Clergy Political Actions and Agendas: New Findings from the National Survey of Religious Leaders

ABSTRACT

We use the National Survey of Religious Leaders (NSRL) to extend prior research on clergy’s political activism and agendas. We find that christian clergy engage in political cue giving at similar rates across religious traditions, though evangelical clergy are less likely than other clergy to engage in direct action. Regarding issue priorities, evangelical clergy focus almost exclusively on a moral reform agenda, with a particular focus on abortion. Both mainline Protestant and Black Protestant clergy often address a social justice agenda, but Black Protestant clergy tend to focus more on community empowerment while mainline Protestant clergy tend to pursue social justice activism that seeks to transcend class and national boundaries. Catholic clergy are more likely than others to advance a wide range of issues. Taken together, these findings update and extend our knowledge about clergy political activities, broadening the traditional two-agenda characterization of clergy political agendas into a four-agenda account.

Religion long has been intertwined with politics, and congregations are sites of political involvement for many religious adherents. Congregations can serve as bases for political mobilization (Beyerlein and Chaves 2020), and the leaders of these congregations have considerable power to motivate their congregants toward political action (Olson 2009). Clergy have been prominent in movements on issues such as civil rights (Billingsley 1999), abortion (Danielsen 2021), immigration (Nepstad 2004), and more. Moreover, their activism and political speech may pressure politicians to make policy changes (Morris 1984), affect the results of elections (Edwards and Oyakawa 2022), and shape their congregants’ views about issues (Brown et al. 2017; Hmielowski, Kim, and Kim 2015; Paterson 2018).

While there is much scholarly interest in clergy politics (for recent examples, see Braunstein et al. 2017; Delehanty 2018; or Malina and Hersh 2021), almost no research on this subject is based on nationally representative samples of clergy, and none is based on nationally representative samples of clergy from across the religious spectrum. National clergy surveys about politics have focused almost entirely on clergy in a small number of predominantly white Protestant denominations, most prominently in the Cooperative Clergy Study (CCS) series of denominational surveys (Guth et al. 1997; Smidt 2004, 2016). The CCS is a high-quality national survey of clergy from several of the largest mainline and evangelical Protestant denominations, but it is not fully nationally representative. The 2000 CCS included surveys of Catholic priests, Jewish rabbis, and African Methodist Episcopal ministers, but the representativeness of those samples is less clear than the representativeness of their white Protestant samples, and the published analyses of those data (e.g. Djupe and Sokhey 2003; McDaniel 2003; Jelen 2003) are more limited than the analyses of data on white Protestant clergy. Consequently, despite the fact that political activism often has been led by Black Protestant churches (Barnes 2005; Billingsley 1999), Catholic churches (Nepstad 2019), interfaith coalitions (Yukich and Braunstein 2014), and Buddhist, Muslim, and Hindu communities (Yukich 2017), we know little about the national prevalence of clergy political activity outside of predominantly white Protestant denominations.

We use the National Survey of Religious Leaders (NSRL)—a nationally representative survey of American clergy conducted in 2019 and 2020— to update and extend knowledge about the political activities and agendas of U.S. congregational leaders. We find that clergy are politically active in several ways, but the nature of their activism, and the issues they address, vary considerably by religious tradition.

LITERATURE REVIEW

Clergy Political Activity

As leaders of community organizations and moral authorities, clergy are well positioned to engage in public political activity through direct action and messaging on important issues. While the clergy’s role in American civic life was of interest even to de Tocqueville (1835: ch. 17), scholarly interest in clergy political activism began in earnest in the 1960s amidst the Civil Rights Movement and protests against the Vietnam War. Black clergy who mobilized their congregations to advocate for equal rights were central to the Civil Rights Movement (Morris 1984). At the same time, some researchers focused on the so-called “new breed” of white clergy—a group of young, theologically and politically progressive clergy in mainline Protestant denominations who were especially active on the important social issues of the day, often to the chagrin of their congregants (Cox 1967; Garrett 1973; Hadden 1969; Johnson 1967; Quinley 1970, 1974). Since then, clergy political activism has been studied in connection with community organizing (Delehanty 2018; Wood 2002), abortion (Danielsen 2021), immigrant and refugee protection (Nepstad 2004; Yukich 2013), same-sex marriage (Adler, Hoegeman, and West 2014), and more.

Scholars have identified two primary means by which clergy engage in public political action: direct action and cue giving (Guth et al. 1997; Smidt 2016). Direct action includes a range of activities, including participating in protests (Beyerlein, Soule, and Martin 2015), community organizing (Wood and Fulton 2015), and encouraging get-out-the-vote efforts (Edwards and Oyakawa 2022). Cue giving refers to clergy signaling to their congregants how to think about a controversial issue or which way to vote in an election. Clergy cannot formerly endorse political candidates in their role as leaders of religious organizations without imperiling their congregation’s tax-exempt status, but many clergy do speak about important political issues both from the pulpit and in public settings (Boussalis et al. 2021; Djupe and Gilbert 2003).

