Social Network Influence on Political Behavior in Religious Contexts
Summary and Keywords
As the most common form of voluntary association in America, houses of worship remain an unquestionably critical component of American civil society. Major approaches to studying religion and politics in the United States are described, and the authors present an argument for focusing more attention on the organizational experience provided by religious contexts: studying how individuals’ social networks intersect with their associational involvements (i.e., studying religion from a “interpersonal” perspective) may actually shed new light on intrapersonal, psychological constructs like identity and religiosity.
Evidence is presented from two nationally representative data sets that suggests considerable variance in the degree to which individuals’ core social networks overlap with their houses of worship. This variance exists within and between individuals identifying with major religious traditions, and such networks are not characterized solely by agreement (as theories of self-selection might suggest).
The United States as a Religious Setting
Tocqueville (2003) famously noted the important role that religion seemed to play in associational life in the early United States. Finke and Stark’s (1992) seminal work documented the development and evolution of this religious marketplace, with denominations rising and falling as clergy have “competed” for congregants over the past 200+ years. Despite growing secularism (e.g., Pew Research Center, 2015; Putnam & Campbell, 2012) and generational changes in how Americans articulate and practice their faith (e.g., Wuthnow, 2007), the United States still ranks as a highly “religious” country relative to other Western, industrialized democracies (e.g., Pew Research Center, 2018). Importantly, however, Americans are not just religious in beliefs and adherence in the abstract but in terms of organized practice as well. One-third of Americans report attending church weekly, and estimates place the number of congregations at between 250,000 and 300,000, with roughly 150 million people belonging to these local organizations (Wald & Calhoun-Brown, 2018). Indeed, houses of worship remain an unquestionably critical component of American civil society.
As congregations are perhaps the most common institution of adult life next to the home and workplace, this article focuses on how Americans experience religion through them and what these experiences might do to their politics.1 In other words, to more fully understand the relationship between religion and politics, the article gives special attention to how social and contextual factors operate within religious settings. To be clear, the approach highlighted here does not discount the importance of beliefs when it comes to understanding political behavior in the mass public. Rather, the point is simply that religion rarely “happens” in true isolation; interpersonal and communal aspects of faith need to be considered in equal measure to beliefs, and such considerations need to weigh prominently as scholars plan their research. The sections that follow begin by briefly discussing major approaches to the study of religion and politics before noting the theoretical and empirical case for focusing more attention on religious contexts and the social networks therein. The article closes by discussing ways in which scholars might design research to better capture patterns of social influence as they relate to religion and political life in the United States.
Michigan Versus Columbia Redux: Approaches to Studying Religion and Behavior
Scholarship on how religion affects public opinion, vote choice, and political participation can be thought of as reflecting the “schools” of research familiar to every student of American political behavior. The studies of the Columbia scholars helped usher in contemporary social science, but their community studies (e.g., Berelson et al., 1954; Lazarsfeld, Berelson, & Gaudet, 1948) were eventually eclipsed by the nationally representative, large-N surveys conducted as part of the American National Election Studies at the University of Michigan (e.g., Campbell, Converse, Miller, & Stokes, 1960). The former studies can be thought of as more “sociological” in their emphasis on behavior being driven by individuals’ embeddedness in their communities, with ties to individuals and groups grounded in socioeconomic characteristics. The latter are often labeled “social-psychological” due to their focus on individuals and how mental constructs like party identification structure vote choice and opinion formation.
Much as studies of party identification came to dominate work on voting behavior, the modal approach to studying religion in the United States has also been largely individualistic, focused on what people believe and do with respect to religious practice (for a discussion, see Sokhey & Mockabee, 2012). “Commitment” models prioritize a person’s “religiosity” (often a combination of church attendance and how important a person reports religion to be in her life, sometimes referred to as “salience”), linking higher levels of this to Republican Party identification and conservativism (e.g., Green, Guth, Smidt, & Kellstedt, 1996; Layman, 1997; Stanley & Niemi, 2004). These measures of how religious a person is are often looked at against the backdrop of religious traditions—categories created by aggregating denominations based on shared beliefs, practices, and history (e.g., Kellstedt et al., 1996; Steensland et al., 2000; for discussions, extensions, and critiques, see Burge & Lewis, 2018). Scholars taking this tack have studied how religiosity and religious tradition structure vote choice (e.g., Green et al., 2005; Kellstedt, Green, Guth, & Smidt, 1994) and public opinion (e.g., Kohut, Green, Keeter, & Toth, 2001; Layman, 2001).2
In contrast, a small (but established and growing) body of research has looked at religion through the lens of how people experience and interact with their congregations. Like more general work aimed at reviving the Columbia scholarship (e.g., Huckfeldt & Sprague, 1987, 1995; Mutz, 2006), these pieces explicitly (re-)introduce context. In some cases this has involved studying people in units like parishes (e.g., Huckfeldt, Plutzer, & Sprague, 1993), and in others within congregations (e.g., Djupe & Gilbert, 2003, 2006; Djupe, Sokhey, & Gilbert, 2007; Gilbert, 1993; Wald, Owen, & Hill, 1988), to better understand how religion can provide a context for communication, activity, mobilization, and influence.
