Tuesday, December 8, 2009

Correlate This

So I listen to Glenn Beck for about 10 minutes once a week. That's the one day a week I drive to work and the drive takes five minutes. So actually I listen to about three minutes because that's how much actual content there is between the commercials. And that three minutes is puh-lenty for me!

The three minutes I caught today had him citing Super Freakonomics. Of couse it was taken completely out of context to support some other point that had nothing to do with the point that Levitt and Dubner were making in the book, but that's Beck and I expect nothing less. And when I write I don't care much to focus on Glenn Beck. Beck can be handled with a simple Media Matters or Jon Stewart link on Facebook. What I'd rather talk about is Levitt and Dubner and what they're actually trying to accomplish with Freakonomics, Super Freakonomics, and their ongoing blog. Because that's actually intelligent.

If you haven't read them yet, run—do not walk—to your nearest bookstore or type—fast as you can with expedited shipping—on Amazon or other to order both of these books.

This is not about liberal vs. conservative, right vs. left, or politics at all. These books are about thinking. How to stretch your brain to think about every day topics such as crime, the environment, abortion, terrorism, names, and on and on and on in a way where the data is actually applied to prove causation, not just correlation. Religious and social conservatives might be outraged by the data showing legalized abortion reduced crime (Freakonomics). And liberal environmentalists go insane at the environmental proposals raised in Super Freakonomics. Which should prove that data is not, and can never be, political at its heart.

Now, one of the points in Super Freakonomics is that we are not entirely rational beings. We all know this. But that doesn't mean we can't, particularly when we go crazy on emotion, ground ourselves in realizing that the data exists. The biggest issue in this context that we have to remember is that correlation does not equal causation. We may not like what causation actually means, but Know Thyself also means Know Thy Data and analyze it according to causation. Correlation can be meaningful in pointing the way, but it still does not equal CAUSATION.

Read the books. Learn to think in a new and mind-stretching way. It won't fit into easily digestible bits, but it really is interesting and a hell of a lot more interesting conversation at cocktail parties.

2 comments:

Gastronommy said...

Dubner was on Beck's show. You can listen to the whole interview at glennbeck.com. You might be surprised.

Gypsy Kaz said...

Even if it is a decent interview (and I'll try to give it a listen though Beck makes me vomit in my mouth after about 3 minutes), Beck is so often guilty of the very point I was trying to make. He uses correlation to create a conclusion not at all based on actual data and context. He's the very antithesis of what Dubner and Levitt are trying to do.