The Volokh Conspiracy
Mostly law professors | Sometimes contrarian | Often libertarian | Always independent
There's a new survey aimed at determining patterns of gun ownership and defensive gun use, from Prof. William English at the Georgetown University. It's much larger than most other such surveys, with over 54,000 adult American respondents, of whom over 16,700 personally owned guns.
It measures some things that others have already measured—for instance, its estimate of the adult firearms ownership rate, 32%, is on par with other surveys. (Some other surveys also try to estimate what fraction of households contains at least one gun; this one didn't, though in response to my e-mail Prof. English wrote, "the individual gun ownership estimates are very much in line with other recent surveys that asked about ownership at the individual level; and some of these also asked about household ownership (Pew, Gallup), generally finding household rates in the 42-44% range.") But the survey also gives us some data that has rarely been gathered before, including (items reordered):
- "Overall, Americans own in excess of 415 million firearms, consisting of approximately 171 million handguns, 146 million rifles, and 98 million shotguns."
- "30.2% of gun owners, about 24.6 million people, have owned an AR-15 or similarly
styled rifle, and up to 44 million such rifles have been owned."
- "48.0% of gun owners, about 39 million people, have owned magazines that hold over 10 rounds, and up to 542 million such magazines have been owned."
There's more interesting stuff in the survey, which I hope to report on in later posts. One general note: Surveys of course have various limitations, including that respondents might misreport information (deliberately or inadvertently), yielding either overcounts or undercounts. Still, for questions such as this, a well-designed survey is likely the best source of information. There is no centralized federal registry of gun owners, and few such registration requirements in the states; plus any such registry would likely be highly incomplete. Gun manufacturing, export, and import data can be a useful way to infer the total number of guns, but it has its own limitations. Perhaps because of this, surveys have long been a standard, though admittedly imperfect, way of estimating gun ownership rates and the like.