Sometimes studies have adopted such a view despite having relatively modest information on patterns of involvement in and features of congregational life (e.g., Burns, Schlozman, & Verba, 2001; Verba, Schlozman, & Brady, 1995). In fact, few studies have been designed to focus on religious contexts specifically. The most common way scholars have gone about studying the intersection of houses of worship and networks has been through the use of name generators administered to survey respondents. Here, individuals (“egos”) are surveyed and typically asked to name three to five “alters” with whom they discuss political or important matters, and to provide information on the contexts—including voluntary organizations—they share with said people (for a review of this methodology, see Sokhey & Djupe, 2014).3 Many of these works—born of efforts to understand social influence more generally (e.g., Mutz, 2006; Mutz & Mondak, 2006)—have looked at how individuals’ social networks intersect with their churches in comparison with other organizational contexts like the neighborhood, the workplace, and so on.
Approaching congregations as another important organizational context embraces the possibility of heterogeneity in how people experience their religion and how it connects with politics. For every congregation within the same denomination that has a “politicized” or perhaps cohesive feel to it (Wald, Owen, & Hill, 1990), there may be one that seems by contrast either sterile, or full of division. For every Catholic church within a diocese that has a vibrant small group activity culture, another may support relatively little programming outside of worship. For every evangelical Protestant church that has a pastor that likes to talk politics, another may be led by someone who tries to avoid current events at all costs (for similar discussions, see Djupe & Gilbert, 2008). The point is that while observing denominations or religious traditions can be useful for some purposes, painting with such broad strokes risks scholars missing large parts of the story of how people experience religion and how religion matters in shaping individuals’ orientations to the political world. Each congregation is different. How and whether these differences matter depends in no small part on how people interface with them via their social networks. Accordingly, the next section discusses why understanding networks as they intersect with religious contexts is critical to understanding religion’s influence.
The Theoretical Case for Examining Social Networks in the Context of Religious Institutions
Networks Structure Access to Resources
Notions of choice and constraint provide a useful starting point for any discussion of how individuals engage the world around them (Sokhey & Djupe, 2011). Broadly speaking, “choice” may be understood as the decisions individuals make about what environments to inhabit and what people to engage—this could mean deciding what neighborhood to live in, who to befriend on one’s street, where to work, and what church to join. Whether because of parental socialization, geographic convenience, reputation, advertising, social interactions/networks, or other factors, individuals who choose to join a religious institution become entwined with networks within a congregation. Many individuals make choices about their lives for reasons unrelated to politics, but these choices have political consequences. Indeed, once choices are made, individuals live within “constraints” that bound the range of available information and resources afforded to them (see also Huckfeldt & Sprague, 1993, 1995). If people choose to join a certain church, the characteristics of that body will bound their options for involvement and the pool of people from which they can construct/extend their social network.
From a theoretical perspective, the networks that people find themselves in—sometimes embedded partially or wholly in religious institutions—are fundamentally important because they in turn dictate exposure to political expertise (Ahn, Huckfeldt, & Ryan, 2014; Djupe & Sokhey, 2014; Huckfeldt, 2001), political disagreement (Huckfeldt, Johnson, & Sprague, 2004; Mutz, 2006), opportunities to engage in political action (e.g., Djupe et al., 2007; McClurg, 2006), and exchanges that can lead to recruitment (e.g., Klofstad, 2007). Studying network effects in the context of any type of association is valuable, and indeed, many institutions of adult life afford opportunities for individuals to make connections and leverage bonds formed with others to overcome information and resource asymmetries (Berelson, Lazarsfeld, & McPhee, 1954; Huckfeldt, 2001; Mutz & Mondak, 2006). However, religious institutions represent an especially important context where individuals form political recruitment networks and may gain civic skills that facilitate participation in politics (e.g., Djupe et al., 2007).
What makes religious institutions particularly valuable when it comes to network formation and usage? Unlike institutions of higher education, the work place, and (for many) the neighborhood, religious institutions afford a context for growth that does not lie behind the same steep pay-walls of monetary or structural advantages (Djupe & Gilbert, 2006; Smith, 2017; Verba et al., 1995). When politically-relevant resources and civic skill acquisition are less stratified by demographics, individuals have greater opportunity to advance their preferences within their communities—and by extension—democratic society more broadly. Because interpersonal interactions within the context of religious institutions foster such growth, understanding how networks within religious institutions function provides essential insights into democratic practice and representation. Given that resources, recruitment, and engagement form a classic recipe for civic engagement (Verba et al., 1995), it is of paramount importance to see congregations as social contexts that provide resources/facilitate exchange, and not simply places where people go that “somehow help communities.”
Networks Can Help Scholars Understand Why Being Religious Matters
“Religiosity” is certainly a powerful individual-level determinant of why some individuals are more involved in church than others (e.g., Putnam & Campbell, 2012). And, different operationalizations of what it means to be “religious” abound, but those who rank higher on this concept tend to be involved in more political activities (at least in the United States—e.g., Putnam & Campbell, 2012; for a review, see Grant, 2013). But while links between religiosity and different types of participation are well established, it is less clear what mechanisms undergird these relationships. For example, it could be that religious individuals are simply more compassionate, altruistic, pro-social (Ellison, 1992), or sensitive to theological doctrines that espouse the merit of doing good deeds. But, perhaps it is networks that actually facilitate action on these impulses through the mechanism of social pressure (e.g., Sinclair, 2012), and more specifically, types of influence including conformity, compliance, and persuasion (see Suhay, n.d., for a review). Likewise, because religiosity leads to more frequent church attendance (this is often by definition, given certain constructions of the measure), it also likely functions to increase exposure to sermons or activities promoting issues that motivate people to action (e.g., Chaves, Giesel, & Tsitsos, 2002). But why should such messaging matter? Here again networks may provide an answer, regulating whether such exposure ends up turning into meaningful activity as they help people process information (e.g., Huckfeldt, 2001). Other mechanisms are also plausible, part of the larger point that such bodies provide a stable context where individuals gather to discuss current events, develop interpersonal relationships, and plan events. Looking at social networks as they intersect with religious contexts may shed light on how greater religious participation actually translates into greater political engagement and activity.