Clergy’s direct action and cue giving can influence both policy and people’s policy positions. Black Protestant clergy and churches mobilizing protestors in the 1950s and 1960s as part of the Civil Rights movement were instrumental in pressuring elected officials to pass impactful legislation (Morris 1984), and church-led souls-to-the-polls efforts help parishioners vote (Edwards and Oyakawa 2022). The messages clergy transmit to congregants through their sermons and other communications may shape the way congregants view important political issues (Brown et al. 2017; Brown, Brown, and Jackson 2021; Paterson 2018; Hmielowski et al. 2015) or how they vote (Campbell and Monson 2003). Prior research on clergy political activism generally has found that clergy are much more likely to engage in cue giving than direct action. Djupe and Gilbert (2003), for example, found that, in 1998, clergy within the Episcopal Church and the Evangelical Lutheran Church in America were at least twice as likely to publicly pray about (56 percent of clergy) or take a stand on a controversial political issue (51 percent) than they were to organize congregational study groups (24 percent), participate in a protest or march (8 percent), or actively campaign for a candidate (6 percent). Similarly, Smidt (2016) found that, among clergy from select predominantly white evangelical and mainline denominations surveyed in 2009, only 11 percent organized a congregational action and only 6 percent participated in a protest or march. By contrast, 83 percent urged their congregants to register and vote, 81 percent publicly prayed about an issue, and 63 percent took a public stand on a controversial issue. We examine whether this difference that has been documented among clergy in predominantly white Protestant denominations persists today, whether it exists within other religious groups, and whether this pattern varies across groups.

Clergy Political Agendas

Scholars also have studied the issues clergy address when they engage politically. Early research on clergy political activism focused on politically and theologically liberal clergy in mainline Protestant traditions engaging in movements for civil rights, poverty reduction, and against the Vietnam War (Cox 1967; Garrett 1973; Quinley 1970). Clergy from conservative Protestant traditions were not active in these movements, a fact attributed at the time to a more other-worldly theology (Stark et al. 1971). This interpretation shifted in the 1980s and 1990s with the rise of the religious right and an emerging “culture war,” with scholars now recognizing that both conservative and liberal clergy might be politically active, but on different sets of issues (Beatty and Walter 1989; Hunter 1992; Koller and Retzer 1980). Guth and colleagues (1997) characterized the situation as one in which mainline and Black Protestant clergy were engaged in a “social justice” agenda addressing issues such as poverty, homelessness, healthcare, and racial justice, while evangelical clergy were engaged in a “moral reform” agenda focused on issues such as same-sex marriage, pornography, abortion, gambling, and changing family structures.

Some have argued that the two-agenda thesis no longer describes clergy activism as well as it once did. Using data from 2009, Smidt (2016) suggested that the two-agenda characterization is less applicable than it was before, with fewer clergy addressing issues related to the moral reform agenda and more addressing social justice issues than in years past. Other research, however, has found that mainline and evangelical clergy continue to pursue distinct political agendas. Some progressive clergy in mainline Protestant traditions remain politically active in social justice movements (Braunstein et al. 2017), and activism among progressive congregations likely has increased in recent years, particularly around immigration (Beyerlein and Chaves 2020). On the other side of the aisle, scholars have documented evangelical Protestant clergy political engagement in a range of “moral reform” campaigns including attempts to restrict abortion and LGBTQ rights (Guth et al. 2003; Jelen 1992).

Given the limited data on clergy from all but a few Protestant denominations, it remains unclear if clergy political activism continues to be well-described by the two-agenda thesis, particularly outside of the well-studied predominantly white Protestant traditions. Black Protestant clergy and congregations have a long history of political activism, from the Civil Rights Movement (Billingsley 1999; Harris 1999; McDaniel 2008; Morris 1984) to efforts aimed at community issues such as hunger, poverty, or education (Barnes 2005; Chaves and Higgins 1992). Consequently, the political agendas of Black Protestant clergy seem similar to the social justice agenda observed among white liberal and mainline Protestant clergy but perhaps with a particular focus on issues that directly address the well-being of Black congregations’ local communities.

Catholic clergy also have engaged a range of political issues, including abortion, immigration, contraception, the death penalty, and more (Jelen 2003, 2004; Nepstad 2019). Some scholars have connected the breadth of Catholic clergy’s political involvement to Cardinal Joseph Bernadin’s call for Catholics to support a consistent ethic that values the protection of life (Jelen 1993; Smith 2008). This ethic supports activism on a number of issues that crosscut the agendas of the major political parties in the United States, such as supporting social service spending, opposing the death penalty, and advocating for abortion restrictions. For this reason, Catholic clergy’s activism may not fit neatly into either a moral reform or social justice agenda.

We build on this literature by using the nationally representative NSRL data to update our understanding of clergy political activism and agendas, and extend knowledge on this subject to a wider range of religious traditions. We examine rates of direct political activism and cue giving among clergy, variations in those rates across religious groups, the extent to which clergy are active on various issues, and whether the two-agenda thesis remains an accurate characterization of contemporary clergy political activism.

DATA AND METHODS

Data

The National Survey of Religious Leaders (NSRL) is a nationally representative survey of congregational leaders from across the religious spectrum. It was conducted in conjunction with the fourth wave of the National Congregations Study (NCS-IV) and the 2018 General Social Survey (GSS). The GSS is an in-person survey of a nationally representative sample of non-institutionalized, English- or Spanish-speaking adults conducted by NORC at the University of Chicago (Smith et al. 2019). The 2018 GSS asked respondents who said they attend religious services at least once a year where they attend. The congregations named by GSS participants constitute a nationally representative sample of U.S. congregations. For the NCS-IV, NORC then contacted those congregations and interviewed a key informant, usually a clergyperson or other leader, about the congregation’s people, programs, and characteristics (Chaves et al. 2020). The religious leaders of these NCS-IV congregations constitute the NSRL sample. This design samples congregations with probability proportional to size, though this feature can be undone using survey weights.¹

The NSRL gathered data from congregational leaders between February 2019 and June 2020, primarily via an online self-administered questionnaire. It gathered data from 1,600 congregational leaders, but we focus exclusively on the 890 primary leaders in the sample. We focused only on primary leaders for two reasons. First, they are a more clearly defined population than the more professionally heterogeneous secondary leaders, which includes people with wide-ranging pastoral responsibilities, such as associate pastors, but also includes people with more specialist roles, such as directors of specific programs or ministries. Second, primary leaders’ political activity seems more likely to be impactful for the congregation than secondary leaders’ political activity.

