A U.S. Private Firearm Ownership And Use Survey

With respect to data on defensive use in the survey results below, I note two main caveats. (And, survey data with well quantifiable systemic error can still be very useful.)

First, there is good evidence that the claimed 302,000 or so instances each year in which shots are fired in defensive uses of firearms is grossly overstated (especially in questions about whether it has ever been done) based upon other data that is "harder" than  survey data, although no shots fired cases are harder to corroborate. 

The number of justified civilian homicides each year 0.1%-0.5% of that number in a typical year, the number of people shot but not killed by civilians in self-defense each year is less than ten times that number, and the case that there are 285,000 warning shot or missed self-defense shots fired each year is extremely implausible given the ratio of warning and missed shots to shots causing injuries and deaths in civilian self-defense circumstances where good data on all of them (e.g. in neighborhoods with automated gunfire locators) is available. Defensive use of firearms in public and in the workplace, for example, is more easily corroborated and doesn't line up with magnitude of the survey responses.

Also, lots of uses of firearms characterized as "defensive uses" of firearms, when analyzed by people familiar with the applicable laws, are in fact illegal. Many constitute illegal menacing, or unjustified uses of deadly force. 

Notably "telling someone you have a gun" is counted as a large share of defense uses of firearms:

It ought to be possible to compare sales data to large magazine, assault rifle, handgun, rifle, and shotgun data to determine how accurate the survey results are in those domains.

The geographic ownership rates by state from the 2021 survey are as follows:

Of course, while statistical sample size sourced uncertainty is easy to estimate, sample selection bias is an issue in every kind of survey data and has gotten worse in recent years. Given the direction of the sample selection bias in election polling (even adjusting for representative samples), it could very well mean that the the bias is towards underestimates rather than overestimates, however. Centiment is not an "A-list" survey operation, especially for such a large survey, but that doesn't inherently mean very much a well done survey is useful no matter who does it.

Both bold and italic emphasis below is mine.

This report summarizes the findings of a national survey of firearms ownership and use conducted between February 17th and March 23rd, 2021 by the professional survey firm Centiment. This survey, which is part of a larger book project, aims to provide the most comprehensive assessment of firearms ownership and use patterns in America to date. This online survey was administered to a representative sample of approximately fifty-four thousand U.S. residents aged 18 and over, and it identified 16,708 gun owners who were, in turn, asked in-depth questions about their ownership and their use of firearms, including defensive uses of firearms.

Consistent with other recent survey research, the survey finds an overall rate of adult firearm ownership of 31.9%, suggesting that in excess of 81.4 million Americans aged 18 and over own firearms. The survey further finds that approximately a third of gun owners (31.1%) have used a firearm to defend themselves or their property, often on more than one occasion, and it estimates that guns are used defensively by firearms owners in approximately 1.67 million incidents per year. 

Handguns are the most common firearm employed for self-defense (used in 65.9% of defensive incidents), and in most defensive incidents (81.9%) no shot was fired. Approximately a quarter (25.2%) of defensive incidents occurred within the gun owner's home, and approximately half (53.9%) occurred outside their home, but on their property. About one out of ten (9.1%) defensive gun uses occurred in public, and about one out of thirty (3.2%) occurred at work.

A majority of gun owners (56.2%) indicate that they carry a handgun for self-defense in at least some circumstances, and about 35% of gun owners report carrying a handgun with some frequency. We estimate that approximately 20.7 million gun owners (26.3%) carry a handgun in public under a ``concealed carry'' regime; and 34.9% of gun owners report that there have been instances in which they had wanted to carry a handgun for self-defense, but local rules did not allow them to carry.

The average gun owner owns about 5 firearms, and handguns are the most common type of firearm owned. 48.0% of gun owners -- about 39 million individuals -- have owned magazines that hold over 10 rounds (up to 542 million such magazines in total), and 30.2% of gun owners -- about 24.6 million individuals -- have owned an AR-15 or similarly styled rifle (up to 44 million such rifles in total). 

Demographically, gun owners are diverse. 42.2% are female and 57.8% are male. Approximately 25.4% of Blacks own firearms, 28.3% of Hispanics own firearms, 19.4% of Asians own firearms, and 34.3% of Whites own firearms. 

In total, Americans own over 415 million firearms, consisting of approximately 171 million handguns, 146 million rifles, and 98 million shotguns.

William English, "2021 National Firearms Survey: Updated Analysis Including Types of Firearms Owned" SSRN (May 18, 2022).

This paper just includes the basically raw results of this particular survey without examining historical trends or comparing it to non-survey data.

Firearm ownership rates (ignoring multiple guns owned by a single person) have historically been lower in urban areas than rural areas and have gradually fallen over time as the population grows more urban and as hunting has grown less popular as a result.

Hat tip to Tom Bridgeland who provided the link to the paper in the comments.