New CPRC Research: How a Botched Study Fooled the World About the U.S. Share of Mass Public Shootings: U.S. Rate is Lower than Global Average

29 Aug , 2018  

Executive Summary

A paper on mass public shootings by Adam Lankford (2016) has received massive national and international media attention, getting coverage in the New York Times, the Wall Street Journal, plus hundreds of other news outlets spanning at least 35 different countries. Lankford’s claim was that over the 47 years from 1966 to 2012, an enormous amount of the world’s mass public shooters — 31% — occurred in the United States. Lankford attributed this to America’s gun ownership.

Lankford claims to have “complete” data on such shooters in 171 countries. However, because he has neither identified the cases nor their location nor even a complete description on how he put the cases together, it is impossible to replicate his findings.

It is particularly important that Lankford share his data because of the extreme difficulty in finding mass shooting cases in remote parts of the world going back to 1966. Lack of media coverage could easily lead to under-counting of foreign mass shootings, which would falsely lead to the conclusion that the U.S. has such a large share.

Lankford’s study reported that from 1966 to 2012, there were 90 public mass shooters in the United States and 202 in the rest of world. We find that Lankford’s data represent a gross undercount of foreign attacks. Our list contains 1,448 attacks and at least 3,081 shooters outside the United States over just the last 15 years of the period that Lankford examined. We find at least fifteen times more mass public shooters than Lankford in less than a third the number of years.

Even when we use coding choices that are most charitable to Lankford, his 31 percent estimate of the US’s share of world mass public shooters is cut by over 95 percent.  By our count, the US makes up less than 1.43% of the mass public shooters, 2.11% of their murders, and 2.88% of their attacks. All these are much less than the US’s 4.6% share of the world population. Attacks in the US are not only less frequent than other countries, they are also much less deadly on average.

Given the massive U.S. and international media attention Lankford’s work has received, and given the considerable impact his research has had on the debate, it is critical that this issue be resolved. His unwillingness to provide even the most basic information to other researchers raises real concerns about Lankford’s motives.

The paper is available to be downloaded here. A copy of Lankford’s paper is available here.

“The power of this report should convince academic and medical groups to reevaluate all prior statistical studies of mass shootings and the overall public health danger of firearms,” said Professor Arthur Z Berg, M.D., DLF APA, Harvard University emeritus. 

Here is a graph that the New York Times put together using Lankford’s data (data on the rate of mass public shooters by country that has not been released to anyone other than the New York Times) and our graph using the corrected data. While the graph with Lankford’s data shows a positive relationship between the rate of mass public shooters and Small Arms Survey measure of gun ownership (it appears to be driven by the outliers of Yemen and the US), our graph implies a slightly negative relationship. Click on Figures to enlarge.

 

“Because of faulty research, it is widely believed that a disproportionate share – 31% – of the world’s mass public shooters occurred in the United States,” said Professor Paul Rubin, Samuel Candler Dobbs Professor of Economics, Emory University.  “In fact, John Lott’s careful analysis of a very large data set – 437 – pages – shows that the proper number is about 2%, less than the U.S. share of world population.  One can only hope that this important research will correct the record.”
Professor Carl Moody, College of William & Mary offered the following: “This is an important paper. The assertion that the US is responsible for 31 percent of worldwide mass shooters is patently absurd. Anyone who doubts the veracity of Dr. Lott’s analysis is welcome to download, for free and in Excel format, the entire Global Terrorism Database (https://www.start.umd.edu/gtd/contact/). There they will find, with a simple back of the envelope calculation, that worldwide since 1970 there have been 58,445 mass firearm attacks. Of these, 402 have occurred in the United States. The US is, according to the GTD. responsible for less than one percent of all mass shootings (0.69 percent) since 1970. Dr. Lott’s calculations are much more carefully done, but Professor Lankford’s analysis is clearly not in the ballpark. Also, social scientists seldom have laboratories. Replication is the only way to verify claims. Any academic who refuses to share his or her data for replication purposes deserves to be shunned.”
UPDATE: So what has been Adam Lankford’s response? From outlets from the Washington Post to Real Clear Politics to Fox News, Lankford has refused to answer their questions about his study. As to our research, Lankford told the Washington Times:

“I am not interested in giving any serious thought to John Lott or his claims,” he said in response to an email seeking comment. . . .