The cooperation rate for the primary leader sample was 70 percent. Considering the NCS-IV’s own 69 percent response rate, the NSRL’s primary leader response rate is approximately 50 percent, with little nonresponse bias on key variables. See Chaves, Roso, and Holleman (2022) for more methodological details about the NSRL.

Measures

Clergy political activities were measured by asking how often within the past two years they had urged people in their congregation to vote; organized a group in their congregation to work on advancing a social political goal; took a stand on a controversial public issue while communicating with their congregation; prayed out loud about a political issue during a worship service; urged people in their congregation to participate in a political action beyond voting; lobbied a public official in their capacity as a religious leader; or participated in a protest or march in their capacity as a religious leader. The response options were “not at all,” “once or twice,” and “more than twice.” We combined the “once or twice” and “more than twice” categories to focus on the percentage of clergy who have engaged in the stated activity at all within the past two years.

Respondents who answered that they did any of the activities listed above, except for urging congregants to vote, were asked if any of their political activities addressed the following issues: abortion, capital punishment, economic issues, education, environmental issues, foreign policy, LGBT issues, gun laws, healthcare, hunger or poverty, immigration, police-community relations, or race relations. Clergy whose only political activity was urging people to vote were not asked these follow-up questions about issues.

Clergy also were asked if they ever had endorsed or opposed a political candidate while preaching, speaking, or writing to their congregation. Those who had not endorsed a candidate were asked if they would endorse a candidate if tax laws allowed it. The response options to this hypothetical question were: “yes, definitely,” “yes, probably,” “I’m not sure,” “no, probably not,” and “no, definitely not.”

The five broad religious traditions in the NSRL are those indicated by the NCS religious tradition variable (TRAD3): Non-Christian, Black Protestant, Roman Catholic, predominantly white mainline Protestant, and predominantly white evangelical/conservative Protestant.² Clergy are classified into one or the other of these traditions based on the denominational affiliation of the congregation they lead. The NSRL sample contains 44 leaders from Jewish, Buddhist, Islamic, Hindu, and other non-Christian traditions, 33 of whom were politically active beyond encouraging congregants to vote. This set of leaders is too small and heterogeneous to serve as a basis for meaningful, reliable inferences, so we do not report results from non-Christian leaders. They are included, however, in analyses of the sample as a whole. See the online NCS codebook for details about how TRAD3 was constructed.

Analysis Strategy

We examined clergy’s political activities and issue priorities, weighting the data so that results reflect the percentages of clergy pursuing each activity or issue without respect to the size of their congregations. Interestingly, the picture is not substantively different if we weight the data so that results reflect the percentages of people in congregations with clergy pursuing each activity or issue. This is because neither clergy political activism nor the issues clergy address in their activism is strongly related to the size of the congregations they serve. The percentages of clergy who pursue each activity or issue are similar to the percentages of religious service attendees in congregations with clergy who pursue each activity or issue.

Any bivariate percentage difference we highlight in the main text is statistically significant at least at the .05 alpha level, using weighted t-tests with bootstrapped standard errors that take into account the use of weighted data. There is one Catholic respondent with a very large weight whose inclusion in weighted analysis sometimes alters the results for Catholic clergy. All analyses on Catholic clergy using weighted data exclude this case; we note when excluding this case meaningfully alters the results.

We used latent class analysis (LCA) to identify clergy political agendas. LCA is an inductive approach that assumes that patterns of responses to survey questions can be explained by respondents belonging to one of several unmeasured categories (Collins and Lanza 2010; Goodman 1974). This approach clusters participants according to their survey responses such that those providing similar answers are grouped together into the same class. We used Latent GOLD 6.0 to fit 10 successive LCA models starting with one latent class and increasing to 10. Using the survey weights in these models made the LCA solutions highly volatile and dependent on relatively minor changes in the model. Consequently, the LCA models were estimated using unweighted data, though the results from comparable LCA estimates using weighted data were substantively similar. In fitting the model, we found that bivariate residuals among several pairs of indicators were correlated, thus violating the assumption of local independence. We relaxed this assumption by allowing these pairs of residuals to co-vary, thus improving model fit.³ We then examined the fit statistics of the models to identify the solution that best explained the data while remaining parsimonious, interpretable, and substantively meaningful.

We then examined how these agendas vary by religious tradition. Using a three-step model that corrects the classification error in latent class estimation (Vermunt 2010), we examined the extent to which a clergyperson’s religious tradition predicts their latent class membership. In doing this, we used a maximum likelihood estimator with proportional classification, implemented in Latent GOLD 6.0 as part of the Step 3 module. We report the distribution of latent classes across religious traditions using the same survey weight used when calculating the percentages of clergy who pursue specific activities or issues.

RESULTS

The Political Activities of Clergy

The vast majority of head clergy engaged in politics in some way in their role as congregational leaders. Figure 1 shows the percentage of clergy from each tradition who have engaged in various political activities as a congregational leader.⁴ Flly 93 percent of clergy reported doing at least one of these activities within the last two years, and many reported doing more than one such activity. This was consistent across religious traditions, with at least 90 percent of clergy from each tradition reporting some level of political engagement.

