Public Deliberation about Gay Rights in Religious Contexts : Commitment to Deliberative Norms and Practice in ELCA Congregations

Many political theorists extol the virtues of deliberation in efforts to reconcile differences in opinion and prevent group fracture. On August 21, 2009, the Evangelical Lutheran Church in America (ELCA) voted narrowly to reverse standing policy by allowing gay and lesbian people in committed relationships to serve openly as clergy. In the aftermath of this decision, numerous congregations began thinking about leaving the denomination. We surveyed a sample of ELCA clergy in the fall of 2009 and spring 2010 to assess their commitment to deliberative norms and practice, their implementation of such practices in congregational meetings designed to discuss the ELCA’s vote, and the outcomes of those more or less deliberative forums. We found considerable commitment both to deliberative practice itself and belief in the efficacy of such practice. Despite the assumption that religious doctrine and public deliberation are incompatible, religious organizations often find deliberative processes essential to their survival.

Can public deliberation thrive in religious contexts?We might expect that reliance on authoritative religious doctrine would leave little room for meaningful deliberation, particularly about contentious issues, in houses of worship.In fact, religion may well be a "conversation stopper" (Rorty 1994(Rorty , 2003)), meaning that the introduction of religious arguments prevents productive deliberation and thus ends debate.
Most political conversations between individuals take place within communities of like-minded people (Mutz 2006) within social institutions, including organized religion (Habermas 1992).Since houses of worship are the most common form of association in the United States (Wald and Calhoun-Brown 2011), the nature of the conversations held within them bears tremendous importance for how democracy, writ large, is practiced.The exchange of information that takes place inside congregations has the potential to reinforce preexisting belief systems and result in uncritical adoption of elite-communicated positions by congregation members.On the other hand, communication within congregations also might serve to engage people with different perspectives in a "liberalism of reasoned respect" (Weithman 1997), wherein citizens deliberate about public issues on shared grounds, generating trust and mutual respect (Gutmann and Thompson 1996;Rawls 1993).Houses of worship exist (and compete) within a vibrant religious economy (Finke and Stark 2005), which leads us to suspect that churches1 are capable of serving as settings for productive deliberative encounters.
The purpose of our inquiry here is to document and analyze meetings guided by deliberative norms inside religious organizations in the contemporary United States.We take as our case the aftermath of the August, 2009, vote by one mainline Protestant2 denomination -the Evangelical Lutheran Church in America (ELCAto allow openly gay and lesbian people in committed relationships to serve as clergy if called by a congregation.Despite the ELCA's relative political progressivism (Wuthnow and Evans 2002) and longstanding effort to grapple publicly with homosexuality (Cadge 2002;Cadge, Olson, and Wildeman 2008), there was considerable controversy surrounding this razor-thin denominational vote.The vote's aftermath has seen individuals and entire congregations leave or consider leaving the denomination.Might diverging opinions about gay clergy tear congregations apart, and under what circumstances might these differences be overcome?To what extent could intentional deliberative practice about the ELCA vote have mitigated conflict within congregations?We apply the question of whether and how congregations used deliberative practice to engage their differences to consider the crucial question of whether congregations can remain together or break apart in light of this controversial new policy.
Two national surveys of ELCA clergy document whether congregations engaged in discussions about the issues involved in this policy shift and whether those discussions conformed to deliberative norms and practice.Our analysis is couched in the context of deliberative practice in reaction to this particular episode of the gay rights debate.We focus on three conceptual angles: the need for deliberation in the congregation, the ability of the congregation to deliberate, and the desire of the clergy and congregation to deliberate. 3

Deliberation in Religious Contexts
Organized religion brings people together for a wide variety of reasons, including a particular worship style, a sense of community, the opportunity to learn about a faith, theological fit, religious education for children, or simply out of tradition, invitation, or marriage (Ammerman 1997(Ammerman , 2005;;Becker 1999;Chaves 2004).Not surprisingly, congregations often assemble strikingly heterogeneous sets of people (Ammerman 1997;Chaves 2004;Djupe and Gilbert 2009;Gilbert 1993;Neiheisel, Djupe, and Sokhey 2009;Wald, Owen, and Hill 1988) that must be managed by their clergy.
It is the essential purpose of deliberative practice to sustain an organization by specifying a process that builds mutual trust and respect, which may allow for the resolution of even the most difficult of moral disagreements (see Gutmann andThompson 1996, 2004).Therefore, deliberation is an organizational maintenance strategy that clergy could employ quite effectively.Most tenets of deliberation are accessible because they are almost intuitive: asking participants to specify a process and a question, encouraging mutual respect, urging participants to take the process seriously, informing people so they are prepared to discuss, and asking that participants not appeal to absolutist sources.It is reasonable to expect voluntary social organizations (such as churches), especially those characterized by disagreement, to rely on deliberative practice to sustain their organization in the face of a diversity of opinion.
Nor may necessity be the only reason why churches engage in deliberative practice.Following developments in the political tolerance literature (see, e.g., Marcus et al. 1995), Djupe and Calfano (2012) explore three ways in which houses of worship can augment support for deliberative democracy.First, religious organizations may directly attempt to instill support for deliberative norms and practice.This is the traditional role assigned to churches as value-instilling institutions (Durkheim 1915(Durkheim /1965;;Leege et al. 2002).Second, church members may learn about deliberative practice as a result of their engagement in the congregation.One likely place to locate such practice is in adult education classes, which sometimes play host to discussions of divisive political issues (Neiheisel, Djupe, and Sokhey 2009; see also Olson, Djupe, and Cadge 2011).Third, congregations may teach individual members how to apply their values to specific issues by working through examples, or what Marcus and colleagues (1995) refer to as "contemporary information" shaping tolerance judgments.How an ELCA congregation chose to deal with its denomination's controversial vote on gay and lesbian clergy constitutes a good example of the ongoing opportunities congregations have to engage in deliberative practice.
A few previous studies have directly examined such deliberations inside religious organizations.Wood and Bloch (1995) examine the United Methodist Church's debate about homosexuality and find that denominational assemblies resemble legislatures in their potential to spread norms of civility and reasoned debate.Likewise, at an elite level, Shields (2007) observes deliberative processes within Christian Right movement organizations in debates about choosing between politically expedient action and principled action.At the local level, Djupe and Neiheisel (2008; see also Djupe and Calfano 2012;Olson, Djupe, and Cadge 2011) find clergy modeling the deliberative process in their public speech on a gay rights ballot initiative, engaging arguments with which they disagreed, and illustrating the importance of entertaining multiple perspectives in public debate.Neiheisel, Djupe, and Sokhey (2009) look within the church to the setting they think is most likely to host deliberation -adult education sessions -and find considerable commitment to deliberative norms across the theological spectrum.Coffin (2005) also notices the potential for deliberation in congregations around controversial issues.Thus, scattered empirical studies have affirmed the deliberative potential of congregations.
Our explanation of whether and how congregations take up controversial issues in a deliberative manner depends on three basic hypotheses: • Churches that need deliberative practice to manage their diversity will be more likely to adopt it.• Churches with the capacity to handle deliberative practice will be more likely to adopt it.• Churches that desire deliberative practice (whether the motivation comes from clergy or congregation members) will be more likely to adopt it.

