Survey-Based Estimates

Although production-based estimates indicate a 25 percent increase in

firearms availability since 1980, survey-based estimates indicate an 11 to

33 percent decrease in households reporting ownership. Three often-used

surveys are the General Social Survey (GSS), the Gallup Poll, and the Harris

Poll.5 According to these surveys, the percentage of respondents reporting

4Production-based data have limitations in that they account for neither additions to the

stock from illegal or other uncounted means nor losses from seized, lost, or nonworking

firearms. These data also exclude firearms manufactured or exported for the military but

include firearms purchased by domestic law enforcement agencies.

5Each survey asks a similar question about gun ownership. Gallup asks “Do you have a gun

in your home?” Harris asks “Do you happen to have in your home or garage any guns or

revolvers?” and the GSS asks “Do you happen to have in your home (or garage) any guns or

revolvers?”

TABLE 3-2 Estimated Number and Per Capita Ownership (rate per

1,000) of Firearms in the United States, 1950 to 1999

Firearms per Handguns per

Year Total Firearms Handguns 1,000 Persons 1,000 Persons

1950 57,902,081 14,083,195 381.3 93.5

1960 77,501,065 18,951,219 430.6 105.4

1970 111,917,733 31,244,813 548.7 153.2

1980 167,681,587 51,707,269 737.9 227.5

1990 212,823,547 72,499,181 853.3 290.7

1999 258,322,465 93,742,357 925.8 336.0

SOURCES: Data for 1950 to 1990 are from Kleck (1997: Table 3.1). The 1999 estimate was

derived by adding the annual net increase in the stock of total firearms and handguns (manufactures

+ imports – exports) to the 1990 estimate using data from U.S. Bureau of Alcohol,

Tobacco, and Firearms (2002: Exhibits 1, 2, and 3).

that they have a firearm in the home has been declining since the late 1950s.

While the estimates vary from year to year, all three surveys indicate a

decline in the percentage of households possessing firearms. From 1980,

when the percentage of households owning a firearm was between 45 and

48 percent, ownership has decreased by 5 to 16 percentage points to a

prevalence of 30 to 43 percent. In discussion with the committee, Cook has

suggested that the decline in ownership per household while individual

ownership remains constant may be due to the increase in female-headed

households during this period. Despite these overall reductions in household

ownership, the relative distribution of firearm ownership across attributes

of gender, race, age, education, income, and region has been remarkably

consistent over time (Maguire and Pastore, 2002: Table 2.70).

Of households owning a firearm, between 59 and 62 percent reported

owning a handgun (Maguire and Pastore, 2002: Tables 2.69, 2.71, and

2.72). All three surveys indicate that gun owners are more likely to be male,

white, and middle-aged or older. Furthermore, gun ownership was higher

among those who live in the South, had less education than a college

degree, and had a higher than average income. Among respondents reporting

household gun ownership, the percentage of blacks reporting handgun

ownership was 6 to 9 percent higher than for whites, and the percentage of

blacks reporting long gun ownership was 11 to 29 percent lower than for

whites (Maguire and Pastore, 2002: Tables 2.71 and 2.72).

Aggregation of Individual Survey Responses

Recent research has aggregated the individual survey responses about

firearms ownership across U.S. communities (Baumer et al., 2002; Rosenfeld

et al., 2001). The GSS is based on a national area probability sample

composed of 100 primary sampling units (PSUs) (in the 1990 sampling

frame) designed to represent the population of people age 18 and older in

the United States. Each PSU is a “self-representing” geographic unit, in the

sense that the respondents are representative of the PSU adult population.

Aggregating the individual survey responses to the PSU level permits

comparisons of the aggregated items, including firearms ownership, across

a representative sample of U.S. geographic areas. Figure 3-1 shows the

geographic distribution of household firearm ownership for the 100 PSUs

in the 1990 GSS sampling frame, covering the period 1993 to 1998.

The figure shows substantial variability in firearm ownership in the

United States. The prevalence of household ownership varies from roughly

10 to 80 percent. Most of the PSUs cluster around the mean ownership level

of 43 percent, with fewer PSUs located near the extremes of the distribution.