Politics at the Pulpit: Elite Religious Cues and Immigration Attitudes

Author(s): Benjamin Knoll and Matthew Baker, Centre College

Abstract

Previous scholarship has demonstrated a link between religiosity and immigration attitudes, often inferring the effect of cues from religious leaders as the motivating source. This study directly examines the “elite cues” linking mechanism with an experiment embedded in a nationally representative public opinion survey. We improve on previous research designs by introducing a pretest that measures immigration policy attitudes among respondents which can then be directly compared to posttest measures after the introduction of the elite cue stimulus. Multivariate analysis of the survey results reveal no support for the elite cues explanation. We discuss the implications of these findings for elite religious cues as an influential factor on immigration policy attitudes in the United States as well as assess the appropriateness of survey experiments to test the elite cues mechanism in driving immigration attitudes.

Introduction

Most scholarly research on immigration has focused primarily on social and economic factors. In terms of social determinants of immigration attitudes, scholars have identified the important effect of cultural anxiety (Citrin 1990; Citrin and Wright 2009; Higham 1955), racial/ethnic attitudes (Brader, Valentino, and Suhay 2008; Burns and Gimpel 2000; Dustmann and Preston 2007), and social context (Hood and Morris 1998; Hopkins 2010; Rocha and Espino 2009). In terms of economic influences on immigration attitudes, research has focused on sociotropic vs. individual economic well-being (Citrin et al. 1997; Espenshade and Calhoun 1993; Wilson 2001), and how opposition to immigration tends to follow predictable cycles in response to macroeconomic boom and bust periods (Daniels 2004).

More specific to our topic, recent research has investigated the effect of religious variables as key determinants of immigration attitudes (Brenneman 2015; Brown 2010; Fetzer 1998; Fitzgerald 2012; Knoll 2009; McDaniel, Nooruddin, and Shortle 2011; Nteta and Wallsten 2012). With the exception of Brown (2010), this research has found a consistently positive relationship between religiosity and/or religious attendance and liberal immigration attitudes. Despite the relative consistently in these findings, the theoretical explanation for this connection still subject to debate. One proposed explanation has been the “elite cues” mechanism: higher levels of religiosity are associated with political attitudes more favorable toward immigration/immigrants because religious leaders broadly tend to oppose more punitive and restrictive immigration policies. Those who have higher levels of religiosity are more likely to be exposed to and internalize these messages from their clergy which are then reflected in their policy attitudes. 

A growing body of literature supports this argument that churches wield significant power to influence their congregants’ attitudes toward “moral” issues. Wald, et al. (1988), for instance, found that a church’s mean level of “moral conservatism” is a strong predictor of the levels of the corresponding individual levels of moral conservatism among church members. This is to say, a church’s position on questions of morality seems to directly impact its congregants. Building off of this earlier study, Bjarnason and Welch (2004) find that church attendance among Catholics is negatively correlated with support for capital punishment. Catholics, the study explains, are distinct among Christians in their vocal opposition to the death penalty. This phenomenon is absent among other Christian denominations, whose church officials often do not express the same moral opposition to the death penalty. The more Catholics attend religious services, where they are likely to hear these messages, the more likely they are to oppose capital punishment. Smith (2008) furthers this line of research by showing that political messaging from Catholic priests has both direct and indirect effects on the opinions of those in their parishes.

Djupe and Gwiasda find a similar effect among Evangelicals’ attitudes toward the environment (2010). The traditional heuristic dictates that Evangelical ministers either ignore environmental issues or address them negatively from the pulpit. This heuristic is broken, however, when Evangelical ministers deliver a pro-environmental message framed as a “moral issue.” This forces church-goers to assess the minister’s position based the soundness of argument, and, as Djupe and Gwiasada (2010) conclude, in-group members will often accept the conclusion itself. This again underscores the church’s position as an influence on moral issues. 

Insomuch as religious elites frame immigration as a moral issue, it may also be subject to these same forces.

