Freakonomics & Correlation/Causation

After the extended media blitz in the past few weeks, along with the recommendations from fellow bloggers like Seth Levine, I decided to pick up a copy of Levitt & Dubner’s “Freakonomics: A Rogue Economist Explores the Hidden Side of Everything.”

While I am almost certain that this book is going to become the over-hyped over-fashionable cocktail-conversation book-du-jour (think: Gladwell’s Tipping Point), it is well worth the read. Seth summarizes,

“The basic idea of Freakonomics is to use statistical analysis to explore relationships and answer some pretty interesting questions about our world (are swimming pools more dangerous than guns; why do drug dealers live with their mothers; how can we tell if sumo wrestlers cheat; etc).”

I was happy to read that Levitt & Dubner directly address (more articulately than I) the relationship between correlation and causation that I passionate wrote about in my post, “My Pet Peeve.”

Page 163 reads,

“A regression analysis can demonstrate correlation, but it doesn’t prove cause. After all, there are several ways in which two variables can be correlated. X can cause Y; Y can cause X; or it may be that some other factor is causing both X and Y. A regression alone can’t tell you whether it snows because it’s cold, whether it’s cold because it snows, or if the two just happen to go together.”

Of course, I would take issue with many of their arguments and conclusions. For example, they try to dispel the conventional wisdom that flying is safer that driving, “The per-hour death rate of driving versus flying is about equal. The two contraptions are equally likely to lead to death.” I ask: isn’t the more relevant metric the per-mile death rate? Afterall, there’s a reason that people fly – to get there faster.

But more importantly, one of their basic arguments is that so-called “experts” in a field (like real estate agents) are often incented to not necessarily present the full truth. And these “expert” authors here run into the same quandary. With strong financial motivations to sell more copies of the book, Levitt & Dubner are compelled to make wild claims, which they do: the decline in the national crime rate in the nineties was a consequence of Roe vs. Wade; parents’ actions have little effect on the outcome of a child. There potentially may be some evidence to support these extremely controversial claims, but we must remember that the motivation of these authors comes down to the same principals which they themselves argue – economics.

David Beisel

David Beisel is a co-founder and Partner at NextView Ventures. He has been focused on early stage Internet startups his entire career, both as an entrepreneur and venture capitalist. As an investor in the digital media space, David was most recently a Vice President at Venrock and previously a Principal at Masthead Venture Partners. Prior to becoming a venture capitalist, David co-founded Sombasa Media, an e-mail marketing company best known for its flagship product BargainDog. Sombasa was successfully acquired by where David served as Vice President of Marketing. David holds an MBA from the Stanford Graduate School of Business and an AB in Economics, magna cum laude and Phi Beta Kappa, from Duke University. He also founded and leads the Boston Innovators Group, an organization which holds quarterly entrepreneur events drawing a thousand attendees.