75

The Onion, a satirical newspaper, re-uses a headline each time a mass shooting occurs in the US:

“No Way To Prevent This,” Says Only Nation Where This Regularly Happens

Most recently for the shooting in Uvalde, TX.

This tradition has its own Wikipedia article, and has been covered by numerous other media outlets.

Similar claims have been made widely; for a prominent example, Sky News reporter Mark Stone questioned Sen. Ted Cruz about it in a widely-reported incident. The Sky News report includes a statement that

Between 2009 and 2018 there were 288 school shootings in the US, the next highest number was in Mexico where there were eight.

But no sources are cited, and in any event “school shootings” are a subset of the “mass shootings” that the Onion headline refers to, and that Mark Stone seemed to be asking after.

The Washington Post report on the interview notes that

Cruz spokesman Steve Guest told The Washington Post that Stone’s line of questioning surrounding mass shootings in the United States was incorrect. Guest pointed to research from the Crime Research Prevention Center, a pro-firearms nonprofit founded by former Trump appointee John R. Lott Jr., that claims “the U.S. is well below the world average in terms of the number of mass public shootings.”

Infuriatingly, the Post doesn’t analyze either claim: they only report that Sky News says one thing, Ted Cruz’s office says another, and that’s it. The response is fairly buried, and the labeling of that response as coming from “a pro-firearms nonprofit founded by [a] former Trump appointee” certainly suggests that the Post sides with Sky News on this, but it’s not backed up. And while the conflict of interest gives a strong reason to be skeptical of the claim, it isn’t in itself a reason to assume that it is false—it’s a reason to dig deeper.

So I’m asking this site to help do that: Is America exceptional in how frequently it experiences mass shootings? Or is it “well below the world average in terms of mass public shootings”? If the difference comes down to definitions—as I suspect it does—I’d like to see analysis of those definitions: does one include incidents that don’t match what the public thinks of when they hear “mass shooting”? Or does the other exclude things that the public would include?

KRyan
  • 1,530
  • 2
  • 12
  • 20
  • 6
    Isn't this a duplicate question? – GEdgar May 27 '22 at 17:08
  • To clarify, are you trying to determine frequency as measured by discreet shooting events, or as measured by deaths (or deaths per capita) from a mass shooting event? – GOATNine May 27 '22 at 17:28
  • 2
    @GEdgar I thought it would be; I tried to find something on this before asking myself. I’d be happy for this to be closed as a duplicate if the question’s been asked and answered before. – KRyan May 27 '22 at 17:42
  • @GOATNine I’m trying to determine the accuracy of these mutually-exclusive claims, where one was made in response to the other. If the discrepancy is explained by competing definitions of what frequency is being measured, then that would be an answer. I would certainly hope that such claims are being made with awareness of different population sizes. – KRyan May 27 '22 at 17:45
  • I recall arguments about whether to include war zones and things like that. Maybe that one was ultimately deleted as impossible to clarify. – GEdgar May 27 '22 at 17:46
  • 11
    @GEdgar If the answer is “They’re both true, because one excludes war zones and the other includes them,” that would be a valid answer. War zones would definitely be something that I don’t think the public has in mind when we make comparisons of the US’s issues with gun violence to elsewhere, personally. – KRyan May 27 '22 at 17:47
  • 6
    Just FYI. Skeptics SE requires that claims come from a notable source. The Onion is a satirical website, posting parodies of news stories. Everything it writes is a joke. It does not claim that ANY information is factually accurate. You need a better source for your question. – Pete May 28 '22 at 01:14
  • 40
    @Pete The Onion headline is a notable, oft-quoted source of a widely-made claim. It’s also not the only source in the question, e.g. the Sky News article. – KRyan May 28 '22 at 02:41
  • 27
    @Pete Satire can be based on truth, it's reasonable to question that basis. This particular headline expresses a frustration that many people feel after every mass shooting event. – Barmar May 28 '22 at 15:46
  • 4
    Does this answer your question? [Does America not lead the world in mass shootings?](https://skeptics.stackexchange.com/questions/42384/does-america-not-lead-the-world-in-mass-shootings) – Robert Columbia May 28 '22 at 15:51
  • 3
    @RobertColumbia Well, it half-answers the question, in that the study that Cruz’s office points was addressed. But the answer states that the thing we really want to know is beyond the scope of that question. If so, then no, that question is not a duplicate, because it would not be beyond the scope of this one. – KRyan May 28 '22 at 16:39
  • 3
    The source of Sky News's figures appears to be CNN analysis: https://edition.cnn.com/2018/05/21/us/school-shooting-us-versus-world-trnd/index.html – dylan-myers May 29 '22 at 14:48
  • @StianYttervik knife attacks in schools are uncomfortably common in China, but they regularly result in either zero deaths or a small fraction of the victims dying (not to downplay how horrible those deaths and the nonlethal outcomes are though, just a lot better than someone doing the same with a gun) – llama May 30 '22 at 19:22
  • 6
    Just an anecdotal data point: I'm 52 and I don't remember of anything that could be deemed a "mass shooting" in schools here in Italy. I interpret "mass shooting" as when a group of one or more individuals enter a school with the intent of harming teachers or pupils with firearms. In the 70s there were attacks by political groups in universities, but these were politically-driven and the targets were not indiscriminate, but opposite political groups of students (fascists vs. communists). – LorenzoDonati4Ukraine-OnStrike May 30 '22 at 20:30
  • It looks like Skeptics has met its match with this one. The answer all comes down to how you distinguish between terrorism and a plot to randomly kill a large number of people due to their race, religion, or where they go to school. – Mike Serfas May 30 '22 at 22:50
  • It would be worth adding a supplemental question: is USA the only land where, when a mass shooting happens, a large number of people start suggesting it didn't really happen, the victims are actors, etc, etc – RedSonja Jun 01 '22 at 09:09

