Nathan Howard/Getty Images

Americans, including political scientists and pundits, are probably overestimating how much their political opponents support partisan acts of violence, a new study from researchers at Stanford University, Dartmouth University, and University of California, Santa Barbara suggests.

After conducting three large survey experiments, the authors found that previous research indicating that up to 44 percent of the American public approves of politically-motivated violence in hypothetical scenarios has vastly overestimated the situation. That estimate may be somewhere between 30 percent to 900 percent too high.

In reality, the authors contend, nearly all respondents, whether Democrat or Republican, support criminally charging suspects who commit acts of political violence. That seems to be more in line with real world data — only a little more more than 1 percent of hate crimes in the United States are politically motivated.

The disconnect likely stems from the way other surveys have been conducted, Dartmouth's Sean Westwood writes. For instance, some of the questions are unclear about what "violence" means and fail to distinguish between violence and political violence. The surveys are also sometimes designed so that most of the available answers indicate support for political violence, which means it's likely that disengaged respondents may overstate their approval.

To gain a more accurate reading, the authors suggest measuring support for political violence by providing respondents with specific examples of such actions, benchmarking "results against general support for all violence," and capturing "support for crimes that vary in severity."

Westwood did clarify that the authors were not arguing political violence is a nonexistent threat; on the contrary, they believe vigilance is necessary. But for that vigilance to be successful, it's important to have a more accurate understanding of the public's attitude. Check out the research here.