Rejoinder :: Exposure is at the core of what we do and most everyone agrees

By Paul A. Djupe

[Image credit: St. Louis American]

I recently wrote a post arguing that religion is not a matryoshka doll, that you do not find smaller and smaller versions of the whole as your proceed down from tradition -> denomination -> congregation -> group-> individual. Instead, you are highly likely to find substantial diversity in a variety of forms. Within evangelicalism, for instance, there are congregations supporting and opposing Trump with individual congregants that take the opposite position. [Add clergy to this mix and you have an extra layer of diversity to contend with.]

Then Tobin Grant wrote a reply soon after rejecting my arguments in multiple ways. I concede that there is no expectation that every coding choice is going to be homogeneous – with few exceptions, groups are going to be diverse and we are interested, in the end, in finding differences of means. My argument is that making these reltrad coding choices leaves on the table so much interesting and useful variation that they are simply not useful for the kinds of questions that political scientists and sociologists study.

But what is novel, even remarkable about Tobin’s argument is that he rejects the common roots of this long standing disagreement. Let him say it in his own words from his post:

My argument is that it is simply untrue that “the reltrad measurement scheme was designed to make assumptions about what people are exposed to.” This may be how Djupe sees reltrad because his research focuses on exposure. But for the originators of the reltrad classification, the goal wasn’t about exposure or even politics.

I was awestruck at this assertion because I knew the Gang of Four[1] had said as much in so many words. A focus on exposure and how to gain religion’s contributions of politically-relevant beliefs and cues has always been at the core. It’s how I have always taught an introduction to religion and political behavior. So, I went to the source of what I’ve long thought of as the authoritative statement of the reltrad approach: “Grasping the Essentials: The Social Embodiment of Religion and Political Behavior.” This book chapter appeared in the 1996 collection of essays by the Gang called Religion and the Culture Wars – it is literally in the center of the book and I’ve always considered it the centerpiece. It’s a how-to, but most importantly tells us why we would want to measure religion in this way. Here’s what they say:


To simplify: Religion is like an educational institution where you learn life-relevant beliefs, arguments, values, and behaviors. “The principal mechanism for the worldly realization of beliefs,” religious communities are the “Word made flesh,” the “critical social context of individuals’ political attitudes and behaviors.”

You can see the same set of assumptions in other pieces adopting this framework. Take, for instance, Olson and Warber (2008: 194): “Historically, members of various religious families have taken distinctive approaches to politics that are rooted in their unique religious traditions and teachings.”

In an earlier conception pitched by Kellstedt and Green (1993: 55), the political knot is drawn tighter: “[P]arties and denominations share certain organizational dynamics….Denominations are primarily organized around their varied conceptions of the good life and have often sought assistance from the parties to defend or achieve their particular goals.” And a few lines later, “While specific denominations have distinct political outlooks, few are large enough to influence politics on their own, and thus practicality dictates that they form alliances with other denominations for political purposes…The basic ‘building blocs’ of these alliances have been religious traditions.”

That is, we (the collective ‘we’ now) have always agreed that exposure in religious communities is essential to capture in order to understand citizen politics, but we disagree about the level of aggregation at which this information is gained. The Gang and followers have asserted religious traditions are sufficient measures of exposure through belonging, while I and others (e.g., Ken Wald, Ted Jelen, Chris Gilbert, Anand Sokhey, Jake Neiheisel, et al.) have advocated for more localized measures of specific content. I think that the evidence provided in my last post (in addition to the long list of posts there with Ryan Burge) speaks clearly on behalf of capturing religious contexts much closer to the individual.

The more important point is that religion and politics scholars need to capture credible measures of exposure. Without it, we are unable to explain why political patterns obtain. If, for instance, evangelicals are more conservative on abortion than other Protestants (or whatever they are calling themselves these days), we just cannot tell why without further evidence. Is the pattern due to hearing pro-life arguments, adopting particular values or beliefs, selection into the denomination, peer influence, or something else? Put differently, finding some differences by religious tradition may be a useful starting place, but it is only a starting place and never an end point.

Digging into this very question raised serious doubts that clergy exercise opinion leadership and revealed that social (interpersonal) influence is widespread. At least in the congregations Djupe and Gilbert (2009) studied, the pattern of results undermined the classic model of religious influence that implies long standing unity revolving around religious reasons. Whether those findings are replicated in other congregations remains to be seen, but we (Jake, Anand, and Paul) have continued to find disagreement present in congregations across the political spectrum that leads some to leave their congregation when that disagreement is salient.

Put plainly, in order to understand religious influence on politics, we need to study two facets of religious communities: (1) what they are exposed to (beliefs, values, arguments, cues), and (2) what religious and secular forces shape adoption of that exposure. Without these elements, we have incomplete explanations and cannot make forward progress in the subfield.

Paul A. Djupe, Denison University Political Science, is an affiliated scholar with PRRI, the series editor of Religious Engagement in Democratic Politics (Temple), and co-creator of (see his list of posts here). Further information about his work can be found at his website and on Twitter.


1In alphabetical order, the self-named “Gang of Four” is: John Green, Jim Guth, Lyman “Bud” Kellstedt, and Corwin Smidt.

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 )

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