3 Answers3

92

Ted Cruz's spokesman cited a non-peer-reviewed analysis by John R. Lott. Lott says he is trying to debunk a study by the criminologist Adam Lankford, which the Washington Post's fact checker describes as follows:

Lankford conducted a statistical analysis of the total number of public mass shooters per country from 1966 to 2012 in 171 countries and controlled for the national population size. He said his data showed the United States had significantly more mass shooters, with 90 between 1966 and 2012, compared with 202 in the rest of the world.

Lankford did not include terrorist attacks in his definition of mass shootings, while Lott did.

When The Fact Checker spotted the Mumbai attack on Lott’s list, we asked Lott to remove terrorism cases from the totals for the four countries listed by Lankford — the Philippines, Russia, Yemen and France. Our hope was to provide as much an apples-to-apples comparison as possible.

Without terrorism cases, Lott’s count of shooters fell dramatically. In the Philippines, the number of shooters fell from 120 to 11, in Russia from 65 to 21 and in Yemen from 65 to 3. Only France did not have a significant decline, going from 5 to 4.

Furthermore, the Washington Post determined that most of the remaining incidents can also be described as militia or terrorist group activity, not lone gunmen. Lankford himself also made this point in his response to Lott.

However, a 2019 academic study found that various definitions of "mass shooting" can lead to vastly different outcomes, even simply using US data.

There is great potential for media reporting bias in mass shootings. People who claim that a mass shooting occurs almost every day of the year are correct only by the standards of Gun Violence Archive. Individuals against the movement toward more comprehensive gun legislation would be more inclined to use the Mother Jones mass shooting data to endorse the rarity of such events, and therefore the lack of urgency needed in mass shooting prevention. Neither of the groups would have to manipulate data to fit their message – they simply need to choose the database with the definition that best fits their agenda. In this way, the absence of a standard mass shooting definition undermines high-quality research and reporting in a field that has been highly politicized.

My conclusion from skimming these articles is as follows:

  1. If the definition of "mass shooting" is broadened to include activity by criminal gangs, terrorist groups, and militias, then the United States is not unusual, even if we are talking about deaths of uninvolved civilians.
  2. If the definition of "mass shooting" is narrowed somewhat to individuals shooting multiple uninvolved civilians on their own agency, it is very hard to quantify the data and there is a lot of disagreement. This includes gangland shootings, for instance. As Lott points out in his article, such incidents are not always reported in foreign news media, especially in countries like China.
  3. If the definition of "mass shooting" is narrowed further to individuals planning events with maximum murder capacity such as school shootings or the Las Vegas shooting, it does seem like the United States stands out; this is closer to Lankford's definition.