Need for Deliberative Practice
Due to the complexity and vicissitudes of the religious marketplace, houses of worship are forced to compete for adherents and thus to innovate in their practices, services, and outreach, which forces them to absorb the practical consequences of attracting a diverse range of people (Finke and Stark 2005).Although the religious economy also promotes differentiation, niche development, and in some instances, exclusion, Stark and Finke (2000) argue that most congregations take an inclusive stance, encourage outreach, and cater to a wide range of needs to permit their survival.From this perspective, the modal condition of American congregations is one of diversity.
To the extent that diversity characterizes congregation members' political views, deliberation may be essential to protect the congregation from unraveling (see Johnson 1996 on "unraveling").Theorists argue that deliberative norms help sustain popular investment in any group (e.g., Gutmann and Thompson 1996).Disagreement will not devolve into exclusion, intolerance, mistrust, or dislike when a group operates explicitly under norms that encourage trust, tolerance, and mutual respect (see Delli Carpini, Cook, and Jacobs 2004;Huckfeldt, Johnson, and Sprague 2004;Mendelberg 2002;Mutz 2006; but see Sunstein 2002).
Although studious avoidance of political issues is also an option for congregations (Becker 1999), we think it is not a viable approach in the long run.The ELCA vote (which we describe below) is a good example of why conflict avoidance lacks long-term viability.In this instance, conflict generated by other, related actors (in this case, the denomination's Churchwide Assembly) resulted in conflict at the congregational level.In fact, from a democratic theory perspective, one of the benefits of being a part of a federal structure is that actions taken at higher levels ensure that issues cannot be avoided, providing mechanisms for interests to be aggregated to the point of salient decision-making (Hunter 1993).We suspect that this is one reason why adult education -that is, structured forums that convene regularly to put adult congregants in formal conversation with one another -is much more common in congregations that belong to denominational bodies than it is in non-affiliated churches (Neiheisel, Djupe, and Sokhey 2009).

Capacity for Deliberative Practice
The functional question of whether a congregation should turn to deliberative practice to manage disagreement is moot if its members are unable to handle the resource-intensive practice of deliberation.Engaging in reasoned, tolerant debate is difficult and information-intensive.It is unlikely that a large majority of citizens possess the required intellectual and temperamental resources to participate in such deliberation on a regular basis.Although countless examples of citizen juries and other forums illustrate the universal applicability of deliberation (Fishkin 1995;Gastil 2000;Gastil and Levine 2005), deliberative forums cannot and do not happen without coordination.Clergy's perception that the congregation has sufficient capacity for deliberation is essential.Deliberation will be less likely to proceed if clergy do not initiate (or at a minimum, sanction) it.Thus it is highly instructive to understand clergy's perceptions of their congregations' openness to deliberative practice.
A clergyperson's perceptions of the congregation's deliberative capacity may be influenced in a variety of ways that dovetail with the political participation literature's treatment of resources (see Verba, Schlozman, and Brady 1995).Socioeconomic status is important, since higher education is tightly linked with liberalism, promoting individualism, tolerance, and the ability to engage in reasoned debate (e.g., Golebiowska 1995;Nie, Junn, and Stehlik-Barry 1996).Deliberative capacity also is influenced by the organization's culture and historical practices that exist apart from its material wealth, just as the civic skills people develop from participating in social organizations have a beneficial effect on individual political participation (Verba, Schlozman, and Brady 1995).That is, the extent to which the congregation has engaged in deliberative practice in settings such as adult education classes and similar forums should augment its capacity for deliberation at any given time, simply because its members have enough experience to know how to deliberate.Furthermore, a culture that values significant social interaction and association among congregation members (measured by widespread involvement in small groups and activities) may engender sufficient confidence in members to allow them to handle a deliberative meeting (see Becker 1999).
Many clergy also play direct -even intentional -roles in preparing the congregation for deliberation.It takes more than clear ground rules for deliberation to work.Participants also need to have an argument repertoire upon which to draw; exposure to discussion and disagreement also increases one's familiarity with a variety of points of view (Price, Capella, and Nir 2002).Clergy do engage public issues on a regular basis (Djupe and Gilbert 2003;Guth et al. 1997;Smidt 2004), but only recently have we understood more about how and with what depth they engage particular issues (Djupe and Calfano 2012;Djupe and Neiheisel 2008;Olson, Djupe, and Cadge 2011).
Some earlier research has tended to assume that clergy are "policy maximizers," trying primarily to convince their congregations of the political implications of religious doctrine (Guth et al. 1997).More recent work, however, reveals that many clergy today display support for strong, talk-centered deliberative practice (Djupe and Neiheisel 2008).Specifically, clergy discuss roughly one argument with which they disagree for every four with which they agree.Remarkably, this ratio holds across issues, communities, and denominations. 4 Djupe and Neiheisel (2008) see this finding as evidence of clergy modeling the deliberative process.By Price, Capella, and Nir's (2002) logic, such a pattern of clergy speech would enlarge the argument repertoire of the congregation.However, such communication also is obviously one-sided since it comes from one (inherently biased) source, which could prevent deliberative norms from thriving.And as scholars of clergy politics have argued for decades, clergy frequently disagree with their congregations; when they do so publicly, it is often because they wish to provide a prophetic voice (Djupe and Gilbert 2003;Hofrenning 1995;Quinley 1974).
Nevertheless, we suspect that the more ELCA clergy engage issues related to the denominational vote on gay and lesbian clergy, the better the deliberative quality of the congregation's discussions is likely to be.The more a clergyperson directly engages the issue in his/her public speech, the more congregation members will be exposed to a variety of frames through which to view the debate.Members of such congregations thus might have an enhanced understanding of how others might view the issue.On the other hand, it is possible that outspokenness on the part of clergy might constrain the types of arguments participants are able to make.The same could be true when the denomination takes a formal stance on an issue (Cadge, Olson, and Wildeman 2008).It is also possible that the generally one-sided nature of clergy communication could detract from deliberative quality by stacking the deck in favor of one position.

