But weather, economics, and education? I’m skeptical that you can just parachute into those fields and add a lot of value. They’re far more complex, are already heavily populated with sophisticated statistical modeling, and generally require some serious subject matter expertise in addition to raw number-crunching skill.
It’s possible that I’m just overreacting to a brief throwaway mention in the Politico piece. If all Silver is trying to do is improve on mainstream news reporting of number-heavy topics, that shouldn’t be too hard. Still, I’d hate to see the basic 538 model get naively overextended into anything that has lots of numbers attached to it. It’s one thing when bloggers (like me) throw up simple wonk-lite analyses of complex topics. After all, no one really takes us seriously as experts. But the 538 brand is all about expertise. It’s inherent in everything that appears there. I hope Silver is careful about what he takes on as he extends his brand.
It’s a tricky problem, statistically. The issue is that while gun ownership rates could plausibly be a cause of fatal crimes and accidents, it can also be a reaction to it, i.e. people purchase guns because they feel unsafe.
I’m not saying that the issue is intrinsically inscrutable. But it’s something that more requires a PhD-thesis-level treatment than a blog post to really add much insight, I think.
You were probably hoping for a clear, concise answer. The problem is that the issue of gun control really isn’t as cut-and-dry as many on both sides of the debate make it out to be.