I'm not a criminologist and I welcome suggestions to modify this answer.

Avery
  • 44,313
  • 16
  • 183
  • 179
  • 1
    Comments are not for extended discussion; this conversation has been [moved to chat](https://chat.stackexchange.com/rooms/136686/discussion-on-answer-by-avery-is-america-the-only-nation-where-this-a-mass-sho). – Oddthinking May 30 '22 at 03:52
18

Snopes.com had an article on this subject in 2018, talking in particular about Lott's CRPC study mentioned in the OP. Besides talking about the issues in defining mass-shooting and the corresponding variability in the data already discussed in in this answer, one aspect they talk about is the use of mean vs median as measure of average. The mean is notoriously sensitive to statistical outliers, as a consequence small countries where one large mass shooting happened in the period looked at by Lott (like Norway) score very high on the "mean deaths from mass shootings per year per million people" metric. The median is a measure of average that is much less sensitive to statistical outliers.

In the Snopes article they calculate the median number of deaths to mass shootings per year from the same dataset used by Lott (thereby sidestepping the definition issue for the moment). They find that according to that dataset, the USA is the only country in the "world" to have a median number of deaths to mass shootings per year in the period 2009-2015 that is not zero, coming in at 0.058 per Million people per year. Said differently, all other countries have 0 mass shooting deaths in more than half of the years. In this sense the US is the "Only Nation Where This Regularly Happens".

So, we see that somewhat paradoxically based on this one dataset (and ignoring any potential issues with it) we can simultaneously conclude that:

  1. “The U.S. is well below the world average in terms of the number of mass public shootings.”
  2. "The US is the only country in the world where mass shootings regularly happen."

Of course, by looking at the median the data does get skewed by the fine-graining of European countries which individually are small enough not to have any mass-shootings in most years. We can of course look at Europe as a whole to counter this effect (since Europe has a larger population than the US this should skew things in the other direction. The median number of deaths to mass-shootings in European countries per year according to Lott's data is 21, versus 18 in the US. Normalized by population this gives 0.028 vs 0.058 per million per year. So while, deaths to mass-shootings are more common in the US than in Europe, this data does imply that it is unfair to say that the US is the only place in the world where this regularly happens.

TimRias
  • 3,677
  • 2
  • 14
  • 16
  • On the flip side if you classify terror attacks properly and remove them from the mass shooting list the numbers in the US would likely drop as well. Some of the recent attacks could easily be considered terror attacks. – Joe W May 31 '22 at 14:32
  • 3
    @JoeW I have checked and it does not make that much of a difference in this case. So I have removed that remark. – TimRias May 31 '22 at 16:36
9

Many of these lists are not per capita. They ignore that USA has a population roughly half of all Europe (not to mention that it's considerably more diverse in most places).

For example, using the table from a recent The New York Times article, America's Gun Problem of mass shootings (public shootings in which four or more people were killed) from 1998-2019:

Graph of mass shootings

Notice that Finland (for example) had 3 where USA had 101. However, Finland population is 1/60th of USA. Adjusting per-capita and this statistic seems to show that mass shootings in Finland are relatively (based on population) 74% more frequent than in USA.

Czech Republic, New Zealand, and Slovakia all work out to around %70 of USA frequency. Some of these countries have less than 40 years of history, allowing for the possibility of further extrapolation.

Wikipedia considers that there were seven major news-worthy mass shootings in Europe last year. (In the same time-span, The Onion reported their headline 4 times.)

(Wikipedia considers that there were four major news-worthy mass shootings in Europe during 2020. The Onion only got to use their headline one time that year. The respective numbers for 2019 were four for Europe and 3 for USA, giving a three year total comparison of 15 (Europe) to 8 (USA). (Trying to draw some equation between the most newsworthy types of shootings.)

Using the frequency of The Onion headline occurrence itself as a guideline, then one would have to say that the USA does not actually experience regular mass shootings, but that they do occur, on average, a few times per year. Between 4 August 2019 and 23 March 2021 (more than 18-month period), the headline only found a single opportunity for invocation. (However, on 4 August 2019, it was used twice!) And before the Buffalo shooting for an entire year, from 26 May 2021 to 16 May 2022, there was no headline from The Onion in that time.)

