Empathy divides Americans by religion, with white evangelicals standing apart

PRRI found a sharp religious divide over empathy. White evangelical Protestants were the only major religious group in which a majority, 54%, agreed that empathy can become dangerous when it leads people away from a society guided by God’s truth. Only 42% said empathy is a foundational value that helps build a healthy society.

Other groups were much more supportive of empathy’s foundational role. At least 66% in every other religious group agreed with that statement. The figures included 73% of white mainline Protestants, 73% of Hispanic Catholics, and 83% of religiously unaffiliated Americans.

This contrast matters because white evangelicals were the religious group most divided over empathy’s role. PRRI reported religious and political results separately, so the survey does not show how much white evangelicals account for the wider partisan difference.

Empathy is the ability to understand another person’s feelings or perspective. In the survey, it was presented as a foundational moral value that can help build a healthy society. That view treats empathy as a tool for recognizing other people’s experiences and responding to them with concern.

Others see a possible danger. White evangelical respondents were asked whether empathy can become dangerous when it leads people away from a society guided by God’s truth. Under that view, compassion may become a weakness if it replaces firm moral standards or makes people accept ideas they believe are wrong.

The disagreement is not simply about whether kindness is good. It concerns which principles should guide decisions when empathy conflicts with religious beliefs or social goals. The survey shows these competing views are especially visible among white evangelicals and Republicans, amid debates over immigration, racial justice, and LGBTQ+ rights.

PRRI found that views of empathy differed strongly by political identity. Eighty-four percent of Democrats agreed that empathy is a foundational moral value that helps build a healthy society. The comparable figure was 54% among Republicans, creating a 30-point gap. Independents fell between them at 70%.

The divide was even wider across PRRI’s Christian nationalism categories. Only 38% of Christian nationalism adherents agreed that empathy is foundational, compared with 94% of rejecters. On the danger statement, six in ten adherents agreed, while only 6% of rejecters did.

These figures show a strong association between political identity, Christian nationalism, and views of empathy. They do not show that party affiliation or Christian nationalism caused those opinions. PRRI also reported religion and partisanship separately, so the survey cannot determine how much white evangelicals contributed to the Republican-Democratic difference.

In this survey, Christian nationalism refers to support for a United States government and laws grounded in Christianity, along with the belief that Christians should exercise dominion over society. PRRI used these ideas as its measure rather than treating religious identity alone as proof of Christian nationalism.

Respondents were classified as Christian nationalism adherents or sympathizers when their answers supported those principles. The article also refers to rejecters, meaning people who rejected the Christian nationalism measure. This classification allowed PRRI to compare views of empathy across different levels of support.

The results showed a major divide. Thirty-eight percent of adherents said empathy is foundational, compared with 94% of rejecters. Six in ten adherents said empathy can become dangerous, while only 6% of rejecters agreed. The categories describe survey responses; they do not establish why respondents hold those views.

An association means two characteristics appear together more often than expected. PRRI found that white evangelical Protestants and Republicans were less likely than many other groups to call empathy foundational. That pattern describes differences in opinion, but it does not explain where those differences came from.

Religion and party can overlap in a population. Other influences may also shape answers, including Christian nationalism, personal experiences, or views about current political issues. PRRI’s results show that Christian nationalism adherents had different responses from rejecters, but the survey does not prove that Christian nationalism created those responses.

The article specifically notes that PRRI reported religious and partisan results separately. It therefore cannot determine how much white evangelicals contributed to the Republican-Democratic gap. A survey can measure relationships in public opinion, but proving cause would require stronger evidence than these reported comparisons provide.

Empathy can influence whether people treat another group’s experiences as morally important. If one side sees empathy as essential, it may favor listening to people affected by immigration rules, racial injustice, or LGBTQ+ policies. If another side sees empathy as dangerous when it departs from God’s truth, it may prioritize religious standards over those appeals.

The key mechanism is a clash over moral authority. One approach emphasizes understanding and shared human concern. The other warns that compassion can be weaponized or can pull society away from what respondents regard as religious truth. Those different starting points can make compromise harder, even when people discuss the same issue.

PRRI President Robert P. Jones said the findings raise questions about the moral foundations that allow people from different backgrounds to share a democracy. The survey does not predict specific policy outcomes, but the reported partisan and religious divides point to continuing tension over cooperation and common values.

A public-opinion survey measures attitudes by asking many people the same questions and reporting the share choosing each response. PRRI surveyed adults ages 18 and older across all 50 states and the District of Columbia. It used 5,200 adults from Ipsos’ KnowledgePanel and 339 additional adults from opt-in panels.

The questions produced percentages for statements about empathy, religion, political party, and Christian nationalism. Researchers can compare those percentages across groups, such as Democrats and Republicans. Those comparisons describe the surveyed population and estimate broader public opinion, subject to sampling limits.

The full sample’s margin of sampling error was plus or minus 1.6 percentage points at a 95% confidence level. In practical terms, repeated samples using the same method would usually produce estimates within roughly 1.6 points of the measured value because of sampling variation. This margin does not remove other possible survey limitations.