Worship Discourse and White Race-based Policy Attitudes

Abstract

The current study relies upon the 2004 National Politics Study to examine the association between exposure to race-based messages within places of worship and White race-based policy attitudes. The present study challenges the notion that, for White Americans, religiosity inevitably leads to racial prejudice. Rather, we argue, as others have, that religion exists on a continuum that spans from reinforcing to challenging the status quo of social inequality. Our findings suggests that the extent to which Whites discuss race along with the potential need for public policy solutions to address racial inequality within worship spaces, worship attendance contributes to support for public policies aimed at reducing racial inequality. On the other hand, apolitical and non-structural racial discussions within worship settings do seemingly little to move many Whites to challenge dominant idealistic perceptions of race that eschews public policy interventions as solutions to racial inequality.

Introduction

Despite the persistence of large racial disparities in all facets of quality of life in the United States, there has been a growing belief among Whites that government programs aimed at providing African Americans with special opportunities for social-economic advancement are no longer necessary (Jones & Saad, 2011; Pew Research Center, 2007; 2009). These sentiments undoubtedly contributed to the passage of state ballot initiatives in California (Proposition 209) in 1996, Washington (Initiative 200) in 1998, Michigan (Proposal 2) in 2006, Nebraska (Civil Rights Initiative 424) in 2008, and Arizona (Proposition 107) in 2010 that have overturned Affirmative Action. In an attempt to preserve this policy, the National Council of Catholic Bishops, the National Council of Churches, and various Mainline Protestant, Catholic, and social justice oriented inter-faith organizations have passed numerous resolutions, lobbied governing bodies, filed amicus curiae briefs, and participated in demonstrations in its defense.

The above groups largely frame racial inequality in structural terms. Framing involves the usage of interpretive schemas to make sense of the world (Goffman, 1974). In doing so, many religious leaders rely upon popular religious stories, songs, and texts to shape a common narrative of the world around them (Patillo-McCoy, 1998). In using structural frames, religious groups assert that racial inequality is the result of a history of unfair practices, apathy towards such disparities, and antipathy towards policies aimed at correcting such disparities. As a consequence, progressive Christian organizations maintain that religious bodies are called to side and act with the “least of these” by challenging corporate, governmental, and other entities that contribute to social inequality and oppression (Swarts, 2008; Warren, 2001; Wood, 2002). Religious institutions, they argue, do God’s work when they advocate for policies aimed at bringing about greater social equality so that all people can live up to their God given potential (Warren, 2001; Wood, 2002). Ignoring or supporting public policies and practices that allow more limited social-economic opportunities for some groups relative to others ignores the work of Jesus in caring for the underserved (Boff & Boff, 1987; Warren, 2001; Wood, 2002).

To be clear, not all faith groups have publically embraced Affirmative Action. Evangelical Protestant leaders have been relatively quiet on this issue. The National Association of Evangelicals and the leadership of the Southern Baptist Convention, the largest Evangelical Protestant denomination, tend to embrace an individualistic frame that lends support, or at most offers a weak challenge, to the status quo of social inequality. These groups largely frame racial inequality as the result of frayed relationships between Blacks and Whites (Bartkowski, 2004; Emerson & Smith, 2001). It is the Church’s role, they argue, to provide a space to atone for the maltreatment of racial/ethnic minorities and to create opportunities for interracial understanding and friendships (Bartkowski, 2004). These groups tend to believe that only salvation through Christ, not public policies, will lead individuals to sincerely atone for past racial prejudices and treat everyone fairly, which, in time, will end racial inequality (Alumkal, 2004; Bartkowski, 2004; Emerson & Smith, 2001). These ideals help explain why theologically conservative Evangelical political activists are more likely than are theologically liberal Mainline and Catholic activists to believe that “if enough people were brought to Christ, social ills would take care of themselves (Green, Jones, & Cox, 2009, pg. 26).”

While Evangelical and non-Evangelical leaders and activists tend to part ways on issues of racial justice, their rank and file is much less distinctive. The overwhelming majority of White Mainliners, Catholics, and Evangelicals, over three-quarters, oppose considerations for racial/ethnic minorities in hiring and over half oppose Affirmative Action in principle (Cavendish, 2004; Verter, 2002). Limited lay support for racial justice likely increases the hesitancy of some clergy to publicly discuss racial issues; nonetheless, many continue to do so. The authors’ analyses of both the 2000 Religion and Politics Study and the 2004 National Politics Study, upon which the present study relies, indicates that over half of worship-attending Whites report having heard sermons/lectures at least once a year about the importance of improving race relations (Jackson et. al, 2004; Wuthnow, 2000). Over forty percent of Whites report that their congregations have hosted or sponsored a program to improve race relations (Jackson et. al, 2004). This leads to the following research question; in instances in which discussions about race occur within worship settings, what role do these discussions play in informing White race-based policy attitudes?