Networks Can Shed Light on What Disagreement Means and What Kinds of Disagreement Are Important
By some measures, the types of networks formed in religious institutions appear to be different than those formed in other institutions of adult life.4 Because individuals can (and do) choose what congregation to attend and leave (e.g., Djupe, Neiheisel, & Sokhey, 2018), there is a perception that churches are homogeneous bodies; this implies that the members of one’s social network who share this organizational context would be wholly agreeable, thereby reinforcing rather than challenging political opinions. And, indeed, there is some evidence to suggest that social network members from church contexts tend to be less disagreeable relative to discussants than from other walks of life like the workplace (Mutz & Mondak, 2006).
However, as Huckfeldt and colleagues remind us (e.g., Huckfeldt & Sprague, 1995; Huckfeldt et al., 2004), individuals often do a poor job completely screening out disagreeable information—this is particularly true as individuals select many involvements (including what church to attend) for a variety of reasons that often have little to do with politics. If one looks at congregations in more detail, individuals may discover that they are home to more disagreement than is commonly recognized, whether gauged in terms of clergy communication, small group dynamics, or interpersonal networks (Djupe & Gibert, 2006, 2008; Djupe et al., 2018). Furthermore, if scholars look at networks using other measurement strategies, or focus on exposure to disagreement based on issues (and not just partisanship), they would see that there is often more heterogeneity within these contexts than one might observe at first glance (for a discussion, see Sokhey & Mockabee, 2012). Looking at social networks as they intersect with houses of worship can help researchers think about different types of disagreement and the potential for deliberative practice and benefit to take place where they may not initially suspect it (Djupe & Gilbert, 2008).
Networks Can Help Scholars Make More Sense of Religious Traditions
Denominations—and more often, the aggregate categories of religions traditions (e.g., Kellstedt et al., 1996)—are often used to highlight differences in religious experience in mass publics. While there is good reason to be careful about such strategies (e.g., Jones-Correa & Leal, 2001), looking at networks as they tie into congregations can help scholars use such categorizations more effectively. For example, the average structure of churches that are Catholic versus those that are evangelical Protestant bear on patterns of interaction and opportunities to build civic skills (e.g., Verba et al., 1995). That is, on its face, attending a Catholic church—with its more rigid hierarchy—may constrain opportunities to participate in skill-building activities compared to many evangelical Protestant churches where the leadership culture is more horizontal in nature.5
Networks Can Help Scholars Better Understand Identity
In an era where theories of polarization—whether political (e.g., Iyengar and Westwood 2015) or religious (e.g., Putnam & Campbell, 2012)—and social identity are at the fore (e.g., Mason, 2018), looking at networks as they intersect with congregations can help researchers understand the forces that shape how individuals see themselves and others. While understanding individuals’ religious beliefs is important, placing people in congregations helps one observe the constellation of factors that contextualize information, maintain and build relationships, and reinforce (challenge) the values that structure how individuals approach the broader world (Djupe & Calfano, 2013a, 2013b).
For example, within congregations there are myriad opportunities to build bonds, connect socially, and exchange knowledge. But what takes place in these formal and informal settings—whether Bible study, the support group, the charity drive, or the coffee hour that dives into politics—does more than disseminate information and help skill development. These and other interactions (not to mention messages from clergy) shape who individuals see as in-groups and out-groups (Tajfel & Turner, 1979), reinforcing—or potentially challenging—identity.
In this way, social networks and other congregational features facilitate understanding of how other demographic characteristics become tied to images of the self, like urban vs. rural divides. Imagine two churches, each Presbyterian, each reading from the same sacred text, but with one sitting as part of a rural community and the other sitting stationed in a college town. While the individuals within each congregation may share a certain common heritage and background, between the two congregations there are likely wide differences in dress code, accepted body language, and age diversity (Ammerman, 1991). Moreover, one might expect the rural congregation to display a different culture—one that is more tight-knit, with talk of faith, politics, and work often bleeding together. In contrast, one might expect the more urban congregation to be host to clearer role definition and more formal encounters between congregants, such as through church governance or organized, small-group activities (Williams, 1984).
Networks Can Provide Perspective on Societal Inequalities
Gender (and other) gaps in political participation have been well documented (e.g., Burns et al., 2001). But why do certain patterns persist, even as churches stand poised to help subsidize resource deficits (Verba et al., 1995)? Looking at how social networks function within religious contexts can help scholars understand how classes of individuals are actually treated within institutions of adult life (Burns et al., 2001) and why institutions may fail to produce certain outcomes. For example, Djupe et al. (2007) find that, despite the fact that women tend to be more active in churches than men, they often do not get the same civic benefits of involvement because their political expertise is devalued and they are less frequently recruited to activity. More generally, looking at social networks can provide understanding of how everyone recreates structures—even in the context of voluntary organizations—that reinforce hierarchies and social orders (e.g., Ridgeway & Smith-Lovin, 1999).
