Since I happened to mention the famous Case/Deaton mortality study in the previous post, here's the latest from Andrew Gelman. As you may recall, Case and Deaton concluded that mortality among middle-aged whites from suicide, alcohol, and drug poisoning had skyrocketed over the past two decades. This set pundits afire with theories about what was going on, but Gelman has done some age adjustment to the cohorts that Case and Deaton used, and then broken up the data by gender, and then by geographic area. Here's what he gets:
After 2005, there's no effect on middle-aged men at all. It's all women. And if you break it down further, nearly the entire effect is concentrated among women in the South. But why? Gelman punts:
I don’t have any explanations for this. As I told a reporter the other day, I believe in the division of labor: I try to figure out what’s happening, and I’ll let other people explain why.
I think that's wise. For one thing, if you slice the data in a different way, you might get a different result. What's more, as I've mentioned several times, the increased mortality affects the young too, not just the middle aged. So if you spun some brilliant theories about why middle-aged whites are so damn depressed these days, you might want to rethink things. Your new theory needs to explain why the young and the middle-aged are dying in greater numbers, and you also need to explain why it's affecting primarily women in the South. Good luck.