The present study aims to add to the literature on religion and racial attitudes in a number of important ways. To our knowledge, no other survey-based study investigates the role that racial discussions within places of worship play on how Americans think about race. The qualitative studies on this subject have been instrumental in informing our understanding of how religious congregations create safe places for individuals to candidly talk about race (Warren, 2001; Wood, 2002). Nonetheless, it is difficult to determine how much one can generalize from such studies. It is also difficult to determine from such studies if exposure to racial messages within houses of worship is associated with a real change in racial attitudes or if participants maintain partisan, ideological, demographic, or other characteristics that are related to both congregation-based racial discussion and racial attitudes.

There are marked differences between this study and the number of survey-based studies on religion and racial attitudes; these studies have largely excluded constructs on exposure to race-based discussions within places of worship (Brown, 2009; Edgell & Tranby, 2007; Hinojosa & Park, 2004). In doing so, other survey-based studies fail to account for the varying theological frames that may lend themselves to racial justice, apathy, or racial prejudice and intolerance. The present study attempts to fill this gap in the literature by examining the relationship between exposure to race-based messages within houses of worship on White attitudes towards Affirmative Action. The next section provides a review of the literature on religion and racial attitudes.

Religion and Racial Attitudes

Despite a variety of theological orientations, research largely suggests that, for Whites, religiosity contributes to support for a dominant ideological outlook. This outlook explains social inequality via the greater work ethic and ingenuity of some groups relative to others. The dominant ideology largely rests upon the premise that the U.S.A. grants everyone the same opportunities (Kluegel & Smith, 1981). It is, therefore, up to the individual to take advantage of such opportunities through their hard work and persistence. This ideology rejects the assertion that certain groups are conferred special privileges that grant them greater access to social-economic and political resources while others face unfair barriers (Kluegel & Smith, 1981). The extent to which racial inequality exists, a greater emphasis is placed upon the poor work ethic of racial/ethnic minorities than on their more limited access to high quality schools, jobs, friendships with influential people, and other social-economic resources (Kluegel, 1990; Kluegel & Smith, 1981; 1983). The civil rights policies of the 1960s arguably contributed to an increasing acceptance of this ideology as growing numbers of Whites began believing that Blacks are no more social or economically disadvantaged than themselves, and viewed Affirmative Action as unnecessary (Kluegel & Smith, 1983).

The emphasis that some religious bodies place upon an unquestioning acceptance of their teachings may lend itself to relatively few adherents challenging the dominant ideological frame on social inequality. Identifying as members of the dominant racial and religious group may also heighten support for an ideology that places a premium on individualistic explanations for social inequality and the subsequent social-economic and political dominance of Whites. Indeed, the more religiously involved Whites are, the less likely they are to believe that racial inequality is a result of structural forces such as racial prejudice and discrimination, unfair laws and institutions, or the lack of access to good schools and social connections (Edgell & Tranby, 2007). The more orthodox Whites are in their religious beliefs, the more likely they are to believe that Blacks are poor because of a lack of hard work (Edgell & Tranby, 2007). The more importance Whites ascribe to religion, the more likely they are to believe that Blacks should do the same as other groups without any special favors; that if Blacks would only try harder they could be just as well off as Whites; and that it is not the government’s job to guarantee equal opportunity for Blacks (Allen & Kuo, 1991).

Along these lines, White Christians (e.g. Evangelical Protestants, Mainline Protestants, Catholics) are less likely than the religiously unaffiliated and persons of Non-Christian faiths to believe that the lack of educational opportunities and racial discrimination account for racial inequality (Hinojosa & Park, 2004; Eitle & Steffens, 2009). Conversely, White Christians are more likely than the religiously unaffiliated and Non-Christians to accept the individualistic explanation of a lack of motivation on the part of Blacks as an explanation for racial inequality (Hinojosa & Park, 2004; Eitle & Steffens, 2009). It follows that Christians are more likely than the religiously unaffiliated and Non-Christians to oppose Affirmative Action and government spending for Blacks (Taylor & Merino, 2011).

Some scholars point to cultural differences among White Christians as contributing to White Evangelicals maintaining a more individualistic outlook on racial inequality than others (Brown, 2009; Emerson & Smith, 2001). Nonetheless, the most consistent and largest differences in such attitudes persist between White Christians and non-believers. Overall, past studies suggest that, among Whites, religiosity is associated with an individualistic interpretation of racial inequality and a rejection of policies aimed at correcting such inequalities.

There is still much to learn about the association between religion and support for race-based policies. Because past survey research on this subject has not included measures of worship-based racial discourse or the inequality frames of congregants, understanding the degree to which religious bodies are able to challenge the dominant perspective on social inequality still remains unclear. The next section provides a discussion on the role that race-based messages within houses of worship play in informing attitudes about racial inequality.

