John Lott is, if not the most influential, certainly the most prolific “academic” in the gun debate. He has authored weekly columns in local newspapers on the horrors of gun free zones, published widely-distributed books on the ostensible benefits of right-to-carry laws, and his newest book The War on Guns has received rave reviews by prominent conservatives, like Sen. Ted Cruz (R-TX), Sen. Marco Rubio (R-FL), and Newt Gingrich.
Before Lott’s flurry of activity, it was difficult to find anybody arguing that widespread gun ownership made societies safer — even the NRA was reticent to make such a bold claim, defending gun ownership with reference to the constitution, not criminology.
But Lott’s recent successes belie a far more shadowy past. A little over a decade ago, he was disgraced and his career was in tatters. Not only was Lott’s assertion that more guns leads to more safety formally repudiated by a National Research Council panel, but he had also been caught pushing studies with severe statistical errors on numerous occasions. An investigation uncovered that he had almost certainly fabricated an entire survey on defensive gun use. And a blogger revealed that Mary Rosh, an online commentator claiming to be a former student of Lott’s who would frequently post about how amazing he was, was in fact John Lott himself. He was all but excommunicated from academia.
Despite his ethical failings, Lott rose from the ashes in the wake of the 2012 mass shooting at Sandy Hook Elementary School to once more become a prominent voice in the gun debate.
Perhaps unaware of Lott’s previous transgressions, or believing he had turned a new page by founding the Crime Prevention Research Center (CPRC), many in the media who were desperate for an authoritative, pro-gun academic voice seized on Lott’s credentials and provided him with a new platform. In the past few years, Lott and his organization have been cited by dozens of media outlets as an authority on gun violence statistics, including the Washington Post, the Wall Street Journal, LA Times, Politifact, CBS, CNN, Fox News, and many others.
However, the media’s newfound faith in John Lott is deeply misguided. Rather than turn a new page, Lott has instead returned to his old playbook and used his platform to deceive the public. Our own multi-year investigation into Lott and his organization has uncovered a startling array of new ethical violations, ranging from the profoundly bizarre to the outright fraudulent.
Here are just five of the most troubling incidents:
Last fall, Lott’s website proudly declared it published a study in a peer-reviewed journal. “CPRC Has New Refereed Publication in Econ Journal Watch: Explaining a Bias in Recent Studies on Right-to-Carry Laws” blared the headline on his website. A link to a downloadable copy of the paper also touts its place in the economic journal.
Having a study accepted in a peer-reviewed journal was a big win for Lott, boosting both his own reputation and that of the CPRC. After all, this would be one of the few publications in recent history that Lott dared subject to peer-review.
The only problem? The paper was never actually published in the Econ Journal Watch.
As the head editor of the journal explained to us, while Lott’s paper had initially been considered for publication, it was ultimately rejected. The issue of the journal Lott said he was published in has no trace of his paper. It is impossible for Lott to have not known his paper was in fact rejected, and it would have taken little effort to correct both the post on the CPRC website and the uploaded paper on SSRN. This is a clear cut case of fraud.
Lott often claims that there is no difference between the frequency of public shootings in Europe and the United States. This is unabashedly false — but he continues to spread the falsehood anyway.
In February, he made the claim before the Tennessee Senate. “Most people may not realize this, but the rate of mass public shootings in Europe is actually fairly similar to the rate in the United States,” he said. “There is no statistically significant difference there, either in terms of the rate or fatalities.”
A couple of months earlier, he said something similar to the Washington Post, which quickly highlighted that his analysis was quite different from that of other experts in the field. As the Post noted, while Lott said the per capita rates of mass shootings in Europe and the United States were approximately the same, another researcher found the U.S. rate to be five times higher. The Post explained that the gulf between the results was due to Lott and the other researcher using different definitions.
But there is an even simpler explanation for the differing conclusions: Lott wasn’t being honest about his own findings.