The Empirical Case for Studying Social Networks in Religious Contexts
Having laid out the theoretical argument for taking a network/congregational tack to the study of religion and politics, the next sections provide further motivation by considering a couple of snapshots of the United States. In looking within and between religious traditions, do we see heterogeneity in the extent to which individuals’ social networks overlap with their church contexts? And to what extent do network members based in a house of worship (with a shared congregational affiliation) tend to agree or disagree with individuals? That is, do the data support the stereotype of churches as homogeneous organizations? Finally, what can the way church networks have been operationalized in these data sets tell us about how we should design more effective studies?
The 2008–2009 ANES Panel Study
We first examine data from the 2008–2009 panel survey conducted as part of the American National Election Studies (ANES); these data were collected outside of the traditional ANES time series and consist of a nationally representative sample interviewed 21 times between January 2008 and September 2009. The ninth wave of this survey (administered in September 2008) included a political network generator. Respondents were asked to name up to six people with whom they had discussed government and/or elections in the preceding six months and were subsequently asked additional questions about the first three individuals named, including—importantly—whether or not the person named belongs to the same church or denomination as the respondent.
We began by coding the religious tradition of evangelical Protestant using the ANES documentation and the appendix materials available in Steensland et al. (2000). We then created measures of the portion of the respondent’s core network that shared her “church or denomination” and present the summary statistics by these religious tradition categories.6 Among all respondents, the average portion of the network sharing a congregation/denomination with the survey respondent is 0.38 (SE = 0.01),7 meaning that on average slightly more than one-third of named “alters” went to same church or belonged to the same denomination as the “ego.” Catholics reported the greatest level of shared church/denomination, with an average of 0.49 (SE = 0.02), followed by evangelical Protestant (mean = 0.47, SE = 0.02), Jewish respondents (mean = 0.45, SE = 0.04), and black Protestants (mean = 0.44, SE = 0.04) respondents. Of the major religious traditions reported, the remaining category of Protestants8 (neither black nor evangelical) reported the lowest level of shared church/denomination discussants.
The tiled histograms in Figure 1 show the distributions of shared church/denomination discussants within networks across the aforementioned religious traditions—vertical dashed lines indicate the average. Here we see a variance in discussant networks within each religious tradition. The modal condition for all respondents is to not share a church or denomination with any of their discussant partners. However, several religious traditions generate near-uniform distributions,9 meaning that respondents belonging to these religious traditions vary dramatically in the extent to which their core social networks intersect with their religious involvements. Simply put, there is a lot of variance in how people experience religion socially, which underscores the importance of acquiring information about congregations.
Are respondents less likely to encounter disagreement among discussants of the same church/denomination than among those named discussants who do not share their religion? To consider this, we look at the partisan and opinion disagreement among named discussants who do share a church/denomination and among those who do not.10 Partisan disagreement is measured as the proportion of respondents’ discussants that do not share their partisanship and thus can be read as percentages. Opinion disagreement is a (self-reported) measure of the extent to which the discussant’s opinions on government and elections are different from the main respondent’s, ranging from “not different at all (1)” to “extremely different (5).” Of the named discussants not sharing a church or denomination with the main respondents, 30% were disagreeable (i.e., did not share the partisanship of the main respondent); the average opinion difference was 2.31 (between “slightly” and “moderately different”). Of those named discussants sharing a church or denomination, 22% did not share the partisanship of the main respondent, and the average opinion difference was 2.13. Both partisan disagreement (diff = 0.8; t = –3.95, p < 0.01) and opinion-based difference (diff = 0.18; t = –4.17, p < –0.01) are significantly greater among individuals not sharing a church of denomination. Thus, respondents do report less disagreement among discussants who share a church or denomination, but disagreement does indeed exist where conventional wisdom might suggest it does not.
The 2011 Faith Matters Survey
To get a second look at the American public using different survey items/measurement strategies, we also examine data from the 2011 wave of the Faith Matters Survey conducted by Putnam and Campbell (2012). In contrast with the name generator technique used in the ANES, respondents were asked a generalized network question about core networks (for a discussion, see Hutchens et al., 2018): “Thinking about your 5 closest friends, how many of them have the same religious affiliation as you, 0, 1, 2, 3, 4, or 5?”11 Figure 1 shows network averages of shared religious affiliation by the familiar tradition categories. For all respondents, the average shared portion of core networks sharing religious affiliation is 0.54 (SE = 0.01; the 0–5 scale was recoded to run 0–1), meaning that just over half of the five individuals a person thought about with this prompt shared their faith. Black Protestant respondents reported the highest level of shared religious affiliation in their networks (mean = 0.69, SE = 0.02), followed by Mormons (mean = 0.64, SE = 0.05), and evangelical Protestants (mean = 0.60, SE = 0.01). In the ANES (previous section), the catch-all category of “Protestant” had the lowest overlap with their core networks. Here, mainline Protestants report the lowest level of shared religious affiliation among “close friends.” The tiled histograms in Figure 2 display the distributions of shared religious affiliation within core social networks, and we again see that there is substantial variance within traditions when it comes to the overlap between social networks and religion.12
Discussion: Designing Research to Capture the Social Facets of Religion
The descriptives from the ANES Panel and the Faith Matters data bolster the argument for looking at church contexts and underscore the need to be careful when speaking in aggregates. However, looking carefully at these particular data sets also makes several points about design which may help scholars conducting future research (and ultimately improve the state of the art).