Worship Discourse and Racial Attitudes

The social capital resources of trust and solidarity that exist among members of religious congregations make it reasonable to expect that discussions about race within worship spaces may inform Affirmative Action attitudes. The trust and solidarity that exists among worship-goers contributes to members valuing the opinions of fellow congregants and clergy. Religious elites often attempt to shape congregant attitudes by utilizing cultural capital in the form of religious texts, songs, and allegories to frame contemporary issues (Patillo-McCoy, 1998). A number of studies point to the importance of clergy and lay leaders in informing political attitudes and behaviors (Brown, 2003; 2011; Djupe & Gilbert, 2008).

If congregational leaders that discuss race are to inform support for Affirmative Action, we maintain that congregants must view social inequality through a structural frame. This contention is based upon a number of studies that find that the more informed Whites are about the structural basis of racial inequality, the more likely they are to support affirmative action (Tuch & Hughes, 1996; Kluegel, 1985; Kluegel & Smith, 1983). Such individuals are likely to interpret the race messages to which they are exposed in a manner that calls for government intervention to address racial inequality.

We further argue that congregation-based racial discussions must be framed as political issues if they are to increase support for corrective policies to racial inequality. The extent to which race and politics are discussed together likely increases a recognition of barriers that disproportionately limit the access of Blacks to quality education and employment, generating the need for public policy solutions. Recently, Brown, Kaiser, and Davis (2010) found some support for this argument. They find that Whites who are exposed to political messages in their houses of worship are more likely than others to believe that they share a common interest with racial/ethnic minorities. This may suggest that exposure to politicized discussions within houses of worship engenders support for policies aimed at benefiting the collective interests of diverse racial/ethnic groups. While plausible, the degree to which exposure to politicized and racial messages in houses of worship is actually associated with support for race-based policies remains untested. It is also unclear if or to what extent the inequality frames of congregants inform their interpretation of racial messages heard within houses of worship. The best studies on this matter are qualitative in nature and based upon the work of faith-based community organizing firms and the Promise Keepers, a conservative Christian organization involved in racial reconciliation efforts during 1990s.

Religion, Structural Frames, and Race

Religious congregations that affiliate with faith-based community organizing firms exemplify the role that social justice oriented religious leaders and laity can play in informing racial attitudes. Faith-based community organizing firms are non-profit organizations that assist local congregations in forging cooperative ties across racial/ethnic and economic lines in an effort to politically act upon their common interests. For these congregations, engaging in relationship building breaks down racial/ethnic barriers and allows for recognition of similar cultural tastes and common interests (Warren, 2001; Wood, 2002). These groups often rely upon religious rituals such as prayer and the study of Biblical texts to emphasize racial/ethnic commonality in Christ as well as the need to engage in cooperative political action to improve the quality of life of marginalized groups and society at large (Warren, 2001; Wood, 2002). Ministers and lay leaders typically say prayers with the intent to call people to social action and refer to the building of a diverse and inclusive community (Warren, 2001; Wood, 2002). Quoted scriptures during worship settings and political meetings often refer to the value in political activism with marginalized groups and the need for socially just policies and legislators.

In religious settings in which clergy and congregants believe that social inequality is linked to structural sins, such as governments that are unresponsive to the needs of the poor, discussions about the need to engage in cooperative activism around shared issues reinforces support for such ideals (Warren, 2001). The congregations affiliated with Metropolitan Organizing Strategy Enabling Strength (MOSES) of Detroit, MI, a faith-based organizing firm, present one such example. Through MOSES, religious congregations joined with other social justice organizations around the state in 2006 to challenge, albeit unsuccessfully, the Anti-Affirmative Action efforts of Ward Connerly and the American Civil Rights Institute to place an initiative on the state ballot to end Affirmative Action in Michigan (MOSES, 2012). It is sensible that White congregants that adopt a more structural frame and attend religious congregations that view political activism as a necessary means to bring about social change are more open to race-based policies.

Religion, Non-Structural Frames, and Race

Conversely, if racial equality is framed in non-structural terms, it is unlikely that exposure to racial messages within houses of worship moves many Whites beyond a dominant ideological perspective on racism and racial inequality. For example, during the mid to late 1990s, the Promise Keepers was the most active Christian group committed to racial reconciliation (Bartkowski, 2004). This group saw racial reconciliation as a key element to their primary mission in re-committing Christian men to Christ, their families, and their communities (Allen II, 2000; Alumkal, 2004; Bartkowski, 2004). They viewed racial separation and conflict as weakening the Christian community because it makes it difficult for Christians to bring others to Christ and to live in peace with all of their Christian brothers and sisters. In problematizing racism, Promise Keepers often view and talk about it as an individual and inter-personal problem (Allen II, 2000; Alumkal, 2004; Bartkowski, 2004). That is, racism is believed to result from an unwillingness to acknowledge one’s racial prejudice and a misinterpretation of the Christian message’s application to all people. Consequently, the solution to racism tends to involve White Christians seeking atonement for their racial prejudices and working to form friendships with racial/ethnic minorities (Alumkal, 2004; Bartkowski, 2004).