Figure 1. Political activities of primary leaders across religious traditions.

Source: National Survey of Religious Leaders, 2019–20.

Note: Appendix Table A1 provides the exact percentages behind this figure.

While most clergy engaged in some political activities in their role as congregational leaders, there is substantial variation in the sorts of activities in which they engaged. In general, the most commonly reported political activities involved talking with their congregation about political issues or voting. The most common political activity among head clergy was encouraging their congregants to vote, with 85 percent of head clergy reporting doing this within the last 2 years—a figure statistically significantly higher than the rate for any other activity (p < .001). The next most common activities involved cue giving on controversial political issues. Seventy percent of head clergy reported taking a stand on a controversial political issue in speeches or writings directed toward their congregation, and 66 percent reported publicly praying about such an issue at least once within the last two years.

Less common were activities related to political organizing or direct action. Fewer than half (45 percent) of head clergy urged their congregants to engage in some political activity beyond just voting. Other kinds of direct action were even less common. Only 30 percent of clergy lobbied a public official, 30 percent organized a political group in their congregation, and 23 percent attended a protest or march in their capacity as a religious leader within the last two years. All of these percentages are statistically significantly lower than the prevalence of each cue giving activity (p < .001), but note that they generally are higher than those found both by Djupe and Gilbert (2003) and by Smidt (2016), particularly when it comes to protesting. In Djupe and Gilbert’s 1998 data, only 24 percent of clergy organized a political group at their congregation and only 8 percent engaged in a protest march. Similarly, Smidt’s 2009 data show that 11 percent of clergy organized an action group in their church, and 6 percent participated in a protest march. This may indicate that clergy are more likely to engage in direct political action than they were in the 1990s and early 2000s, but it also may reflect the wider range of religious groups represented in the NSRL. In general, differences in samples and question wording make these NSRL results not directly comparable to earlier results.

Rarest of all activities was explicitly endorsing a political candidate. U.S. tax laws formally prohibit congregational leaders from endorsing candidates if they wish to retain their congregation’s tax-exempt status. In practice, however, some leaders do endorse candidates: 9 percent of head clergy reported that they endorsed a political candidate, a percentage that is statistically significantly lower than the rates for each other practice (p < .001). In a follow-up question, clergy who reported that they had not endorsed a candidate were asked if they would do so if U.S. tax laws allowed it. A further 14 percent of all clergy reported that, hypothetically, they would probably or definitely endorse a candidate if U.S. tax laws allowed them to, meaning that one in four (23 percent) head clergy either have endorsed a candidate or would do so if it did not put their congregation’s tax status at risk.

There were important differences across religious groups. Perhaps the most important difference is that, while evangelical leaders engaged in cue giving at about the same rate as leaders from other traditions, they were substantially less likely to encourage or engage in direct activism. Among evangelical leaders, only 35 percent urged their congregants to get engaged in activism, 14 percent lobbied a public official, and 9 percent attended a protest. By comparison, 52 percent of non-evangelical primary leaders urged political activism in their congregations, 42 percent lobbied an elected official, and 35 percent attended a protest—percentages that are statistically significantly different from evangelicals at least at the .001 alpha level. While evangelicalism as a political force has garnered much attention in recent years, it seems that pastors of evangelical congregations are substantially less politically active than their counterparts in other traditions.

There are ways that other traditions stand out as well. Catholics were especially likely to lobby elected officials (40 percent⁵ compared to 28 percent among non-Catholic clergy, a difference that is statistically significant at the .01 level). However, they were particularly unlikely to endorse candidates, as only 1 percent of Catholic head clergy reported endorsing a candidate, significantly (p < .001) lower than the 9 percent of non-Catholic clergy who had done so. A further 13 percent of Catholic leaders said they would endorse a candidate if tax laws allowed, meaning that 14 percent of Catholic leaders either had endorsed a candidate or would do so if allowed, significantly (p < .05) lower than the 24 percent of non-Catholic clergy who would do so.

Mainline clergy were statistically significantly (p < .001) more likely than others to engage in lobbying (43 percent compared to 26 percent of non-mainline head clergy) and protesting (37 percent compared to 20 percent). Black Protestant clergy were particularly likely to mobilize within their congregations, with statistically significantly higher (p < .001) rates of organizing a congregational group (43 percent compared to 26 percent among other head clergy), and encouraging their congregants to engage in political activism beyond just voting (63 percent compared to 39 percent). They also stand out for being the most likely to say that they had endorsed or would endorse a political candidate, with a third (37 percent) saying as much compared to only 19 percent of other clergy—a difference that is statistically significant (p < .001).

The Political Issues Clergy Pursue

Religious leaders engage in a wide range of issues. Figure 2 shows the issues head clergy addressed when they were politically active and how that varied across religious traditions.⁶ The 18 percent of clergy who did not report engagement in any political activity beyond encouraging their people to vote are excluded. Overall, the issues on which clergy were most politically active were race relations, abortion, hunger or poverty, and LGBT issues, with over 50 percent of politically active clergy addressing each of these issues in their political activities. The percentages of clergy addressing each of these issues is statistically significantly (p < .05) higher than the percentages addressing any of the other issues. Note that NSRL data were collected almost entirely before the murder of George Floyd on May 25, 2020, and the subsequent widespread protests, so these results suggest that racial justice was an important issue for many clergy even before the summer of 2020. Other issues addressed by at least 40 percent of politically active head clergy include immigration (43 percent), economic issues (42 percent), and education (41 percent). Statistically significantly (p < .01) fewer clergy addressed healthcare (34 percent), police-community relations (34 percent), gun laws (32 percent), and environmental issues (31 percent). Of the issues we asked about, the least likely to be addressed were capital punishment (17 percent) and, even less frequently (p < .05), foreign policy (14 percent). Both of these issues were addressed at statistically significantly lower (p < .001) rates than any other issue.