This is an amazing response from someone who has refused to provide even the most basic data to other researchers. After reading our study or looking at Appendix 1 listed below, people can judge for themselves the appropriateness of Lankford’s response.

Appendices 1 and 2 for our research on mass public shootings around the world are available here.

Appendix 1 Foreign Cases (451 pages)

Appendix 2 US Cases (10 pages)

Excel file for International Mass Public Shootings

 

, ,


32 Responses

  1. Sam Costanza says:

    They lied or distorted the data?

    I’m shocked. Shocked!

    Well, not that shocked.

  2. joefoam says:

    Lankford’s response reminds me of the global warming theorists ‘believe us or we won’t even talk to you’. Thank you Dr. Lott for your tireless work to set the record straight. Now if we could just get this data widely publicized.

  3. John Milliken says:

    Anytime one refuses to defend ones own work, it certainly brings to light ones own doubts about its validity and veracity. It begs the question of his sources, his citations, and his motives – – which obviously did not include accuracy.

  4. MarkPA says:

    A couple of thoughts occur to me.

    First, let’s look at the actual US numbers of victims of mass-shootings. A couple hundred deaths per year. On a par with the number of accidental drownings. Each death is a tragedy, whether by gunshot or drowning. But, these numbers are not on a scale worthy of attention when we have 10,000 1-on-1 or 1-on-few homicides or 22,000 suicides.

    Second, never mind the fact that the number of US victims of mass-shootings – or – the number of drownings are negligible compared to other causes of death. The real jeopardy we should be concerned with is the POTENTIAL for much larger numbers.

    How concerned ought we to be that the number of drownings might rise from 200/year to 2,000 or 20,000? Can you figure this one out? Can we extrapolate history into the future to form a rational prediction? How concerned ought we to be that the number of mass-shootings might rise from 200/year to 2,000 or 20,000? Can you figure this one out? Is there a potential for this number to rise at an exponential rate?

    Perhaps we really ought to be worried about the potential for an epidemic – even a pandemic – of mass shootings (even though we aren’t justified about drownings.) Why could this be so? Should we be concerned by the jeopardy of a huge rise in the number of crazy individuals? How about terrorists? How about individuals who – while acting alone – are motivated by some common concern – political, religious, economic?

    Little imagination is required to see the potential for a rapid rise in mass shootings. What could be done about it? Is it possible to eliminate the artifact – the scourge of man’s inventive nature? How did that work with alcohol Prohibition? Drug prohibition? How difficult/easy is it to hand-craft guns? To manufacture guns on a small scale? How well developed is the black-market for guns? As long as there is a demand, will guns be produced, imported or stolen to supply this demand? If the black-market for guns is well-lubricated, is it available to meet a surging demand by terrorists?

    What might we – American society – do about this jeopardy?

    The answer MIGHT be found in Lott’s study. Why is America’s rate so much lower than that of the other countries he has studied? This is counter-intuitive. Our market is awash in guns for civilians; there is nothing like it in any other country! Maybe it’s something in our water; or air; or diet. Culture?

    Or, is it captured in Akinyele Omowale Umoja’s book title: “We will shoot back”. Anyone contemplating a mass shooting in America recognizes the fact. Americans will shoot back; Americans CAN shoot back. This simply isn’t the case in most other countries (with few exceptions such as Israel).

    Can Americans see through the jungle of facts and arguments to identify the antidote to mass-shootings? That prophylactic that serves to keep America’s rate low; that which is most likely to prove successful in preventing an epidemic, and more importantly, a pandemic.

    • SM says:

      Can you explain to me why you are referring to shooting as if it was a virus or bacteria that can be “cured”? Shootings are NOT an illness of any kind. Using an unappriopriate dictionary like this undermines the sense of discussion, becouse it gives off completely wrong idea.

      And if you wish to find your “antidote” simply ask police in cities like Boston : “Why are you letting shooters go free?” Most of the people shooting or attacking are not arrested, as long as the victim is not critically wounded. Executing the law is the best method to fight crime.

      • Henry says:

        He’s not really “referring to shooting as if it was a virus or bacteria that can be cured.” He’s postulating that as a possible opposition argument (one that we have already often seen made), then destroying it.