At Promise Keeper conferences during the mid to late 1990s, public events of atonement often involved Whites and Blacks embracing as Whites sought forgiveness from Blacks (Bartkowski, 2004). Attendees were also called to take the message of racial reconciliation back to their home churches and neighborhoods in an effort to form friendships across racial/ethnic lines. The individualistic frame for racial problems was consistent with the opposition of many Promise Keepers to public policies aimed at ameliorating racial disparities (Alumkal, 2004; Bartkowski, 2004). While the Promise Keepers are not a coalition of local religious congregations, their approach to racial inequality provides a compelling case for how a non-structural and apolitical frame to racial inequality among religious Whites may shape their racial attitudes. It is plausible that worship-going Whites are unlikely to support race-based policies if they adopt a non-structural frame of social inequality and attend religious bodies that do the same. This leads to the following hypotheses;

1. The more structurally aware Whites are, the more supportive they are of race-based polices.

2. Whites that attend religious congregations in which they are exposed to political discourse become more supportive than others of race-based policies as they are exposed to racial messages within their houses of worship.

3. Whites with a greater recognition of structural inequality become more supportive than others of race-based policies as they are exposed to racial messages within their houses of worship.

Sample

This study relies upon the 2004 National Politics Studies (NPS), which is based upon a national sample of individuals, aged 18 years or older. Interviews occurred throughout the United States in urban and rural centers of the country where significant numbers of Black Americans reside. In 2004, a total of 3,309 Americans were interviewed, 919 of whom were Non-Hispanic Whites; the study group for the present study. Each of the NPS racial/ethnic groups were sampled from separate weighted sampling frames. The Program for Research on Black Americans in the University of Michigan’s Institute for Social Research went into the field in September of 2004 and concluded in February of 2005. All of the interviews were conducted over the telephone. The interviews were conducted in either English or Spanish, depending on the preference of the respondent. The overall response rate in both years is 31%.² A more detailed description of the study is provided by Jackson et. al (2004).

Measures

Dependent Variables: Race-based Policy Attitudes

Race-based policy attitudes were assessed by an Affirmative Action and a Job Preference measure. The Affirmative Action measure assesses the extent to which respondents believe that “affirmative action is a good thing.” Responses are scored as one if individuals believe that Affirmative Action is a good thing and a zero if they do not hold this opinion. The Job Preference measure assesses the extent to which individuals have a favorable opinion of some groups in society being given preference in hiring and promotion because of past discrimination. Responses are scored as one if individuals favor job preferences and a zero if they do not³.

Independent Variables: Politicized Worship Discourse and Structural Orientation

The lay and religious leader political discourse variables are meant to assess political discourse within worship settings. Clergy politics assesses the degree to which, in the past year, respondents were encouraged by their clergy, or someone in an official position to take some action on a political issue such as sign a petition, write a letter, attend a protest, march or demonstration, or contact a public official. The lay politics variable assesses the extent to which, in the past year, respondents participated in political discussions with others at their place of worship. Finally, the racial-discourse congregation variable is a two-item index that assesses exposure to discussions about race relations within houses of worship. The first item assess the extent to which respondents heard a sermon and/ or lecture in the past year at their place of worship that focused upon improving relations between members of different racial or ethnic groups. The second item assesses the extent to which respondents attended a religious congregation that, in the past twelve months, helped sponsor a program or meeting to encourage greater understanding among different racial or ethnic groups. This index ranges from 0-The respondent is not exposed to sermons/lectures/ discussion on race relations within a house of worship and does not attend a house of worship that helps sponsor race relations programs- to 2-The The respondent is exposed to sermons/lectures/ discussion on race relations within a house of worship respondent and attends a house of worship sponsors programs on race relations. The alpha scores for this index is .67.^4,5

The structural orientation measures consist of two separate variables. The first assesses the degree to which individuals disagree with the notion that America is a land of opportunity in which you only need to work hard to succeed. The second variable is a two-item system blame index. This index is constitutive of disagreement with the following assertions; 1. Because Irish, Italians, Jewish and many other minorities overcame prejudice and worked their way up, Blacks should do the same without any special favors, and 2. If racial and ethnic minorities don’t do well in life they have no one to blame but themselves. This index ranges from 0-The respondent does not disagree with the either of those statements to 2-The respondent disagrees with both of those statements. The alpha score for the index is .67.

Interaction Terms

The interaction terms are meant to assess the moderating effect of politicized worship discourse and structural orientation on the relationship between attending a racial-discourse congregation and support for race-based policies. The politicized worship discourse interaction variables are the products of the clergy or lay political discourse variables multiplied by the racial-discourse congregation variable. The structural orientation interaction variable is the product of the structural orientation variables multiplied by the racial-discourse congregation variable.