Figure 2. Political issues addressed by clergy across religious tradition.

Source: National Survey of Religious Leaders, 2019–20.

Note: Appendix Table A2 provides the exact percentages behind this figure.

There were substantial differences across traditions in the issues politically active clergy address. These differences suggest that the two-agenda split has persisted, at least among Protestant clergy. Catholic clergy were active on a range of issues. The most common issue addressed by politically active Catholic clergy was abortion, with fully 83 percent of politically active Catholic clergy addressing abortion, statistically significantly (p < .05) more than any other issue. In clear second place was immigration (70 percent), followed by hunger or poverty (58 percent).⁷ Catholic priests also were the only clergy group with substantial activism on capital punishment. Forty-one percent of Catholic priests addressed capital punishment, compared to only 15 percent of non-Catholic clergy—a difference that is statistically significant (p < .001). The only issues on which Catholic clergy were statistically significantly (p < .05) less active than non-Catholics were LGBT issues, race relations, and police-community relations.⁸

While Catholic priests were highly active on many issues, the picture for evangelical clergy is very different in that it indicates a clear preference for a moral reform agenda. Politically active evangelical head clergy were statistically significantly (p < .05) less likely than non-evangelical clergy to address eleven of the thirteen issues the NSRL asked about. The two issues evangelical head clergy addressed at high rates are the “culture war” issues that highlight the political priorities of evangelical religious leaders (Hunter 1992). Evangelical leaders were second only to Catholic leaders in their activism on abortion, with three-quarters (77 percent) addressing abortion within the past two years. The lone issue that evangelical leaders were statistically significantly (p < .05) the most active on was LGBT issues, with 66 percent of politically active primary leaders reporting that they addressed LGBT issues within the past two years. While much has been made about the political power of American evangelicals, it seems that their activism is largely confined to these two issues.

In contrast, Black Protestant leaders had very low levels of activism on LGBT issues and abortion but were highly active on several other, mainly social justice, issues. Unsurprisingly, given the long history of the Black Church’s involvement in advocating for civil rights (Edwards and Oyakawa 2022; Morris 1984), Black Protestant leaders are highly active on race relations and police-community relations. On police-community relations in particular, Black Protestant leaders stand out from clergy in other groups. Fully 67 percent of politically active Black Protestant primary leaders addressed police-community relations in the past two years, compared to only 25 percent of politically active head clergy from other traditions—a very large and statistically significant (p < .001) difference. The story with race relations is similar, with 71 percent of Black Protestant clergy reporting activism on this issue, significantly more than the rate of activism among other clergy (p < .001). Black Protestant head clergy also were highly active on many social welfare issues, with statistically significantly (p < .001) higher rates of engagement on economic issues (64 percent), education (65 percent), and healthcare (56 percent) than other clergy. They were comparatively inactive on the “culture war” issues, with statistically significantly (p < .001) lower rates of engagement on abortion (26 percent) and LGBT issues (23 percent) than other clergy.

Mainline Protestant leaders also were active on a range of social justice issues. Much like Black Protestant leaders, nearly three quarters (72 percent) of politically active mainline Protestant primary leaders addressed race relations in their activism, a percentage that is statistically significantly (p < .001) higher than clergy in other predominantly white traditions. Mainline Protestant primary leaders were also just behind Catholic leaders in addressing immigration (64 percent of politically active leaders) and environmental issues (48 percent), in both cases addressing the issue at statistically significantly (p < .01) higher rates than non-mainline clergy. Unlike their Catholic and evangelical peers, mainline primary leaders were seldom active on the issue of abortion during this period (26 percent of politically active leaders; statistically significantly lower than non-mainline clergy at the .001 level).

Assessing the Two-Agenda Thesis

Are the differences across religious traditions described above accurately characterized by the thesis that clergy essentially follow either a social justice or a moral reform agenda? We used latent class analysis to assess this. For these analyses, we included all clergy who reported that they engaged in at least one political activity other than encouraging people to vote, thus excluding the 121 clergy who were not asked about the political issues they pursued in their activism. Because we are interested in how these agendas vary by religious tradition, and the non-Christian grouping is too small and heterogenous for reliable analysis, we estimate the latent classes just from the sample of 670 Christian clergy who were politically active beyond just voting, and who have complete data on the political issue items. Figure 3 shows the Bayesian information criteria (BIC) of the LCA models, and reveals that a model with four latent classes appears to be the best solution, as it minimizes the BIC.⁹

Figure 3. Fit of LCA models.

Note: Appendix Table A3 provides the full fit statistics for this analysis.

Closely examining the four-class solution shows that the classes are interpretable in substantively meaningful ways. Figure 4 shows the probabilities that clergy categorized within each agenda address each issue, as estimated by the model.¹⁰ We call the first class (containing 11 percent of clergy) Universally Engaged, as it includes clergy who were active across nearly all the political issues the NSRL asked about. The second class of clergy (comprising fully 55 percent of the sample) might be characterized as those pursuing a Moral Reform agenda, but it is dominated by activity on a single issue: abortion. Clergy categorized as pursuing this agenda also are occasionally active on issues such as LGBT issues and race relations, but not at especially high rates compared to all clergy. The high rate of abortion activism here highlights how central abortion was to the political activism of religious leaders before the June 2022 Supreme Court decision to overturn Roe v. Wade.