While Lott claims the per capita rate in the United States and Europe are approximately the same, his own data tables tell a different story. Accepting his data at face value, between 2009 and 2015, the United States had 25 mass shootings versus 19 in the E.U. and 24 in Europe as a whole. This comes out as a rate of .078 shootings per million individuals in the United States, .038 for the E.U., and .032 for Europe as a whole. The United States has more than double the mass shooting rate of the E.U. and Europe, directly contradicting Lott’s statements about his own data.
Further, Lott’s carefully crafted criteria to include an incident as a mass shooting is highly suspect. Lott goes to great lengths to exclude mass shootings that are the result of burglaries and gang violence, but he includes terrorist attacks. This choice means that while the Texas biker gang gunfight last summer is excluded in his statistics, the November Paris attacks, which accounted for more than one-third of Europe’s mass shooting fatalities, are included.
However, when scholars study these mass shootings, they frequently exclude terrorist attacks from the analysis, for much the same reason Lott excludes burglaries and gang violence: the motivations are different. When researchers use a more appropriate set of criteria, the chasm between the rate of mass shootings in Europe and the United States widens even further. Researchers can also include all incidents of mass shootings (regardless of motivation) or use complex statistical analysis to determine whether the mass shooting difference between the United States and Europe is significant. The result remains the same — the United States fares far worse.
All of these methods point to the same conclusion: even if Lott wasn’t lying about his own results, his analysis would still be deeply flawed.
In their paper “The Impact of Right-to-Carry Laws on Crime: An Exercise in Replication,” Carlisle Moody, a CPRC board member, and three co-authors examine the impact of right-to-carry (RTC) laws on violent crime and critique an earlier study by John Donohue and his colleagues.
Donohue and his colleagues had concluded that the most significant effect of concealed carry laws is an increase in aggravated assault, but Moody et al. reported that: “the most robust result, confirmed using both county and state data, is that RTC laws significantly reduce murder. There is no robust, consistent evidence that RTC laws have any significant effect on other violent crimes, including assault.” This result fits well with Lott’s long established hypothesis that concealed carry significantly decreases crime, and the authors interpret it as a direct repudiation of Donohue’s results.
But there’s just one problem. Moody and his co-authors misread their own analysis.
As Table 3 on page 7 (pictured below) clearly demonstrates, the increase in aggravated assault for county level data is statistically significant, yet is not bolded by the authors like all the other statistically significant findings. In statistics, a result is usually considered significant if there is a less than 5 percent chance that the result is due to random chance, meaning it has a “t-statistic” greater than 1.96. A significant result in turn means that the authors of a study can put a higher degree of confidence in their finding. As the table below shows, the “stat” for the “post-law trend” for “Assault” (highlighted with a red box) has t-statistics of 2.8 and 2.25 for the general and specific model respectively. Further, the result itself is a positive number, indicating an increase in assault.
Nowhere in the Moody paper does it explain why significant T-stats are un-bolded, and it remains undiscussed in the conclusion, despite the fact that it directly undermines the thrust of their entire paper. Ironically, their paper actually supports Donohue’s finding that RTC laws significantly increase aggravated assaults.
Had Moody and his co-authors reported their own results correctly, they would have been left with the puzzling conundrum of concealed carry laws both reducing murder and increasing aggravated assaults. This finding flies in the face of well-established criminological facts and indicates the paper is likely crippled by bad statistical modeling choices.
This isn’t the first time that Lott and his allies have pushed studies with convenient errors that make the results appear to fit their more guns, less crime hypothesis.
As Ian Ayres and Donohue described in a brutal takedown of Lott and his allies’ research, there were at least two previous cases where Lott used this tactic. The first time, Lott presented a series of graphs to the National Academy of Sciences, which David Mustard, one of Lott’s allies, then decided to include in a comment for a 2003 Brookings Institute book. When Donohue demonstrated the results were the product of fatal coding errors, Lott’s ally was forced to withdraw those graphs from the book. Also in 2003, Lott supported (and initially co-authored) a paper appearing in the Stanford Law Review by Plassman and Whitley that also appeared to support the more guns, less crime hypothesis. Again, Donohue proved that their results were based on coding errors, undermining the authors’ central claim.
