Methods

Data come from two national random digit dial surveys conducted by Fact Finders, Inc of Del Mar, New York for the Harvard Injury Control Research Center in the spring of 1996 and the spring of 1999. The samples comprise, respectively, 1905 and 2521 adults living in the 50 US states.

Samples were stratified by state (proportional to population size according to the 1990 census). All households with a single telephone line, including those with unlisted numbers, had an equal probability of inclusion. Households without a telephone were excluded.

The surveys were designed to be representative of households and only one adult from each household was interviewed. Rather than interview whichever adult who happened to be home at the time of the call, interviewers alternately asked to speak with a man or with a woman living in the household. If there was no adult of the requested gender, the initial respondent was interviewed.

Once a telephone number was randomly selected, as many as 10 repeat phone calls were made until a final disposition was reached. In 1996, 27% of contacted households refused to participate; in 1999, 35% refused. Both surveys had a greater than 98% completion rate, comparable to other recent national firearm surveys.^{14, 15}

The key question in the 1996 survey asked: “Some states have recently changed their laws concerning gun carrying. If more people in your community begin to carry guns, will that make you feel more safe, the same, or less safe?”

The key questions on the 1999 survey asked, “Do you think regular citizens should be allowed to bring their guns into (a) restaurants, (b) college campuses, (c) sports stadium, (d) bars, (e) hospitals, and (f) government buildings. To conserve interviewer time, the first 1500 respondents were asked only about the first three locations, and the remaining 1021 adults were asked only about the last three locations.

In both surveys, eight independent variables were examined: (1) gender; (2) race (white versus non-white); (3) age (aged 40 or over); (4) income (greater or less than $35 000); (5) community (city, rural, or suburban); (6) children (under 18) living at home; (7) gun in the household; (8) region (South or not). In addition, the 1996 survey included three additional variables: (1) gun carrying (“In the past 30 days have you carried a gun on your person?”); (2) safety of the area (“If you are out alone in your neighborhood at night, do you feel very safe, somewhat safe, somewhat unsafe, or very unsafe?”); and (3) confidence in police (“Are you confident or not confident that your local police will respond quickly if you call them about a crime?”).

We used multiple logistic regression to analyze correlates of respondents' attitudes, holding other factors constant.

Results

In the 1996 survey, 59% of respondents said they would feel less safe as more people in their community began to carry guns and 12% said they would feel more safe (24% said they would feel the same and 5% didn't know or didn't respond) (table 1). The majority said they would feel less safe, independent of their age, race, gender, income, region, or whether they lived in a city, suburb, or rural area. The only subgroup for which a plurality would feel safer were those who had carried a gun on their person in the previous month.

In the bivariate and multivariate analyses presented in table 1, we eliminated more than 100 respondents because they were police officers, security guards, or because they did not respond to one or more of the questions dealing with the 11 independent variables. Of the remaining 1789 respondents, the groups significantly more likely than others to believe they would feel less safe were non-whites, urban dwellers, adults with children in the household, and people who do not own or carry guns.

In the 1999 survey, an overwhelming majority said “No,” they did not think regular citizens should be allowed to bring their guns into restaurants (88%), college campuses (94%), sports stadiums (94%), bars (93%), hospitals (91%), or government buildings (92%). Ten per cent or fewer of respondents said “Yes” to any of the locations (table 2). For all independent variables a majority of respondents did not think regular citizens should be allowed to carry their firearms into any of the six specified locations (not shown). Women and non-gun owners were especially likely to believe that regular citizens should not carry guns into the six listed venues. For example, 96% of women, compared with 80% of men, and 95% of non-gun owners compared to 74% of gun owners, believed that guns should not be brought into restaurants (not shown).

Discussion

We find that most Americans believe they would feel less safe as more people in their community carry firearms. Our results are similar to other survey results that find that Americans are more likely to feel less safe rather than more safe if more people in their community acquire firearms.^{16, 17}

Our results are also consistent with previous polls that find that the large majority of Americans are not favorably disposed to civilian gun carrying. For example, a 1991 CBS News/New York Times national telephone poll of 1430 adults asked “Do you think that when ordinary people carry weapons like guns or knives or mace they make the streets safer, or do you think carrying weapons creates more problems than it solves?” Sixty nine per cent answered “more problems,” 15% said “safer”, and 16% said either “don't know” or did not answer.