    • Duane says:

      you lay it out so well , keep preaching brother keep preaching

  5. James M says:

    If you read the book “Understanding Post-Modernism,” you will understand why he doesn’t share his data. The new philosophy does not believe in data, it is considered a biased weapon used in the struggle for power, not a collection of objective facts. He sees himself in a competition for the domination of his ideas over yours, with no regard for conventional norms of reality.

  6. Jim Delton says:

    One way to thank Dr. Lott is to support his work financially… just sayin….

  7. Richard White says:

    It is an aphorism in the scholarly community that you can prove nearly anything you want if you are allowed to cherry-pick your data, but you can prove anything at all if you are allowed to make you data up entirely. Dr. Lott’s data are available to everyone to see, and they are as inclusive as is possible, so he can hardly be accused of either sin.

    How about Prof. Lankford? We can’t know if he doesn’t share. And he hasn’t.

    Hmmmm.

    • Henry says:

      From Kellermann to Bellesiles to Lankford, anti-gun researchers seem to have this curious behavior of refusing to let people see their data. The reasons are obvious.

  8. soljerblue says:

    I spent an entire career — nearly 40 years — in broadcast journalism. I do not care whether the reportage on crime, especially gun crime, comes from print or broadcast media. In the general media the information will almost always be slanted against the 2nd Amendment to the U.S. Constitution, the right of individual Americans to keep and bear arms for sport or personal defense, and against those individuals who exercise that right as being somehow deficient in proper citizenship.

  9. Run this again with numbers through 2017.

    I’d like to see how what happened in Las Vegas will change your numbers.

    • Fat Hubie says:

      Statistics, surveys, polls, and your opinion have no bearing on the inalienable rights of the people.

    • Thomas Wicklund says:

      Las Vegas will do little to affect number of shootings — it was one shooting. It will affect number killed by mass shootings.

    • Duane says:

      there were 50+ killed in Vegas and 500 wounded, in mass killings around the world since 2015 there have thousands of people killed, just add up the killings in Europe. Go into countries in the middle east with all the bombings they do in markets or anyplace large #’s of people gather and you will see it rise into the thousands real quick. These are thyou can easily look up .

  10. Brian Carroll says:

    I’m looking at your data and it’s mostly terrorism / armed insurgency activity. Good on you for sharing your data, but I don’t think it gets to the thrust of the “crazy guy shoots up normal citizens” argument. America is so awash in guns that crazy folk can access guns and take lethal action with ease. The violence shown in your stats isn’t of that variety. Who knows what’s in Lankford’s data. He should release his too.

    Either way, both of your data sets end in 2012, excluding all of the record-setting mass shootings of the last eight years.

    • johnrlott says:

      Dear True Brian:
      You might want read the paper. Do you want to exclude the Pulse Nightclub, Ft. Hood (first attack), San Bernardino, Chattanooga, Sikh Temple, and other attacks would also clearly be classified as terrorist attacks? If you do want to exclude them, could you please explain why? The paper also says what the results are even if you cut out all the terrorist attacks: 709 shooters in foreign countries over the 15 years, compared to the 202 that Lankford says he was able to find out over 47 years.

  11. Nigel VIvash says:

    Why don’t you read the report, the countries with worse rates then the US are either war zones e.g. Afghanistan or countries with a homegrown terrorist/insurgent problem, e.g Algeria, Nigeria, Pakistan or countries with a rampant crime problem e.g South Africa.. do you really want to compare those countries with the USA. When you compare the statistics with the rest of the Democratized world the statistics make for chilling reading..

    Not that anybody on here or Fox News will care to read the report!

    • johnrlott says:

      There are 61 countries with higher murder rates from these mass public shootings. This includes Russia, Norway, Finland, and Switzerland, with rates that are all at least 45% higher than the US rate. While some of these are war zones, most of them are not, right?

    • Equus pallidus says:

      You haven’t a clue as to what your talking about. Russia alone, though have gun laws more strict than England, has a murder by gun rate that is almost 5 times higher than the United States. Do some actual reading then come on back and we can talk…,

  12. True American Patriot says:

    Dr. Lott,

    Why does your study include data from terrorist activity? Your 451-page list of “mass shootings” from around the world is almost entirely composed of terrorism, while the list of American fatalities is only directly related to actual mass shootings – angry white American (probably mentally ill) men who open fire on civilians in public places.