Other scholars have attempted to determine why elites engage in this behavior as well as which types of clergy are more likely to involve themselves and in what ways. These findings show that pastors see themselves as “spiritual representatives” of their churches and are particularly likely to engage in political cueing when they are either geographically or ideologically isolated from their broader communities outside of their church (Djupe and Gilbert 2002). 

Additionally, both Guth, et al. (1997) and Putnam and Campbell (2012) find that politically liberal pastors and congregations engage more frequently in the process of political cueing than their conservative counterparts. Smith (2005) also finds that liberal Catholic clergy not only engage in cueing more frequently but also that they wield more actual influence than do their conservative counterparts, though, as Putnam and Campbell (2012) note, their influence is exhibited in different ways. Churches with individuals who “connect faith and politics” are more likely to be identify as Republicans, despite the fact that traditions with “more political activity at church” have fewer Republicans (Putnam and Campbell 2012, 440). This highlights the different ways in which conservatives and liberals give and receive cues in church communities.

Though this theory of elite religious cueing is widely proposed, there is little direct evidence validating it as a driver of immigration attitudes. Instead, previous research has inferred this relationship based on the correlation between elite cueing and attitudes among religiously active congregants. In other words, previous studies have answered the questions of how and to what extent religious behavior influences immigration attitudes but have not been able to address clearly the fundamental questions of why and by what means. Knoll (2009) writes of his own piece, but also summarizing the state of the literature: “Even though we have herein demonstrated with confidence that religion exerts an independent effect on immigration preferences, the argument for elite cues presented in this article is merely implied by these results” (328). 

Nteta and Wallsten (2012) attempt to account for this by asking individuals directly whether or not they received a message on immigration from their pastor. They offer additional support for elite cueing theory as they show that American religious leaders are communicating support for liberal immigration policy and that the cues are impactful in changing the immigration attitudes of their congregants. While this research is a more direct test of the elite cues mechanism, it relies on survey data that does not directly indicate whether the elite cues given are in support or in opposition to liberal immigration policy, limiting the conclusion’s applicability. 

Our current objective is to provide a more direct assessment of the elite cues mechanism that has either been inferred by previous research or limited in its methodology. We take advantage of a survey experiment that directly measures attitudes toward a particular immigration policy (President Obama’s 2015 executive order) and how these attitudes are affected by an elite religious cue introduced in the survey. This will provide the most direct test to date of the elite cues mechanism linking religiosity with immigration policy preferences.

Hypothesis

Based on the previous research described in the literature review and the elite cues mechanism discussed above, we propose the following hypothesis:

Those exposed to an elite religious support cue for Obama’s executive immigration actions will become more approving of the executive order than those not exposed to the cue.

Additionally, there are reasons to expect that this effect will be stronger for some individuals than for others. For example, Knoll (2009) argued that the elite cue mechanism worked for individuals with higher levels of religious service attendance because they are more likely to receive these elite religious cues on immigration more frequently. Theoretically, more frequent religious service attendance would increase the likelihood that an individual would be exposed to an elite religious cue. More frequent attendance is also often an indicator of more devout observance of one’s religious traditions and teachings, meaning that they would be more likely to take seriously and internalize the cues given by their religious leaders. 

The effect of elite religious cues on approval for Obama’s executive immigration action will be stronger for individuals who attend religious services more frequently than for those who attend less frequently.

There is also good reason to suspect that partisanship and political ideology may mediate the effect of elite religious cues on immigration policy attitudes. Social psychology research has consistently shown that political conservatives are more sensitive to hierarchy and deferential to authority (Graham, Haidt, and Nosek 2009; Haidt 2012). We also know that conservatives tend to be more religious and invested in their congregations (Layman 1997). We may thus expect conservatives to be more likely to defer to cues given by religious authorities with whom they identify. When religious authorities indicate support for a particular position on immigration policy, we may then expect conservatives to internalize and manifest that position in their stated policy opinions. On the other hand, we expect an already-high degree of support from pro-immigrant policies from Democrats and liberals and thus do not expect a further endorsement from a religious authority to substantively affect their stated levels of support.

