The “link” between gun possession and gun assault

Twitter is a useful place to sample the arguments made by those in this country—and in other countries, often—who would deny us our gun rights.  One study that I’ve seen cited often comes from the American Journal of Public Health‘s November 2009, Vol. 99, No. 11 edition, entitled, “Investigating the Link Between Gun Possession and Gun Assault,” written by Charles C. Branas, et al.  Dr. Branas is a professor of epidemiology at the University of Pennsylvania.

The authors proposed to treat gun assaults as a public health matter and sought to determine whether being in possession of a gun made a person more at risk of being shot. To that end, they surveyed participants from Philadelphia, about half of whom had been shot while around half had not.  The conclusion was that a person with a gun is at least 4.46 times more likely to get shot in an assault than is someone who is unarmed.  Unsurprisingly, the Brady Center cited this study in its amicus brief in the Peruta v. San Diego case regarding carry license issuance.

The results sound alarming, especially to those of us who choose to carry firearms for self-defense, but a detailed reading of the study shows a number of significant flaws.

First, the authors chose Philadelphia as a representative city and claimed that a person has a chance of being assaulted at any time and in any location in that area.  Crime data from the City of Brotherly Love tells us otherwise, though.  The crime index for Philadelphia puts the city as safer than ten percent of U.S. cities.  I’m an English teacher, but that seems to mean that ninety percent of cities are safer than Philadelphia.  And the map of crime distribution given by shows a clear difference among the various neighborhoods with regard to one’s chances of suffering violent or property crimes.

The study also acknowledges that the control group—participants who were not shot—was much more unemployed than the general population.  This is a problem that studies of human behavior often face, since people who work for a living have better things to do than answer a bunch of questions for a slice of pizza and a few bucks.  But certainly, this admission shows that the participants in the control group are not representative of all of us, especially given the costs of guns and ammunition.

A key flaw in this study, though, is in the group of participants called the case group—namely, those who had been shot.  The following was found:

…. compared with control participants, shooting case participants were significantly more often Hispanic, more frequently working in high-risk occupations, less educated, and had a greater frequency of prior arrest.  At the time of shooting, case participants were also significantly more often involved with alcohol and drugs, outdoors, and closer to areas where more Blacks, Hispanics, and unemployed individuals resided.  Case participants were also more likely to be located in areas with less income and more illicit drug trafficking.

The racial disparities there are a complex issue, as are matters of employment and education, but questions of criminal association and activity aren’t hard to understand.  If you commit crimes or hang out with those who do, you’re more likely to suffer a crime committed against you.

Another point, related to the above, is the subject of whether the victim had the chance to resist the attack.  The authors state that in 2-sided situations—in other words, situations in which both parties knew each other and came knowingly to the fight—the participants’ chances of getting shot increased.  In other news, water is wet on a warm day.  I’ve discussed the concept of a fair fight elsewhere, but for today’s purpose it seems obvious to me that the less we attend a gunfight, the less shot we get.

One last admission, found near the end of the study, I find interesting.  The authors state that they did not control for any “prior or regular training with guns. . . .”  Indeed.  There are many possible responses to this admission, but the most pointed is to wonder if the authors believe the study of epidemiology can be done well without any training or practice.  Using a gun is a skill.  As Jeff Cooper told us, you’re no more armed because you own a gun than you’re a musician if you merely own a guitar.

But there’s a bigger flaw in all of this.  Rigorous, valid scientific study of human beings is fraught with peril.  There are ethical concerns to address before even beginning, and more than that, there is the need to make sure the people conducting the study have eliminated all possible other explanations for the gathered data.  We aren’t elementary particles or animals guided solely by instinct.  When studying the migration of birds or the combination of molecules, one is as good as any other, all things being equal.  Humans, by contrast, have huge variation in behaviors, in skills, and in life histories.  When studies treat us as commodities, rather than as individuals, they risk arriving at sweeping conclusions on the basis of scant evidence.  I am by no means a science-denier of any kind, but I’m also not willing to buy any old bafflegab offered to me on the basis of a pleasing conclusion without reading the details.

When it comes to studies that policy-makers wish to use in regulating our rights, always look under the hood.  That engine may sound powerful from outside, since three hamsters and a squirrel can raise quite a racket, but the argument all too often is going nowhere.

 The views and opinions expressed in this post are those of the author’s and do not necessarily reflect the position of