U.S. Outlier in Mass Public Shooting Events, or Mass Shooters?

Image from wikimedia commons 

In 2016 Adam Lankford of the University of Alabama, published a paper examining the distribution of mass shooting/mass shooters around the world. The published result was the United States had a disproportionate number of these rare events.  Lankford's study showed the United States had 5% of the world's population, but 31% of the the mass shootings/mass shooters, as defined by Lankford.  It has become clear that Lankford was looking almost exclusively at single perpetrators, though he included about 2% where two people were involved. Lankford excluded terrorist attacks he identified as "sponsored" terrorism, but included some terrorist attacks.

John Lott questioned Lankford's finding based on the definitions used in the paper, which treated mass shootings and mass shooters as equivalents, although it was not precisely clear which definition was of a higher priority.

There is a discussion by both sides featured in Econ Journal Watch from March of 2019. Here is a link to John Lott's and Carlisle E. Moody's critique and Adam Lankford's reply.

Lott and Carlisle's critique relies heavily on definitions. They show, if the definition used by the FBI in its 2014 active shooter report, referenced by Lankford, is rigorously followed, then Lankford's paper is misleading about mass public shootings in the rest of the world. They agree the United States has more single perpetrator public mass shootings than other parts of the world. They show many other parts of the world have many more public mass shootings than the United States, with multiple perpetrators. One of Lott and Carlisle's critiques is that Lankford refused to release the data his study is based upon.

Lott and Carlisle provide links to the data they used in their study, which was not limited to single perpetrator public mass shootings. Using the FBI's definition, which does not exclude terrorism, Lott and Carlisle find the USA has 2.9% of public mass shootings for 4.5% of the world's population.

Lott and Carlisle examined data from 1998 to 2012. They state finding data of any earlier crimes is extremely difficult, especially for events before use of the Internet became common.

Lankford looked at data from 1966 to 2012. Lankford has been regularly criticized for not releasing his data and for the difficulty of finding rare events in non-English speaking countries that occurred decades ago.

Adam Lankford, in reply, admits his definitions could have been more precise. He makes the argument it makes no practical difference because the definitions are about 98% congruent. That appears to be true in the United States. It is not so clear in the rest of the world. He claims the Lott and Carlisle data supports his claims for single perpetrator mass shootings.

The definition of a public mass shooting has a long history of confusion and contention. Multiple definitions have been used. Definitions have political implications. Former Prime Minister John Howard used a changed definition of mass shooting to claim Australia had no mass shootings after the highly restrictive gun laws in Australia were put in place in 1997, after the public mass shooting at Port Arthur in 1996.

The definition used in Australian studies before 1996 was four victims shot and killed, and included private murders. In 2014 Geoff Hunt killed himself and four family members. It was a domestic killing, with Hunt killing his disabled wife and children. The Australian definition did not exclude domestic, gang, crime or terrorist murders. The definition was changed to five victims shot and killed, apparently to exclude the Hunt murders. From smh.com.au:

 “I don't believe we were on the cusp of going down the American path,” Howard said. “But I do think the gun laws have had the practical value of reducing the mass slaughters. There were 13 before the new laws; and if you define such an event as five or more victims,  there have been none since.

The National Institute of Justice in the United States has recognized the problem of definitions when studying mass shootings. They considered incidents, not perpetrators.  From nij.ojp.gov:

The literature does not define “mass shooting” consistently, or even in similar contexts. The federal criminal code lacks a distinct mass shooting offense; this may help explain why researchers use different terminology, or types of criminal offense, in their analyses of the same phenomenon.[5]

Among the 44 studies analyzed, the most common definition of a mass shooting is an incident in which four or more victims are killed with a firearm in a public place (48%). Several studies defined the offense as an event during which as few as two (5%) or three (9%) victims are killed, whereas more than one-third of the studies more broadly defined the term as an incident in which multiple victims are killed (38%). Others either defined a mass shooting incident as having a minimum of five victims or did not specify a victim threshold.

The definitions in the analyzed studies include incidents that take place in publicly accessible spaces such as schools, workplaces, places of worship, and businesses. The incidents are also defined as a single, continuous event within a short time frame, but the specific time frame can vary. The definitions often exclude ideologically motivated terrorist acts as well as gang, drug, and other shooting incidents that resulted primarily from the commission of other crimes, such as aggravated robbery, familicides, and domestic violence. Some of these studies, however, do not specify whether certain types of offenses were excluded from the definition.

Some definitions of "mass shooting" include all incidents where 4 or more people were injured or killed, regardless of the context.

In Lankford's response to Lott and Carlisle's critique, he uses several examples from their data set, which is publicly available. Lott and Carlisle are hindered, somewhat, because they do not know exactly what incidents Lankford included in his data set, and which incidents he excluded.

A major contention is over what proportion of mass shootings, where the number of perpetrators is unknown, could be assumed to be 1, or perhaps 2 perpetrators, in the international data. Data gathered from open sources, such as online newspapers from foreign countries, is spotty. Better reporting in the United States is likely to show a bias against the United States.

