President Donald Trump continues to be White Christians’ preferred candidate for the November election, but support among voters in three major traditions – White Catholics, White Protestants who are not evangelical and even White evangelical Protestants – has slipped since August, according to a new Pew Research Center poll.

In 2020 election, deep divisions between White Christians and everyone else

Democratic candidate Joe Biden, by contrast, is leading the presidential contest among every other religious group analyzed in the survey, including Black Protestants, Hispanic Catholics, Jews and the religiously unaffiliated. The poll of 10,543 U.S. registered voters nationwide was conducted Sept. 30 to Oct. 5, as Trump spent four days in the hospital amid a White House COVID-19 outbreak.

Among White Catholic voters, Trump is ahead of Biden by 8 percentage points: 52% in this group say they would vote for Trump (or lean that way) if the election were held today, while 44% favor Biden. This gap has narrowed significantly – Trump was 19 points ahead of Biden (59% to 40%) the last time this question was asked in a poll conducted in late July and early August.

Support from White Protestants who do not consider themselves to be evangelical or born-again has dropped at a similar pace: 53% say they would vote for Trump if the election were today, down from 59% who said this in the summer poll. Even White evangelical Protestants have softened slightly in their support for Trump, though they overwhelmingly remain on his side: 78% of White evangelicals intend to cast ballots for Trump, compared with 83% who said this in August.

Pew Research Center conducted this study to understand how Americans from various religious groups view the upcoming 2020 presidential election and the presidential candidates. For this analysis, we surveyed 11,929 U.S. adults, including 10,543 registered voters, during the last week of September and the first week of October 2020. The survey was in the field when Trump announced, early on the morning of Oct. 2, that he and first lady Melania Trump had contracted COVID-19.

Everyone who took part in this survey is a member of Pew Research Center’s American Trends Panel (ATP), an online survey panel that is recruited through national, random sampling of residential addresses. This way nearly all U.S. adults have a chance of selection. The survey is weighted to be representative of the U.S. adult population by gender, race, ethnicity, partisan affiliation, education and other categories. Read more about the ATP’s methodology.

Here are the questions used for this report, along with responses, and its methodology.

Biden, meanwhile, is currently favored by 90% of Black Protestant registered voters, 70% of Jews and 67% of Hispanic Catholics. Among people without a religious affiliation, Biden is the preferred candidate for 83% of atheists and agnostics, and 62% of people who describe their religion as “nothing in particular.” In all of these groups, support for Biden is on par with that seen in the August survey.

The survey was conducted days after the candidates held their first presidential debate on Sept. 29. On the morning of Oct. 2, Trump announced that he had tested positive for the coronavirus, and that evening he was hospitalized at Walter Reed Medical Center for treatment.

While support for Trump has dropped among all three groups of White Christians analyzed in the survey, support for Biden did not see a corresponding (statistically significant) jump. One partial factor in Trump’s declining support from White Christians might be that the new survey, for the first time, gave respondents the option of saying they would vote for Libertarian candidate Jo Jorgensen or Green Party candidate Howie Hawkins. Only Trump and Biden were listed by name in the August survey.

Overall, 6% of registered voters in the new survey say they would vote for Jorgensen (4%), Hawkins (1%), or some other candidate or none of the candidates listed (1%). By comparison, 2% of registered voters in the summer poll indicated they would vote for someone other than Trump or Biden, or that they would vote for neither candidate.

White Christians are a key segment of the electorate because they make up roughly 44% of  U.S. registered voters. Roughly 7% of registered voters are Black Protestants, 5% are Hispanic Catholics, 2% are Jewish and 28% are religiously unaffiliated.

The survey’s findings about 2020 voting intentions are in line with long-term trends in party identification. White evangelical Protestants – Trump’s strongest supporters – are the most solidly and consistently Republican among major religious groups in the electorate, and they have grown even more uniformly Republican in recent decades. White Catholics and White Protestants who are not evangelical also have shifted in a Republican direction in recent years. By contrast, Black Protestants, religiously unaffiliated voters, Jews and Hispanic Catholics have long been solidly Democratic.

And while the new survey did not ask respondents to rate Trump’s handling of his job as president, past polls show that approval of Trump’s performance has generally hovered within a fairly narrow range among most religious groups.

The share of White evangelical Protestants who said they approve of Trump’s job performance was 72% in the August survey, identical to the share who said this in Trump’s first summer on the job three years ago. Similar shares of White Catholics (54%) and White Protestants who are not evangelical (53%) said this summer that Trump is doing a good job, within the typical band of 50% to 60%.

Trump has consistently received his lowest approval ratings from Black Protestants, Hispanic Catholics, Jews and religiously unaffiliated voters.

Trump's job approval rating consistently highest among White Christians, far lower among other groups

Note: Here are the questions used for this report, along with responses, and its methodology. Also, on Oct. 15, Center researchers discussed these and other relevant findings at a public virtual event titled “Faith and the 2020 Elections: A Look at Public Opinion.” The recording can be found here.

Gregory A. Smith  is an associate director of research at Pew Research Center.