By Ryan P. Burge and Paul A. Djupe

[Image credit: The New Yorker]

Since the beginning in this country, it has been socially desirable to be religious. Americans have positive feelings toward religious people, on the whole, and negative feelings toward those who reject religion, which puts pressure on individuals to at least pretend to be religious. That was the landmark claim by Hadaway, Marler, and Chaves in 1993 (“What the Polls Don’t Show”), that survey estimates were above what actual attendance figures show. A PRRI study examined this in another way, showing higher attendance rates for telephone polls, when respondents talk to people, versus self-administered surveys done online. The overwhelming impression one gets from reading scholarly research on this question is that the non-religious exist outside of society. The nones effectively surrender the ability to participate in polite society on account of their beliefs.

There’s another story, though, that suggests just the opposite. Djupe’s grandparents (Rev. Walter and Esther Johnson) refused to eat at restaurants that served alcohol, even when traveling and it was the only establishment for miles. Indeed, for some religious groups alcohol served as a barrier to society, one they would not cross. There’s a very interesting study in the BJPS showing that the repeal of Sunday Blue Laws (prohibition of liquor sales on Sunday) ate into church attendance (which then drove down voter turnout). Religious groups have at least some tendency to cloister together socially, some much more so than others. We can view behavioral rules about dancing, drinking, and clothing as ways in which religious groups create social boundaries with society. They may have many friends, but they are bounded by church membership.

Can we find any evidence for these two stories? Fortunately, a set of questions from the General Social Survey fit the bill perfectly, allowing us to map a bit of the social worlds of different religious groups. These data give us a sense of how social people are as well as how they are social.

The GSS asked about four different contexts for socializing: spending an evening with friends outside the neighborhood, with neighbors, with family, and at a bar. Responses were frequencies ranging from never to almost daily. The following figure shows the various frequencies for spending an evening with friends outside the neighborhood. The variation is not tremendous, but does vary by religious group. Jews are the most social by this measure, with roughly 70 percent reporting doing so at least monthly (nones report about the same level). That figure is about 10 percentage points lower for evangelicals and mainline Protestants.

There are multiple potential reasons why. Perhaps the social worlds of Protestants are more family-oriented, which might follow living in the same community as family and as well as involvement within the church that reinforces family (and neighborhood) bonds. The figure below compares trends across the span of the GSS in these four dimensions of socializing by religious tradition.

There are indeed quite different patterns across groups. Family is the most frequent group to socialize with for religious majorities – Christians – while the importance of friends is elevated for religious minorities. It is relatively rare for evangelicals to spend time at a bar, though all groups but the nones hover around 25 percent. Nones have changed quite a bit over time, dropping from 50 percent reporting at least monthly trips to a bar down to 37 percent. It is not surprising to see the slide of hanging out with neighbors over this period, which a majority did in the 70s at least monthly, and now has slipped to less than a majority. Effectively, the only socializing context to remain stable or increase is the family.

There’s one other pattern that is perhaps worth remarking on. While spending time with family remained stable for most all groups and perhaps has shown some increase, the nones time socializing with family dipped through the 1980s before steadily rebounding. Could this be a result of the culture wars?

Another reason why these groups may differ, among many, is that they differ by age. Older people are less social than younger people, more focused on family (especially when they have kids). So, once we control for age, as the figure below does, the gaps between groups shrink considerably (such as between evangelicals and nones). The graph also confirms the sizable gap between younger and older Americans in their socializing.

One could chalk up the low levels of socializing by evangelicals to be due to aging, but they are aging just like the white population and mainline Protestants are older. A more likely culprit is that evangelicals are much more likely to hold and advocate for exclusivism – that it is good and right to socialize with and shop at stores owned by people from your own religious group. Djupe and Calfano have discussed the importance of this concept in multiple peer reviewed articles discussed in multiple posts. The kinds of socializing covered in the GSS measures skew toward broad interactions with society, which exclusivism would undercut. This is not to say that evangelicals are not social, just social with a narrower, more homogeneous slice of Americans (though see chapter 11 by Djupe, Neiheisel, and Sokhey in The Evangelical Crackup). This has implications for the diversity of information they receive – the echo chamber that results.

It is also interesting to see the decline in socializing by nones over this time period. Whereas they were the most social in the 1970s, their average from these measures has declined to be on par with everyone else. That is surely evidence of the normalization of the religious nones and the decline of the religious social desirability effect among many Americans, at least in areas of the country where nones are more highly concentrated.

We began this post with the long-standing social bias toward being religious in mind. Drawing that out, we thought that the social desirability bias might push the nones out to the edge of society. Over time, religious minorities are clearly more sociable according to these measures, which shade toward diverse interactions, suggesting that religious minorities cannot rely on random connections with their neighbors and coworkers for social support. Friendships with those tolerant of their beliefs must be made and kept intentionally. So, to see sociability drop steadily for religious nones, especially given how young they are, is surely a sign that the ingrained bias against the non-religious is fading. We should remember, however, that nones are a diverse group and are not synonymous with atheists, so some of the drop is likely to be the diversification of the nones too.

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

Ryan P. Burge teaches at Eastern Illinois University in Charleston, Illinois. He can be contacted via Twitter or his personal website.