Control Variables

In an attempt to replicate the analytical approach of past studies on religion and racial attitudes, this study controls for the frequency of worship attendance, religious faith⁶, the extent to which individuals attend a majority White religious congregation, education, family income, gender, employment status, and region. In an effort to minimize selection bias, we also account for political partisanship, the degree to which individuals believe that they live in a mostly White neighborhood, the percentage of Whites that live within the respondent’s census tract, and the degree to which respondents have mostly White friends.⁷ Lastly, we account for the perceived racial background of the interviewer.

Statistical Analyses

Because both Job Preferences and Affirmative Action are dichotomous dependent variables, we employ logit regression analyses to test our hypotheses.

Results

Worship Discourse, Structural Orientation, and Race-based Policy Attitudes

The data presented in Table 1 may suggest that Whites have different opinions of Affirmative Action based upon how this policy is presented. Whites were seemingly more supportive of the broad idea of Affirmative Action than they were of Job Preferences. Whereas roughly a fifth of Whites supported job preferences for racial/ethnic minorities, roughly half held a favorable opinion of Affirmative Action. It is important to note that the reported attitudes are quite similar to those reported in other national studies, such the General Social Survey (Schuman et. al, 1998), the Gallup Poll (Jones, 2005), American National Election Study (Schuman et. al, 1998), and the Pew Charitable Trust (2007).

Table 1.

Percentage Distribution and Mean Scores of Religion, Structural Racism, and Race-based Policy Attitudes

Race-based Policies
Support Preferences for Minorities in the Workplace	23.39
Support Affirmative Action	53.32
Worship Discourse
Lay Politics	26.12
Clergy Politics	22.74
Racial Discourse Congregation
0. Does not Hear Race Messages in  House of Worship AND Does not  Attend a House of Worship with a  Race-Relation Program	50.82
1. Hears Race Messages in House of  Worship OR Attends a House of  Worship with a Race-Relation  Program	24.16
2. Hears Race Messages in House of  Worship AND Attends a House of  Worship with a Race-Relation  Program	25.03
Structural Orientation
U.S. is Not a Land of Opportunity	18.61
System Blame
0. Blacks do not deserve special favors & Minorities are Solely to Blame if they do not Succeed	54.84
1. Blacks do not deserve special favors OR Minorities are Solely to Blame if they do not Succeed	23.29
2. Blacks deserve special favors & Minorities are not solely to blame if they do not succeed	21.87
N=	919

Consistent with past studies, these findings may reflect the greater problems that Whites have with the idea of racial preferences than perhaps government sponsored job skill development and college readiness programs for racial/ethnic minorities (Kravitz, 1995; Summers, 1995). At this point, however, it is difficult to say. It is fairly clear, however, that Whites hold a greater disdain for job preferences than they may for other aspects of Affirmative Action.

Table 1 also indicates that between a quarter and a fifth of all Whites discussed politics in their places of worship and was encouraged by their clergy to take some form of political action. And, at least half of whites attend a place of worship that either hosts a program or delivers a sermon on the importance of improving race relations. Lastly, the structural orientation measures suggest that roughly a fifth of Whites reject the idea that the U.S. is a land of opportunity in which you only need to work hard to succeed. The system blame measures indicates that roughly a fifth of Whites believe both that the discriminatory experience of Blacks is distinctive from that of White ethnics and they therefore are worthy of special favors and that if minorities don’t do well in life, they are not solely to blame. We now move to assessing the relationship between religion, structural orientation, and race-based policy attitudes.

Main Effects: Worship Discourse, Structural Orientation, and Race-based Policy Attitudes

On the whole, worship discourse is unrelated to race-based policy attitudes. Individuals that attended racial-discourse congregations, were encouraged by their clergy to take political action, and talked about politics with their place of worship were no more or less likely than others to support Job Preferences or Affirmative Action. We have evidence, however, that the more system blame oriented Whites are, the more likely they are to support both Job Preferences and Affirmative Action.

These analyses also indicate that Non-Republicans are more supportive of both Job Preferences and Affirmative Action than are Republicans. Evangelical Protestants and Catholics are less supportive of Job Preferences than are secular individuals. Finally, female and younger Whites are more supportive of Affirmative Action than are other Whites. We now move on to discussing the moderating effect of politicized worship discourse and structural orientation on the relationship between attending a racial-discourse congregation and race-based policy attitudes.

Interaction Effects: Worship Discourse, Structural Orientation, and Race-based Policy Attitudes

The analyses presented in Tables 3 and 4 largely support our hypotheses that Whites who are exposed to political discourse within houses of worship and are more heavily structurally oriented became more likely than others to support race based policies as they became exposed to racial discussions within places of worship.

Table 3.