Given the extensive history of Lott supporting erroneous research, one is forced to wonder whether Moody and his colleagues were influenced at all by the thank you note at the beginning of their paper: “The authors thank The Crime Prevention Research Center for its support.”
After the mass shooting at a gay nightclub in Orlando, Florida in June, Lott published a piece in which he wrote, “Since at least as far back as 1950, all but three U.S. mass public shootings (with more than three fatalities) have occurred in places where citizens are not allowed to carry their own firearms.”
This claim has been a staple for Lott, who has repeated it in various forms in numerous articles, usually phrasing it as areas “where citizens were banned from carrying guns.” To support his contention, Lott cites his own report analyzing different aspects of mass shootings.
However, what Lott repeats in public is quite different from what his report actually shows. While Lott’s public statements equate gun-free zones with areas that prohibit concealed carry, his mass shooting report expands the gun-free zone definition to include areas where Lott feels it might be difficult to obtain a permit or where there might not be many permit holders despite being able to legally carry. Indeed, Lott admits in the report that more than six mass public shootings in the past six years have occurred in areas that legally allow citizens to carry their firearms, a direct contradiction of his public statements.
And not only does Lott mischaracterize his own research, but the research itself is also filled with significant errors.
In October 2015, after a student at Umpqua Community College in Oregon opened fire in a classroom, killing nine others, the CPRC website immediately proclaimed: “Umpqua Community College is yet Another Mass Public Shooting in a Gun-Free Zone.” As evidence, Lott cited the student handbook and the fact that the campus guards were not allowed to carry.
However, while it is true that campus guards were unarmed, Lott’s claim that concealed carry was prohibited is definitively false. Public colleges in Oregon are prohibited from banning guns on campus, thanks to a 2011 state court decision. The Umpqua Community College student handbook also expressly states that there is an exception to the prohibition of firearms “as expressly authorized by law or college regulations.” This includes concealed carry permits.
“UCC was never designated as a ‘gun-free zone’ by any signage or policy,” Umpqua Community College spokeswoman Anne Marie Levis told Politifact shortly after the shooting. “Umpqua Community College does comply with state law by allowing students with concealed carry licenses to bring firearms on campus.”
Not only was Umpqua not a gun-free zone by policy and law, it also wasn’t a gun-free zone in practice. Multiple reports at the time revealed that there were several armed students on campus at the time of the shooting.
In June 2010, a gunman in Hialeah, Florida targeted his estranged wife who was working at the Yoyito Cafe-Restaurant, killing her and three other women before taking his own life. And again, Lott classified the shooting as taking place in a gun-free zone.
As Lott noted, under Florida law, guns are not allowed in establishments that primarily serve alcohol. As proof that this shooting took place in a gun-free zone, Lott argued that the Yoyito Cafe Restaurant was a popular destination for parties where alcohol was served, and because it primarily served alcohol, the restaurant was a gun-free zone.
That logic is absurd. Serving alcohol at parties is in no way indicative that an establishment is primarily devoted to selling alcohol. Even a cursory glance at the restaurant’s reviews clearly indicate that Yoyito is a small Cuban restaurant devoted to selling traditional dishes.
Furthermore, Lott completely ignores the pertinent Florida law regarding restaurants with bars. A letter from the concealed weapons division of the Florida Department of Agriculture clearly notes that the law is written in such a way as to “allow the carrying of firearms in restaurants or similar businesses that primarily serve food but that also happen to serve alcohol as well.” In other words, the serving area where patrons are dining in a restaurant does not constitute the part of the establishment primarily devoted to the sale and consumption of alcohol.” By law, the Yoyito Cafe was clearly not a gun-free zone at the time of the shooting.