Some other points for perspective:

  • In 2015, militants killed 174 students in Kenya

  • In 2014, militants killed 145 (mostly children) in Pakistan

  • The above TNYT article explains:

    Most shootings in America never appear in national headlines. The majority of gun deaths in 2021 were suicides. Nearly half were homicides that occurred outside mass shootings; they are more typical acts of violence on streets and in homes (and most involve handguns). Mass shootings were responsible for less than 2 percent of last year’s gun deaths.

    That includes suicides which are almost 2/3 of the gun deaths.

A couple other links that examine these questions:

https://www.maciverinstitute.com/2022/05/debunking-every-major-mass-shooting-myth/

https://worldpopulationreview.com/country-rankings/mass-shootings-by-country (Using the map at the top, can be seen that Finland, Switzerland, Slovakia, Czech Republic, France, Norway, Serbia, Albania, Belgium, exceed USA in "per capita mass shooting deaths by 2022" - sources provided there. Of all the countries included on that map, I see only Italy and UK with numbers in that category that are an order of magnitude lower than USA.)

fred
  • 178
  • 7
  • Comments are not for extended discussion; this conversation has been [moved to chat](https://chat.stackexchange.com/rooms/136687/discussion-on-answer-by-fred-is-america-the-only-nation-where-this-a-mass-shoo). – Oddthinking May 30 '22 at 05:33
  • 7
    So your answer uses the number of entries in the Wikipedia category for Europe, and compares it to the number of Onion headlines. Wouldn't it be a better comparison if you also used the number of entries in the Wikipedia category for the United States instead? Then, you'd be sure to compare apples to apples ([seven apples](https://en.wikipedia.org/wiki/Category:2021_mass_shootings_in_the_United_States) to [24 apples](https://en.wikipedia.org/wiki/Category:2021_mass_shootings_in_the_United_States), to be precise). – Schmuddi May 31 '22 at 17:27
  • 5
    One of the problems with this answer is that "newsworthy" is country dependent - many shootings in the US barely making local news would be national or international news in Europe, exactly because mass shootings are so much rarer there... – Denis Nardin May 31 '22 at 21:16
  • @Denis Nardin It seemed to me that the opposite was true. That we have better data for USA than we do for foreign countries. – fred Jun 01 '22 at 00:46
  • @Schmuddi I could also have used The Onion to answer the question all by itself. "This" in the phrase "this regularly happens" refers to the kinds of events that The Onion chooses to affix their headline. It happens a few times per year, on average – fred Jun 01 '22 at 00:50
  • 5
    Personally, I believe that the frequency with which a satirical newspaper uses a particular headline is a very poor measurement of anything, so I don't think that your answer would be improved if you actually did what you imply. I like my suggestion more, to be honest. Oh, incidentally, the USA _is_ a foreign country to many, if not most people in the world. Just saying. – Schmuddi Jun 01 '22 at 05:48
  • 3
    Of the seven wikipedia incidents you reference, only three are outside the former soviet union and one of those didn't involve guns. Total western european deaths: 13 (including the non-gun attack), substantially less than the *single* US incident at Uvalde. What was your point again? – matt_black Jun 01 '22 at 13:20
  • @Schmuddi Yeah, but it's the simplest, most literal answer, considering the framework of the question. (That's what I started with.) (It's the actual answer - "THIS".) – fred Jun 02 '22 at 06:24
  • Comparing USA to most other countries is a false comparison. Comparing USA to Europe, or at least to half of Europe, may be a more valid kind of comparison. However, mass shootings are so uncommon that it is possible that no direct comparison is legitimate. – fred Jun 02 '22 at 06:34
  • @matt_black That's only for one year. (Admittedly, ) See TimRias answer where he actually does what I wanted to do, compares Europe (as a whole) with USA. – fred Jun 02 '22 at 06:36
  • 2
    You quote only one metric from your final link which the report criticised for statistical reasons. Their alternative metric had the USA at the top. Not least because the USA had events every year whereas most other countries had only one or two in the 7-year period. – matt_black Jun 02 '22 at 16:08
  • @matt_black That's from the map at the top of the page in that link. The USA is so much larger than these other countries, that relatively, if they have one event in 7 years, it is more per capita than if USA had 7 in one year. (Indeed, this is the main point of my answer.) It's misleading to show a list with USA at the top and a triple digit number, followed by a string of single digit numbers from countries with populations orders of magnitude less. – fred Jun 02 '22 at 16:46
