By Ryan P. Burge and Paul A. Djupe

I have to be honest, I think that COVID is one of the most fascinating polling topics in recent memory. One idea that I have been rolling around in my head is: how much of the death total do people blame on President Trump’s handling of the pandemic? I’m a political scientist, so I am well aware that party identification is the strongest force in the social world, but COVID may make it difficult for even the strongest partisan to place all the blame on Trump. To help answer that, we were lucky enough to put a question in the field two weeks ago where we asked about it. Here’s the question:

If Hillary Clinton were president during the pandemic, how many Americans do you think would have died from COVID-19 by this point? (for comparison, as of October 11, roughly 213,000 Americans have died of COVID-19)

Those taking the survey could respond by moving a slider with a range of available values from no deaths to 600,000 deaths. It’s probably a good idea to begin with the simplest cut at the data: political partisanship.

Democrats, on average, believed that if Hillary Clinton had won the presidential election in 2016, then 78,000 fewer Americans would have died from the Coronavirus. Independents did not have as much faith in a Clinton administration’s handling of COVID, only believing that 27,000 would have been saved by Clinton’s response. Republicans in our survey, however, believe that a Clinton administration would have bungled the handling of COVID so badly that 48,000 more people would have died compared to the Trump team’s policy response.

But, what about religion? I broke the sample down into religious traditions and again calculated the average number of hypothetical deaths under a Clinton administration. There are some results that comport with our assumptions about how religious traditions interact with political partisanship. But there are some surprises, as well.

Firstly, it makes sense that the religiously unaffiliated estimate a lot fewer deaths than other religious traditions given that vast majorities of the nones identify with the Democratic Party. At the top of the scale are non-denominational Protestants who tend to be politically conservative and are closely aligned with the Republican Party.

However, there are some surprises in this graph. Black Protestants on average, believe that 179,000 would have died from COVID-19 under Hillary Clinton’s watch, which is just 34,000 fewer deaths than have occurred while Donald Trump has been in the White House. Given that nearly nine in ten Black Protestants are set to vote for Joe Biden in 2020, it seems notable that this group doesn’t feel that a Democrat would have handled the situation much better.

The other finding that jumps out is the gap between evangelicals and mainline Protestants and Catholics. Clearly, these mainliners and Catholics are not as politically coalesced around President Trump as evangelicals, it’s worth considering that the latter two groups seem to be especially convinced that Trump bungled the response to a great degree. That may play a role in 2020 given their concentration in the Midwest battleground states.

How does the intersection of religion and politics impact people’s estimates of COVID deaths under President Clinton’s watch? To test that, I estimated a regression model with the number of COVID deaths as the dependent variable. I interacted church attendance, race, and political partisanship alongside controls for age, gender, income, and education. The results are below.

To be honest, I don’t really know what to make of these results. For instance, a white Democrat who never attends church predicted the number of deaths at about 90,000 under Clinton. However, for a white Democrat who attends church weekly, the predicted number of deaths doubles to 180,000. For white Republicans, as church attendance goes up, the projected number of COVID deaths goes down from about 280,000 among never attenders to 250,000 among the most frequent attenders.

For non-white respondents the effects are much more modest. In fact, church attendance is not statistically significant for Democrats or Republicans. Non-white Democrats estimate that about 150,000 people would have died from COVID with Hillary Clinon in the Oval Office. For Republicans, it’s much higher at around 250,000 deaths. For non-white political independents, church attendance does drive down death estimates from 200,000 to 140,000.

I have to say that these results do not comport with my assumptions. Because of the power of partisanship, I would have assumed that Republicans would have believed that hundreds of thousands more would have died under Clinton’s watch, while Democrats would have had much more optimistic expectations. The sense that I get here is of resignation: that a different person in the White House would have maybe lessened the death count, but not as dramatically as I would have supposed. It does seem that most people have realistic expectations about the ability of the government to solve the problem of COVID-19 in the American federal system where individuals have outsized notions of what freedom entails.

Paul A. Djupe, Denison University, is an affiliated scholar with PRRI, the book series editor of Religious Engagement in Democratic Politics (Temple), and co-creator of religioninpublic.blog (posts). Further information about his work can be found at his website and on Twitter.

Ryan P. Burge teaches political science at Eastern Illinois University in Charleston, Illinois. He can be contacted via Twitter or his personal website.