“Dear Dartmouth, I am one of your students, I am being stalked, please let me carry a gun to protect myself” read the headline of a piece on Fox News in August 2014.
The first person account was a harrowing story about teenager Taylor Woolrich’s desperate attempts to escape and protect herself from a persistent stalker who was ruining her life. The article blasted Dartmouth for not allowing her to carry a gun, and noted that carrying a gun was the only way she could remain truly safe.
As a BuzzFeed investigation later revealed, Lott, who is neither a young female nor a stalking victim, was the one who penned the piece. Indeed, Woolrich’s article is almost a copy and paste rendition of a previous article published by Lott on the Daily Caller.
“It’s his op-ed… Word for word, except the chunks that match what’s said in my speech,” Woolrich later told BuzzFeed. “It’s not like John Lott held a gun to my head and told me to talk to the media… I wanted to talk to the media, if it could mean something positive. But I wanted to talk to the media about stalking.”
Despite reservations about her message being co-opted, Taylor agreed to have him help her write for Fox, worrying: “I don’t know if I should just say yes and not piss him off.” Eventually, Woolrich changed her number and completely broke off contact with Lott.
“I was trying to be brave and just speak up,” she said. “I didn’t realize I was being turned into an NRA puppet.”
While Woolrich may have been eager to share her story at first, this doesn’t excuse the fact that Lott wrote a first person narrative on behalf of someone else, using his own words. When a Fox editor later thanked Lott for the piece, Lott replied, “It was actually easier for me to write this in the first person for her than the way I had originally written it.”
This isn’t the first time Lott has written in the first-person female voice. Back in the early 2000s, Lott and his research were coming under increasing fire from the academic community. Mary Rosh, claiming to be a former student of Lott’s, rose to his defense in online chatrooms and comment sections. She praised Lott as the best professor she had ever had and took deep offense whenever somebody questioned Lott’s research. A few online commenters found her passion rather bizarre, consoling her: “I’m sorry if you’re taking this personally, but you are not John Lott.”
Except she actually was. A blogger matched Lott’s IP address with that of Mary Rosh, and a humiliated Lott was forced to admit that he and Mary were the same person.
As conservative journalist Michelle Malkin emphasized at the time, “Lott’s invention of Mary Rosh to praise his own research and blast other scholars is beyond creepy. And it shows his extensive willingness to deceive to protect and promote his work.”
Why does the media still rely on John Lott?
In an attempt to appear fair and balanced, news outlets have offered John Lott a platform to debate a subject for which there really is not two sides. Gun violence is decidedly uncontroversial among scholars: more guns cause more suicides, homicides, and accidents.
These are the arguments being made by serious academics in peer-reviewed journals from Harvard, Stanford, Yale, and Johns Hopkins. On the other side of the debate, you have John Lott, a handful of conservative academics on the board of the CPRC, Gary Kleck, and a few others.
Much like the public debate over climate change, the journalistic quest for neutrality in discussing gun control has led to a sacrifice of intellectual integrity and honesty. Over the past two decades, John Lott has routinely demonstrated an unwillingness to engage honestly in the gun violence debate. Lott is not a credible source, and it’s time the media stop treating him as such.
Update: Since the publication of this article, the description of the study Lott claimed to have published in the Econ Journal Watch has been corrected on the Social Science Research website. An archived version of the paper touting its publication in the journal is still available here. The news of the study has also been changed on the CPRC website, removing the reference of it being published in the Econ Journal Watch. The original headline touting this publication is still evident in the URL, and an archived version of this news on the CPRC website is still available here.
Evan DeFilippis is a master’s student at Princeton’s Woodrow Wilson School. He is currently a Summer Fellow at Harvard’s Government Performance Lab.
Devin Hughes is the founder of Hughes Capital Management, LLC, a registered investment adviser.
Evan and Devin write on gun violence issues at Armed With Reason.