What Mass Shootings Tell Us About Gun Control

Yesterday morning, an Illinois man named James Hodgkinson approached a baseball field in Alexandria, Virginia, and opened fire on members of the Republican Congressional baseball team. Hodgkinson, said “to be distraught over President Trump’s election,” indiscriminately shot between fifty and sixty rounds but — miraculously — only managed to injure four people before being shot and killed by Capitol Police.

To no one’s surprise, this incident has reignited the gridlocked debate about gun control. The Washington Post, relying on data from the Gun Violence Archive, reported that this shooting was the 154th mass shooting in America so far this year. Per WaPo:

The archive considers an incident a mass shooting if four or more people are shot, not including the shooter. Some definitions are broader: If the shooter is included in the tally, the number of mass shootings rises to 195. Some, however, are much more narrow: If a mass shooting is defined as four or more victims killed in a public location, excluding robberies and gang violence, the number falls to four.

I am not personally interested in debating how we should define the term “mass shooting.” I am, however, inexplicably drawn to analyzing large datasets, so I spent the morning digging into the Gun Violence Archive’s mass shooting database.

A few notes: the GVA archive is downloadable as an Excel file, but not all of the pertinent information comes with the spreadsheet. I spent the better part of the morning investigating and categorizing each individual incident. In fifty cases, or about a third of the dataset, there was either no suspect in custody, police were not releasing details about the suspect’s motives, or no motive information was available. I removed these incidents from parts of my analysis. Further, some might dispute how I categorized certain shootings. An example: when a shooting was described as a drive-by shooting with two or more shooters, I categorized it as a gang shooting. Someone else might call that terrorism. With respect to the broader implications, I don’t think the distinction matters much.

Of the 104 shooting events I could categorize, I used eight broad categories (listed from most deaths to least): Gang-related, Domestic Violence, Interpersonal Disputes, Robberies/Home Invasions, Workplace Violence, Terrorism, Shootouts (where two or more people exchange fire), and Hate Crimes. Here is a chart of those incidents:

This is where the debate often goes off course. Republicans and Second Amendment advocates tend to point to gang shootings, acts of terrorism, and other crimes as evidence that criminals, by definition, are people who are willing to break the law, and as such are willing to acquire guns illegally. And this point is duly noted: of the incidents where a gun was illegally acquired — or an illegal gun was used — the vast majority were gang-related shootings. One the other hand, in the incidents involving interpersonal disputes, workplace violence, and domestic violence, Democrats and Gun Control advocates have an equally compelling point: most of these crimes would not have happened if the perpetrator, temper running high, did not have immediate access to a gun.

With this idea in mind, I compared the locations of the shootings with the local gun control laws. To do this, I borrowed from Gun Law Scorecard who, through a somewhat opaque process, rates each state by the gun control legislation they’ve enacted. Cross referencing each shooting by the state’s Gun Law Score, I put together this table:

States that received an “F” grade likewise had the most mass shootings, the most deaths, and the most injuries. The Gun Law Scorecard appears to be vindicated, at least in terms of the states they fail. (On the other hand, the states with the fewest mass shootings and least deaths had B, C, and D grades, so the rest of the grading system might require some tweaking.)

But not all gun control measures are created equal. There is plenty of academic debate over the efficacy of allowing or prohibiting concealed carry. However, one measure that seems to be generally effective is universal background checks. In states where universal background checks are required, impulse or “hot temper” mass shootings make up less than a quarter of the overall incidents, compared 58% of such shootings in states that do not require universal background checks. Moreover, mass shootings are more common and lethal in non-background check states.

The single most effective gun control measure appears to be universal background checks. Though Republicans are right that there are a number of situations where a person intent on committing a felony act of violence will acquire a gun by any means necessary, there are many incidents where the mere presence of a gun fatally escalates an situation that would otherwise only result in bruised fists and egos. The evidence also shows that mass shootings are more rare in states that require a background check, the shootings are less lethal in those states, and there are fewer injuries per shooting. Although I generally support the Second Amendment right to gun ownership, a nationwide universal background check on all public and private gun purchases seems to be a clear and effective strategy for reducing, though not eliminating, mass shootings.