I just wanted to point you all to a long essay I wrote for the Deseret. It’s called: My church is closing, and I don’t know what comes next — for me, or America. No graphs there – just a lot of reflection about what it feels like to be the last pastor of a church that was founded in 1868.

All credit to the tremendous Landon Schnabel for a great paper that was published at the Journal for the Scientific Study of Religion. The title tells the story, “A Search for Liberalizing Religion: Political Asymmetry in the American Religious Landscape.” The conceit is pretty simple – there are lots of examples of highly religious people being really conservative. Schnabel wants to conduct a search for any evidence that being highly engaged in a certain religion pushes someone to the left side of the political spectrum.

Let me summarize the conclusion by just quoting from Schnabel’s last paragraph:

Even “liberal” religion is typically conservatizing and the fact that we can call it “liberal” at all appears to be a function of selection.

This is something that I have seen threads of in other work that I have put out there. For instance, I wrote a post using data that focused on measuring beliefs about the social gospel for Religion News Service a couple of years ago. It’s no shocker that evangelicals are less likely to subscribe to the tenets of the social gospel, but this last regression was really surprising to me. It’s the likelihood of someone saying that “God is more concerned about individual morality than social inequalities.” I estimated a regression for Democrats, Independents and Republicans based on religious attendance.

What is really stunning is the fact that even among Democrats, going to church more makes them more likely to believe that individual morality is more important to God than societal problems. In other words, the more Democrats go to church, the more they look like Republicans. I wanted to pull on that thread just a bit more.

This is the political ideology of the sample but it’s broken down by frequency of religious attendance.

You can see that trademark cascade pattern. The top row (which is never attenders) has a whole lot of blue and not much red. Just 21% of never attenders are conservative, while 46% identify as liberal. As church attendance goes up, those ratios begin to shift from left to right.

Among yearly attenders, the conservatives start to take over compared to liberals (36% vs 25%). Among weekly attenders, 52% are conservative, while just 16% are liberal. It’s even more extreme among the most frequent attenders. For folks who are attending religious services multiple times a week, about 60% are conservative and 10% are liberal.

But, race has to play a role here, right? This probably only holds for white respondents and breaks down for people of color. Nope. The finding is consistent for every racial group.

Among white respondents, 45% of never attenders are liberals. It’s just about 12% of white weekly attenders. A Black person who never attends religious services is twice as likely to identify as a liberal compared to one who attends multiple times per week (48% vs 23%). The Hispanic trend line isn’t so smooth compared to the previous two, but the trajectory is the same – there aren’t many Hispanic liberals in the pews each Sunday.

There’s just no two ways about it – people who are more religiously active are significantly less likely to describe themselves as liberals. That’s the case for every racial group as well – this transcends these categories.

Let me put a finer point on this by showing you how this relationship works for the twenty largest Protestant denominations included in the data. Same basic analysis – calculating the share who identify as liberal at each level of church attendance.

That’s a whole lot of trend lines pointing downward – meaning those who attend the most frequently are the least likely to identify as liberal. That’s true for Southern Baptists and a lot of non-denominational folks, too. Which tracks – those are evangelical denominations. There are several where the line is pretty flat like the Assemblies of God and Church of Christ. In both cases, there’s no relationship between religious attendance and political ideology.

However, what may be most striking is that even folks who are members of what are perceived to be left leaning or moderate denominations like the United Methodist Church and the Presbyterian Church USA are showing a similar pattern to Southern Baptists – higher attendance means less liberalism.

There are only two denominations that are clearly pointing upward – the Evangelical Lutheran Church of America (and that’s likely because of the weekly+ group being so liberal) and members of the Episcopal Church.

That’s the only two out of these twenty where the most frequent church goers are more liberal than those who attend with less frequency. Those two denominations represent about 8% of the sample that was used to make the graphic above. The Southern Baptists by themselves are 15%.

But what about non-Protestant groups? I tested that, too. I had to expand my sample to include the last four years of data to make sure that my N size was large enough.