Polling and market research are twin disciplines. Polling is a crucial way of understanding what the public thinks. This is obviously central to developing any strategy to give the voters what they want. Yet, we all know that asking people questions can be tricky. We need to get questions right or else we risk supplying an incorrect view of the public. Given how important question design is, it is useful to understand what market research can teach politics.
The US Public Worries About Political Violence
The US public says that the threat of political violence is a scourge in the US. A Pew Research poll just before the US election (October 2024) highlighted that two thirds of the public thought it was a major problem. It seems very reasonable to worry. Recently we have had, amongst other things, the murders of Charlie Kirk and Minnesota House of Representatives Speaker and her husband. We had seen the attack on Paul Pelosi and attempted assassinations of Donald Trump.
It is a bipartisan problem; Republicans and Democrats have been killed. The US has plenty of disaffected people, as well as challenges treating mental illnesses, and lots and lots of guns. A small number of people, even a single terrible person, can cause untold suffering.
What Market Research Can Teach Politics
One thing we shouldn’t do is worry that everyone in the US is itching to murder their political opponents. This is where an understanding of survey design can help. Nate Cohn has an excellent piece on this topic. He explains the challenge in understanding the public’s support for political violence. The challenge is a familiar one in survey-based market research. Words mean different things to different people and at different times.
But while a small percentage of Americans do endorse violence, that percentage is much smaller than either the polls or social media make it seem. Only a vanishingly small number of Americans support political violence in any meaningful sense.
Nate Cohn, 2025
Are The Polls Wrong?
The polls aren’t wrong per se. The challenge is that support for political violence is really hard to judge, as are many concepts in survey research. We have the results of the question but don’t really know what the results mean.
But support for political violence is not easy to quantify. The measurement issues are so serious that The New York Times/Siena Poll has never once asked about it over the last decade (plenty of people have asked us to do so). The usual poll questions — for instance, “Do you think it is ever justified for citizens to use violence to achieve political goals” — aren’t quite up to the task.
Cohn, 2025
Words Are A Real Problem
There is a lot happening in a question like, “Do you think it is ever justified for citizens to use violence to achieve political goals”? As such, we don’t know what the respondents thought they were answering.
- Do they mean US citizens? Citizens in a democracy? Citizens in a dictatorship? Does the distinction matter?
- What does justified mean? Does it mean you would do it personally? You sympathize with those doing it? Does justify mean that you acknowledge that people have justifications for being upset, but you don’t accept their reasons for violence?
We haven’t turned to the big question of what people heard when they were asked about the word violence?
- Does violence just mean murder?
- Does violence mean actions in the moment to stop people actively attacking the vulnerable?
- Does it mean fighting the Nazi’s in the second world war?
- Does it mean the US War of Independence?
- Is destruction of property violence?
A while ago “silence is violence” became a popular slogan on the left. That always seemed a bit odd. In my life I have been punched and kicked, which I think of as violence, and not stood up for, which I do not think of as violence. Silence and violence always seemed quite different to me but what I think doesn’t matter. It is what the people who answer the question think. Do people when asked think violence is harsh words?
What words means to the respondent really matters to the meaning of a survey.
Surveys Of Opinion Are Tricky
It is really hard to understand what the public think. We can’t really tell what question the public thought they were answering. That said, I think we can be confident that the vast bulk of people in your neighborhood don’t want to murder each other over political differences. Yes, they are serious problems but there is also hope.
For more on surveys and important questions What Caused The Classic Polling Disaster?, Survey Methodology And The Future Of West Ham United, People Aren’t Getting Worse, and Does Doing Good Generate Loyalty?
Read: Nate Cohn (2025) Why Support for Political Violence Is Not as High as It May Seem: A tiny percentage of Americans support it in any meaningful sense, despite what some polls may show, The Tilt, New York Times, Sept. 20, 2025