Each year from 1996 to 1999 the National Opinion Research Center (NORC) asked a national sample of 1200 adults, “Do laws allowing any adult to carry a concealed gun in public provided that they pass a criminal background check and gun safety course make you feel more safe or less safe?” In 1996, the year of our survey, 56% replied less safe, compared with 36% who said more safe. Since then the percentages have gone up and down. The NORC surveys also asked “Most states require a special license to allow people to carry a concealed firearm. Should licenses to carry concealed firearms be issued to any adult who has passed a criminal background check and a gun safety course, or only to people with a special need to carry a concealed gun, such as private detectives?” Each year 54%–60% wanted to restrict the licenses to people with a special need to carry; about 40% preferred the other extreme—permissive “shall issue” laws, which require licensing authorities to issue permits to any applicant who meets specified criteria (for example, is not a convicted felon).¹⁸

An April 1999 NBC News/Wall Street Journal poll of over 1000 adults nationwide asked “Do you approve or disapprove of the idea of passing new laws to make it easier for people to carry concealed weapons?” Seventy three per cent disapproved, while 22% approved.

Voting results appear consistent with the survey findings. In April 1999, Missouri held the first state referendum on gun carrying. (Missouri had initially banned the carrying of concealed weapons in 1875 when Jesse James was still at large.) Although the gun lobby spent some $3.8 million on the campaign, an amount far greater than their opponents, the measure went down to defeat. Pre-election polling found that women were even more opposed than men to the permissive gun carrying proposal.¹⁹

Gun carrying is often opposed for specific sites. In the 1999 NORC survey, 76% of respondents believed that “public places, such as stores, movie theaters, and restaurants” should be allowed to prohibit people from bringing guns onto their premises.¹⁸ During the 1999 Missouri referendum, professional sports teams, fearing for the safety of players, fans, and referees, strongly opposed the relaxation of the gun carrying laws in their state. Union chiefs for baseball and football players were similarly opposed. Don Fehr, executive director of the baseball players association, wrote in a letter to baseball commissioner Bud Selig that the proposition was “one of the most dangerous ideas we have had to confront”.²⁰

Colleges often have strict regulations concerning guns. In 1994, the Association for Student Judicial Affairs unanimously adopted a resolution urging colleges to support tough rules to keep guns off campuses.²¹ The Mormon church, although based in Utah, where most households own guns, says that guns, even legally concealed firearms, do not belong in houses of worship.²²

Permissive gun carrying laws have potential advantages. If carrying guns became more prevalent, more crimes might be thwarted and others not even attempted. On the other hand, permissive gun carrying laws may cause problems. Increased civilian gun carrying may cause some criminals to arm themselves, or to resort to violence more quickly when dealing with their victims. Moreover, increases in permitted gun carrying may make it more difficult for police to prevent illegal gun carrying.²³ Guns can make arguments more lethal. People who are tired, angry, drunk, or afraid, and who are not trained in dispute resolution, may use guns inappropriately. Police officers, who receive large amounts of training, still are often inadequately prepared to handle ambiguous but potentially dangerous situations. Heavy stress, confusion, and fear are inherent in most possible shooting situations. Heart rates skyrocket, and it is difficult to think clearly and to act deliberately. Even police officers make serious mistakes. Not surprisingly, evidence indicates that much self described self defense gun use by regular citizens in America may be illegal and socially undesirable.^{14, 24–26}

Our study focuses on people's beliefs and attitudes concerning gun carrying. A limitation of our study is that respondents were asked only a hypothetical question about their feelings of safety as more people begin to carry guns, and whether they believe “regular” citizens should be allowed to carry in six specific locations. We do not know why respondents said they would feel safer or less safe, nor why they do not believe “regular” Americans should be carrying guns into restaurants, bars, etc. Nor do we know how strongly respondents hold their opinions. As with all telephone surveys, ours is subject to sampling error (3% for each question due to chance alone when n=1000; under 2% when n=2500), and to systematic error²⁷ such as under-representing individuals who are institutionalized (for example, criminals) or without telephones (for example, people who are poor).²⁸ Finally, self report data are subject to potential inaccuracies due to social desirability responses, intentional distortions, or non-candid responses.²⁹ None the less, the size of the differences in our study suggest that eliminating all potential survey problems would not change our general results.

Implications for prevention

In the past 15 years, many state legislatures have made gun carrying laws more permissive, taking discretion as to who should be granted a gun carrying license away from the police. While people carry guns for safety reasons, our results indicate that most Americans believe that increased gun carrying by others makes them less safe rather than more safe and that concealed firearms should not be allowed into a large number of public venues. These findings suggest that existing legislation regarding gun carrying may not adequately take into account the feelings and preferences of the large majority of United States adults and that, at least on these grounds, such legislation should be revisited.