In any group of several thousand people, you will be able to find some distinctions — people of different races, religions, ideologies, even in what might otherwise be a sea of homogeneity. In some crowds, though, that’s less the case — such as the crowd that stormed the Capitol on Jan. 6. There, the uniformity of gender (male) and race (White) were surpassed only by the uniformity of ideology: supportive of Donald Trump and his false claims about the 2020 election.

New analysis from researchers at the University of Florida highlights an interesting aspect of the people from that state who had been charged with federal crimes related to the violence that day. They were not only not representative of Republicans in the state overall demographically, but they were also not representative because they were more likely to have voted in recent elections.

Sara Loving, a student at the university, worked with Professor Daniel Smith to compare the accused with the party overall. Using USA Today’s database of Capitol riot arrestees, she compared those from Florida with a state voter file. In most cases, the suspects were registered to vote. Then she categorized them by race, gender and age.

As might be expected, those arrested in the riot were far more likely to be male than registered Republicans in the state. The party is mostly White in Florida, but even given that, the alleged rioters were more likely to be White. They were also younger; Republicans in Florida skew heavily older, but the Jan. 6 suspects did not.

The more remarkable finding from Loving and Smith is that the riot suspects were more likely to have voted in each of the past three general elections in the state and in the 2020 primary. Across the board, the suspects turned out to vote at higher rates than Republicans overall.

There’s an important caveat here: We’re talking about a relatively small group of people. About 60 riot suspects were included the analysis, meaning that small changes in numbers lead to big changes in percentages. It’s also too few to be considered representative of Florida overall — but of course, they are entirely representative of the group of Floridians charged with participating in the storming of the Capitol, which is the point.

Shortly after the riot, CNN did a similar analysis, finding that many of those allegedly involved in the Jan. 6 riot hadn’t cast ballots in 2020. This seemed to reinforce that many of those involved were focused on operating outside of the political system, which the day’s violence obviously represented. At that point, far fewer people had been arrested.

The research from Loving and Smith offers a more complex picture. Instead of being suddenly compelled to action by Trump’s false claims about the election, this suggests that those suspects were mostly politically engaged all along. At the risk of reading too much into these small numbers, the increased participation in the primary in an election year in which the Republican presidential nominee was a fait accompli seems particularly telling.

In recent polling, three-quarters of Republicans said they thought President Biden’s win was not legitimate. In other words, the view that motivated the rioters Jan. 6 was itself not a fringe view but a common one. And at least for the cohort of alleged rioters from Florida, accused participants were more fervently political than their peers, not less.