Desire for Deliberative Practice
We must not lose sight of the possibility that practical considerations may give way to a simple lack of desire to implement deliberative norms.True deliberation requires that no source be regarded as superior to any other to ensure neutrality among all argument sources.Yet in many religious contexts, doctrine inherently supersedes all other sources of knowledge -and clergy are the primary professors of the faith.This hardly means, however, that clergy necessarily expect their congregations to accept doctrine at the expense of genuine deliberation.We expect to show that the more clergy value the implementation of deliberative norms, the more likely the congregation will be to follow these norms in ongoing discussions about controversial issues.
Desire for deliberation on the part of the congregation matters as well.The most appropriate measures of such desire would be congregation members' support for deliberative norms, but we lack adequate data on this front.What we do have as a proxy is information about members' degree of involvement in congregational activities, which should be related to the demand for citizen participation in decisionmaking processes (Putnam 1993).Moreover, greater exposure to debate about controversial issues in settings such as adult education forums should create a demand for (or at least a tolerance of) more debate, especially when the issue at hand is highly salient to both the denomination and congregation.

Factors Undermining Deliberation
Simply holding a meeting to discuss a contentious issue hardly equals productive deliberation.Besides an absence of forces that promote deliberative practice in the congregation, there are several factors that might impede deliberation.For one, the presence in the congregation of a member who is especially affected by the policy debate could inhibit deliberation that includes all available stances.Specifically, the presence in the congregation of gay or lesbian members or people struggling with their sexuality could inhibit the range of arguments participants feel free to present, thus inhibiting the quality of the deliberative process (Neiheisel and Djupe 2008).
We might also be concerned about the emotions people bring to deliberation about matters as charged as homosexuality.Passionate argument is not necessarily incompatible with deliberation (Gutmann and Thompson 2004: 50-51), but there is a difference between expressions of passionate argument and anger.In short, angry outbursts and expressions add little to deliberative practice.
Finally, we suspect that deliberative quality will suffer in meetings that are arranged by the participants themselves.The problem here is that participants, rather than an elite whose self-interest is tied to the maintenance of the organization (like a clergyperson), are in charge.Such meetings may be less likely to have an elite playing a moderating role and thus might be more likely to permit or even encourage emotional outbursts. 5In this way, lay-called meetings may suffer from some of the same problems as direct democracy (Cronin 1999 as cited in Gutmann and Thompson 2004) even while they satisfy the principle of equality of participants in a more comprehensive way (Gutmann and Thompson 1996).

The August 2009 ELCA Vote on Gay and Lesbian Clergy
Our case is set in the aftermath of a high-profile vote about whether to ordain openly gay and lesbian clergy that was taken at the ELCA's Churchwide Assembly on August 21, 2009.The biennial Churchwide Assembly is the highest legislative authority in this hierarchically organized mainline Protestant denomination.The clergy and laity who comprise the Churchwide Assembly are elected by, and represent, the denomination's 65 regional synods.There were 1,045 voting members of the 2009 Churchwide Assembly, 60 percent of whom were laity.The Churchwide Assembly has the authority to amend the ELCA's constitution and bylaws, elect denominational officials and board members, set denomination-wide policy, and adopt a budget, among other functions (ELCA 2009b).
It was not unusual that the ELCA held a formal vote about ordaining openly gay and lesbian clergy, nor was it necessarily surprising that it passed.The ELCA (which was created by a merger of three smaller Lutheran church bodies in 1987) is a progressive denomination that prioritizes social justice and inclusion in its formal policies (Cimino 2003).Clergy and laity in the denominations that merged to form the ELCA were active supporters of the Civil Rights Movement (Findlay 1993;Walz and Montreal 2007), and ELCA clergy today skew left in their political orientation (Djupe and Gilbert 2003;Guth et al. 2007;Smidt 2004;Walz and Montreal 2007).
The ELCA also has a particularly strong commitment to open discussion of difficult issues at the congregational level (Cadge, Olson, and Wildeman 2008;Cimino 2003).Since 1991, the denomination has formally declared itself "a community where open, passionate, and respectful deliberation on challenging and controversial issues of contemporary society is expected and encouraged" (Bloomquist and Duty 1999: 1).One of the publications the ELCA provides to its congregations is essentially a howto manual on deliberation, titled Talking Together as Christians about Tough Social Issues (Bloomquist and Duty 1999).
After many years of intentional study (Cadge 2002), the 2009 Churchwide Assembly voted to approve a series of resolutions that had the effect of committing the ELCA "to finding ways to allow congregations that choose to do so to recognize, support, and hold publicly accountable couples who wish to have lifelong, monogamous, same-gender relationships" (ELCA 2009a;Salmon 2009).The vote was so contentious that the Churchwide Assembly passed a preceding collegiality resolution specifying that ELCA members should "bear one another's burdens, love the neighbor and respect the bound consciences" of all (ELCA 2009a).The very fact that it formalized all of these positions falls in line with the denomination's view that it should "arrive at positions to guide its corporate witness through participatory process of moral deliberation" (Bloomquist and Duty 1999: 1; see also Cimino 2003).
Reactions to the close vote on gay clergy (559 to 451) ranged from jubilation to predictions of impending denominational schism, both inside the Churchwide Assembly meetings and across the ELCA's thousands of American congregations.An illustration from a 2010 news story (Condon 2010) relates the story of Rev. Gail Sowell, who served as pastor of two ELCA congregations in Wisconsin until members of both were outraged by the vote; one of the churches went so far as to fire her and leave the denomination."'It was pretty gruesome, Sowell said, recalling shouting matches inside the sanctuary; the mass resignation of one church's council, save one member; even whispers around town that she was a lesbian" (Condon 2010).By a year after the vote, an estimated two percent of ELCA congregations had left the denomination (Spencer 2010); others had endured substantial congregational turmoil, primarily because many ELCA laity had come to see the denomination as too liberal for their tastes.
To address this divisive, high-profile change in ELCA policy, many congregations held special meetings or forums to enhance their understanding of the vote and air their opinions about it.Unlike other mainline Protestant denominations, the ELCA already was primed to discuss homosexuality in a formal, ongoing fashion (Cadge, Olson, and Wildeman 2008).The denomination has convened various task forces and undertaken formal, national and regional study programs about homosexuality.One important element of the study programs was sending discussion materials and guides to all ELCA congregations for their consideration.Previous research documents a high level of clergy satisfaction with these materials (Cadge, Olson, and Wildeman 2008).

