Late last week Megan McArdle of the Daily Beast posted an online interview she did with criminologist James Alan Fox, in part taking issue with our mass shootings investigation at Mother Jones. I was glad to see McArdle continue the conversation with Fox; as he and I agreed in a recent exchange, mass shootings deserve continuing inquiry.
Based on data that includes all gun crimes in America involving four or more fatalities, Fox has found that mass shootings are not on the rise. He has also applauded our more narrowly focused investigation, which found a rise in a specific type of mass shooting, and acknowledged its value to studying the problem.
McArdle, however, maintains that our in-depth project is of no value: It is, in her view, "not a database of mass shootings; it's a database of mass shootings that Mother Jones wanted to include in their database."
In fact, we stated explicitly at the outset that our investigation was not all-inclusive, but rather a deep study of a particular kind of mass gun violence. Because no clear definition of "mass shooting" exists, we consulted with federal law enforcement officials and academic experts—including Fox—to develop criteria. Our goal was to dig deeper into the specific phenomenon of seemingly incomprehensible attacks in public places—shopping malls, religious buildings, workplaces, schools—to try to make more sense of these "senseless" tragedies. Thus, we excluded other types of clear-cut cases in which the primary motive involved gang activity, armed robbery, or domestic violence.
Given the complexity of mass gun violence, we suspected that we might encounter a couple of exceptional cases that would require discretion in applying our criteria. We chose to include the Columbine massacre, as well as the Westside Middle School shooting, because they fit the criteria despite that two shooters were involved in each case, an exception we flagged for readers. In other words, we included all school shooting cases from the last 30 years clearly fitting this type of attack, as defined by fatality count, location, and motive. (It bears noting that if we were to have excluded those two cases, it would not have meaningfully altered any of our broader analytical findings.)