In a positive sign for Joe Biden days away from the presidential election, likely voters who preferred a candidate other than Donald Trump or Hillary Clinton in 2016—about 7 percent of all likely voters—are breaking three to one for the former vice president, according to a new study by researchers from Northeastern, Harvard, Northwestern, and Rutgers.

The key driver of these voters’ choice: whether they believe that the federal government took the pandemic seriously enough.

Photo by Adam Glanzman/Northeastern University

Among the voters in this group who now favor Biden, 87 percent say the federal government didn’t take enough action against the coronavirus, compared to only 11 percent of those who favor Trump. 

Those voting for Biden also reported much higher adherence to public health recommendations, the survey found. And they had much higher trust in Dr. Anthony Fauci, a leading member of the White House coronavirus task force, whom Trump has publicly challenged.

The survey also found that voters who supported Trump in 2016, but now favor Biden,  outnumber Clinton-to-Trump switchers two to one, netting roughly a 2 percent margin shift in the Democratic nominee’s favor. Again, the key factor is belief that the federal government inadequately responded to the pandemic.

“That could easily be the margin between winning and losing the election,” says David Lazer, University Distinguished Professor of political science and computer and information sciences at Northeastern, and one of the researchers who conducted the study. 

Since April, the researchers have tracked nearly 80,000 likely voters online to gauge how their choice for president may have changed over time. Nearly 20,000 people took the survey more than once. 

Voters’ preferences remained largely stable. To the extent that their preferences changed, the shifts could be largely predicted by the subjects’ public health-related behaviors and beliefs on whether the federal government had acted aggressively enough on the pandemic. These switches yielded a small uptick in support for the Democrat.

“Across all of these analyses, we see people assessing that we should have responded more aggressively, and that’s driving a shift toward Biden,” says Lazer.

The findings could be a warning sign for the president’s hopes for a second term in the White House. With days remaining until Election Day, and with the number of COVID-19 cases rising since the summer, there is likely little Trump can do to change voters’ evaluations of his pandemic response, researchers wrote in the study.

“If Trump had been a bit more aggressive in pushing mask-wearing, we may have still have had a spike of cases in the fall, but it just would have happened a bit later in the election cycle, and that would have been politically advantageous,” says Lazer.

The study found that COVID-19 could influence voting in several ways, including its prevalence in a voter’s county. It could be that people who live in areas seeing greater rates of infection blame the president for failing to contain the virus more effectively. Furthermore, some people might hold Trump responsible, regardless of how much their daily lives have been directly affected, the study added.

Lazer characterized former Trump supporters who have switched their allegiances as people who routinely wear face masks, sanitize their hands, and avoid large gatherings to limit the spread of SARS-CoV-2, the coronavirus that causes COVID-19.

While there aren’t a substantial number of these defectors, they outnumber former Clinton supporters who have switched to Trump and aren’t taking pandemic safety measures, Lazer adds.

The survey showed that among likely voters who initially preferred Trump before switching to Biden, belief that the federal government has not taken the COVID-19 outbreak seriously enough more than doubled, from 25 percent to 51 percent.

That figure increased from 59 percent to 74 percent among initially undecided voters who now say they will vote for Biden, according to the survey’s latest round. 

By contrast, voters who moved away from Biden don’t report significant changes in pandemic-related attitudes or behaviors relative to other groups of voters, suggesting that they changed their minds for other reasons.

“Maybe they got a tax break. Maybe they care a lot about the Supreme Court and they really like Amy Coney Barrett,” Lazer says. “There are a lot of other issues other than COVID-19.”

For media inquiries, please contact media@northeastern.edu.