Statisticians Study Gun Control

Leah Libresco Sargeant (@LeahLibresco) is a statistician, a former newswriter for the data journalism site FiveThirtyEight, and Yale alumna. She openly admits to being anti-gun, an opinion that motivated her and her colleagues to study deaths in the United States where a firearm was used to find a statistically viable approach to champion. Despite holding anti-gun opinions, she decided to let data rather than emotions drive results. She summarized her findings for The Washington Post, “I used to think gun control was the answer. My research told me otherwise.”

“[M]y colleagues and I at FiveThirtyEight spent three months analyzing all 33,000 lives ended by guns each year in the United States, and I wound up frustrated in a whole new way. We looked at what interventions might have saved those people, and the case for the policies I'd lobbied for crumbled when I examined the evidence. The best ideas left standing were narrowly tailored interventions to protect subtypes of potential victims, not broad attempts to limit the lethality of guns.

“By the time we published our project, I didn't believe in many of the interventions I'd heard politicians tout. I can't endorse policies whose only selling point is that gun owners hate them. Policies that often seem as if they were drafted by people who have encountered guns only as a figure in a briefing book or an image on the news.

“I found the most hope in more narrowly tailored interventions... We save lives by focusing on a range of tactics to protect the different kinds of potential victims and reforming potential killers, not from sweeping bans focused on the guns themselves.”

From her personal Twitter account:

I was all for banning silencers, assault weapons, etc. until I researched gun deaths for FiveThirtyEight. Now, I'm better informed and frustrated at how politicians sold me soundbitable solutions that are (at best) not backed by data, & at worst are incoherent and could only be written by people ignorant about basic gun mechanics. There are better solutions to tout, but they're not simple bans.

My colleagues and I spent months researching all gun deaths in USA (https://fivethirtyeight.com/features/gun-deaths) and our work kept convincing me that there's not as much plausible policy to reduce gun deaths as I had believed. I say plausible not re political opposition but if the anti-gun folks had relatively free reign, there still wasn't that much I was confident would help. Assault weapons aren't a real category.

I have the most confidence in smaller, more focused interventions: gang violence de-escalation work, protecting domestic abuse victims, and support for those with suicidal ideation (2/3 of all US gun deaths). I'm grief-stricken by #LasVegas murders, and I wish data had pointed me toward greater hope in a Big Gun Policy that I believed would work, if we all mustered the will to pass it. But it didn't. The next step isn't a single solution, but sustained, personal outreach.

Interestingly, though not surprising, John Pfaff at Slate reached the same conclusions:

"Philadelphia Councilmember Curtis Jones Jr. published a report titled “100 Shooting Review Committee Report.” The report received far less attention and was covered only by the local Philadelphia press. That’s a shame. The report’s authors examined more than 2,000 shootings. What they found is that gun violence is much more complicated than Adam’s blueprint suggests, arguing that a better way to focus on gun violence is to target the violence more than the guns.

"The Philadelphia report—written by a wide range of sometimes contentious stakeholders, including the Philadelphia Police Department, the district attorney’s office under reformer Larry Krasner, the Department of Public Health, and the Defender Association of Philadelphia—suggests that such interdiction [targeting access and removal of guns] is likely futile. The authors provided analyses and policy recommendations for a city suffering from a record 559 homicides in 2021. While the proposals from the Philadelphia police broadly track with the Adams plan, the recommendations from the other stakeholders, including the city’s district attorney, caution strongly against an approach that centers on gun interdiction."

Along this same path, the RAND Corporation conducted an analysis of 27,900 research publications on the effectiveness of gun control laws. Of them, only 123 (0.4 percent) passed the rigor needed to be a valid research study. While some of those 27,777 studies may have had some useful insight for non-empirical discussions, most were “deeply flawed”.

The 123 studies that met RAND's criteria still had serious statistical defects, such as a lack of controls, too many parameters or hypotheses for the data, undisclosed data, erroneous data, misspecified models, and other problems. These glaring methodological flaws are not specific to gun control research; they are typical of how the academic publishing industry responds to demands from political partisans for scientific evidence that does not exist. Further, because violence with firearms in the United States is a statistically rare phenomenon, the sampling error rates that have to be accounted for in all types of research are greater than any measured “results” due to policy.

RAND found that not only is the social science literature on gun control broadly useless, but it can also be detrimental by providing fodder for advocates wanting to say "studies prove" for their particular favored policy that is unlikely to have beneficial outcomes. This matters because gun laws, even if they don't accomplish their goals, have large costs. They can turn otherwise law-abiding citizens into criminals, they increase prosecutorial power and incarceration, and they exacerbate the racial and socioeconomic inequities in the criminal justice system.

Amazing! It took multiple research studies to realize that violence isn't caused by access to tools. I've been on Camp Perry firing lines with 1,200 shooters armed with AR-15 and M14/M1A Service Rifles and hundreds of rounds each. Injury rates since 1907 when the National Matches were first hosted there have been so non-existent that they aren't even tracked. But nobody asks real shooters for opinions about firearms...