Making Choices About Network Items
It was previously noted that the most common method for collecting data on social networks in religious contexts has been to use name generators embedded in surveys (e.g., the 1992 CNEP; Beck, Dalton, & Huckfeldt, 2006). Such batteries, when used with the standard “important” or “political matters” prompts, generally return a report on an individual’s “core” social network (Marsden, 1987). Respondents provide information on the contexts the named discussants (“alters”) share with the main respondent (“ego”), with some portion being fellow church members.
The ANES item adheres to the more conventional way in which “ego-centric” network data have been collected in the social and behavioral sciences (for reviews, see Borgatti, Everett, & Johnson, 2018; Knoke & Yang, 2008)—respondents name distinct individuals and report on their characteristics one by one. The Faith Matters item, by contrast, asks respondents to provide a summary across a similar set of people (“5 closest friends”), without calling upon the respondent to think about the persons individually (for a discussion of “summary” vs. “dyadic” measures, see Hutchens et al., 2018; Sokhey & Mockabee, 2012).
Both items are intended to get at an individual’s “inner” network of strong/intimate ties, but beyond this they diverge into pros and cons. On the one hand, the name generator produces information specific to individuals (which can be summarized/aggregated, or looked at in ego-alter pairs), while the summary item provides only summary information (which cannot be looked at piecemeal). On the other hand, the summary measure saves valuable survey time, and—contingent on the phrasing—can get us a report on an individual’s larger network (beyond the three to five named persons). By contrast, name generator techniques likely produce better quality data,13 but are most certainly time and effort intensive and therefore often constrain researchers from asking about as many network members as they would like. This results in an underestimate of network size if researchers treat the network battery as “the network” (which many do).
Both items suffer from less than ideal wording for scholars interested in the intersection of networks and (religious) contexts. For these purposes, the ANES item is more valid in that it asks respondents whether named discussants share their “church or denomination” (the Faith Matters item asks for a report on the number that share one’s “religious affiliation”). However, even here the imprecise wording of the question (“or denomination”) introduces unnecessary ambiguity about whether named discussants actually share the same local religious experiences.
To get better information on social networks as they pertain to religion, scholars should be explicit when asking respondents to report on the organizational contexts they share with individuals. And while a name generator is preferable to a summary item, including both can help the researcher to get a sense of how much of a person’s “broader” network she is missing by focusing on detailed reports on only three to five individuals. If survey time is not an issue, providing multiple name generators to respondents may produce the most useful data of all (e.g., Makse et al., 2019). That is, administering respondents a traditional “compound” name generator14 (see Sokhey & Djupe, 2014, for a discussion) as well as a second, church-specific network battery can produce a more complete picture of what the church experience looks like for an individual, as well as a better sense of where religion sits relative to the rest of her social life.
Capturing Other Information on Congregational Life
Scholars interested in generalizing to the American public would be unlikely to design and conduct a study that makes congregations a unit of analysis. However, such an approach would enable the collection of (large enough) samples of congregants within the selected houses of worship to make reasonable conclusions about perception vs. reality when it comes to interpersonal networks, small group dynamics, and broader distributions of opinion in the church (though see Djupe & Gilbert, 2003, 2008, for such a study focused on two major mainline Protestant denominations).15 Still, researchers can gather valuable information on church context while administering more traditional surveys to the mass public. On this note, the Faith Matters survey provides a good template, for it asks respondents to answer a detailed set of questions on their activities within their place of worship. In designing such questions, scholars should pay particular attention to response options and recall instructions to ensure that the picture of religious activity painted by a respondent reflects a typical period. Researchers should also ask as many questions as possible about an individual’s “organizational history”—that is, what congregations she has attended and when. While many studies collect some data on religious exit, more detailed histories would help scholars to better understand the forces of stability and change that characterize the religious marketplace in the United States.
Looking at religiosity and placing people into traditions is useful for characterizing the religious landscape of the United States. However, while it may be debatable as to whether politics is still local (e.g., Hopkins, 2018), it is undeniable that a large part of religious practice remains very much so. To fully understand the consequences of this diversity of association, scholars need to look at how people experience it both as individuals and together with their friends, family, and acquaintances. Embracing the social components of religion can advance social scientists’ understanding of how faith matters when it comes to politics and help address questions of inequality, participation, and representation—issues of fundamental importance to the democratic enterprise.