    It does not take a paper (albeit self-published through your own non-profit foundation & pandered to Fox News & I’m sure many other similar echo chambers) to know that there is less terrorism in the United States than in many other developing nations.

    Have you submitted this paper to any journals for peer review & publication, other than the obviously friendly critics you’ve cherry-picked to include on this site?

    • johnrlott says:

      Dear True American Patriot:
      I realize that you haven’t read the research, but you might want to do so. Do you really want to exclude the Pulse Nightclub, Ft. Hood (first attack), San Bernardino, Chattanooga, Sikh Temple, and other attacks would also clearly be classified as terrorist attacks? If you do want to exclude them, could you please explain why? The paper also says what the results are even if you cut out all the terrorist attacks: 709 shooters in foreign countries over the 15 years, compared to the 202 that Lankford says he was able to find out over 47 years. Thanks.

      • True American Patriot says:

        What are your criteria to differentiate between a “terrorist attack” & a “mass shooting”?

        Again I am just a normal civilian providing armchair criticism, so I am curious to know what other academics have reviewed your paper & if you seek to publish it in any journals.

        • johnrlott says:

          Again, I am not sure why you simply don’t want to read the paper. Here are two of the relevant discussions.

          “Our primary source is the University of Maryland Global Terrorism Database (GTD), which collected data on over 170,000 terrorist attacks from 1970 to 2016 (Global Terrorism Database, 2017 and LaFree et al, 2015). The GTD defines terrorist attacks as “the threatened or actual use of illegal force and violence by a non-state actor to attain a political, economic, religious, or social goal through fear, coercion, or intimidation.” The database lists attacks that were carried out using everything from firearms, incendiary, knifes, bombs, vehicles, chemical, biological, or radiological weapons. They divide their attacks into six categories: 1) Terrorism; 2) Insurgency/Guerilla Action; 3) Other Crime Type; 4) Intra/Inter-group conflict; 5) Lack of Intentionality; and 6) State Actor.
          The only categories that sometimes meet our criteria for mass public shootings are “terrorism”, “other crime type,” and “intra/inter-group conflict.” Government sponsored or directed/ordered killings (the “State Actor” category) are completely excluded. We then reviewed each case using Nexis and web searches to determine whether they met our definition (exactly the same as Lankford used, minus insurgency-related shootings). Less than 50% of the terrorism shooting cases identified by the GTD met the definition of mass public shootings. Including insurgency-related shootings identified by GTD would have increased the number of foreign mass public shootings by 208, from 1,448 to 1,656, and would have reduced the United States share of these attacks to 2.53%.”

          “Lankford argues that are “functionally” the same. Both types involve premeditated attacks that aim to kill and wound as many people as possible because they know that the more people they harm the more media attention they will receive. They also involve the same type of planning, such as picking targets that aren’t able to defend themselves. “

          • Henry says:

            “Again, I am not sure why you simply don’t want to read the paper. ”

            Possible because actually doing the homework tends to debunk one’s favorite “common sense solution.”

  13. DRcc says:

    Please pardon my amateur analysis of your data (less than 30 minutes effort)… but here I go. (abbreviation RoW == Rest of World)

    Examined data provided in Excel sheet. If you eliminate the cases where the number of perpetrators > 2 {which eliminates militant groups, terrorist organizations (roughly)} and focus on individual or lone wolf type perpetrators, the U.S. rate is 21.8% of the world (331 U.S. vs. 1185 RoW). This points to two things, 1) U.S. is a safe place in terms of armed groups being able to effectively operate and cause damage, 2) the ability of 1-2 people or to cause massive damage is massively out of proportion in the U.S compared to RoW. Following this through considering population density, within the U.S. the rate is 5.8x higher than RoW (331 deaths/264.5Million people (U.S.)vs. 1185 / 6185Million people (RoW) which is 1.116 per million (U.S.) vs. 0.192 per million (RoW)).

    I realize the point Lott is making is ‘mass-shootings’ but the discussion is really more pointed towards attacks by smaller groups. As I don’t know where gun control is pervasive across the world, I wonder if you looked at those countries – again eliminating militant if the proportion gets worse or better. Also, in countries where there is gun control, is there a higher percentage of militant / terroristic activity.

    Thanks for the data and your analysis.

