How Often Do People Defend Themselves With A Gun?

Self Defense Guns Thumbnail Updated
Katherine Nagasawa / WBEZ
Self Defense Guns Thumbnail Updated
Katherine Nagasawa / WBEZ

How Often Do People Defend Themselves With A Gun?

WBEZ brings you fact-based news and information. Sign up for our newsletters to stay up to date on the stories that matter.

This week’s question asker, Rady, wrote in because she and a pal have this recurring argument it’s about guns.

“I have this friend Mike, and in the last couple of years he has become a gun person,” she says.

Rady is pretty uncomfortable with her friend owning guns.

“He owns them and shoots them, and basically I am worried that he’s going to [accidentally]  shoot himself or his girlfriend, who is my friend,” Rady says.

Mike is a store owner, and Rady says that’s one of the reasons he thinks having a gun is a smart move.

“His argument,” Rady says, “is ‘[that by owning a gun], I am making it more likely that I am going to protect my home and protect my store from thieves.”

So Rady asked Curious City:

How often do people really defend themselves with a gun?

Her question is one that lies at the heart of the American gun debate. And folks on both sides — those who want more gun control and those who oppose tighter gun regulation insist they know the answers. But the truth is: Rady and Mike are both a little right and a little wrong with a lot of squishiness in between. That’s because the data on “defensive gun use” relies heavily on self-reported recollections that leave a lot of room for interpretation and debate. We’ll get to that debate below, but we’ll also try to answer Rady’s other question: “Will that gun really make Mike safer?”


How often do people defend themselves with guns?

The short answer is that there’s no general agreement. Why? Because it’s a tough thing to measure, according to Daniel Webster, who directs the John Hopkins Center for Gun Policy and Research.

“The sad truth is that it is almost impossible to measure self-defensive gun use,” Webster says. “The reason is that what is self-defense is a very subjective thing.”

That subjectivity is illustrated in the decades-long debate over two influential sets of numbers that attempt to quantify how often guns are used for self-defense. Two prominent researchers, Gary Kleck, a professor emeritus at Florida State University, and David Hemenway, a professor of public health at Harvard University, have argued the issue for decades. They’ve debated almost everything, including how researchers should ask questions, how many people to survey, what people might consider self-defense, and even whether people will actually remember defending themselves with a gun.

So, as you might imagine, when they surveyed Americans about guns and self-defense, they came up with very different findings.

Kleck arrived at an estimate of about 2.5 million defensive gun uses each year in the U.S. His researchers called about 5,000 people and asked them if they’d defended themselves with a gun in the last year, along with follow-up questions. Sixty-six respondents (or 1.3 percent) said yes. And from there, Kleck’s team extrapolated for the entire adult population of the U.S. and arrived at their number, more than 2 million.

This number has been cited by the U.S. Supreme Court and is often used by gun advocates to demonstrate how often guns are used by people to defend themselves against “bad guys.”

Hemenway, on the other hand, got a much smaller number than Kleck, about 80,00 times a year. He takes his data from something called the National Crime Victimization Survey (also known as the NCVS), which is derived from U.S. Census Bureau interviews. The NCVS polls 80,000 people and then calls them again six and 12 months later to make sure they are only reporting something that happened in the past year.

NCVS researchers asked the questions differently than Kleck’s team. Instead of asking directly about defensive gun uses, they walked respondents through a series of more than 40 questions about their experiences with crime in the last year and about what happened. For example, they ask questions like, “During the last 6 months (other than any incidents already mentioned), did you call the police to report something that happened to you which you thought was a crime?” But they never ask, “Did you defend yourself with a gun in the last year?”

Kleck says this lack of direct questioning is a problem.

“No respondent is ever specifically asked about defensive gun use,” he says. “Just a generic question about what they might have done for self-protection while an incident is occurring.” 

Consequently, Kleck thinks the NCVS ends up undercounting defensive gun use from respondents who, he thinks, may have forgotten the experience or may not have considered themselves “victims” of a crime. Other people, he argues, might have been wary of telling a researcher about their experience. 

“People don’t like to admit to strangers that they have done something that is controversial or possibly illegal,” he says. “For example, if you were illegally carrying a gun but engaged in a lawful act of self- protection, you have something to conceal. Or just the fact that it’s a controversial act that many people would disapprove of on moral grounds.”

Hemenway, however, thinks Kleck’s methods overcount due to exaggeration, miscalculating the time period, or misunderstanding what really constitutes self-defense. In a separate study, Hemenway asked about 4,000 people directly in the same way Kleck had if they’d used a gun to defend themselves in the last five years. And he says he did get a lot of reports of defensive gun use.

“But then when you ask them what happened, [they said] things like, ‘I was talking to my neighbor and then he threw a beer and I got my gun,’” Hemenway recalls. “And that’s not self-defense. That’s escalating violence.”