Ahn, T. K., Huckfeldt, R., & Ryan, J. B. (2014). Experts, activists, and interdependent citizens: Are electorates self-educating? New York, NY: Cambridge University Press.Find this resource:
Ammerman, N. T. (1991). North American fundamentalism. In M. E. Marty & R. S. Appleby (Eds.), Fundamentalisms observed (pp. 1–65). Chicago: University of Chicago.Find this resource:
Anderson, M. R. (2009). Beyond membership: A sense of community and political behavior. Political Behavior, 31(4), 603.Find this resource:
Beck, P., Dalton, R. J., & Huckfeldt, R. (2006). Cross-National Election Studies: United States Study, 1992. Ann Arbor, MI: Inter-university Consortium for Political and Social Research. Study, 6541.Find this resource:
Berelson, B., Lazarsfeld, P. F., & McPhee, W. (1954). Voting: A Study of Opinion Formation in a Presidential Campaign. Chicago, IL: University of Chicago Press.Find this resource:
Borgatti, S. P., Everett, M., & Johnson, J. (2018). Analyzing social networks, 2nd ed. SAGE.Find this resource:
Brady, H. E., Verba, S., & Lehman Schlozman, K. (1995). Beyond SES: A resource model of political participation. American Political Science Review, 89(2), 271–294.Find this resource:
Burge, R. P., Lewis, A. R. (2018). Measuring evangelicals: Practical considerations for social scientists—CORRIGENDUM. Politics and Religion, 11(4), 920–925.Find this resource:
Burns, N., Schlozman, K., & Verba, S. (2001). The private roots of public action. Cambridge, MA: Harvard University Press.Find this resource:
Campbell, A., Converse, P. E., Miller, W. E., & Stokes, D. E. (1960). The American voter. Oxford: John Wiley.Find this resource:
Chaves, M., Giesel, H. M., & Tsitsos, W. (2002). Religious variations in public presence. In R. Wuthnow & J. H. Evans (Eds.), The quiet hand of God: Faith-based activism and the public role of mainline Protestantism (pp. 108–128). Berkeley: University of California Press.Find this resource:
Davis, N. T. (2018). Religion and partisan‐ideological sorting, 1984–2016. Social Science Quarterly, 99(4), 1446–1466.Find this resource:
De Tocqueville, A. (2003). Democracy in America (Vol. 10). Regnery Publishing.Find this resource:
Djupe, P. A., & Calfano, B. R. (2013a). Religious value priming, threat, and political tolerance. Political Research Quarterly, 66(4), 768–780.Find this resource:
Djupe, P. A., & Calfano, B. R. (2013b). Divine intervention? The influence of religious value communication on US intervention policy. Political Behavior, 35(4), 643–663.Find this resource:
Djupe, P. A., & Gilbert, C. P. (2003). The prophetic pulpit: Clergy, churches, and communities in American politics. Rowman & Littlefield.Find this resource:
Djupe, P. A., & Gilbert, C. P. (2006). The resourceful believer: Generating civic skills in church. Journal of Politics,68(1), 116–127.Find this resource:
Djupe, P. A., & Gilbert, C. P. (2008). The political influence of churches. Cambridge University Press.Find this resource:
Djupe, P. A., & Sokhey, A. E. (2014). The distribution and determinants of socially supplied political expertise. American Politics Research, 42(2), 199–225.Find this resource:
Djupe, P, A., Sokhey, A. E., & Gilbert, C. (2007). Present but not accounted for? Gender differences in civic resource acquisition. American Journal of Political Science, 51(4), 906–920.Find this resource:
Djupe, P. A., & Neiheisel, J. R. (2012). How religious communities affect political participation among Latinos. Social Science Quarterly, 93(2), 333–355.Find this resource:
Djupe, P. A., Neiheisel, J., & Sokhey, A. (2018). Reconsidering the role of politics in leaving religion: The importance of affiliation. American Journal of Political Science,62(1), 161–175.Find this resource:
Ellison, C. G., & George, L. K. (1994). Religious involvement, social ties, and social support in a southeastern community. Journal for the Scientific Study of Religion, 33, 46–61.Find this resource:
Ellison, C. G. (1992). Are religious people nice people? Evidence from the national survey of black Americans. Social Forces, 71(2), 411–430.Find this resource:
Finke, R., & Stark, R. (1992). The churching of America 1776–1990. Rutgers University Press.Find this resource:
Gilbert, C. P. 1993.The impact of churches on political behavior: An empirical study (Vol. 319). Greenwood.Find this resource:
Grant, J. T. (2013). Religion, politics, and civic engagement in the United States. Oxford University Press.Find this resource:
Green, J. C., Guth, J., Smidt, C., & Kellstedt, L. (1996). Religion and the culture wars: Dispatches from the front. Lanham, MD: Rowman & Littlefield.Find this resource:
Green, J. C., Smidt, C. E., Guth, J. L., & Kellstedt, L. A. (2005, November). The American religious landscape and the 2004 presidential vote: Increased polarization. In Washington, DC: Pew Forum on Religion and Public Life (Vol. 372, pp. 143–60).Find this resource:
Hopkins, D. J. (2018). The increasingly United States: How and why American political behavior nationalized. Chicago: University of Chicago Press.Find this resource:
Huckfeldt, R. (2001). The social communication of political expertise. American Journal of Political Science, 45(2), 425–438.Find this resource:
Huckfeldt, R., Johnson, P. E., & Sprague, J. (2004). Political disagreement: The survival of diverse opinions within communication networks. Cambridge University Press.Find this resource:
Huckfeldt, R., Plutzer, E., & Sprague, J. (1993). Alternative contexts of political behavior: Churches, neighborhoods, and individuals. The Journal of Politics, 55(2), 365–381.Find this resource:
Huckfeldt, R., & Sprague, J. (1987). Networks in context: The social flow of political information. American Political Science Review, 81(4), 1197–1216.Find this resource:
