Political partisanship has been on the rise in the United States and across the world. Many hesitate to reveal their political identity, and others view opposing parties more unfavorably than ever before. But how does partisanship impact behavior during a pandemic?
Yanna Krupnikov, professor in the Department of Political Science at Stony Brook, researches the potential for power in information in hopes of identifying the points at which new information can have the most profound effect on the way people form political opinions, make political choices and, ultimately, take political actions.
During the past year, she has done large surveys that look at how political beliefs shaped responses to the COVID pandemic. Her research, co-authored with John Barry Ryan, associate professor of Political Science at Stony Brook and researchers from Northwestern University, the University of Arizona and the University of Pennsylvania, found — as you might expect — that Republicans and Democrats responded differently to the pandemic. They took on different behaviors, had different levels of concern and preferred different policy approaches. At the same time, those differences were nuanced.
There was no COVID in prior years, of course, and Krupnikov said partisan difference is something that has been part of politics since large surveys of the American public became popular in the 1950s.
“Research suggests that these partisan differences may have grown in the 1990s as politicians of both parties became better ‘sorted’ — liberal politicians became more reliably Democratic and conservative politicians became more reliably Republican,” Krupnikov said. “This sorting created clearer political, partisan cues for the American public, which has further helped to establish partisan gaps.”
Krupnikov said that these gaps on issues do not emerge on their own, but require the aforementioned cues from political leaders.
“We found the biggest political divides between Democrats and Republicans who are ‘affectively polarized’ — Democrats and Republicans who have the most dislike for ordinary voters from the other party,” she said. “If you were a Republican who strongly disliked the other party, you were more likely to report low levels of concern about COVID and be opposed to most COVID policies. If you were a Democrat who strongly disliked the other party, you were more likely to support all COVID policies and have very high levels of concern about COVID. But Democrats and Republicans who had more neutral feelings about the opposing party actually behaved very similarly.”
Krupnikov’s research, in fact, found that in areas with high case counts, the differences between Democrats and Republicans do not even reach statistical significance. Partisans have similar rates of worry and similar support for policies.
“We focus on people’s willingness to follow politicians’ cues,” she said. “There are two types of cues that are important, cues of how similar partisans behave, and cues about how the other side behaves. Both are important, but the negative cues — ‘this is what the other side does’ — may actually be more important. This means that the messages politicians send, and how clearly they send those messages, are critical.”
Krupnikov said her research indicates that for partisans on either side who strongly dislike the opposing party, Donald Trump and the United States became one and the same.
“Once that happens, everything becomes extremely politicized,” she said. “But again, this is not the case for everyone, just the most polarized people.”
However, Krupnikov says it’s difficult to connect outcomes to one particular cause or condition.
“People are in informational environments where lots of different things are happening at once,” she said. “They’re embedded in networks of friends and family, and they bring their own beliefs and attitudes to every issue. It is difficult to point to one particular aspect or piece of information and suggest that this is the pivotal cause behind a phenomenon.”
Instead, the research suggests that people generally want to be part of their group and behave in ways that follow the norms of the group. The stronger a person’s connection to a group, the more likely they are to follow.
“If those norms are ‘we don’t wear masks’ and you really want to be part of the group and be very different from the opposite group, you would be more likely to follow the norms of the group and not wear a mask,” she said.
Krupnikov said the need to conform to group norms is a very powerful force in modern politics. She pointed to research by Stony Brook Political Science alumna and current assistant professor at the University of South Carolina, Elizabeth Connors ’19, suggesting that people are often uncertain about the political opinions they should hold. “They look for cues from politicians and social groups about the way they should behave in complicated political circumstances,” said Krupnikov.
The question becomes, if attitudes toward COVID and masks and vaccines are based on belief, how do you change someone’s belief? Krupnikov says that changing beliefs is difficult and complicated, especially when they are an entrenched part of someone’s worldview.
“Often there’s no one right way to interpret something. For any person it may be very contextual or specific to a certain issue,” Krupnikov said.
She added that research has long suggested a politicization of science. Her co-author on the COVID research, James Druckman of Northwestern, has been working on exactly that, arguing that because there is uncertainty in science, it gives political actors an entry into politicization and an opportunity to leverage scientific uncertainty to their advantage.
“People aren’t just denying scientific arguments because they decided to,” said Krupnikov. “They are denying scientific arguments because they received a political cue to do so. Public opinion is a function of political and social context, not an outcome that randomly emerges.”
Looking back at the evolution of the pandemic, an obvious question might be: If former President Donald Trump had immediately endorsed masks and shutdowns, would his base have followed, and would that have changed Democrats’ response to these same precautions?
“In our paper we suggest that different cues by Trump would probably have led to different behaviors from the Republicans,” she said. “There is research that suggests that Trump is a really important cue for Republicans. It’s more difficult with the Democrats because the Democratic politicians also provided very strong cues about protective behaviors. As long as Democratic politicians still provided those strong cues, it’s probably unlikely that Trump embracing precautions would have changed how Democrats in general behaved.”
Krupnikov notes, however, that this is largely speculative. The ongoing challenge is making health issues non-partisan. Unfortunately, she said that’s extremely unlikely as it’s impossible to prevent politicians from politicizing various events, including a pandemic.
“There are incentives in place for political leaders to send divisive cues to people,” she said. “In light of that, it’s difficult for politicians to make issues non-partisan.”
Public opinion, however, may be a different case. Krupnikov’s research indicates that the gaps between Democrats and Republicans, in terms of policy perceptions and concern for the virus, declined as cases increased.
“This suggests that people are responsive to the conditions around them,” she said. “This is not the case for everyone, but there’s evidence to suggest that people take what is happening around them into account despite partisanship. I think that news coverage would benefit from more nuance, rather than just a focusing on the gaps. In fact, coverage of gaps may actually lead to more division in the future.”
— Robert Emproto