A study published on Friday appears to confirm what news reports suggested long ago: President Trump’s campaign rallies were associated with a rise in violence when they came to town.
A city that hosted a Trump rally saw an average of 2.3 more assaults reported on the day of the event than on a typical day, according to the study, led by researchers at the University of Pennsylvania and published in the journal Epidemiology. The authors found no corresponding link between assaults and rallies for Mr. Trump’s Democratic rival, Hillary Clinton.
“It appeared to be a phenomenon that’s unique to Donald Trump’s rally,” said Christopher Morrison, a postdoctoral fellow at the University of Pennsylvania and the lead author of the study.
It may come as little surprise that the rallies were associated with increased violence, as the often volcanic clashes between Mr. Trump’s supporters and opponents were widely covered at the time.
In March 2016 alone, a Trump rally in Chicago was called off after violent clashes broke out, while an anti-Trump protester was punched at a rally in North Carolina and another was punched and kicked at a rally in Arizona. The following month, several Trump supporters were assaulted at a California rally.
Mr. Trump himself repeatedly seemed to endorse attacks on his detractors, too.
“Maybe he should have been roughed up,” he said of one protester who was reportedly punched and kicked in November 2015. “I’d like to punch him in the face, I’ll tell ya,” he said of another a few months later. He even offered to pay legal fees for his supporters if they became too aggressive.
The supporters also often aimed offensive and violent rhetoric at Mrs. Clinton, suggesting she be killed.
To determine whether those words and news reports corresponded with an actual shift in violence, the researchers compiled a list of 31 Trump rallies and 38 Clinton rallies held in cities with assault data available online.
They compared the number of assaults reported on the day of the rally to the number reported on the corresponding day of the week, for each of the four weeks before and after the event.
On a typical day, cities saw an average of 19.4 assaults, they found. On the day of a Trump rally, that number rose to 21.7.
The pattern held even when the researchers controlled for the influence of factors like population size, data sources and the day used for the comparison.
The researchers offered two explanations for the increase in assaults. Either they were the result of clashes at or near the rallies, or they occurred elsewhere in the cities after the aggressive mood on display by Mr. Trump, his supporters or his opponents had spread through “social contagion.”
There were some limitations to the findings, the authors noted. They may not apply to the rallies or cities that weren’t studied, and a greater police presence during the rallies may have made it more likely for an assault to be reported.