Hemenway worries that when you ask someone directly about an experience with a relatively rare event, you can end up with deceptively high numbers made worse through extrapolation.

Kleck, however, pushes back on all these points.

Why are researchers and the public so divided on guns?

When it comes to opinions on guns, Americans tend to, well, stick to their guns.

A 2017 Pew Research study called America’s Complex Relationship With Guns examined how Americans feel about guns. This survey shows big differences in perceptions and opinions about guns between men and women, gun owners and non-gun owners, and often Democrats and Republicans.

Some researchers say scholars who study guns and self-defense are also not immune to the politics and ideology that drives the gun debate.

“Among a lot of people, you can pretty much guess where they come out on the gun issue based on what party they vote for,” says economist and gun-rights advocate John Lott, who identifies himself as a Republican. “I don’t think that was necessarily true 25 years ago, but I think nowadays, especially for the gun-control side, I bet every single one are Democrats.”

Webster, the expert at Johns Hopkins who’s often at odds with Lott on gun policy, agrees that science isn’t immune to ideology — especially on such a hot-button issue with hard to quantify data.

“Certainly there’s an ideological piece,” he says. “It should not influence the research and interpretation of the data. But clearly you do see different interpretations and which data people think best reflect what’s going on with respect to how people use guns. … You can measure homicides … but so much else [in gun research] is measured in very imperfect ways. So different researchers will make their own decisions on what to use and what assumptions to make or not make about them.”

Jens Ludwig, who directs the University of Chicago’s Crime Lab and researches gun control policy, has watched these debates play out for years. He says that there’s some merit in the scholarship on both sides. But because of the passionate feelings on both sides of the issue, at some point the debate became more about showing why the other side was wrong rather than gleaning useful insights from a rival’s research.

Will that gun really make Mike (or anyone who owns a gun) safer?

When people try to answer this question, they’ll often point to big analyses showing more gun deaths in homes with guns. Or they’ll bring up studies about the the rise or fall (depending on which scholar you talk to) of violent crime rates in communities that permit people to carry concealed guns.

But when you’re talking about the safety of a specific gun in an specific household, that’s a different story.

“The risk and potential benefits of having a gun are going to vary dramatically from household to household and community to community,” Webster says.

These variances are based on well-documented risk factors that experts agree increase the chance a gun will be used for a homicide or suicide in the home. These include how safely the gun is stored and whether someone in the home has a history of criminal behavior, domestic violence, substance abuse, impulsivity, or certain types of depression or mental illness. All of these factors can affect the likelihood a gun will cause harm inside the home.

Even a scholar like Kleck, who many consider “pro-gun,” believes this is a nuanced issue. When asked how he would advise Rady and Mike on the safety of Mike’s gun, he said there were many factors to consider.

“I’d want to know if those two or anybody living in a household with them have a criminal record or a record of violent behavior in the past,” he says. “Because if the answer is ‘no,’ then I’d side with Mike. On the other hand, [if] the answer is ‘yes,’ then I’d say you’d really have to look into it a lot more detail because it’s going to be a closer call.”

Kleck adds one more thing.

“For a person who’s likely to freeze in the face of a threat, it’s not going to do them much good,” he says. “And you know most people basically should try the standard alternative: First they should call the cops if they can get to a phone and if they have time. You know, if the police can arrive in time, then that’s great. That’s a good alternative. Let the professionals deal with it.”

So, will that gun really make Mike safer? A lot of that depends on how well he can use a gun, how safely he stores the gun, and whether anyone in his home has a host of behavioral, criminal, and psychological characteristics. 

But will Rady and Mike (and the rest of the country) ever come to some agreement on this issue? That doesn’t seem likely any time soon. Studies show that gun perceptions and opinions often fall along party, gender, and gun ownership lines. Even researchers aren’t totally immune from these conscious or unconscious biases.

So in a country that seems to grow increasingly divided ideologically by the day, we’ll likely be dueling out this issue for many years to come.

More about our questioner

Rady is a data scientist in Chicago. She says she knows people can become very passionate about guns. That’s one of the reasons why she asked that we use just her first name and give few other details about her.

When I showed Rady all the research that’s been done on defensive gun use, she was disappointed but not surprised especially about the ideological lens that can color research.

“That’s true for almost any science because if you have a brain, you have cognitive biases,” she says. “We all do. So I’m more likely to believe evidence that supports my point of view, and I’m more likely to look very closely [at] evidence that doesn’t support my point of view in order to find something wrong with it.”

These were not the answers Rady had been looking for. She’d hoped Curious City would find some rock solid data that would convince her friend once and for all to get rid of his guns. But now that she knows how complicated gun data is, and how set we are in our views on guns, she says her hope is starting to wane. 

“My family has the opposite political beliefs of what I do and we have these kinds of arguments all the time, too,” she says. “And it’s true that either you believe guns are going to protect you or you believe they’re not, and it doesn’t seem like the evidence is going to sway you.”