The effect of elite religious cues on approval for Obama’s executive immigration action will be stronger for political conservatives and Republicans than for political liberals and Democrats. 

While our focus in this research is on elite religious cues, there is also a good deal of research showing the effectiveness of elite political cues on attitudes and behaviors (Lodge, Steenbergen, and Brau 1995; Miller and Krosnick 2000; Zaller 1990). It is possible that the combination of religious and elite cues may interact in such a way as to strengthen and reinforce one another. Thus, those who approve of President Obama may be more likely to internalize a religious cue endorsing the president’s executive order.  

The effect of elite religious cues on approval for Obama’s executive immigration action will be stronger for those who approve of Obama’s job performance than those who disapprove. 

Data and Method

To directly test the effect of elite religious cues on immigration policy attitudes, we used survey data collected from the 2015 Colonel’s Canvass Poll, a nationally representative sample of 715 adult Americans, although due to missing data not all are included in each analysis. This survey was conducted in March of 2015 and sampled both landline (62%) and cellphone (38%) respondents. Post-stratification weighting is used to account for underrepresentation among racial minorities and younger respondents.

Methodology

As is standard for experimental research designs, we assess the effect of the treatment stimulus with a series of difference-of-means t-tests between the control and treatment groups, including among the various demographic and political subgroups as described in the previous section. Because of the usual levels of correlation between the various independent variables, we also use a standard multivariate regression analysis to determine the independent effect of each variable on changes in favorability toward President Obama’s executive immigration action. 

Dependent variable

Respondents were first asked early in the survey to indicate their approval of “President Obama’s executive order that expands the number of undocumented immigrants who are allowed to stay and work in the country.” They indicated responses on a 0 to 10 scale with higher values corresponding to higher levels of support. Half of these respondents were later randomly selected to receive a second “treatment” question which was disguised as one of five “religious awareness” questions in response to a sampling of headlines. This question informed them that American religious leaders have “recently express[ed] support for President Obama’s executive action on immigration, including the U.S. Conference of Catholic Bishops and leaders of many Protestant denominations and other religious groups.” This prompt was intentionally designed to be both suitably broad, cueing both Catholic and Protestant respondents, and intentionally narrow so as to make respondents believe that the “religious leaders” referenced would likely include be their own. The control group received the other four religious awareness questions but were not given the executive immigration action prompt. Both the control and treatment groups were then asked again to indicate their 0 to 10 level of approval of President Obama’s executive order on immigration. This question was placed toward the end of the survey in order to mask the research design as much as possible and also to minimize the potential effects of social desirability. 

The pretest score was then subtracted from the posttest score to produce a new value for each respondent measuring the change in approval from the first time the question was asked to the second time. A positive value of this new variablewould indicate that an individual answered the posttest question more favorably than the pretest question (that is, they become more favorable toward the executive action at the end of the survey as compared to the beginning) and a negative value would mean the opposite. Theoretically, if the elite cue mechanism works as hypothesized, we should expect to see the favorability change variable higher for the treatment group than for the control group, indicating that the reception of the elite religious endorsement increased levels of approval among those who received it. 

We note that it is possible that that the explicit cuing of “President Obama” in the treatment prompt may prime respondents to associate their perceived favorability of the specific immigration policy described with their preexisting feelings toward President Obama. We chose to do so, however, in order to reflect as closely as possible the way in which respondents would likely encounter information about specific immigration policies in the “real world”—that is, with specific partisan framings. In this way, the question wording attempts to achieve as much external validity as possible which is often difficult in an experimental design.

