American Religion is Becoming More Gay-Accepting

By Paul A. Djupe

This past week, the Supreme Court issued a landmark ruling in Bostock v. Clayton County that provided protection from discrimination in employment based on sexual orientation as covered by Title VII of the Civil Rights Act. The Court’s ruling follows its 2015 ruling in Obergefell legalizing same-sex marriage, and one of the most radical shifts in public opinion in the history of measuring it – by March 2020 PRRI reported that 69 percent supported non-discrimination laws for LGBT Americans, including majorities of every state in the US and majorities of every religious tradition, including white evangelicals.

Courts establishing rights are one (tremendously important) thing, but that does not mean that everyone in the US accepts the groups gaining rights. So, I thought I would approach this from another direction, capitalizing on questions asked in the 2007 Baylor Religion Survey (wave 2) that I learned about from sociologist Sean Bock’s work, “By your best guess, how would your current place of worship feel about each of the following behaviors?” One of those listed was “homosexual behaviors.” This is not ideal phrasing from my point of view, but for comparability’s sake I used that same language in our March 2020 survey.[1] The response options ranged from “forbids” to “isn’t concerned about it.” The figure below shows the distribution of responses for church-attending (more often than ‘never’) members of religious traditions. The shifts are stark.

The perceived acceptability of ‘homosexual behaviors’ has changed radically in this thirteen year period. In 2007, fully 90 percent of evangelicals (of all races) perceived their house of worship to forbid (63 percent) or strongly discourage (27 percent) such behaviors. By 2020, that figure had dropped to 65 percent, with the portion perceiving them as forbidden dropping almost in half – from 63 to 34 percent. That sort of change is not limited to evangelicals, but can be seen across the board, though those already supportive in 2007 (e.g., Mainline Protestants) showed less movement. Even a strong majority (60 percent) of nones in 2007 who still attended a church indicated their house of worship forbade or strongly discouraged such behaviors – that figure was cut in half by 2020 to 29 percent. From the reports of our survey respondents, American religion has rapidly changed its orientation toward LGBT Americans, becoming more accepting. This finding is in accord with noted shifts in religious elite calls for LGBT acceptance, even among some evangelicals.

That does not necessarily mean a direct translation to America’s partisans, but in this case, we can see that partisans have changed considerably across this time period too. Remember that this analysis only includes those who attend houses of worship, so it does not include all Americans. The figure below shows that the change from 2007 to 2020 was effectively proportional across partisan identities. Strong Republican reports that their house of worship forbids or strongly discourages ‘homosexual behaviors’ dropped over 20 percentage points, which is just a few points shy of how much strong Democrats reported a drop. As a result of the growing acceptance of LGBT Americans, this means that Republicans and strong Republicans are the only partisan groups in 2020 with a majority reporting that such behaviors are forbidden or strongly discouraged in their houses of worship.

One of the important implications of the Court’s ruling in Bostock is for so-called religious freedom – the ability for businesses to refuse to offer their services to LGBT Americans because of their religiously motivated disagreement. Some elites suggested that Bostock was the end of religious freedom in the US or at least the end of legislative bargaining on the matter, though not all observers share these views. We of course have no data to address these points, but I can wade into public support for service refusals – the core behavior at issue in religious freedom cases. Do those whose religion forbids ‘homosexual behaviors’ also support service refusals? How does that compare with partisan differences?

The figure below shows support for service refusals (which means the business can deny service to a LGBT person) by religious acceptance of ‘homosexual behaviors’ as well as partisanship.[2] It is clear that while greater religious acceptance of ‘homosexual behaviors’ increases opposition to service refusals, there are also rather wide partisan differences across the board. At each step of religious experience, Democrats differ from Republicans by about a scale point (25 percent of the full range). But notice that the religious gap from ‘forbid’ to ‘isn’t concerned’ is about the same gap as the partisan gap – about one scale point. The influences are roughly symmetrical so that religion does not close partisan gaps and partisanship does not enforce uniformity over religious differences.

No one would suggest that people are cookie cutouts of their religious imprint; perhaps no one would argue the same about partisanship, though that claim appears to be more common these days. It is important to see that Americans’ lives are still complicated with some cross-cutting messages reaching them, affording them some degree of choice about how they guide their political and social lives.

Moreover, the results bear on the claims about the enduring, eternal claims that religions are opposed to same-sex marriage and rights. We already knew that adherents’ views are changing and, by their report, now we know that so are the views of their houses of worship. To academic observers, at least, this is no surprise – religious organizations have to adjust their views to fit the culture to an extent in order to maintain and grow their memberships. We can find evidence of this over time with respect to race, gender, religious particularism, and now sexual orientation. This is not to say that the impulse to discriminate against LGBT Americans will end soon; large minorities still stand opposed to LGBT equality, buoyed by their religious groups, and there will always be holdouts. But the social foundations for that view appear to be crumbling across the board.

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 (posts). Further information about his work can be found at his website and on Twitter.


1. There are a few things to note about this comparison. First, I removed the ‘don’t know’ responses from the Baylor survey before calculating the percentages; we (Ryan Burge, Andrew Lewis, and I) did not include a don’t know option in our 2020 survey. We used a slightly modified, but otherwise traditional “reltrad” formulation, while the Baylor survey included an already coded variable that follows a different path using congregations. The two approaches are not strictly, but loosely comparable. As noted in the text, I limited the analysis to those who reported attending a house of worship more often than never. Baylor used a hybrid sampling procedure and a mail-in distribution method to garner about 1,648 respondents. We used Qualtrics Panels for survey respondent recruitment of our 3,100 respondents and imposed quotas so the sample demographics resembled the nation in gender, age, region, and racial dimensions. It is possible that some differences between them are due to the way both sets of researchers sourced their respondents, though the amount of those differences is unknowable.

2. The question text was, “Do you favor or oppose allowing a small business owner in your state to refuse to provide products or services to gay or lesbian people, if doing so violates their religious beliefs?” The responses ranged from “Strongly favor (they can refuse service)” to “Strongly oppose (they cannot refuse service).”

One comment

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s