Figure 4. Estimated probabilities of addressing each issue for clergy categorized as pursuing each agenda.

Note: Appendix Table A4 provides the exact probabilities behind this figure.

The third and fourth classes both indicate a social justice agenda, but with important differences between them. The third class (18 percent of the sample), which we dub Social Justice—Global Citizenship, contains a number of progressive priorities and involves some activism that seeks to transcend class and national boundaries. Clergy in this class were among the most likely to address environmental issues, immigration, economic issues, and LGBT issues. The fourth class (16 percent of the sample), which we call Social Justice—Community Empowerment, is characterized by clergy who primarily address issues of concern in their local communities, such as education, healthcare, and police-community relations.¹¹

Investigating how these agendas vary across religious traditions reveals that the third and fourth LCA classes correspond to differences in issue priorities between pastors of predominantly white liberal Protestant congregations and pastors of Black Protestant congregations. Figure 5 shows the distribution of clergy across these four classes within each religious group.¹² Over half of Black Protestant clergy (54 percent) fall in the Social Justice—Community Empowerment class, while at most 7 percent of clergy from any other Christian tradition do. Some Black Protestant clergy are categorized within the Moral Reform (24 percent) or Universally Engaged (12 percent) classes. Interestingly, although the two social justice agendas share many common issue priorities, such as race relations, economic issues, and hunger, only 9 percent of Black Protestant clergy are categorized as pursuing a Social Justice—Global Citizenship agenda. Instead, the Social Justice—Global Citizenship agenda is favored among mainline Protestant clergy, with 35 percent advancing this agenda compared to at most 27 percent of clergy from other traditions. While Black Protestant and white liberal clergy are aligned in many ways, we find that there are important ways in which their issue priorities differ. A plurality of mainline Protestant clergy also are engaged in a Moral Reform (46 percent) agenda, though substantially fewer than the 78 percent of evangelical leaders categorized as such (p < .001). Only 13 percent of mainline clergy are categorized as Universally Engaged, roughly in line with the population as a whole.

Figure 5. Prevalence of political agendas within religious traditions.

Source: National Survey of Religious Leaders, 2019–20.

Note: Appendix Table A5 provides the exact percentages behind this figure.

In contrast, evangelical activism is overwhelmingly characterized by a single agenda: Moral Reform. As we already noted, a large majority of evangelical clergy (78 percent) were placed in this category, a far greater share than any other tradition (p < .001). Evangelical clergy also were the least likely to be categorized within the Universally Engaged class, with only 5 percent of politically active clergy categorized in this way—statistically significantly fewer (p < .05) than any other tradition. This finding underscores the singular focus of evangelical clergy’s political activism. Though a minority of evangelical leaders advance either a Social Justice—Community Empowerment (5 percent) or a Social Justice—Global Citizenship (13 percent) agenda, they were much less likely to do so than non-evangelical clergy (p < .01).

As was evident in the analysis of individual issues, Catholic priests collectively address a wide range of issues. Priests were more likely than Protestants to address a broader agenda, as 15 percent are categorized in the Universally Engaged class, compared to only 8 percent of Protestant clergy (p < .05). Note, though, that this Protestant-Catholic difference is driven by the contrast between Catholic and evangelical leaders, as Catholic leaders are not significantly more likely to be categorized as Universally Engaged than Black Protestant or mainline Protestant clergy.¹³ Many Catholic priests advance a Moral Reform agenda, with 52 percent doing so, roughly similar to the percentage of mainline clergy categorized in this way. Most of the remainder advanced a Social Justice—Global Citizenship agenda, with 27 percent of politically active Catholic priests advancing this agenda—more than either evangelical or Black Protestant leaders (p < .01). Overall, the higher prevalence of broad-based involvement among Catholic priests is consistent with prior work highlighting the importance of a universal ethic of life that crosscuts conventional political differences (Jelen 1993; Smith 2008). Abortion remains an important issue for Catholic clergy, but, unlike many evangelical leaders, their activism is not singularly focused on that issue.

DISCUSSION AND CONCLUSION

These results advance knowledge about clergy members’ political activism and the agendas they pursue. Consistent with prior research, we found that clergy were politically active in several ways, but there was considerable variation across religious groups in the type of political activities clergy pursued and in the issues clergy addressed in their activism. Across the board, clergy were more likely to engage in political cue giving (such as publicly discussing or praying about controversial political topics) than they were to engage in direct action (such as attending marches or protests). While leaders from all traditions engaged in cue giving at similar rates, leaders of predominantly white evangelical Protestant congregations were substantially less likely to engage in direct action.

Regarding the issues clergy typically address when they are politically active, we found that the long observed two-agenda thesis mostly holds true today when looking at the oft-studied white Protestants, but it needs to be amended when taking a broader view of the American religious landscape. When they are politically active, evangelical leaders are overwhelmingly likely to pursue a moral reform agenda, with a particular focus on abortion. While a plurality of white mainline leaders also pursue a moral reform agenda, they are much more likely than evangelical leaders to pursue a social justice agenda. A majority of Black Protestant leaders pursue a social justice agenda as well. Therefore, the two-agenda thesis remains a broadly accurate characterization of the political agendas of Protestant clergy.