We also call attention to the fact that our operationalization of the dependent variable is a direct measure of change in approval from the beginning to the end of the survey, not objective favorability overall. This is important because most experimental survey designs lack a pretest measure of attitudes. Respondents are ordinarily split into control and treatment groups with the latter receiving some sort of stimulus, and then measures of the outcome of interest are compared between the two groups to determine the effect of the stimulus. Without a pre-treatment pretest, however, it is impossible to definitively know whether any apparent difference is due to the treatment or existed in the treatment group as a result of the random assignment process. In contrast, our experimental design has both pretest and posttest measures of favorability toward the executive immigration order allowing us to directly track any change in attitudes from Point A to Point B and determine how they change in response to our elite religious cue prompt.

Independent Variables

Frequency of religious service attendance is a six-point ordinal variable ranging from “never” to “more than once a week,” which is collapsed to a binary variable for the bivariate analysis between those who attend at least “once or twice a month” and “a few times a year” or less. Theological traditionalism is measured by agreement that one’s religion or church should “preserve traditional beliefs and practices” while theological progressivism is measured as agreement that it should “adopt modern beliefs and practices” or “adjust beliefs and practices in light of new circumstances.” We also include measures of the % Latino and % foreign-born in a respondent’s zip code as per the 2010 Census and 2013 American Community Survey, respectively, which are collapsed at their means into binary variables for the bivariate analysis. We also examine relationships for partisans and ideologues (leaners are included with the partisan and ideological group). For the multivariate analysis, we also include standard demographic controls for age, race/ethnicity, education, income, as well as dummy variables representing Catholic and Evangelical Protestant affiliation.

Results

Figure 1 and Table 1 display the distribution of change in approval of Obama’s executive immigration action from the pretest to the post-post. Figure 1 shows how approval scores changed among both the control and treatment groups while Table 1 reports the distribution of scores separately for each group. As can be seen in Figure 1, nearly three-fourths (74.7%) of all individuals had a value of zero, which occurs when the pre-and post-test responses are identical, indicating no change. Table 1 also shows that the distribution of attitude change is nearly identical between the control and treatment groups.

Figure 1. Distribution of change in favorability toward Obama’s executive immigration order posttest vs. pretest

Table 1. Distribution of change in favorability toward Obama’s executive immigration order posttest vs. pretest, by control and treatment group

	Control group	Treatment group
Approval decreased	11.7%	11.6%
No change	73.2%	76.1%
Approval increased	15.1%	12.3%

Bivariate analysis

Looking at the results as a whole, a difference of means t-test reveals no statistically significant difference in the scores between the control (M=0.17, SE=0.13) and treatment (M=-0.04, SE=0.11) groups; t(398)=1.24, p = 0.215. At first pass, these results indicate that the elite religious cue had no effect on an individual’s attitude toward Obama’s immigration executive action as there is no statistically significant difference in attitude change between the control and treatment group.

Demonstrating that there is no discernible effect for the treatment group as a whole does not, however, eliminate the possibility that there is an effect for a subset of respondents as we originally hypothesized might be the case. Table 2 displays the results of a series of bivariate difference of means tests among the various subgroups discussed in the hypothesis section. As can be seen, there are no significant differences between the control and treatment groups for any of the hypothesized subgroups when analyzed at the bivariate level.

Table 2. Bivariate analysis of effect of elite religious endorsement on immigration policy approval.

	Control mean (SE, N)	Treatment mean (SE, N)	t-statistic
Overall
Frequent church attendance	0.06 (0.20, 80)	0.02 (0.17, 117)	0.150
Infrequent church attendance	0.26 (0.12, 110)	-0.13 (0.11, 83)	1.701
Democrats	0.27 (0.20, 94)	-0.12 (0.13, 112)	1.703
Independents	-0.02 (0.22, 19)	0.27 (0.24, 23)	-0.876
Republicans	0.00 (0.21, 73)	-0.07 (0.24, 62)	0.205
Liberals	0.25 (0.18, 80)	0.17 (0.11, 87)	0.395
Moderates	0.00 (0.30, 45)	-0.35 (0.25, 50)	1.206
Conservatives	0.10 (0.20, 53)	-0.08 (0.29, 50)	0.521
Obama approve	0.19 (0.16, 92)	-0.06 (0.10, 105)	1.438
Obama disapprove	0.19 (0.16, 92)	0.02 (0.21, 70)	0.076