Data and Design
Our data collection strategy unfolded in two steps.Using a strategy employed in previous studies (Beyerlein and Chaves 2003;Chaves et al. 1999;Djupe and Olson 2010; see also Schwadel and Dougherty 2010), we surveyed a random sample of ELCA clergy in the fall of 2009 and spring of 2010.The fall survey, which we undertook in November and December, 2009, captured congregational experiences with issues related to the August vote, while the spring survey, conducted in March, 2010, was designed to collect evidence about whether the passage of time might have changed congregational dynamics around the subject of gay clergy.The spring survey also allowed us to double our overall sample size.
In both survey instruments, questions asked (1) if and how gay rights and homosexuality had come up in the clergy respondent's congregation; Our response rate was fairly typical for surveys of clergy.In the fall, we sampled 1,045 congregations from the online ELCA directory using a 1/10 interval and a random seed, and proceeding through the states in alphabetical order.Of those congregations, 916 listed an email address or were found to have one through subsequent web searches.We administered our survey online through the Survey Methods website.Churches with email addresses were sent email invitations with instructions to forward the email to the head pastor if the recipient did not occupy that position.Churches without listed emails were sent letters addressed to the head pastor that specified a web address at which to take the survey online.Of the 221 responses we received, 214 originated from the email invitation (clergy received three email reminders) and seven resulted from the mailed invitation (which came with no reminder).The overall response rate was 21.1 percent.
For the spring survey, we obtained the full list of congregations directly from the ELCA.We first removed all congregations that we had sampled in the fall, as well as those that were included in Calvin College's 2009 Cooperative Clergy Study, before taking a simple random sample of all remaining congregations.We compared our fall and spring samples to the total population of ELCA clergy, and both were representative. 6In total, 1,100 clergy were sampled in the spring, of whom 953 listed email addresses.As in the fall, the former were sent email invitations (and three reminder emails) while the latter received invitations by mail (and one reminder letter).Of those invited by email, 200 individuals (21.0 percent) responded, while 21 individuals (14.3 percent) of the mail invitees responded, yielding an overall response rate of 20.1 percent. 7In the end, there were so few differences between the two samples that we combined them for further analyses.

