In Philadelphia, researchers at the University of Pennsylvania find, possessing a gun is strongly associated with getting shot. Since "guns did not protect those who possessed them," they conclude, "people should rethink their possession of guns." This is like noting that possessing a parachute is strongly associated with being injured while jumping from a plane, then concluding that skydivers would be better off unencumbered by safety equipment designed to slow their descent. "Can this study possibly be as stupid as it sounds?" asks Stewart Baker at Skating on Stilts. Having shelled out $30 for the privilege of reading the entire article, which appears in the November American Journal of Public Health, I can confirm that the answer is yes.
The authors, led by epidemiologist Charles C. Branas, paired 677 randomly chosen gun assault cases with "population-based control participants" who were contacted by phone shortly after the attacks and matched for age group, gender, and race. They found that "people with a gun were 4.5 times more likely to be shot in an assault than those not possessing a gun." Branas et al. suggest several possible explanations for this association:
A gun may falsely empower its possessor to overreact, instigating and losing otherwise tractable conflicts with similarly armed persons. Along the same lines, individuals who are in possession of a gun may increase their risk of gun assault by entering dangerous environments that they would have normally avoided. Alternatively, an individual may bring a gun to an otherwise gun-free conflict only to have that gun wrested away and turned on them.
The one explanation the researchers don't mention is the one that will occur first to defenders of the right to armed self-defense: Maybe people who anticipate violent confrontations—such as drug dealers, frequently robbed bodega owners, and women with angry ex-boyfriends—are especially likely to possess guns, just as people who jump out of airplanes are especially likely to possess parachutes. The closest Branas et al. come to acknowledging that tendency is their admission, toward the end of the article, that they "did not account for the potential of reverse causation between gun possession and gun assault"—that is, the possibility that a high risk of being shot "causes" gun ownership, as opposed to the other way around. While the reseachers took into account a few confounding variables related to this tendency (including having an arrest record, living in a rough neighborhood, and having a high-risk occupation), they cannot possibly have considered all the factors that might make people more prone to violent attack and therefore more likely to have a gun as a defense against that hazard. To take just one example, not every criminal has an arrest record. Yet it seems fair to assume that criminals in Philadelphia are a) more likely than noncriminals to be armed and b) more likely than noncriminals to be shot. That does not mean having a gun increases their chance of being shot. Certainly they believe (as police officers do) that having a gun makes them safer than they otherwise would be. Nothing in this study contradicts that belief.