Every once in a while I sneak the odd book review into my blog, and this time the selection is Freakonomics: A Rogue Economist Explores the Hidden Side of Everything by Steven D. Levitt and Stephen J. Duber. This is not exactly a timely review; the book was published in 2005 and, as of 2007, it had sold over 3M copies.
As it turns out, Freaknomics is the first book that I’ve read cover-to-cover (so to speak) on my Kindle. While I’ve purchased a few Kindle books (e.g., a useless tax guide, a David Baldacci novel, and Against the Gods) and several Kindle French newspapers (you can score what’s usually tomorrow’s issue of Le Monde for $0.75), for whatever reason, I’d never before managed to get all the way through a work on my Kindle. It could be the device, it could be the books. Time and experience will tell.
Levitt describes himself as a rogue economist, but I think of him more as a rogue data junkie or data miner. The guy loves to pour through data and look for interesting and often non-obvious correlations, sometimes over very long periods of time. Among other issues, the book investigates questions such as:
- What is the effect of information asymmetry between a client and a Realtor?
- How can you discover cheating through mining large data volumes — either in schools or amongst Sumo wrestlers?
- Why do crack cocaine dealers tolerate such low pay and poor (i.e., often deadly) working conditions?
- What explains the relatively sharp drop-off in violent crime in the 1990s? What percent of the drop is explained by more police? What explains the rest? (And, boy, will you find the answer controversial.)
- Does “good parenting” matter? Do Mommy-and-me and other head-start programs actually increase test scores later in life? What about habits like daily oral reading?
- What is the sociological impact of a child’s name?
One of the bittersweet tales surrounding Levitt is that he and his wife lost a baby to Meningitis several years ago, prompting him to join a support group for parents who’d lost children. During his involvement with that group, he was shocked at the number of children who drowned in the family pool, prompting him to start up his data mining engine, eventually arriving at the following conclusion: if you have both a gun and a swimming pool, your child is 100x more likely to die from the pool than the gun.