You Digest Data With Your Gut

Listening to this week’s econtalk with Ed Leamer in which Russ and Ed discuss (among many other things – recommended) this article (weird, I first found this article. WTF?) about Tiger Woods:

Mr. Woods is such a dominating golfer that his presence in a tournament can make everyone else play significantly worse. Because his competitors expect him to win, they end up losing; success becomes a self-fulfilling prophecy.

The article then progresses to the ridiculous:

Ms. Brown argues that the superstar effect is not just relevant on the golf course. Instead, she suggests that the presence of superstars can be “de-motivating” in a wide variety of competitions, from the sales office to the law firm.

Ed and Russ appropriately cut this conclusion up, but I want to focus on their response to the former point. When presented with a counter-intuitive assertion, these two economics professors start talking about their private experiences and those of their friends to validate it. That’s not science, is it?

But that’s the problem that Ed Leamer talks about in his book and on a previous Roberts podcast. In social science, data can cast a new light on a situation, but if the only narrative you have is your own you’re not going to change your mind or learn anything. The likelihood is, of course, the weekend warrior experience is totally irrelevant when considering the psychological context of playing in a tournament with Woods and so, without a narrative (ie, a plausible setting and mechanism of causation), you have no empathy and so no ability to understand the conclusion.

Narrative without data is bullshit. Data without narrative is incomprehensible.