journal and the New York Times. It is now available online for free at:
http://www.amconmag.com/2005_05_09/feature.html
"Freakonomics" had a lot of journalistic stuffing that irritated me. But I understood the authors weren't writing for people with nerdish interests like me. If I wanted to know whether Levitt used multivariate regression or some better statistical procedure, whether he controlled for particular factors I thought might be confounds, etc. I recognized I'd have to go to the original research report. Likewise with Sailer's analysis, you can't tell if his argument holds without him telling you what he did and where he published it -- the magazine piece is about as informative as "Freakonomics" (but meaner and more polemical in tone).
Post a Comment
<< Home