Moderating Impact of Worship Discourse and Structural Orientation on the Relationship between Worship-based Race Discourse and Support Job Preferences: Logit Regression Analyses: Odds Ratios¹⁹

Interaction Effects
Racial Discourse Congregation * Clergy Politics	3.410**	---	---	---
	(1.089)
Racial Discourse Congregation * Lay Politics	---	1.970*	---	---
		(0.555)
Racial Discourse Congregation * System Blame	---	---	1.404*	---
			(0.193)	2.115**
Racial Discourse Congregation * U.S. is not Land of Opportunity	---	---	---	(0.520)
Worship Discourse
Racial Discourse Congregation	0.586**	0.634**	0.572**	0.673**
	(0.096)	(0.111)	(0.117)	(0.103)
Clergy Politics	0.228**	1.002	0.959	1.026
	(0.111)	(0.246)	(0.239)	(0.254)
Lay Politics	1.298	0.611	1.299	1.226
	(0.320)	(0.247)	(0.316)	(0.300)
Structural Orientation
System Blame	2.340**	2.373**	1.850**	2.352**
	(0.289)	(0.292)	(0.273)	(0.289)
U.S. is not Land of Opportunity	1.379	1.323	1.347	.790
	(0.303)	(0.289)	(0.293)	(0.224)
Log Likelihood	−385.750	−390.940	−390.752	−389.183
N	919	919	919	919

^*<.05;

^**<.01 (Standard Errors are in parentheses)

Table 4.

Moderating Impact of Worship Discourse and Structural Orientation on the Relationship between Worship-based Race Discourse and Support for Affirmative Action: Logit Regression Analyses: Odds Ratios

Interaction Effects
Racial Discourse Congregation * Clergy Politics	1.174	---	---	---
	(0.268)
Racial Discourse Congregation * Lay Politics	---	1.546*	---	---
		(0.340)	1.119
Racial Discourse Congregation * System Blame	---	---	(0.128)	---
Racial Discourse Congregation * U.S. is not Land of Opportunity	---	---	---	2.106**
				(0.556)
Worship Discourse
Racial Discourse Congregation	0.966	0.880	0.944	0.916
	(0.116)	(0.109)	(0.116)	(0.100)
Clergy Politics	0.700	0.833	0.824	0.827
	(0.224)	(0.163)	(0.161)	(0.163)
Lay Politics	1.314	0.813	1.321	1.273
	(0.253)	(0.252)	(0.255)	(0.248)
Structural Orientation
System Blame	1.758**	1.782**	1.616**	1.768**
	(0.188)	(0.192)	(0.221)	(0.190)
U.S. is not Land of Opportunity	1.381	1.362	1.384	0.868
	(0.292)	(0.289)	(0.293)	(0.227)
Log Likelihood	−561.652	−559.924	−561.415	−557.547
N	919	919	919	919

^*<.05;

^**<.01 (Standard Errors are in parentheses

Job Preferences

As expected, the analyses presented in Table 3 largely suggests that the relationship between exposure to racial discourse in worship settings and support for job preferences is statistically different among those that are exposed to political discourse within worship settings and those that are not and between those maintaining a structural orientation and those that do not. In the racial-discourse congregation / clergy politics interaction term of Table 3, the main effect coefficient of Racial-Discourse Congregation (e.g. congregation that preaches sermons / hosts programs on race relations) represents the odds of attending a Racial-Discourse Congregation increasing support for Job Preferences when the Clergy Politics variable is zero (e.g. does not attend a religious congregation where clergy encourages political activism). This main effect is significantly different from zero and has an odds ratio of less than one, indicating that attending a racial-discourse congregation where individuals are not politically encouraged by clergy contributes to Whites being less likely than others to support job preferences. On the other hand, attending a racial-discourse congregation where individuals are encouraged to take political action contributes to Whites being more likely to support Job Preferences. To that end, these analyses suggests that Whites that are politically encouraged by clergy become more likely than others to support Job Preferences as they become exposed to racial discussions in a place of worship. Similarly, the racial-discourse congregation / lay politics interaction term suggests that Whites who discuss politics in houses of worship become more likely than others to support Job Preferences as they become exposed to racial discussions in their place of worship. The racial-discourse congregation / U.S. is not a land of opportunity interaction term suggests that Whites that do not believe that the U.S. is a land of opportunity become more likely than others to support Job Preferences as they become exposed to racial discussions in their place of worship.

Finally, the congregation / system blame interaction term suggests that Whites that hold a system blame orientation become more likely than others to support Job Preferences as they become exposed to racial discussions in their place of worship. Figure 4 illustrates that individuals with the two lowest levels of system blame orientation maintain a downward trend in their probability of supporting job preferences as they are exposed to more race messages in houses of worship. However, this trend begins to reverse course among individuals with the highest level of system blame orientation. The predicted probability estimates listed in Figure 1 more clearly illustrate these interaction effects.

Figure 1.

