Far fewer Americans support political violence than recent polls suggest

A series of recent events has sparked alarm about rising levels of political violence in the U.S. These episodes include the assassination of political activist Charlie Kirk on Sept. 10, 2025; the murder of a Democratic Minnesota state legislator and her husband in June 2025; and two attempts to kill Donald Trump during the 2024 presidential campaign.

Some surveys have reported that a large number of Americans are willing to support the use of force for political ends, or they believe that political violence may sometimes be justified.

My research is in political science and data analytics. I have conducted surveys for almost 25 years. For the past three years, I have studied new techniques that leverage artificial intelligence to conduct and analyze interviews.

My own recent surveys, which use AI to ask people about why they give their answers, show that the surprisingly high level of support in response to these questions is likely the result of confusion about what these questions are asking, not actual support for political violence.

A failure to communicate

Why would multiple surveys get the answers to this important question wrong? I believe the cause is an issue called response error. It means that respondents don’t interpret a question in the way the researcher thinks they will.

As a result, the answers people provide don’t really reflect what the researcher thinks the answers show.

For example, asking whether someone would support the use of force to achieve a political goal raises the question of what the respondent thinks “use of force” means in this context. It could be interpreted as violence, but it could also be interpreted as using legal means to “force” someone to do something.

Such response errors have been a concern for pollsters ever since survey research began. They can affect even seemingly straightforward questions.

What did you mean by that?

To avoid this problem, I used an AI interviewing system developed by CloudResearch, a well-known survey research company, to ask respondents some of the same questions about political violence from previous surveys. Then I used it to ask what they were thinking when they answered those questions. This process is called cognitive interviewing.

I then used AI to go through these interviews and categorize them. Two short reports that summarize this process as applied to both polls are available online. These analyses have not been peer-reviewed, and the results should be considered very preliminary.

Nonetheless, the results clearly demonstrate that respondents interpret these questions in very different ways.

Nuance matters

For example, in my survey, about 33% of Democrats agreed with the statement that “use of force is justified to remove President Trump from office.” However, when asked why they agreed, more than 57% gave responses like this: “I was not thinking physically but more in the sense that he – the president – might need to be ‘fired’ or forced out of office due to rules or laws.” Still others were envisioning future scenarios where a president illegally seizes power in a coup.

Once you account for these different interpretations of the question, the AI only coded about 8% of Democrats as supporting use of force in violent terms under current conditions.

Even here, there was substantial ambiguity – for example, this type of response was not unusual: “The language ‘use of force’ was a bit too broad for me. I could not justify killing Trump, for example, but less extreme uses of force were valid in my eyes.”

Similarly, 29% of Republicans agreed that “use of the military is justified to stop protests against President Trump’s agenda.” However, almost all of the respondents who agreed with this statement envisioned the National Guard interceding nonviolently to stop violent protests and riots. Only about 2.6% of Republicans gave comments supporting use of the military against nonviolent protests.

Almost all those who agreed that use of the military was justified expressed thoughts like this: “I see the military coming and acting as a police force to stop or prevent the demonstrations that become violent. Peaceful protesters must be allowed to exercise their right to free speech.”

When is political violence justified?

Even questions that explicitly ask about political violence are open to wide interpretation. Take, for example, this question: “Do you think it is ever justified for citizens to resort to violence in order to achieve political goals?”

The lack of a specific scenario or location in this question invites respondents to engage in all kinds of philosophical and historical speculation.

In my survey, almost 15% of respondents said violence could sometimes be justified. When asked about the examples they were thinking of, respondents cited the American Revolution, the anti-Nazi French Resistance and many other incidents as a reason for their responses. Only about 3% of respondents said they were thinking about actions in the U.S. at the current time.

Moreover, almost all respondents stated that violence should be a last resort when all other peaceful and legal methods fail.

One respondent illustrated both problems with one sentence: “The (American) colonists tried petitions and negotiations first, but, when those efforts failed, they resorted to armed conflict to gain independence.”

A call for understanding

Even these numbers likely overestimate Americans’ support for political violence. I read the interviews, checking the AI system’s labeling, and concluded that, if anything, it was overestimating support for violence.

Other factors may also be distorting reports of public support for political violence. Many surveys are conducted primarily online. One study estimated that anywhere from 4% to 7% of respondents in online surveys are “bogus respondents” who are selecting arbitrary responses. Another study reported that such respondents dramatically increase positive responses on questions about political violence.

Respondents may also be willing to espouse attitudes anonymously online that they would never say or do in real life. Studies have suggested that “online disinhibition effects” or “survey trolling” can impact survey results.

In sum, my preliminary research suggests that response error is a substantial problem in surveys about political violence.

Americans almost universally condemn the recent political violence they have witnessed. The recent poll results showing otherwise more likely stem from confusion about what the questions are asking than actual support for political violence.