However, although many mainline and Black Protestant clergy pursue a social justice agenda, there are important differences in the sorts of social justice issues they address. Black Protestant activism mainly focuses on addressing the needs of their communities, with high rates of engagement on issues such as education, healthcare, and police-community relations. Mainline Protestant activism, on the other hand, more often focuses on addressing issues outside their own communities, such as immigration or the environment. While there is clear overlap in these agendas—both often address race relations, economic issues, and hunger—they nevertheless represent distinct foci. Looking beyond Protestants, Catholic priests stand out for a higher likelihood of addressing a broad range of issues, including addressing one issue (capital punishment) on which non-Catholic clergy seldom were active.

Overall, then, these results suggest that broadening our view of clergy activism beyond white Protestants prompts amending the traditional two-agenda characterization of clergy political activism to the four-agenda characterization we offer here. That is, rather than characterizing clergy political activists as falling into just two camps—those pursuing moral reform and those pursuing social justice—it seems more appropriate to describe clergy activists as falling into four camps: those pursuing moral reform, those pursuing social justice mainly through community empowerment, those whose social justice work contains more activism that seeks to transcend class and national boundaries, and those who are more universally engaged.

There is every reason to think that clergy will remain politically active, and that, through both political cue giving and direct action, they will continue trying to influence how both their members and public officials view political issues. The findings we presented in this article update and expand knowledge about clergy politics by documenting the rates at which they pursue various activities and address various issues. These findings also show the importance of studying a religious landscape that includes more than white Protestants. We hope these results inspire more research about clergy political activities and the influence clergy wield in their congregations and communities.

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The National Survey of Religious Leaders (NSRL) was funded by the John Templeton Foundation. Data were gathered by NORC at the University of Chicago. The NSRL dataset and codebook are available from the Association of Religion Data Archives (https://www.thearda.com/data-archive?fid=NSRL) and from the Inter-university Consortium for Political and Social Research (https://www.icpsr.umich.edu/web/ICPSR/studies/38576). The codebook also is available at the NSRL website (https://www.nationalsurveyreligiousleaders.org/).

Notes on contributors

Joseph Roso

Joseph Roso is a assistant professor of sociology at Ambrose University. His research examines trends in the culture and politics of American religious congregations and congregational leaders. His work has been published in the Journal for the Scientific Study of Religion Review of Religious Research, Sociology of Religion, and Politics and Religion.

Mark Chaves

Mark Chaves is Anne Firor Scott Distinguished Professor of Sociology, and Professor of Religious Studies and Divinity, at Duke University. Much of his research focuses on the social organization of religion. He directs the National Congregations Study (NCS), a wide-ranging survey of a nationally representative sample of religious congregations, and also the National Survey of Religious Leaders, a similarly wide-ranging survey of leaders of congregations that participated in the fourth wave of the NCS.

Turgut Keskintürk

Turgut Keskintürk is a PhD student in the Department of Sociology at Duke University. His research explores the formation and organization of cultural, political, and moral belief systems. His work has appeared in Poetics and Social Science Research.

Notes

¹ Specifically, when conducting analysis on weighted data we used the variable WT_NSRL_PRIMARY_DUP that limits analysis to just primary leaders and weighs cases inversely proportionate to the number of regularly attending adults at their congregation. For more details, consult Chaves, Roso, and Holleman (2022) and the NSRL codebook and documentation available at the Association of Religious Data Archives: https://www.thearda.com/data-archive?fid=NSRL.

² Among predominantly white Protestant denominations, the mainline-conservative/evangelical distinction captures differences in views about the Bible and attitudes about adapting religious traditions to cultural change. Mainline denominations such as the United Methodist Church, Episcopal Church, Presbyterian Church (USA), Evangelical Lutheran Church in America, United Church of Christ, and Christian Church (Disciplines of Christ) are more liberal in these areas by interpreting the Bible in light of the historical and social conditions in which it was written and expressing openness to other religions and the secular world. Evangelical/conservative denominations such as the Southern Baptist Convention, Assemblies of God, Lutheran Church—Missouri Synod, and most nondenominational Christian congregations are more conservative in these areas by interpreting the Bible literally and viewing other religions and secular institutions as targets of conversion rather than as partners in efforts to make a better world.

³ The pairs of indicators with correlated residuals are capital punishment and abortion, gun laws and abortion, race relations and LGBT issues, abortion and LGBT issues, hunger or poverty and economic issues, and race relations and police-community relations.

⁴ Appendix Table A1 provides the exact values behind Figure 1.

⁵ When the Catholic outlier case is included, the percentage of Catholics who lobbied an elected official increased to 55 percent.

⁶ Appendix Table A2 provides the exact values behind Figure 2.

⁷ Including the outlier Catholic case increases the percent of Catholic leaders active on immigration and hunger or poverty to 79 percent and 71 percent, respectively.

⁸ When the outlier case is included, Catholic clergy have close to average activism on race relations (52 percent of politically active Catholic leaders) and police-community relations (43 percent).

⁹ Appendix Table A3 provides the full set of fit statistics.

¹⁰ Appendix Table A4 provides the exact values behind Figure 4.

¹¹ An LCA model constrained to only three classes combined these third and fourth classes into a single social justice agenda. The vast majority of clergy sorted into either of the social justice classes in the four-class model were placed in the combined social justice class in the three-class model.

¹² Appendix Table A5 provides the exact percentages behind Figure 5.

