Freakonomics: A Rogue Economist Explores the Hidden Side of Everything by Steven D. Levitt
My rating: 4 of 5 stars
People like the bitch about Freakonomics because of the loosey-goosey, back of the hand, unscientific data analysis. As a statistician and data analyst, I suppose I should be more bothered by this than I am. The fact of the matter is, I actually think Freakonomics is great. They think of things in terms of hypotheses to be tested, the back up their findings with data, and if there's a perverse negative feedback loop that occurs where people are incentivized to do bad data analysis in order to get recognition by the freakonomics people, well, a) that's the sort of thing that's supposed to get caught in peer review, and b) it kind just proves the Freakonomics authors right (http://www.freakonomics.com/2011/11/22/beware-this-blog-apparently-causes-academic-fraud/).
This cartoon (source: xkcd.com, obviously) is about mythbusters, but it pretty much sums up how I feel about Freakonomics as well.