Probability Estimates of the Relationship between Exposure to Racial Discourse in Houses of Worship and Support for Job Preferences by Clergy Political Encouragement, Lay Political Discussion, System Blame Orientation, and Perceptions of Opportu

The probability estimates of the association between exposure to racial discourse within houses of worship by exposure to political discussions in houses of worship (e.g. clergy political encouragement / lay political discussion) warrant further discussion. Interestingly, these estimates suggest that Whites that attend religious congregations that expose them to racial discussions (e.g. race sermon and programs) and political messages (e.g. clergy political encouragement / lay political discussion) are as likely to support job preferences as Whites that do not attend such religious congregations (e.g. no clergy political encouragement / no lay political discussion and no race sermon and programs). Both groups are seemingly most likely to support such a policy.

The proportion of both groups that support job preferences look very similar to the proportion of Whites as a whole-23%-(see Table 1) that support this policy. However, the difference between Whites that are exposed political discussions (e.g. clergy political encouragement / lay political discussion) within houses of worship and those that are not is how their support for job preferences changes as they are exposed to racial messages (e.g. race sermon and programs) within places of worship. As Figure 1 indicates, among Whites that do not attend political congregations (e.g. no clergy political encouragement / no lay political discussion), exposure to racial messages in places of worship is associated with support for job preferences falling below mean support for this issue. Conversely, among Whites that attend political congregations (e.g. clergy political encouragement / lay politics) exposure to racial messages in places of worship is associated with support approaching mean support for this issue. In sum, these analyses suggest that attending political congregations that avoid racial discussions are associated with Whites being more conservative on this policy than are Whites on average, while attending political congregations that discuss racial issues are associated with Whites more closely mirroring general support for this policy.

Affirmative Action

The Affirmative Action analyses presented in Table 4 largely mirror the Job Preference Analyses of Table 3. The racial-discourse congregation/lay politics interaction term suggests that Whites that talk about politics in houses of worship become more likely than did others to support Affirmative Action as they became exposed to racial discussions in places of worship. The racial-discourse congregation / U.S. is not a land of opportunity interaction term suggests that Whites that do not believe that the U.S. is a land of opportunity become more likely than others to support Affirmative Action as they become exposed to racial discussions in places of worship. The remaining interactions are non-significant.⁸ The predicted probability estimates listed in Figure 2 more clearly illustrate these interaction effectsⁱ.

Figure 2.

Probability Estimates of the Relationship between Exposure to Race Discourse in Houses of Worship and Support for Affirmative Action by Lay Political Discussion and Belief that the U.S. is the Land of Opportunity

Discussion

The present study offers a new way of thinking about religion and racial attitudes. While previous studies may lead many to assume that, for Whites, religiosity is often associated with racial prejudice and/or intolerance, we find, as Gordon Allport (1979) argued roughly sixty years ago, that religion can make and unmake prejudice. This study suggests that unmaking prejudice involves houses of worship combining race talk with political strategy. That is, we find Whites who attend political congregations to be more supportive than others of race-based policies as they are exposed to racial messages within worship settings. This may suggest that when politically conscious clergy and lay members talk about race, they frame racial inequality as resulting from structural problems that require public policy solutions. Our findings also suggest that unmaking prejudice involves congregational leaders discussing the significance of improved race relations to congregants with structural frames. More specifically, Whites with a heightened awareness of structural inequality are more supportive than others of race-based policies as they are exposed to racial discourse within worship spaces. This may indicate that congregants with structural frames likely interpret racial messages in a manner that lends support to race based policies. These findings are consistent with qualitative studies on religious leaders and laity affiliated with faith-based community organizing firms, such as MOSES, that frame racial inequality as a moral indignation that requires religious bodies to act as the hands of a just God (Warren, 2001; Wood, 2002).

Conversely, our findings indicate that religion offers a rather weak challenge to racial inequality when religious bodies and individuals frame racial issues in apolitical and non-structural terms. Exposure to apolitical racial conversations moves relatively few Whites to challenge a dominant ideological worldview that eschews public policy solutions to racial inequality. Political houses of worship that avoid racial discussions may implicitly reinforce beliefs that racial inequality is unrelated to structural barriers to opportunity and, therefore, policies aimed at correcting such inequality are unwarranted. Along these lines, relatively few Whites with non-structural frames are moved to support race-based policies upon hearing about racial issues within their places of worship. Consistent with the Promise Keeper perspective on race, such individuals likely interpret racial messages with a focus on racial reconciliation as opposed to racial justice (Alumkal, 2004; Bartkowski, 2004). In the absence of discussions aimed at influencing public policy, racial discussions likely remain at the level of atonement for racial prejudice, rarely broaching the subject of public policy solutions to racial inequality.

In sum, the present study challenges the notion that, for White Americans, religiosity is often associated with racial prejudice. Rather, we argue, as others have, that religion exists on a continuum that spans from reinforcing to challenging the status quo of social inequality (Lincoln & Mamiya, 1990; Roozen, McKinney & Carroll, 1984). The extent to which Whites discuss race along with the potential need for public policy solutions to address racial inequality within worship spaces, worship attendance contributes to Whites supporting public policies aimed at reducing racial inequality. On the other hand, apolitical and non-structural racial discussions within worship settings do seemingly little to move many Whites to challenge dominant idealistic perceptions of race that eschews public policy interventions as solutions to racial inequality.

