Tag: The ARDA

A Research Note: Evangelicals in Name Only

This week Ryan Burge wrote a guest essay for the New York Times titled Why ‘Evangelical’ is Becoming Another Word for ‘Republican’. He was responding to a Pew Report from mid-September that showed that during the Trump presidency, those who were his supporters became more likely to self-identify as evangelicals regardless of their religiosity. In addition, Ryan observes that people of other faiths are increasingly likely to claim to be evangelicals. Ryan’s subsequently shared data on twitter showing that the percentage of evangelicals who said “religion was very important to them” had dropped by nearly 10% between 2008 and 2020. I responded, “We really need a deep dive on these self-ID evangelicals who don’t think religion is particularly important and never go to church.”

What follows isn’t the “deep dive” I imagined. Consider it splashing about in the kiddie pool,

So I went to the Association of Religion Data Archives to find a recent data set I could explore. Settling on the 2019 American Values Survey from PRRI, I began work. I selected White Evangelicals. I screened by race and then by self-identification as evangelicals. I separated the WE pool by how often they attended church. I labeled those who attended a few times a year or less as “Evangelicals in Name Only” (EINO). This let me compare those WEs who regularly attend with those who do not.

This is cross-sectional data and not longitudinal, so I can’t directly get at what the Pew report suggested. But by using 2019 data, I figured I could possible pick up the political identity embedded in the EINOs. At the outset, I should note that 31% of the White Evangelicals in the AVS are EINOs! I compared EINOs (n=115) with the rest of the WE sample (n=264) through a series of cross tabulations, looking for significant differences between the two groups. My preliminary and somewhat simplistic analysis explored three broad categories: demographics, social attitudes, and political attitudes. (The lower the percentage below the table, the more significant the difference between the groups.)

As this example shows, EINOs are much more likely to be working or lower class. They are also less likely to have a bachelor’s degree, more likely to be male, more likely to be single, more likely to be younger, and less likely to be able to handle a $400 emergency. They are roughly the same as other WEs in terms of region of the country, living in a rural area, or renting vs owning a home.

When I look at social attitudes, significant differences between EINOs and WEs are evident. For example, 46% of EINOs think abortion should be legal in all or most cases compared to only 14% of WEs.

These same difference appears with reference to Medicare for All, Legalization of Marijuana, and Same-Sex Marriage. On the most relevant culture war issues for evangelicals, EINOs aren’t especially interested. They are much more likely to favor free college and less likely to support corporate tax cuts.

As we move more into the political realm, some interesting contrasts appear. First, EINO’s were more likely to have supported Hillary Clinton for president in 2016.

They were also more favorable in evaluating Obama’s job performance as president and were more likely to identify as Independents.

Where the EINOs line up with WEs is especially evident when it comes to issues of race, immigration, and nationalism. The AVS has a series of questions where respondents are asked how well a particular word or phrase applies to them. As this cross-tab shows, there is almost no difference in the distributions of the two groups.

Other areas where similar patterns hold for EINOs and WEs include belief that whites are discriminated against, that immigrants are replacing our culture, that renaming/removing confederate memorials is bad, that the confederate flag represents Southern historic pride, and that society punishes men.

This analysis is consistent with the work of Andrew Whitehead and Sam Perry. It echoes what a new Pew report found exmplirng the role of Christian Nationalism, especially as it overlays on faith issues.

As I said at the outset, this is a very preliminary analysis and is only indicative. I don’t know that these EINOs said they were evangelicals just because they held nationalist views. But if these positions are what respondents are identifying as evangelical positions, it raises serious concerns about how the evangelical church can respond to the changes going on in contemporary society.

Exploring Evangelical Complexity

As I’ve written before, there is a well-developed cottage industry organized around the question “who are the evangelicals and what are they thinking?”. While I’m pretty sure we aren’t getting closer to any definitive answer, it feels like we’re beginning to grasp why the question remains such a conundrum.

This past week, Ryan Burge and Paul Djupe addressed the variety of answers to the question on the Religion in Public blog. Written in partial response to a recent book edited by Mark Noll, David Bebbington, and George Masden — Evangelicals, Who They Are Now, Have Been, and Could Be — they attempt to explore the “blind men and the elephant” problem in studying evangelicals.

