Democrats and Republicans are reacting very differently to the coronavirus pandemic. And the divide goes far beyond whether to wear a mask.
A majority of Republicans say they would feel comfortable flying on an airplane, eating indoors in a restaurant or seeing a movie in a theater. Large majorities of Democrats and political independents say they would not.
Those findings, from a survey conducted in early July for The New York Times by the online research firm SurveyMonkey, show how opinions about the pandemic increasingly fall along partisan lines. Separate data on mobility shows the same partisan split in actual behavior — and it can’t be explained by differences in the prevalence of the virus itself.
“The degree to which Republicans are more comfortable than both Democrats and independents is quite jarring,” said Laura Wronski, a research scientist for SurveyMonkey. “It appears that people are living in quite different realities.”
That divide has implications for both public health and the economy. Public health officials have repeatedly urged Americans to cover their faces, practice social distancing and avoid large indoor gatherings. Economists across the ideological spectrum have echoed those messages, arguing that it will be impossible to restore the economy to health until the virus is under control.
But as mask-wearing, business closings and other public health measures have become politicized, virus cases have surged in much of the country, leading to a pullback in economic activity.
Republicans are less worried about the virus.
On many issues, like tax policy and health care, voters fall along a continuum, with Republicans on one side, Democrats on the other and independents falling roughly in the middle. On the pandemic, however, Republicans — and particularly conservative Republicans — stand on one side of a wide gulf.
Just 10 percent of Democrats said they would feel very or somewhat safe seeing a movie in a theater. A somewhat larger share of independents said they would feel safe, but they were still a distinct minority, at 24 percent. Among Republicans, the share grew to 55 percent.
The gaps were somewhat narrower, but still large, in places where the virus is prevalent.
Partisanship is the starkest divide, but not the only one. Women are generally more concerned about the virus than men, and Black, Hispanic and Asian survey respondents are substantially more worried than whites — not surprising, given that those groups have also been hit harder by the pandemic. Perhaps more unexpectedly, older Americans don’t report being significantly more concerned about the virus, despite being at substantially higher risk of death or severe illness.
There is a split in actions, not just words.
Given high levels of partisanship in American society, researchers sometimes worry that surveys won’t capture whether people’s actions conform to their stated views. For example, a staunch Republican, in order to signal support for President Trump, might report feeling comfortable going to an indoor restaurant while staying away in practice.
But partisan differences show up in real-world data, not just in surveys. Cellphone mobility data analyzed by the Federal Reserve Bank of Dallas shows that people in Republican-dominated parts of the country are leaving home more often and traveling around more than those in predominantly Democratic areas.
The partisan pattern holds up even controlling for the fact that Covid-19 cases were initially more prevalent in Democratic strongholds like New York and Seattle. The mobility data shows that people are being more cautious in areas that were harder hit by the virus or that have had recent flare-ups. But even in such cases, Republican-leaning areas show more activity, on average, than Democratic-leaning ones.
The divide is just as wide on the economy.
Confidence in the economy has dropped sharply among all groups during the pandemic, but the partisan gap is as wide as ever. In the latest survey, 60 percent of Republicans said they expected “continuous good times economically” over the next five years. That wasn’t just far higher than the 15 percent of Democrats who said the same in July; it was also twice as high as the 31 percent of Democrats who gave that answer in February, before the coronavirus began to inflict significant damage on the American economy.
In other words, Republicans are more optimistic about the economy now than Democrats were before the pandemic began.
“It just seems like there are two separate economies or two separate pandemics,” Ms. Wronski said. “There’s no way to reconcile these different experiences when you think about how there’s record unemployment or you hear about businesses that are closing.”
Economic optimism has improved somewhat since April among all partisan groups. But it was essentially flat in July compared with June. That’s consistent with other measures of economic progress, which have stalled recently as the virus has surged in many parts of the country and some states have begun to reimpose shutdown orders.
Behavior has become more partisan.
Public opinion on the coronavirus has never been immune from the partisan divide that characterizes so much of the nation’s discourse. In a survey conducted for The Times in early March, 77 percent of Democrats said they were worried that there would be a coronavirus outbreak in the United States, compared with 48 percent of Republicans.
Behavior was less starkly partisan early in the pandemic, however. The March survey showed that comparable numbers of Republicans and Democrats had taken steps like canceling travel plans, stocking up on food or working from home. And a survey in April found that majorities of Republicans and Democrats alike believed that the government response to the pandemic had been “about right” — although Republicans were more likely to say the response had “gone too far.”
About the survey: The data in this article came from an online survey of 6,155 adults conducted by the polling firm SurveyMonkey from July 6 to July 12. The company selected respondents at random from the nearly three million people who take surveys on its platform each day. Responses were weighted to match the demographic profile of the population of the United States. The survey has a modeled error estimate (similar to a margin of error in a standard telephone poll) of plus or minus two percentage points, so differences of less than that amount are statistically insignificant.