If Not Data, What?
My colleague Russ Roberts writes:
Over the years, I have become increasingly skeptical of the power of statistical techniques to measure causation in complex systems. … I happen to believe that concealed handguns do deter crime and allowing concealed handguns is a good thing. And you can claim that the evidence that shows I’m right is "good" statistical analysis. The other side disagrees. They claim it’s "bad" statistical analysis. Who’s right? I have no idea. But what’s clear to me is that my belief in the virtues of allowing concealed hand guns has little to do with the empirical evidence. And I would argue that the opponents are really in the same boat. They just don’t like guns and they’ve dressed up their prejudices in fancy statistical analysis. …
If Russ relies little on data to draw his conclusions, then on what does he rely? Perhaps he relies on theoretical arguments. But can’t we say the same thing about theory, that we mainly just search for theory arguments to support preconceived conclusions? If so, what is left, if we rely on neither data nor theory?
Try saying this out loud: "Neither the data nor theory I’ve come across much explain why I believe this conclusion, relative to my random whim, inherited personality, and early culture and indoctrination, and I have no good reasons to think these are much correlated with truth." That does not seem a conclusion worth retaining. If this is really your situation, you should move to a nearly intermediate position of uncertainty. Either you should believe that truth-correlated data or theory has substantially influenced your belief, or you should retain only a very weak belief.
HT to Arnold Kling.
Peter Boettke has a reply:http://austrianeconomists.t...
Dear Rosser:
1) If you look across Western Europe or Europe as a whole, there is not the relationship that you imply. Switzerland has a much higher gun ownership rate than the US, but it has one of the lowest murder rates in Europe. Norway and Finland also have gun ownership rates that are just a little less than that in the US. If you read either MGLC or The Bias Against Guns, I have extensive discussions on the problem with purely cross sectional data. Let me note that the countries that currently have low murder rates in Europe tended to have even lower murder rates before gun control. England is a good example. It had no gun control prior to 1920 and yet London, a city of millions of people, had a total of two gun murders in 1900. It had five armed robberies. After stricter gun regulations in the 1950s and 1997, the UK say increases in murder and violent crime. You really need panel data to answer this question properly.
2) If you look around the US, murders are extremely geographically concentrated. 50 percent of US counties have zero murders in any given year. 25 percent have one murder. just over 3 percent of counties account for over 70 percent of all murders. Those 3 percent also happen to be the lowest gun ownership rate counties in the US. Again, though, I wouldn't put a lot of weight on purely cross sectional data.
3) If you look across all countries for which the data is available and control for income, there actually appears to be more crime where there are stricter gun regulations. The top ten countries in terms of murder have either complete or virtually complete gun bans. Even in the past during the 1970s and 1980s in the former USSR, despite the totalitarian state, they had a murder rate much higher than ours in the US. Again, however, panel data is necessary. I have just tried to frame this discussion in terms of the cross sectional data that you reference.
If you are looking for a detailed discussion of these points, please see The Bias Against Guns.