Holding Congregational Meetings
If deliberation has an organizational component, we should be most likely to find it in open meetings.Thus, we begin by asking whether the congregation held any public meetings about the ELCA's vote on gay and lesbian clergy.In our fall survey, 46 percent of clergy reported that their churches had held meetings to discuss the vote, with the percentage increasing very slightly to 48 percent (an insignificant difference) by the spring.Congregations that were perceived by their clergy to be opposed to the outcome of the ELCA's vote were much more likely to have held meetings, as Figure 1 demonstrates.Figure 1 also illustrates the distribution of disagreement with the denomination's decision to allow gay clergy.In congregations that were perceived by their clergy to disagree with the decision, 55 percent held an open meeting, compared to just 32 percent of congregations perceived to agree with the decision.Roughly two thirds of clergy in both the fall (65 percent) and the spring (69 percent) perceived their congregations to disagree with the decision, and - consistent with one the ELCA's formal goals (Bloomquist and Duty 1999)evidently such discord gave rise to more deliberation in the form of congregational meetings.
Separately, we asked each clergyperson whether he or she personally would have voted in favor of the new policy on gay and lesbian clergy at the 2009 Churchwide Assembly.These results closely resemble the actual vote, with 56 percent saying they were in favor (54 in the fall and 58 in the spring; this difference is insignificant) and 44 percent opposed.Next we asked how strongly committed respondents were to their position on the vote; those opposed were significantly less committed to their position than were those in favor.The reason for this difference in intensity of preference may lie in responses to the next question, which asked for the clergyperson's position on the vote in principle (not thinking about any potential ramifications for the denomination).Given that condition, the proportion saying they would support the policy increased by almost ten percent, such that roughly two thirds would have agreed with the decision in both the fall (63 percent) and spring (66 percent).Thus, in the aggregate clergy see themselves as the mirror image of their congregations regarding support for gay clergy (at least in principle): two thirds of clergy favor the policy, while fewer than one third of them think their congregations agree with that position.
Among clergy whose congregations had held meetings to discuss the vote by the spring, a plurality (46 percent) had one such meeting; a quarter had two; 12 Proportion in Agreement percent had three; and 9 percent each had four or five meetings.About two thirds of these meetings were called by the clergyperson, with the remainder called by the congregation.No pattern appears regarding who called these meetings on the basis of clergy-congregation disagreement about the vote.Clergy reported relatively low attendance at these meetings, with most (about three quarters) of the meetings attracting a small minority (25 percent or less) of congregation members.Not surprisingly, attendance was proportionately lower in larger congregations.Attendance was markedly higher, however, in congregations that hold substantial numbers of adult education sessions,8 suggesting that congregational contexts where face-to-face meetings are the norm are especially likely to promote deliberation about difficult topics.Attendance also was highest when the congregation (as opposed to the clergyperson) called the meeting.
We model whether the congregation held a meeting to discuss the ELCA vote in Table 1, testing our conceptual framework of deliberation being driven by need, capacity, and desire.We find that meetings were more likely to happen in congregations that were perceived by their clergy to oppose the decision (28 percent more likely, moving from the mean-minus-one standard deviation to the mean-plusone standard deviation).In some instances, these meetings would have been part of a process of determining whether the congregation will stay in the ELCA, as almost all (90 percent) of the congregations that were considering leaving the denomination (9.8 percent of the sample in the spring) had held an open meeting about the policy on gay clergy by spring 2010.However, we should recall Figure 1, which shows that a reasonably large proportion (about a third) of congregations perceived by their clergy to support the vote held open meetings as well.Thus, these meetings should hardly be thought of primarily as forums for the discontented.
It is noteworthy how perceived agreement between clergy and their congregations bears upon whether meetings were held to discuss the ELCA vote.When clergy believe they and their congregation agree on the outcome of the vote, there is a twelve percent decrease in the probability of holding a meeting on the vote.Here we assume that convergence in opinion reduces the need for an open forum.On the other hand, clergy-congregation disagreement is significant in several other ways.First, our results show that such disagreement boosts the likelihood of holding a meeting.Moreover, congregations described by their clergy as "politically divided" are more likely to meet.Taken together, disagreement with the vote and with others in the congregation about the policy it generated would seem to create a need for deliberation.Whether such meetings actually do calm conflict, however, is beyond the scope of our study.Model Statistics: N=366, -2LL=425.18,Nagelkerke R 2 =.262, % correctly predicted=69.4,% error reduction=29.0∆ refers to the average effect of a +/-1 standard deviation shift from the mean of the independent variable.
One of our central contentions is that when a group has experience dealing with controversial issues, the group will be better equipped to address brewing conflict.This should be the case for congregations grappling with disagreement about their denomination's policy on gay clergy.Indeed, our data indicate that congregations with a strong commitment to adult education are more likely to have held a meeting to discuss the vote.The particular norms that govern the congregation's adult education sessions 9 seem to matter less than the simple fact that such sessions are held in the first place.However, there is a peculiar relationship between broader norms about the value of exposing arguments to disagreement and the propensity to hold a meeting.This relationship is negative, meaning that commitment to deliberative norms is higher among clergy whose congregations did not hold a meeting.Clergy who endorse deliberative norms tend to be more politically liberal and thus more likely to perceive disagreement with their congregations about the vote outcome and political questions in general.However, these variables are controlled in our model, 10 so our hypothesis is that liberal clergy might be worried that a meeting to discuss the vote would not be sufficiently open and deliberative.Hence, no meeting at all would be better than a meeting that undermines broader deliberative norms.
We also assess the effect of clergy's public speech about various issues that are relevant to the vote on gay clergy.These measures are designed using an approach to analyzing clergy speech pioneered by Djupe and Neiheisel (2008), who presented clergy with a series of arguments about an issue and asked (1) whether they had mentioned the argument and (2) whether they agreed with the argument.This strategy measures both the amount and content of clergy speech on a given issue.From a battery of six arguments, we use two measures: a "volume of speech" measure (the mean number of arguments mentioned was 3.2; s.d.= 1.5), and a measure capturing whether clergy presented at least one argument with which they disagreed (19 percent mentioned a disagreeable argument).The volume of speech measure has no effect on the likelihood of a congregational meeting.The effect of mentioning a disagreeable argument depends on the individual clergyperson's feelings of job-bound political efficacy, as measured by this statement: "Ministers have great capacity to influence the political views of their congregation."Previous research shows that in general, this efficacy measure correlates well with religious tradition, with higher agreement likely in evangelical Protestant churches (Djupe and Gilbert 2009).The efficacy measure also leads to different effects of clergy speech, with highly efficacious clergy more likely to influence the opinions of church members (Djupe and Gilbert 2009).
Figure 2 reveals that efficacy works similarly in our study: congregations with highly efficacious clergy who present a disagreeable argument are about 20 percent more likely to hold a meeting, whereas congregations with low-efficacy clergy who 9 We asked whether adult education sessions held in the church were governed by deliberative norms.Clergy could agree or disagree whether (a) "they would explicitly encourage participants to think seriously about the views of others;" (b) "it would be essential for all attenders to participate;" (c) "it would be essential that a range of views are presented;" (d) "it would be essential for participants to learn how our values apply to issues;" and (e) "it would be essential for participants to learn how to talk through their differences." 10 Adding a political ideology variable to the model does not change the democratic norms result; ideology is also nowhere near significant and changes no other effect.present a disagreeable argument are about 10 percent less likely to hold a meeting.These effects are among the strongest in our model.Clergy are not going to wade into territory where they feel uncomfortable, especially if it could undermine the health of their congregations.On the other hand, confident, efficacious clergy who are willing to tackle positions with which they personally feel uncomfortable may model effective deliberation for their congregations.
What about various contextual effects specific to the composition of the congregation?The extent to which a congregation is composed of lifelong members of the ELCA (i.e., those who have the most "brand loyalty") increases the likelihood of a meeting on the vote having taken place.Lifelong members are more likely to care enough to spend their time addressing a purely denominational issue, both because of their knowledge of the denomination and their longstanding investment in it. 11We also find a sizable effect for church size: larger congregations are more likely to hold meetings.This is so in part because larger congregations are more diverse, and meetings help to process this diversity.Also, the basic principles of organizational sociology suggest that smaller congregations would have less need for meetings because informal conversation is likely to suffice (see Becker 1999).In 11 More brand-loyal congregations are smaller, more rural, founded earlier, less engaged in outreach activities, and have declining memberships, suggesting some degree of isolation from the broader climate of opinion.larger congregations, however, it would be difficult to process disagreement about contentious issues informally, so meetings become necessary.All told, there is considerable variance in whether ELCA congregations held a meeting to discuss the denominational vote to allow gay clergy, and our model explains a reasonable amount of it.Perceived clergy-congregation disagreement -on multiple levels -plays a considerable role.In such settings, meetings likely give voice to dissension and absorb and process conflict.This pattern emphasizes the importance of organizational maintenance, which serves to maintain harmony of the congregation, if not the denomination.