Huckfeldt, R., & Sprague, J., (1993). Citizens, contexts, and politics. In A. Finifter (Ed.), Political Science: The State of the Discipline II (pp. 281–303). Washington, DC: American Political Science Association.Find this resource:
Huckfeldt, R. R., & Sprague, J. (1995). Citizens, politics and social communication: Information and influence in an election campaign. Cambridge University Press.Find this resource:
Huckfeldt, R., & Sprague, J. Indianapolis-St. Louis Election Study, 1996–1997. Ann Arbor, MI: Inter-university Consortium for Political and Social Research [distributor], 2007-05-07.Find this resource:
Hutchens, M. J., Eveland, W. P., Morey, A. C., & Sokhey, A. E. (2018). Evaluating summary measures of heterogeneous political discussion: The critical roles of excluded cases and discussion with people holding extreme views. Communication Methods and Measures, 12(4), 276–294.Find this resource:
Iyengar, S., & Westwood, S. J. (2015). Fear and loathing across party lines: New evidence on group polarization. American Journal of Political Science, 59(3), 690–707.Find this resource:
Jones-Correa, M. A., & Leal, D. L. (2001). Political participation: Does religion matter? Political Research Quarterly, 54(4), 751–770.Find this resource:
Kellstedt, L. A., Green, J. C., Guth, J. L., & Smidt, C. E. (1996). Grasping the Essentials: The Social Embodiment of Religion and Political Behavior. In J. C. Green, J. L. Guth, C. E. Smidt, & L. A. Kellstedt (Eds.), Religion and the Culture Wars: Dispatches from the Front. Lanham, MD: Rowman and Littlefield.Find this resource:
Kellstedt, L. A., Green, J. C., Guth, J. L., & Smidt, C. E. (1994). Religious voting blocs in the 1992 election: The year of the evangelical? Sociology of Religion, 55(3), 307–326.Find this resource:
Klofstad, C. A. (2007). Talk leads to recruitment: How discussions about politics and current events increase civic participation. Political Research Quarterly, 60(2), 180–191.Find this resource:
Klofstad, C. A., Sokhey, A. E., & McClurg, S. D. (2013). Disagreeing about disagreement: How conflict in social networks affects political behavior. American Journal of Political Science, 57(1), 120–134.Find this resource:
Knoke, D., & Yang, S. (2008). Social network analysis (Vol. 154). SAGE.Find this resource:
Kohut, A., Green, J. C., Keeter, S., & Toth, R. C. (2001). The diminishing divide: Religion’s changing role in American politics. Brookings Institution Press.Find this resource:
Layman, G. C. (1997). Religion and political behavior in the United States: The impact of beliefs, affiliations, and commitment from 1980 to 1994. Public Opinion Quarterly, 61(2), 288–316.Find this resource:
Layman, G. C. (2001).The great divide: Religious and cultural conflict in American party politics. New York: Columbia University Press.Find this resource:
Lazarsfeld, P., Berelson, E. B., & Gaudet, H. (1948). The People's Choice: How the Voter Makes Up His Mind in a Presidential Campaign. New York: Columbia University Press.Find this resource:
Lewis, V. A., MacGregor, C. A., & Putnam, R. D. (2013). Religion, networks, and neighborliness: The impact of religious social networks on civic engagement. Social Science Research, 42(2), 331–346.Find this resource:
Marsden, P. (1987). Core discussion networks of Americans. American Sociological Review, 52, 122–131.Find this resource:
Makse, T., Minkoff, S., & Sokhey, A. (2019). Politics on Display: Yard Signs and the Politicization of Social Spaces. Oxford University Press.Find this resource:
Mason, L. (2018). Uncivil agreement: How politics became our identity. University of Chicago Press.Find this resource:
McClurg, S. D. (2003). Social networks and political participation: The role of social interaction in explaining political participation. Political Research Quarterly, 56(4), 449–464.Find this resource:
McClurg, S. D. (2006). The electoral relevance of political talk: Examining disagreement and expertise effects in social networks on political participation. American Journal of Political Science, 50(3), 737–754.Find this resource:
Mutz, D. C., & Mondak, J. J. (2006). The workplace as a context for cross-cutting political discourse. Journal of Politics, 68(1), 140–155.Find this resource:
Mutz, D. C. (2006). Hearing the other side: Deliberative versus participatory democracy. Cambridge University Press.Find this resource:
Nadeau, R., Niemi, R. G., Stanley, H. W., & Godbout, J. F. (2004). Class, party, and south/non-south differences: An update. American Politics Research, 32(1), 52–67.Find this resource:
Pew Research Center. (2015, May 12). America’s changing religious landscape. Washington, DC: Pew Research Center.Find this resource:
Pew Research Center. (2018, July 31). Americans are far more religious than adults in other wealthy nations. Washington, DC: Pew Research Center.Find this resource:
Putnam, R. D., & Campbell, D. E. (2012). American grace: How religion divides and unites us. New York: Simon & Schuster,Find this resource:
Ridgeway, C. L., & Smith-Lovin, L. (1999). The gender system and interaction. Annual Review of Sociology,25(1), 191–216.Find this resource:
Robins, G. (2015). Doing social network research: Network-based research design for social scientists. SAGE.Find this resource:
Scheufele, D. A., Nisbet, M. C., Brossard, D., & Nisbet, E. C. (2004). Social structure and citizenship: Examining the impacts of social setting, network heterogeneity, and informational variables on political participation. Political Communication, 21(3), 315–338.Find this resource:
Scheufele, D. A., Nisbet, M. C., & Brossard, D. (2003). Pathways to political participation? Religion, communication contexts, and mass media. International Journal of Public Opinion Research, 15(3), 300–324.Find this resource:
Schmidt, E. R. (2018). The influence of religious–political sophistication on US public opinion. Political Behavior, 40(1), 21–53.Find this resource:
