Why did face masks become a political statement?

First off, I need to tell you that our research is apolitical. That said, I think the mask as a political statement has a lot to do with the political divide in our country. Just about everything in our culture today becomes politicized. Once one party takes something seriously, the other side pushes back.

Is mask-wearing a partisan issue, according to your research? If so, can you explain why Democrats are more likely to wear them than Republicans?

It has become a partisan issue, and more so over time. But there are many other critical factors that were overshadowed by the media’s focus on politics. The actual data on this is complicated. What drives mask-wearing in any location are local issues affecting that area. So if a certain city or state is experiencing a spike in COVID cases, people will start wearing masks. Also, if an area is densely populated, more people will wear masks regularly.

At the time that we conducted our study back in May, the data seemed to indicate that local factors were affecting mask-wearing behavior more than politics. The composition of big cities with higher population densities tends to be more Democratic. Higher-density regions tend to have higher rates of COVID infection. Rural areas with fewer people, on the other hand, tend to be more Republican. What follows is that people who live in low-density areas with lower COVID-19 infection rates, who are more likely to be Republican, will be less likely to wear masks. People who live in crowded areas with more COVID cases will be more likely to wear masks. We showed that when COVID density is statistically controlled for, political party affiliation was no longer a significant predictor of mask-wearing.

What are the true predictors of who will wear face masks? Why?

People’s age is one predictor. The older they are, the more likely they are to wear a mask. Higher income, higher education and county population are all associated with a higher likelihood of wearing a face mask. However, when the density of COVID infection rates is taken into account, all of those predictors go away. COVID infection rates were the most important factor in May. 

When we noticed this trend, we became alarmed, because people who were living in areas less affected by COVID were being lulled into complacency. They felt safe and weren’t taking precautions—and this turned out to be dangerous. Areas where people felt safe were, in some ways, the most vulnerable. Consider Texas and Florida, where people initially felt safe and the rate of COVID-19 infection rose dramatically over the summer, probably due to lack of preventive health and safety precautions. Because people were so focused on a political explanation, that story was missed. No one looked at the most important issue driving mask-wearing behavior, which is where people lived. If people had been paying attention to that, the media could have amplified the danger. People felt too safe and those were places where they were in danger of developing major outbreaks and that’s what seems to have happened.

You also recently did another study of “shy voters.” Why don’t people tell pollsters their true preferences? Is this more prevalent among Trump or Biden voters? Why?

There are many reasons people don’t feel comfortable disclosing their opinions on phone polls, but it boils down to two main ones—fear and distrust. They are fearful of the potential lack of anonymity of the data being collected. They know the polling company has their phone number and voice recording and responses, so they believe the company can identify who they are by connecting their phone number and voice to actual responses and they don’t think their response is truly anonymous. Additionally, in today’s climate, when we read about company databases being hacked on a regular basis, they don’t trust that their data is secure.

Reflecting on the fear factor in our divided political world, there is an overwhelming amount of data that shows people are reluctant to share political opinions with neighbors, colleagues, friends or even family members. People self-censor and, for some, these issues spill over into concerns about telephone polls. As a result, some people tend to either avoid such polls or if they have to participate, they may not be honest.

In our study, we looked at over 2,000 people’s opinions about participating in telephone polls. About 10% of Democrats reported not being comfortable disclosing their opinions on phone polls compared to 15% of Republicans who reported that they wouldn’t disclose their opinions. Opinions that are on the political right are considered more politically incorrect and people who hold those opinions reported feeling more ostracized.

How does this affect media reporting of the election? Do you think there’ll be a surprise election outcome?

Both sides amplify whatever results support their position. People shouldn’t be lulled into complacency, thinking the outcome of the election is a foregone conclusion. All kinds of things could still happen and all kinds of factors that are not accounted for may come into play. People should continue to be energized and take voting seriously because there is still time to affect the outcome.

When people try to explain the outcome of the 2016 election, the notion of hidden Trump voters is commonly invoked. On the one hand, many scientists dismiss the idea, because it’s not easy to measure, and there are other measurable factors that are known to have played a role—like the undercount of voters without a college degree. On the other hand, “shy voters” are not hard to find. Anyone who doubts the existence of “shy voters” can ask 10 of their friends how they feel about participating in phone polls. In all likelihood, one or two will say that they intentionally avoid them due to confidentiality and other concerns.

We don’t know to what extent this can or will affect the election, but our research suggests that it’s a phenomenon that should be taken seriously and studied more. In highly contested states with razor-thin margins, anything can make a difference.