Meeting Process Quality
It is not enough for us to know whether or not congregations held meetings to discuss the ELCA vote on gay and lesbian clergy, since meetings can be civil or not, productive or not, and well attended or not, among other things.To gauge the deliberative quality of these meetings, we asked clergy about the presence or absence of three norms: (1) whether all those attending the meeting(s) participated, (2) whether most attendees expressed their opinions about the vote, and (3) whether attendees took seriously the arguments expressed by their peers.Although these three items do not capture all aspects of deliberation (see Mendelberg and Oleske 2000 for a full list), they do capture its essential elements. 12We averaged the three measures (each coded from 1 = strongly disagree to 4 = strongly agree) into an index of deliberative quality (α = 0.69).The distribution of the resulting variable (Figure 3) has a mean of 2.8 (averaging just below "agree") and a standard deviation of 0.6, suggesting that clergy tend to think the congregation's meetings were, on average, both participatory and respectful.The skew statistic is small and negative, but the standard error is larger than the statistic and the kurtosis statistic is small (suggesting it does not deviate from normal).
Results of our OLS regression analysis of deliberative quality appear in Table 2.The evidence shows that congregations perceived by their clergy to oppose the outcome of the vote held less deliberative meetings.This is the largest effect in the model, about twice as large as any other effect and covering about 10 percent of the dependent variable range.We easily can conclude that emotions were running highest in congregations that opposed the new policy on gay clergy, especially since the correlations between opposition to the vote and anger and frustration being voiced at the meeting are quite strong (r = 0.328 for opposition and anger; r = 0.467 for opposition and frustration).Expressions of anger almost certainly interfere with reasoned debate and reciprocity.However, the effect of opposition in the model is not determinative; other factors are at play as well.For one, the clergyperson's own stance on the vote interacts with the congregation's stance to affect perceptions of process quality.In Figure 4, we see that when clergy and congregation agree with the vote's outcome (solid line, left side), as well as when the clergy themselves disagree with the vote (dashed line), clergy's evaluations of meeting process quality are uniformly high.It is only when the clergyperson supports the vote and the congregation opposes it (solid line, right side) that we see low evaluations of meeting process quality. 13We must bear in mind that it the clergyperson reporting the meeting 13 This finding raises the specter that it is not the congregation that cannot deliberate, but the clergyperson who cannot see through a fog of disagreement.It is also possible that congregants are aware of their disagreement with clergy and thus hold back from expressing their opinions to the group.Fortunately, we can test these notions with available data, although our focus on clergy's (mis)perceptions is a limitation.We use t tests assessing the differences in perceptions of the meeting based on clergy support for the ELCA vote among congregations that did not support the vote, thus assessing whether congregations that disagreed with the vote are different when headed by a clergyperson who supports the vote.We find that congregations are just as likely to express anger regardless of their clergyperson's stance on the vote.Frustration, however, is more common in congregations where clergy disagree with the vote.Disagreement was more commonly perceived between members in congregations with clergy who support the vote, but more disagreement was expressed in churches with clergy who opposed it.An equal amount of disagreement was expressed about the vote among meeting participants regardless of their clergyperson's position.Lastly, churches headed by clergy who supported the vote were perceived as more politically divided.The evidence is mixed at best, but does not seem to confirm either counter-hypothesis.The evidence does seem to counter the idea that congregants participate less because of disagreement with the clergy.In fact, disagreement between meeting participants, which creates a more demanding form of participation, process quality here, not congregation members, so it may be the case that clergy who support gay rights felt especially frustrated if their congregations disagree.Several variables measuring the congregation's participatory culture find statistical purchase as well.Whereas the number of available adult education appears more common in such churches.Since those churches are also more politically divided in the first place, according to the clergyperson, reduced process quality could be due to the difficulty inherent in processing considerable disagreement.opportunities predicts holding a meeting (recall Table 1), having such opportunities in place seems not to affect how meetings about the ELCA vote were run.Instead, what matters most are the congregational norms by which adult education sessions are governed.At the same time, greater availability of congregational groups and activities apart from adult education boosts process quality by about the same amount as adult education.It is not surprising that the extent of formal social relationships among congregants seems to promote reciprocity and civility in meetings about a controversial issue.That said, congregations described by their clergy as "politically divided" appear to have reduced process quality, although this effect is statistically marginal.And when members of the congregation called the meeting, the clergy take a dimmer view of the proceedings (by about 0.15 points) than if they called the meeting themselves.In that situation, the clergy obviously would have played a less prominent (or no) role, which likely contributes to their feelings about the meeting and how it actually was run.How clergy have dealt with issues related to the ELCA vote in their public speech also affects the perceived quality of the congregation's meeting about the vote.Congregations headed by clergy who address many political issues have better quality meetings.The relative effect of whether clergy were willing to mention an argument with which they disagree depends, as it does in Table 1, on the clergyperson's feelings of job-bound political efficacy (see Figure 5).However, in this case efficacy has a negative effect, and the mention of a disagreeable message by a highly efficacious clergyperson deepens that negative effect.This result says something about both the clergyperson and congregation.According to Djupe and Gilbert (2009), low-efficacy clergy are so for two reasons: either they tend to disagree with their congregations or their congregations are comprised of more newcomers (as opposed to lifelong ELCA members).Still, when low-efficacy clergy engage both sides of the debate, they appear to be able to prepare their members for productive deliberation.On the other hand, due to a particular style of argumentation or who receives the message (or both), high-efficacy clergy undermine the process quality of meetings when they address disagreeable arguments.
The deliberative quality (at least as we have measured it) of meetings to discuss the vote depends on a combination of history and currency.How the congregation has dealt with controversial issues in the past (as approximated by norms governing adult education) and its general participatory culture combine to affect how new controversies are processed.In addition to these historical norms and practices, several current forces dictate the quality of meeting process.Disagreement with the ELCA's decision tends to undercut meeting process quality, as does internal division within the congregation over general political matters.Nonetheless, clergy can help lay the groundwork for quality deliberation both by engaging the issue themselves and by providing a diversity of views for congregation members to consider.In this way, they serve not only as opinion leaders, but also as models of productive deliberation.