Sinclair, B. (2012). The social citizen: Peer networks and political behavior. Chicago: University of Chicago Press,Find this resource:
Smith, A. E. (2017). Democratic talk in church: Religion and political socialization in the context of urban inequality. World Development, 99, 441–451.Find this resource:
Stanley, H. W., & Niemi, R. G. (2004). and Group Support, 1952–2000. Models of Voting in Presidential Elections: The 2000 US Election, 123.Find this resource:
Steensland, B., et al. (2000). The measure of American religion: Toward improving the state of the art. Social Forces, 79(1), 291–318.Find this resource:
Stetzer, E., & Burge, R. P. (2016). Reltrad coding problems and a new repository. Politics and Religion, 9(1), 187–190.Find this resource:
Sokhey, A. E., & Djupe, P. A. (2011). Interpersonal networks and democratic politics. PS: Political Science & Politics, 44(1), 55–59.Find this resource:
Sokhey, A. E., & Djupe, P. A. (2014). Name generation in interpersonal political network data: Results from a series of experiments. Social Networks, 36, 147–161.Find this resource:
Sokhey, A. E., & Mockabee, S. T. (2012). Reexamining political discussion & disagreement in church networks: An exit poll assessment. Politics & Religion,5(2), 253–279.Find this resource:
Suhay, E. (n.d.). Political conformity: Evidence and mechanisms. Working paper. Prepared for Political Persuasion Conference, Laguna Beach, CA, January 9–10, 2016.Find this resource:
Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. In W. G. Austin & S. Worchel (Eds.), The social psychology of intergroup relations. Monterey, CA: Brooks-Cole.Find this resource:
Taylor, R. J., & Chatters, L. M. (1988). Church members as a source of informal social support. Review of Religious Research, 30, 193–203.Find this resource:
Tocqueville, A. De (2003). Democracy in america (Vol. 10). Regnery Publishing.Find this resource:
Tourangeau, R., Rips, L., & Rasinski, K. (2000). The psychology of survey response. Cambridge University Press.Find this resource:
Verba, S., Schlozman, K. L., & Brady, H. E. (1995). Voice and equality: Civic voluntarism in American politics. Cambridge, MA: Harvard University Press.Find this resource:
Wald, K. D., & Calhoun-Brown, A. (2018). Religion and politics in the United States (8th ed.). Rowman & Littlefield.Find this resource:
Wald, K. D., Owen, D. E., & Hill S. S. (1988). Churches as political communities. American Political Science Review, 82(2), 531–548.Find this resource:
Wald, K. D., Owen, D. E., & Hill, S. S., Jr. (1990). Political cohesion in churches. Journal of Politics,52(1), 197–215.Find this resource:
Williams, M. D. (1984). Community in a black Pentecostal church: An anthropological study. Waveland.Find this resource:
Wuthnow, R. (2007). America and the challenges of religious diversity. Princeton University Press.Find this resource:
Wuthnow, R. (2010). After the baby boomers: How twenty- and thirty-somethings are shaping the future of American religion. Princeton, NJ: Princeton University Press.Find this resource:
(1.) Throughout this article we use the terms “congregation,” “church context,” and “religious context” interchangeably.
(3.) We discuss approaches to the study of social networks (in religious contexts) in more detail later in the article.
(4.) The focus here is on disagreement, though one could think about difference in other ways. For example, Scheufele et al. (2003) note that networks tied to religious contexts may take on a more participatory tone than more secular networks.
(5.) For example, many evangelical churches have “elders” who act as a congregation-sourced decision making body. Of course, at the same time, some evidence suggests that evangelical churches may be more insulated from the community than Catholic parishes (e.g., Djupe & Neiheisel, 2012). And here, too, taking a network perspective is valuable—it urges the examination of consequential social ties that flow throughout the congregation and community.
(6.) Averages were constructed by summing the number of discussants the respondent named as belonging to the same church or denomination and then dividing by the total number of discussants the respondent named.
(7.) The precise question wording was: “Does (NAME) belong to the same church or denomination that you belong to?”
(8.) The ANES provides a derived classification of “Protestants” (along with “Catholics,” “Jews,” “Other” and “No Religion”) based on a series of denomination items asked of respondents. We followed the literature to split out evangelical and black Protestants (e.g., Steensland et al., 2000), but have not actively separated out mainline Protestants (e.g., ELCA and Anglican respondents) in these descriptives.
(9.) Network averages of 0.5 indicate that respondents only named two discussants and that, of those two discussants, one person shared their church or denomination.
(11.) The term “religious affiliation” is also problematic, as it introduces ambiguity: that is, in the mind of a respondent, does the person that is thought of share her religious context, religious tradition, and/or denomination?
(12.) These data also demonstrate the potential for discussants of the same denomination to be disagreeable: of those reporting that all five individuals shared their faith, the average number reported to share their partisanship was four individuals (and this average drops to 3.8 among those reporting 4 discussants sharing their religious affiliation).
(14.) A standard wording—here, from Huckfeldt and Sprague’s (1996–1997) Indianapolis–St. Louis Study—is as follows: “From time to time, people discuss [government, elections and politics/important matters] with other people. I’d like to know the people you talk with about these matters. These people might or might not be relatives.” A context-specific generator might ask about discussion “with people who attend your place of worship.”