Conclusion

As the only survey-based study of which we are aware that assesses the association between worship discourse and race-based policy attitudes, the current study provides important insight on the role that racial discourse within worship settings may play in informing White race-based policy attitudes. Our findings lead us to believe that progressive religious bodies can heighten support for racial justice among White Americans. This study may speak to the success that faith-based community organizing firms have had in mobilizing some suburban White congregations to join coalitions aimed at pushing for policies that expand social-economic opportunities for Blacks and Hispanics (Warren, 2001). The capacity of these religious congregations to inform White racial attitudes has important public policy implications given the numerical supremacy of Whites and their greater propensity, relative to others, to involve themselves in the political process (Verba, Schlozman, & Brady, 1995).

At the same time, however, no more than twenty percent of the Whites in our study attend religious congregations that contribute to heightened support for race-based policies. The reality is that many social-justice oriented White clergy and lay leaders face challenges in critiquing social inequality in a manner that does not alienate other congregants. The mobilization experiences of faith-based community organizing firms (FBCOs) well exemplify this point. FBCOs represent no more than five percent of religious congregations in the metropolitan areas in which they are located and are under-represented among White congregations (Wood & Warren, 2002). The difficulty of FBCOs in attracting and retaining White suburban congregations is partially rooted in a membership body that finds difficulty recognizing how their largely suburban and middle class interests are connected to the concerns of social-economically disadvantaged racial/ethnic minorities (Warren, 2001). The fact that many Whites view the opportunities afforded to them as being available to all groups makes it difficult for clergy and/or lay leaders to convince congregants that racial/ethnic minorities, Blacks specifically, face unique hardships that require policy solutions (Kluegel, 1981; 1983; 1990).

Many White congregational leaders who have attempted to align their congregations with movements pushing for policies aimed at increasing job and/or educational opportunities for racial/ethnic minorities have faced resistance (Warren, 2001). In some cases, this has involved religious leaders being threatened with their removal (Warren, 2001). As a consequence, some religious leaders likely refrain from encouraging their laity to support and involve themselves in racial justice efforts. Some scholars point to the weak support for racial justice among White laity as contributing to the reduced infrastructure, financial resources, and the subsequent impact of Mainline and Catholic programs, such as Project Equality, aimed at promoting racial/ethnic diversity in the workplace over the past thirty years (Verter, 2002). Our study suggests that congregants that are exposed to politicized racial messages and interpret racial inequality within social justice frames tend to support race-based policies. This does not negate the fact that congregational leaders who call Whites to question an ideology that rationalizes their dominant status may confront the dilemma of doing justice at the expense of losing members and relevance.

Table 2.

Impact of Worship-based Discourse on Support for Race-based Policies: Logit Regression Analyses: Odds Ratios

	Support Job Preferences	Support Affirmative Action
Worship Discourse
Racial Discourse Congregation	0.816	1.007
	(0.110)	(0.105)
Clergy Politics	1.007	0.836
	(0.244)	(0.163)
Lay Politics	1.303	1.318
	(0.311)	(0.254)
Structural Orientation
System Blame	2.306**	1.759**
	(0.279)	(0.188)
U.S. is not Land of Opportunity	1.355	1.383
	(0.294)	(0.293)
Control Variables
Worship Attendance 9	1.066	0.993
	(0.121)	(0.088)
White Congregation10	1.264	1.168
	(0.278)	(0.198)
Evangelical Protestant11	0.368**	0.702
	(0.134)	(0.203)
Mainline Protestant	0.587	1.005
	(0.202)	(0.295)
Catholic	0.469*	1.029
	(0.154)	(0.282)
Non-Christian	0.755	0.686
	(0.225)	(0.182)
College Graduate12	1.493*	0.848
	(0.295)	(0.136)
Employed13	0.977	1.030
	(0.203)	(0.174)
Family Income	1.000	1.000
	(0.000)	(0.000)
Female	1.391	1.329
	(0.272)	(0.207)
South 14	0.648*	1.034
	(0.137)	(0.173)
Age	0.998	0.987*
	(0.008)	(0.006)
Democrat Party15	3.087**	2.768**
	(0.851)	(0.548)
Independent Party	2.270**	2.025**
	(0.644)	(0.387)
Other Party	3.191**	1.938*
	(1.311)	(0.635)
Perceived Neighborhood Racial Homogeneity16	0.780	1.067
	(0.173)	(0.192)
Percent of Whites in Census Tract	0.999	1.007*
	(0.004)	(0.004)
White Friends17	0.948	0.829
	(0.182)	(0.129)
White Interviewer18	1.114	0.994
	(0.205)	(0.147)
Log Likelihood	−393.933	−561.899
N	919	919

^*<.05;

^**<.01 (Standard Errors are in parentheses)