I read the Noll book last month and found it very helpful in understanding the development of the intellectual history approach to evangelicalism. The book reflects some coherence in that approach while still exploring the challenges inherent therein. Bebbington’s contribution focusing on four theological beliefs has merit but its applicability remains somewhat challenging in today’s marketplace. It is a very good book that involves some significant dialogue among the contributors.

There is real value in locating evangelicalism in a historical vein but there is often a disconnect between that view and how social scientists explore the question. I remember n the mid-80s being allowed to sit in as the token sociologist in a group of historians — including Joel Carpenter and the recently passed Don Dayton — at the Institute for the Study of American Evangelicals. The intervening decades have not brought us closer to consensus.

Ryan and Paul explore three different approaches social scientists might use as definitional schemas. First, they look at “organizational attachment” most often measured by the RELTRAD variable in surveys. Second, they try to use theological variables (measured by agreement with some standard (although largely inadequate IHMO — see my previous post) survey formulations from respectable polling groups. Examining some data, they do not find major differences between evangelical and non-evangelical Christians. Their third approach focuses on the “born-again” identification. While those in evangelicals denominations are more likely to claim the identity than mainline denominations (but only marginally different from Black Protestants), one is left to wonder what exactly that means. In my years as an administrator in Christian Colleges, I found I had to prep prospective faculty from non-evangelical traditions. They had deep faith commitments but didn’t use the born-again language search committees wanted to hear.

Early in their blog post, they share the following insight:

Perhaps sadly, the citizenry does not conform to consistency and academic rules of classification, which leads to some strange combinations of religious attributes. That is, religion is not like a matryoshka doll.

The same day Ryan and Paul wrote their piece, Peter Wehner wrote a reflection on the Noll book for Cardus — I think they landed on twitter within minutes of each other. Peter quickly moves from contemporary politics to Bebbington and then to scripture. He writes of people whose lives were transformed by the Gospel which then gave them the motivation to address power and injustice. Instead we see faith used as a means to gain power and control over others. Yet today:

We are much more tribal than we care to confess, and far too quick to manipulate faith to support our worldly desires. Rather than having our sensibilities shaped by the ethic of Jesus, too many of us use Christianity to validate our preexisting attitudes, what we already believe, what we already want to do.

He then discusses Michele Margolis’ From Politics to the Pews which suggests that we are political first and religious second.

The difficulty in all of these approaches is that we still know far too little about what is happening in people’s minds when they are making decisions as evangelicals. Are they, in fact, acting as evangelicals or, as Peter suggests, are they simply validating prior positions with religious language. (There’s been a debate this weekend on whether abortion is a motivating force in evangelical voting or a rationalization covering other policy preferences).

Because these issues are so multidimensional, it becomes very difficult to make sense of causal order, intervening variables, and triggering factors. In a different series of posts this weekend, Ryan Burge was exploring the relationship between partisan ideology and denominational affiliation (in response to the “religious left” twitter discussions). He showed that there were very few religious traditions in which liberals outnumber conservatives, one of which was the United Church of Christ. Most show more of a mixed pattern. Then there are those like the Southern Baptist Convention was are more heavily on the politically conservative end of the scale.

But that made me think about how hard it is to unpack that descriptive data. I asked myself, where are UCC congregations located? So I went to my trustworthy source, The Association of Religion Data Archives (ARDA), and looked at the geographic distributions according to the 2010 congregations survey. What I found was that UCC congregations predominate in the Northeast and Midwest. I’ll let the reader figure out where the Southern Baptist congregation are.

If you consider what the infamous Blue and Red State maps look like, you’ll see the ways in which these maps would overlay. So are UCC folks politically liberal or do they reflect the dominant values of their region. By the way, the UMC — my own denomination — shows up in Ryan’s data as 25% liberal, 25% moderate, and 50% conservative. The congregational map for the UMC is dominant in the Eastern half of the US but more evenly distributed North and South. (I also looked at these maps by adherents per 1,000 population but it didn’t change much).

One could do the same analysis by age distribution, social class characteristics, or educational level. In any case, it’s very difficult to figure out where “evangelical” fits in the myriad factors influencing political identity and voting behavior.

I don’t have an answer, unfortunately. I simply keep wrestling with the gaps in our theoretical formulations and trying to figure out whether any classification system will give us a handle on this ever-puzzling phenomenon.