Meeting Outcomes
We now turn our attention to two potential learning outcomes of congregational meetings about the ELCA vote. 14We asked respondent clergy whether they thought meeting participants "came to understand their differences and each other better" and "came to a better understanding of the issues surrounding the ELCA vote."Most clergy agreed with both statements, with more agreeing that participants learned about the issues surrounding the vote (73 percent) than about their differences and each other (55 percent).Figure 6 shows how these two questions fit together in a graphic display of a two-way contingency table; it shows the distribution of learning about each other (in the columns) for each level of learning about the ELCA vote (in the rows).It is much more common for clergy to disagree strongly that participants learned anything about the ELCA vote and each other (75% in the bottom row) than it was for them to agree strongly to both statements (31% in the top row).Moreover, there is a good deal of slippage in the middle categories (agree and disagree), with less than two-thirds giving the same judgment to both questions.
Table 3 contains ordinal logit estimates of our models of these two questions about meeting outcomes.We will discuss both models simultaneously, noting differences between the two as we go.In each model, conduct at the meeting is an important determinant of learning outcomes.When anger was expressed in the meetings, clergy were less likely to think participants learned about each other and about the ELCA vote.However, a more deliberative process (the dependent variable in Table 2, used here as an independent variable) promotes learning at roughly the same rate as the expression of anger in the meetings suppresses it.Surprisingly, these two independent variables (deliberative quality and expressions of anger) are not correlated (r = -0.03,p = 0.65, n = 190).The level of attendance at the meeting has no effect on learning, nor does who called the meeting, at least with regard to learning about each other.When the congregation called the meeting, clergy are more likely to report that meeting participants learned about issues surrounding the ELCA vote.
The stances on the vote held by the clergy and the congregation have little influence on perceived learning about the vote, but both have potent effects on clergy's perception that participants learned about one another.The averaged prediction is presented in Figure 7 for high (mean+sd) and low (mean-sd) combinations of the two variables.When the congregation disagrees with the new ELCA policy, clergy are likely to report a score of 2.5, which is dead center on the scale), meaning that congregations are neither more nor less likely to learn about each other in these meetings.However, differences emerge among congregations that agree with the ELCA vote outcome.When the congregation agrees with the clergyperson's position, the clergyperson is likely to perceive learning, but the (very few) clergy who disagree with their congregations supportive of the policy report lower levels of interpersonal learning.Ignoring the latter results for a moment (because of their limited application), the remaining results shown in Figure 7 illuminate the difficulty of reciprocity under conditions of moral disagreement.And moral disagreement applies not only to opinion dynamics inside the group, but between members of the group and an opinion-object -in this case, the ELCA Churchwide Assembly.Even agreement among meeting participants is not sufficient to promote learning when such disagreement with the ELCA is present.Of course, the presence of agreement (both within the discussion group and between the group and the external decision-making body) undermines some of the necessary conditions for deliberation to occur in the first place.
That said, deliberative experience and norms regarding the conduct of adult education sessions have little influence on meeting participants' learning about each other.However, they do have significant (if small) effects on learning about the    3, column 1.
ELCA vote.More adult education, especially when deliberative norms are guiding it, promotes learning about the vote (at least according to clergy's perceptions).Thus we see the payoff of nurturing a participatory culture within an organization.On the other hand, clergy adherence to broader deliberative norms appears to diminish perceived learning about the ELCA vote in congregational meetings, an effect for which we have no explanation.Meeting participants nevertheless appear to learn more about each other when clergy adhere to deliberative norms.
Nothing else in our model predicts learning about the ELCA vote, but two effects remain for learning about other meeting participants.When clergy perceive agreement between their own political views and those of the congregation, meeting participants learned more about each other.Also, the more clergy engage issues related to the ELCA vote in their public speech, the less learning about one another is perceived to take Recall that congregations in which clergy address a wider variety of issues were more likely to hold a meeting -and to experience relatively high-quality deliberation at their meetings.These two effects indicate the complicated nature of preparing a group for meaningful political deliberation.For example, clergy who speak frequently about a range of viewpoints about controversial issues seem to prepare their congregations for deliberation by strengthening their argument repertoires (Price et al. 2002).However, such preparation also can strengthen resistance to new ideas (Chaiken, Giner-Sorolla, and Chen 1996;Petty and Cacioppo 1979).If most members know they already share political views in common, they would expect concurrence at meetings about contentious issues; whether they actually hear concurring views or project those expectations on participants is immaterial.Hence, the more that participants are forewarned and have a priori expectations about what will occur in a meeting (and how it will occur), the less likely they may be to learn about other participants in the context of such meetings.
A straightforward relationship between a deliberative process and social learning is highly desirable (Barabas 2004;Price, and Nir 2002), but education often is an explicit component of preparation for deliberative meetings (e.g., Fishkin 1996;Gastil and Levine 2005).At the same time, there is a relationship between deliberation (process quality) and learning, but it is not determinative.Many of the factors that predict a high-quality deliberative process either undermine learning or fail to facilitate it.In the end, we realize that deliberation and learning are two distinct activities, perhaps especially in the context of a long simmering issue like the practice of openly gay clergy in this denomination; learning may be more likely to result from deliberation involving novel issues.In fact, while clergy may hope learning will be a result of congregational meetings, they probably hope most for a civil process that will let the organization live to fight another day.That is, organizational maintenance is the number-one goal, and deliberative practice is a way to protect the organization by giving voice to, and thus tolerating (and maintaining), disagreement.