  • @matt_black I think that comparing USA to most other countries is a false comparison. Comparing USA to Europe, or at least to half of Europe, may be a more valid kind of comparison. However, mass shootings are so uncommon that it is possible that no direct comparison is legitimate. – fred Jun 02 '22 at 16:47
  • @fred My point was that the metric used in the map was one that was criticised as being unrepresentative later in the article. It is the metric that is the problem, not the list of comparison countries. – matt_black Jun 02 '22 at 16:49
  • @matt_black map says "up to 2022" ? (In any event, even that is just one example, as you can see, similar results from other lists, like the one in TNYT or from CNN. There are many such similar.) (Also, I think that the list of comparison countries IS a problem.) – fred Jun 03 '22 at 13:03
  • @matt_black If we make a list of Western Hemisphere countries, also, for example (our closest neighbors), then USA also does not stand out. – fred Jun 03 '22 at 13:10
  • @fred The label on the map is clearly wrong. The Norway total is from a single incident that happened in 2011 and has not repeated. The French total includes a carefully planned, multi-site, multi-shooter terrorist attack. Averaging the deaths over multiple years does not represent the "typical" occurrence of these events in other countries. And using the mean rather than the median as the metric seriously misrepresents "typical". The alternative metric used further down the article show the USA at the top on a better metric... – matt_black Jun 03 '22 at 21:21
  • 1
    @fred And the list of comparison countries IS a problem, I agree. We certainly should not include either Russia or SErbia in typical western developed or democratic countries. The overall point being that the USA **does** stand out unless you choose odd metrics and dodgy comparison countries. – matt_black Jun 03 '22 at 21:23
  • @matt_black It's not a fair comparison. European countries are much smaller than America, and also generally less diverse. They are more analogous to states within the USA, for this purpose. Comparing USA to Europe as a whole gives more comparable results. USA doesn't stand out. – fred Jun 04 '22 at 20:54
  • @matt_black Other aspects of the map seemed to me to have been updated. I don't think that the map is mis-labeled. – fred Jun 04 '22 at 21:03
  • @fred you are completely ignoring the *specific* issues I pointed out with the map (it uses a bad statistic and clearly milabels its time period). The later part of the *same* source points this out and offers an alternative statistic that puts the US at the top of the list. Plus the top 6 western countries in Europe have a total population *larger* than the US: that isn't "much smaller". If you don't like the other conclusions in your own source then don't quote it just because you like the first map. That is twisting the source for motivated reasoning. – matt_black Jun 05 '22 at 12:06
  • @matt_black The map is there, and it is used. You can choose to use other arrangements of data. The map is representative of other lists that are often used to demonstrate the point. The map on that page happens to be at the top of it, is labeled as current, and seems to be updated (at least a little). It's the last item in my answer. Another has already answered (since mine), comparing as I suggested. That source is also more evenhanded than you present. The critique is not specifically about the map at the top of the page. The body concerns a similar arrangement of data and they explain it. – fred Jun 11 '22 at 15:58
  • @fred I'm sorry to have to point this out. But your argument seems to be that "I Like the conclusion the map shows and so do others so we will ignore criticism of the metric in the *same source*" That Is motivated reasoning not a skeptical argument. And the issue isn't about updates or currency, the map is clearly misleading on critical statistics (I've Mentioned the problem with norway before). You are simply trying to ignore criticism that undermines your source, not dealing with the criticism using evidence (even from *the same source*). – matt_black Jun 12 '22 at 20:58
  • @matt_black Of course, I have similar thoughts about your interpretation. My thesis was that assertions showing lists of simple frequencies per country do not reflect population differences. This page has a very nice PER CAPITA map, apparently current to 2022. (That's why it was relevant and I included it.) – fred Jun 13 '22 at 21:54
  • @matt_black The page on worldpopulationreview.com showcases, HEADLINES, the map, which would indicate that this is their primary answer. The writers are surely aware that this is what most will look at. To be fair, the central statement of the article is "Does the United States have more mass shootings than other countries? It depends on the data." I suppose that any reader can look at the article and decide for themselves; fortunately, I linked it, to make it easier. – fred Jun 13 '22 at 21:54