You may have seen the USA Today article released on December 3, 2013. It is an informative piece addressing mass shooting in the United States and includes helpful criticism of the methodology the FBI and local law enforcement has employed collecting these data and how it can be improved. If you are going to dive in, be sure to click through the interactive links – they contain some of the most revealing information. The article can be found here:
Dr. Gene Deisinger, our colleague at SIGMA, characterized the article as one of the most complete public media analyses of mass violence he has seen, and I agree.
Looking at the issue longitudinally, per capita rates of mass killings really have not changed. What has changed is a shift from private/family killings to public/stranger killings which doesn’t quite come out in the article but is outlined in Grant Duwe’s work as well as that of others. Overall there is a slight uptrend in the number of victims per incident though it doesn’t appear statistically significant with the mode remaining very close to 4 - the definition of a mass incident. However the majority of catastrophic mass killings, e.g., 10 or more, have occurred in the last 30 years. It is interesting that the authors do not focus solely on gun violence as I anticipate there is more information forthcoming on non-gun mass killing.
Two important areas remain to be addressed (there are likely others):
1) Disrupted or prevented attacks
2) Attacks in which the perpetrator intended to kill multiple victims but fewer than 4 died (these incidents are not counted). This is harder to get a handle on but has likely skewed the analysis by excluding cases when, for example, the reason fewer than 4 died was not a function of perpetrator intent but of rapid delivery of trauma services or some other intervention.