Updated

Between 1996 and 2013, 991 police officers have been killed in the line of duty. These public servants put their lives on the line to protect Americans. Their deaths should concern everyone, including public health researchers. But researchers’ dislike of guns may be clouding their judgment, and a new study, which claims that more police were feloniously killed in states with more guns, mistakenly concludes that private gun ownership causes police deaths.

The study, which was announced last week and is forthcoming in the American Journal of Public Health, received extensive national and international news coverage.  But if the researchers hadn’t left out controls used by everyone else for this type of empirical work, they would have gotten the opposite results from what they claimed.

Previous research has done just that.  And it has found that concealed handgun permits lead to fewer police deaths. The authors offered no explanation for the new study’s unorthodox approach.

There is a big benefit to using so-called panel data, where you follow changes in crime rates across many different states over a number of years.  Doing that allows you to have many different experiments and makes it possible to more accurately explain for differences in crime rates across states or over time.

A couple of simple examples show why other studies on crime take into account these factors.

Take a common comparison of different countries. As many people point out, the UK has both a lower gun ownership rate and a lower homicide rate than the US does. Yet, it does not logically follow that reducing gun ownership leads to a reduction in crime. And, in fact, after the UK’s 1997 nationwide handgun ban, their homicide rate actually increased by 50 percent over the next eight years.  The UK still had a lower homicide rate than the US, but this wasn’t because of the handgun ban.  Other factors must have played a role.  The ban itself raised their homicide rate.

A similar point applies over time.  Suppose a state passes a gun control law at the same time that crime rates are falling nationally. It would be a mistake to attribute the overall drop in national crime rates to the law that got passed.  To account for that concern, researchers normally see whether the drop in crime rate for the state that had the change is greater or less than the overall national change.

Looking at data by state over many years allows researchers to account for both of these potential biases, but the American Journal of Public Health study doesn’t account for this bias over time and the authors offer no explanation for this lapse.

If they had done what everyone else does, it would have reversed their results.  Instead of their claim of a one-percentage point increase in the percent of suicides committed with guns increasing the total number of police killed by 3.5 percent, they would have found it reducing police killed by 3.6 percent.

Previous work has shown that letting law-abiding citizens carry guns reduces the rate that criminals carry guns, thus making it safer for both civilians and police alike.

But perhaps the craziest thing about this study is how it “measures” gun ownership.

While the media talks about gun ownership being related to police deaths, what they are actually measuring is the percentage of suicides committed with guns.  While this may have some relationship to gun ownership, this much more likely picks up whether the population is relatively more male, as men are more likely to use guns for suicide, as well as other demographic and geographical differences.  For example, even when women own guns, they are more likely than men to use other methods of committing suicide.

Despite all the extensive news coverage of the American Journal of Public Health study, there were no interviews with anyone who might have been critical of the controversial study.

In reality, criminals have ways of getting guns even when guns are banned.  For example, drug gangs will get their guns to protect their drugs just as easily as they get their drugs to sell.  Thus gun control primarily disarms the citizens who obey the laws.  There are lots of good law-abiding citizens who not only protect themselves and their fellow citizens, but even help protect the police.