Lankford's paper showed no correlation between the homicide rate or the suicide rate in a country, and the number of mass shootings. It showed a correlation between firearms ownership in a country and the number of mass shootings. This implies there would be no correlation between firearm ownership and either the homicide rate or suicide rate in different countries.

Other researchers have found there is no correlation between firearms ownership and homicide or suicide rates among all countries. There are studies which show a correlation of gun ownership and gun homicides or gun suicides, but this is irrelevant. It makes no difference if homicides with guns or suicides with guns are reduced, if the overall homicide or suicide rate is unaffected or increases. There are studies which bias the result by carefully picking which countries to compare, excluding some and choosing others.

This popular paper in Medium takes a relatively unbiased look at international statistics and how they are manipulated with relationship to gun data. One of the primary methods is to only look at "gun deaths" instead of homicide or suicide rates.

Without Lankford's data set, it is impossible to know if his data would confirm that gun ownership levels are not correlated to homicide or suicide rates in different countries.

It is of interest, because Lankford cites several studies which claim a correlation between suicide and homicide and gun ownership inside the United States. It is difficult to determine the number of gun owners in each state. Typically, proxies are used. Lankford was apparently not interested enough to determine if those papers would be contradicted by the international data.

Looking at mass shootings only is a way to look only at gun deaths.

It is overall homicides that matter, not only homicides with guns. Similarly, it is overall suicides that matter, not suicides with guns.

The international data is fairly clear that gun ownership, by itself, does not increase homicide or suicide rates.

It may be the United States has a significantly higher rate of single perpetrator mass murder shootings than other countries. It is irrelevant if the overall homicide rates are not affected by gun ownership.

In his paper, Lankford makes a statement about the Australian experience with extreme gun control. He states:

It may take more cases of unambiguously successful gun control, such as Australia’s, to begin to gradually change America’s gun culture. Or it may take more scholarly research which provides empirical evidence of the link between firearms availability and public mass shootings and thus shows that policymakers and legislators may be able to directly influence the prevalence of these high-lethality crimes.

The Australian gun control scheme has been far from unambiguously successful. Several papers have shown no significant effect on homicides or suicides. Using Lankford's apparent definition of public mass shooting, Australia had four in the 25 years previous to 1996, and one in the 23 years after 1996 (not the Hunt murders. This event occurred after the John Howard statement, in Darwin, in 2019). The numbers are so small as to be statistically meaningless.

Even the small, statistically meaningless numbers may be explained by the change in the coverage by Australian media. Before the 1996 mass murder in Port Arthur, the Australian media had intense coverage of mass shootings. The Port Arthur killings were an example of media contagion. The perpetrator, during the trial, repeatedly asked if he had "broke the record".  After Port Arthur, having attained the desired policy change, the Australia media coverage of mass shootings changed considerably. Having lived in Australia for 9 months in 2017, 2018, and 2019, I am relying on my personal observations from Australian media, and from records of the media before Port Arthur.

Lankford has shown the United States has a culture which uniquely values "fame", and whose media elite glorify mass killing by gun beyond most other cultures.

Unfortunately, many American public mass shooters also seek fame and glory—but they obtain it through killing—and the media coverage they receive in the United States seems to give them exactly what they want (Langman, 2015a; Lankford, 2013; Lankford & Hakim, 2011; Larkin, 2009; Newman et al., 2004).

It seems plausible the media contagion effect drives the American mass shooting phenomena more than the American high level of gun ownership.

Lankford notes the level of American gun ownership is difficult to change. Media contagion may be more amenable to change than levels of American gun ownership. The American media, collectively, changed their policy on publicizing celebrity suicides, when it was learned that suicide clusters were fostered by media contagion.

The policy was voluntary. Given major media's commitment to promoting restrictive gun policies, it is unknown if voluntary cooperation in reducing mass shootings is possible, given the current major media ideological orientation.

Definitions and data sets are important. One of the keys to replicating scientific research is to have the data sets publicly available.  Papers which restrict access to the data used in the paper are subject to deserved skepticism.

Lankford does not appear to consider any of the costs of gun restrictions. Costs are important.

It is bad policy to change society based on extremely rare events, events covered by an ideologically driven media to obtain a specific policy which media elites desire.

It is as if, having exhausted homicide and suicide to justify restrictive gun controls, those who desire an unarmed society have turned to extremely rare events to justify their policy choices. It is a common way to manipulate the democratic process.

In April of 2020, a Canadian who was prohibited from legally obtaining firearms went on a murder rampage, killing at least 22. Nine of the deaths were reported to have been by arson.

It does not appear any of the firearms used by the Canadian were legally obtained, or were semi-automatic firearms demonized by the Canadian Prime Minister as "assault weapons".

The Prime Minister used the horrific event to launch his desired ban on the so-called "assault weapons", which do not appear to have been used in the crime.

©2020 by Dean Weingarten: Permission to share is granted when this notice and link are included.

Gun Watch