Conclusion
A wide range of political theorists, theologians, and other observers often are troubled by the possibility and presumption that religious groups value religious dogmas and traditions over and above any other source of knowledge.Fidelity to religious sources of knowledge may encourage disobedience, incivility, and a lack of reciprocity for citizens who do not adhere to the same doctrine.In many ways, we disagree with the proposed a priori tension between religion and deliberation and see a way through this tension in the normal practices of organizational religion.Our research shows that disagreement is a commonplace characteristic of congregations, yet civility is a prized norm within most of them.Thus we suspect that a commitment to deliberative norms in organized religious contexts is much more common than scholars often surmise.In fact, a few previous studies have found as much (Coffin 2005;Djupe and Neiheisel 2008;Djupe and Olson 2010;Neiheisel, Djupe, and Sokhey 2009;Olson, Djupe, and Cadge 2011;Wood and Bloch 1995).
We took as our case study a situation of profound moral disagreement -the question of whether to allow gay and lesbian clergy to serve openly in the ELCAand looked inside congregations for commitment to deliberative norms, practices, and outcomes.Results from two rounds of surveys of nationally representative samples of ELCA clergy indicate widespread commitment to deliberative norms in meetings convened to discuss the controversial ELCA vote.
Our results reinforce, and give further shape to, the argument advanced by Huckfeldt and Mendez (2008), who detail a dynamic of moths drawn to flames in political discussion.That is, some people are drawn to political discussion even when it may result in disagreement, which then tends to draw down discussion levels.Huckfeldt and Mendez posit that an individual's political motivation drives this dynamic, but we insert a different mechanism by arguing that organizational leaders (such as clergy) inspire deliberation out of a need to confront and create space for disagreement.The deleterious effects of disagreement may be managed if differences are brought out into the open and placed within a structure that is equipped to handle conflict (see also Huckfeldt, Johnson, and Sprague 2004).Deliberative practice is precisely such a structure because it creates a safe space for disagreement.When congregations encourage discussion between and among people who disagree, they implicitly allow disagreement to be overcome -and the congregation to survive or even thrive.Some of this dynamic surely is compatible with Huckfeldt and Mendez's (2008) argument that individuals seek out discussion with disagreeable partners, but it is important to realize as well that exposure to disagreement is due at least in part to elite mobilization within organizational contexts.Mutz (2006;Mutz and Mondak 2006) has formulated the compelling tension between deliberation and political participation.Those who are exposed to disagreement in their political discussions are more likely to avoid participating in politics than people who encounter only consonance among their political contacts.At the same time, those exposed to disagreement tend to have more tolerant attitudes and more moderate opinions (Mutz 2002).The result is a "democratic dilemma" wherein one can either deliberate and encounter desirable tolerant and moderate opinions, or participate and bring more intolerant and extreme opinions to the table.
Voluntary organizations, such as congregations, are an important potential mediating force in Mutz's democratic dilemma.As we have found, organizations comprised of people who disagree with each other often resort to deliberative practice to sustain the organization.Thus open meetings rooted in explicit norms that tolerate (even reify) disagreement are one means of ensuring safety for factions within the organization.We suspect that deliberative practice in civil society enhances tolerance and moderates extreme opinions by bringing wide-ranging information and arguments to the attention of those who otherwise would not have been exposed to them.And engaging significant organizational and personal resources to cope with disagreement within an organization is likely to detract from the will and ability to engage a broader political agenda.
Fortunately, we have evidence to address this point.Clergy of congregations that held meetings to discuss the vote indicate that they would like the ELCA to be less involved with political issues, less involved in political deliberations within the denomination, and less involved in advocating the denomination's positions before government. 15These findings are all the more compelling because the questions we asked about desired denominational engagement were some of the first items on our survey instrument, appearing well before anything was asked about meetings about the ELCA vote on gay clergy (so these items do not suffer from survey-order effects that might remind the respondent to hold a particular attitude).It therefore would seem that disagreement within the organization (congregation) compels organizational elites (clergy) to manage disagreement by confronting it through a structure designed to safeguard the organization, thereby detracting from the resources and desire necessary to mobilize congregants to participate further in politics.Disagreement between individuals may still create participatory dilemmas, but one significant mechanism sustaining Mutz's democratic dilemma is organizational maintenance in civil society.
To be sure, organizational maintenance is hardly a pernicious use of deliberative practice; it is one of its goals.Deliberation is employed instead of having to resort to force to resolve disagreements in any society (Gutmann and Thompson 1996), whether in a state or an organization.A healthy deliberative process should result in societal survival rather than violence, chaos, and dissolution.It is for this reason that religious congregations draw upon deliberative practice to sustain themselves (as do any other forms of social organization).For those organizations that are confronted with disagreement, whether fueled from the outside (perhaps by denominations) or from within (perhaps by new members or the reemergence of dormant issues or divisions), deliberation may constitute the best hope of coping and thus surviving.Moreover, in the words of the ELCA itself, "Open and inclusive processes of deliberation … contribute toward the up-building of the common good and the revitalizing of public life" (Bloomquist and Duty 1999: 1). 15The t-test for the effect of holding meetings and denominational political involvement is 3.4, p = 0.00; the t-test for the effect of meetings on denominational political deliberation is 3.1, p = 0.00; and the t-test for the effect of meeting on denominational lobbying is 4.0, p = 0.00.The mean differences range from 0.33 to 0.44 (on a four-point scale).The effects of holding a meeting interact with the clergy's stance on the vote in a regression model of the desire for political action from the denomination.Clergy who disagree with the vote are less supportive of the denomination engaging in politics (an index of the three involvement variables), but retreat a bit from that stance to be more supportive when their congregation has met.On the other hand, clergy who support the vote's outcome are more likely to want the denomination to engage in politics, but retreat from that stance when their congregation has met to discuss the vote.This is precisely the effect that deliberation is supposed to have: it should moderate opinions by exposing them to a diversity of information and arguments.
raised Lutheran, 3 = most but not all raised Lutheran, 2 = less than a majority raised Lutheran, 1 = No, most are newcomers to Lutheranism.Congregation class status "Would you say that members of your congregation are primarily of what class?" 1 = working, 2 = lower-middle, 3 = middle, 4 = uppermiddle, 5 = upper Declining membership trend Over the past five years, has your membership been.
.. 1 = increasing, 2 = stable, 3 = decreasing.Outreach activities Is an index ranging from 0 to 4 composed of the following items: "In the last 12 months, has your church done any of the following?Sponsored an outreach event intended to bring people into your congregation; Used different worship materials (e.g., music) to appeal to non-members; Had a special committee to work on recruiting new members; Mailed or distributed newsletters, letters, or flyers to recruit new members."Church size, logged "What is the approximate average weekly attendance at all worship services?"Church involvement level "About what percentage of adult members are active in the congregation beyond attending weekly services in any activity (governing boards, small groups, programs, etc.)?" Years serving congregation "How many years have you served your present congregation?"Female clergy 0 = male, 1 = female.Fall vs. spring survey 0 = fall, 1 = spring.Member is openly gay We first asked clergy "How did the issue of homosexuality and/or gay rights *first* come up in your congregation?(please choose the *one* most accurate response)."And then followed up by asking "Since that first occurrence, what has motivated the discussion of homosexuality and/or gay rights in your congregation?(please choose *as many* as apply)."Out of a list of 16 items, we coded responses the following responses to suggest that a member was openly gay ( = 1; otherwise = 0), "Member struggling with own sexuality; Member came out as gay; Family of a gay/lesbian individual brought it up."Anger expressed at meeting After an introduction, "Please tell us what happened at the meeting by agreeing or disagreeing with the following statements."We asked if, "The expression of views by some attenders could be seen as angry."Coded 4 = Strongly agree, 3 = Agree, 2 = Disagree, 1 = Strongly disagree.Level of attendance at meeting "Please describe the attendance at the meetingwas it... 1 = most of the congregation, 2 = half of the congregation, 3 = quarter of the congregation, 4 = only sparsely attended." (2) what clergy thought about the August decision and its ramifications; (3) what clergy thought their congregations' opinions were about the August decision; (4) whether the congregation had held any open meetings to discuss the vote and what such meetings were like; (5) whether clergy had publicly addressed gay rights, homosexuality, or the denominational vote; and (6) a series of questions about the clergyperson and his/her perceptions of the congregation.

Figure 1 :
Figure 1: Distribution of Congregational Meetings across Congregational Agreement with the ELCA Vote (also showing the distribution of perceived congregational agreement with the vote).

Figure 2 :
Figure 2: Interactive Effect of Clergy Efficacy and Mentioning a Disagreeable Argument on Holding a Congregational Meeting about the ELCA Vote.

Figure 3 :
Figure 3: Distribution of Meeting Process Quality Assessments

Figure 4 :
Figure 4: Interactive Effect of Clergy and Congregational Stance on the Vote on the Process Quality in the Congregational Meeting on the Vote

Figure 5 :
Figure 5: Interactive Effect of Clergy Efficacy and Clergy Mentioning a Disagreeable Argument on the Perceived Process Quality in the Congregational Meeting on the Vote

Figure 6 :
Figure 6: Whether Meeting Participants Learned about Each Other and their Differences (columns) by Whether they Learned about the ELCA Vote (row %)

Figure 7 :
Figure 7: Interactive Effect of Clergy and Congregational Stance on the Vote on Whether Participants Learned about Their Differences and Each Other

Table 2 .
19 Djupe and Olson: Public Deliberation about Gay Rights in Religious Contexts Produced by The Berkeley Electronic Press, 2013

Table 3 :
Whether the Meeting Helped Attendees Learn More about Each Other and the ELCA Vote (Ordinal Logistic Regression Estimates)