¹³ When including the single Catholic case with very large weight, Catholic leaders are statistically significantly more likely to be categorized as Universally Engaged than clergy in any other tradition.

Appendix

Table A1. Percentages of Head Clergy Engaging in Political Activities

	All Clergy	Catholic	Conservative Protestant	Black Protestant	Mainline Protestant
Any activity	93.1 %	93.7 %	93.6 %	90.9 %	96.1 %
Any activity beyond urging voting	82.0 %	78.3 %	81.9 %	83.3 %	86.9 %
Urged voting	84.8 %	82.0 %	87.1 %	89.1 %	74.6 %
Took public stand	70.0 %	66.7 %	74.0 %	66.2 %	73.3 %
Prayed publicly about issue	66.4 %	60.0 %	69.3 %	70.4 %	72.1 %
Urged activism	44.6 %	50.6 %	34.5 %	63.4 %	47.6 %
Lobbied	29.6 %	39.8 %	13.5 %	37.8 %	42.6 %
Organized political group	29.5 %	44.1 %	18.2 %	42.9 %	35.2 %
Protested	23.4 %	40.6 %	8.9 %	27.5 %	37.1 %
Endorsed candidate	8.9 %	1.1 %	10.4 %	15.5 %	3.7 %
Endorsed OR would endorse if allowed	22.9 %	13.5 %	22.0 %	37.1 %	11.3 %
N	890	182	317	117	229

Source: National Survey of Religious Leaders, 2019–20.

Table A2. Percentages of Politically Active Head Clergy Addressing Each Issue in Their Activism

	Overall	Catholic	Conservative Protestant	Black Protestant	Mainline Protestant
Race relations	55.3 %	30.9 %	39.1 %	71.4 %	71.7 %
Abortion	54.0 %	83.0 %	77.0 %	26.3 %	26.0 %
Hunger or poverty	50.8 %	58.0 %	37.2 %	49.1 %	62.3 %
LGBT issues	50.6 %	24.1 %	65.7 %	23.1 %	53.7 %
Immigration	43.4 %	69.8 %	32.9 %	25.5 %	63.5 %
Economic issues	41.6 %	48.1 %	20.9 %	63.5 %	46.5 %
Education	41.0 %	44.3 %	13.5 %	65.4 %	36.0 %
Healthcare	33.6 %	38.4 %	14.2 %	55.7 %	38.3 %
Police-community relations	33.5 %	17.6 %	22.0 %	67.2 %	18.5 %
Gun laws	32.2 %	23.7 %	15.4 %	31.4 %	46.9 %
Environmental issues	31.4 %	38.2 %	13.5 %	28.6 %	47.5 %
Capital punishment	17.4 %	41.4 %	8.2 %	28.0 %	6.0 %
Foreign policy	13.6 %	16.1 %	9.6 %	12.1 %	14.7 %
N	750	162	255	98	201

Source: National Survey of Religious Leaders, 2019–20.

Table A3. Fit Statistics of LCA Models

Number of Classes	LL	BIC	Parameters	L^2	df	Error	Entropy
1	−5,442	10,968	13	3,117	657	0.00	1.00
2	−4,961	10,098	27	2,156	643	0.06	0.79
3	−4,880	10,027	41	1,994	629	0.10	0.74
4	−4,824	10,005	55	1,881	615	0.13	0.73
4*	−4,760	9,918	61	1,754	609	0.13	0.72
5	−4,783	10,015	69	1,799	601	0.13	0.75
6	−4,746	10,033	83	1,726	587	0.14	0.78
7	−4,724	10,078	97	1,681	573	0.15	0.77
8	−4,703	10,128	111	1,639	559	0.16	0.77
9	−4,686	10,185	125	1,605	545	0.16	0.79
10	−4,664	10,232	139	1,561	531	0.14	0.81

*Residuals for Capital Punishment and Abortion, Guns and Abortion, Race and LGBT, Abortion and LGBT, Hunger and Economics, and Race and Police were allowed to covary.

Table A4. Estimated Probabilities of Addressing Each Issue for Clergy Categorized As Pursuing Each Agenda

	Agenda
	Universally Engaged	Moral Reform	Social Justice—Global Citizenship	Social Justice—Community Empowerment
Race relations	.885	.372	.714	.738
Abortion	.691	.667	.439	.304
Hunger or poverty	.975	.390	.807	.663
LGBT issues	.858	.370	.631	.220
Immigration	.931	.304	.847	.341
Economic issues	.928	.159	.698	.627
Education	.965	.173	.440	.830
Healthcare	.932	.060	.411	.676
Police-community relations	.804	.146	.173	.670
Gun laws	.870	.074	.469	.256
Environmental issues	.760	.080	.646	.159
Capital punishment	.678	.078	.235	.191
Foreign policy	.604	.028	.166	.001

Source: National Survey of Religious Leaders, 2019–20.

Table A5. Distribution of Politically Active Clergy Across Political Agendas, by Religious Tradition

	All Clergy	Catholic	Conservative Protestant	Black Protestant	Mainline Protestant
Universally Engaged	10.8 %	15.1 %	4.6 %	12.4 %	12.7 %
Moral Reform	55.0 %	52.2 %	77.8 %	24.1 %	45.6 %
Social Justice—Global Citizenship	18.0 %	26.8 %	13.0 %	9.4 %	35.1 %
Social Justice—Community Empowerment	16.2 %	5.8 %	4.7 %	54.1 %	6.7 %

Source: National Survey of Religious Leaders, 2019–20.