    • johnrlott says:

      Dear DRcc:
      You might want to look at the paper. If you eliminate all cases with over 10 shooters, you end up with 3,081 shooters, or an average of 2.3 shooters per attack. This might be helpful. Out of our 1,448 cases, news reports provide of the number of killers involved in the attack in only 380 instances. In 98 cases, a lone killer was identified, that is 26% of the cases that list a number of attackers. Another 42 attacks had two killers and 27 had three, so that indicates 44% of the cases where the number of killers was identified had between one and three shooters.

    • Mark N says:

      Are you saying being killed by 1-2 people is somehow worse than being killed by 3 or more people?

  14. Tony says:

    “For Lott, the actual data shows that the mass shooter problem in the U.S. is far from being a unique American experience. His question remains how the media and a president was able to get away with pushing a botched study published in a “low-rank journal” despite criticism.” (From news article)

    Lott’s question – indicated in the paragraph above – is actually very easy to answer. The first thing one has to point to is Lott’s unstated premise: that the media and the then president were seeking to disseminate facts and truth. That premise, at best, is very naïve. In fact, it is laughable.

    That premise is completely false. Neither the media nor the then president had any interest in facts or truth. They had (and have) an agenda. It is, essentially, a socialist agenda. And the completion of that agenda requires that the American population be disarmed. Therefore, ANYTHING that can be used to forward that agenda, whether true, partially false or completely false, is acceptable.

    This is so because what passes for an “ethical standard” among socialists is whether an action “serves” the socialist cause. If it serves the socialist cause, it is considered “good”, even if it is based on a complete lie. In fact, lying is encouraged if it “serves” the socialist agenda.

    And this explains the second part of the question. A bogus AND criticized “study” is published in a “low-rank journal” because it is a “study” that was “made to order”. The “purpose” of the “study” was to generate completely false “statistics” that could then be used by the propagandists (the media and the then president) to “push” a line condemning the American practice of gun ownership. As part of that effort, they (the propagandists) would be in charge of publicizing the “made to order” fake study after the stooge got it published in a second-rate journal that either lent itself or sold itself to the propaganda effort.

    This is completely in synch with the methodology used by the socialists to disseminate propaganda, especially propaganda based on outright lies. The process begins in “academia”. I use the term loosely. It actually refers to charlatans who are stooges of the people who pull the strings. They generate the “studies” and the “papers” that push a particular propaganda line and “document” them with bogus “facts” and “statistics” to create the impression that their garbage is somehow “valid”.

    These “studies” are then published in different venues. Some, with not so obvious falsehoods, can be published in “reputable” venues controlled by “academics” who share the socialist world-view. Others, especially those with outright unsustainable lies and fabrications, are published in second-rate journals who suffer little from a disclosure that they published (as in the case of the Lankford “study”) garbage. That is what they normally publish anyways, so nothing much changes.

    Once published, the media picks up the cue. The socialist politicians, notified in advance of what was coming, are also ready for their part of the dissemination effort. They “turn on the heat” and start pushing the line, “citing” the alleged “study” as the “source” and imbuing it with a false credibility. The idea is to use the disguised garbage to push a specific policy or legislative initiative.

    The next step is to activate the “street dogs”, the fascist “Sturmtruppen” or stormtroopers on the streets, to start generating violence and destruction based on the propaganda line unleashed by the media (and in this case, the then president). This is where the fascist lapdogs of Antifa, for example, and other similar organizations, come into play as part of the propaganda continuum.

    And so, we see the sequence: academia – media/politicians – fascist “Sturmtruppen”, all lined up by a democratic party that has been taken over by the “democratic socialists”, the new “cover name” for communists.

    Nothing is new under the sun. Same dog, different collar. Take away the swastika, replace it with a jackass. Same thing. Next!

  15. Jen Wainwright says:

    Perhaps Adam Lankford suffers from ADDS, thus we should all pray for his mental health. ADDS, Agenda Driven Data Syndrome, is becoming all too common. The possible reasons include poisoning by chemtrails, flat-Earth mentality, 5G exposure, or an overall dysfunction of one’s neurotransmitters as a result of uncontrolled hate and fear. 31% of ADDS sufferers are addicted to conspiracy theories. 31% of that 31% learn everything from Youtube “documentaries”.
    This data is based upon extensive research, which I am unwilling to share due to fear of retaliation by Illuminati.

Leave a Reply

Your email address will not be published. Required fields are marked *