Multivariate analysis

As a further test of these bivariate results, we performed a multivariate regression analysis estimating change in approval for the immigration executive order using the various hypothesized factors as predictors. We interacted each of these predictors with a dummy variable indicating whether or not the respondent received the elite religious cue as well as the standard 

demographic control variables described earlier. (We use robust standard errors to correct for heteroscedasticity.) The results presented in Table 3 confirm those presented in Table 2 and indicate that, once controlling for each predictor as well as the various demographic variables, the elite religious cue still does not seem to exert a discernible effect on any of the hypothesized subgroups.

Table 3. Multivariate analysis of effect of elite religious endorsement on immigration policy approval.

Variable	B (SE)
Treatment group	0.54 (0.85)
Obama approval	-0.14 (0.41)
Obama approval × treatment	-0.19 (0.59)
Partisanship	0.01 (0.11)
Partisanship × treatment	-0.07 (0.17)
Ideology	-0.06 (0.12)
Ideology × treatment	0.02 (0.15)
Religiosity	0.09 (0.07)
Religiosity × treatment	-0.09 (0.11)
Age	-0.01 (0.00)*
Income	-0.03 (0.00)
Education	0.10 (0.07)
Black	-0.27 (0.22)
Latino	-0.08 (0.21)
Asian	-0.05 (0.15)
Evangelical	0.17 (0.21)
Catholic	-0.02 (0.22)
N	296
R-squared	0.046

Discussion and Conclusion

Our analysis has uncovered no support for our hypotheses related to the effect of elite religious cues and immigration policy attitudes at either the bivariate or multivariate level. Not only were we unable to uncover any evidence that our elite religious cue produced a measurable change in opinions toward Obama’s executive immigration action in the aggregate, we were also unable to discern any effect of the religious cue among most of our hypothesized subgroups. 

Why might this be the case? There are a number of possible explanations. Perhaps elite religious cues do work in the way described previous literature on the topic, work only when given by religious leaders themselves in an actual real-world setting and not by a telephone surveyor in the respondent’s home (as per Nteta and Wallsten 2012, e.g.). This experiment simply informed respondents of the general position of a broad group of religious elites and did not attempt to convey their words exactly. We designed the question in this way so as to be sufficiently broad to fit the vast majority of American churchgoers. In the absence of direct contact by actual religious elites, perhaps the cueing effect is diminished or eliminated entirely in our brief telephone survey context. 

It is also possible that the research design muted the potential effect of elite religious cue because the cue was given just minutes before the posttest question. It is possible, however, that any meaningful cueing effect requires time to “sink in” among those in the population and perhaps also requires multiple repetitions. Offering individuals one cue and only a few minutes to process and internalize it may be inadequate to produce the desired effect. It is possible that our experimental telephone survey research design simply lacks the necessary external validity to appropriately test the elite religious cues mechanism.

Another explanation for our non-findings is that elite religious cues are ineffective in the presence of an overwhelmingly partisan cue like the one in the pre-and post-test questions. Our question explicitly mentioned President Obama by name, triggering a strong partisan framing, which might have drowned out the potential effect of elite cues. We designed the question in this way so as to maximize external validity as much as possible given the reality of political discourse on immigration policy where it is almost always discussed in a partisan environment. The inability to effectively disentangle religious and partisan cues in a “real world” context has meaningful implications about the effectiveness of elite religious cues to change attitudes on immigration policy.

Despite these concerns, it is also very possible that the above results do reflect reality. Perhaps elite religious cues are simply ineffective to change attitudes on immigration and are not the key linking mechanism between religiosity and immigration attitudes as inferred by previous research (Knoll 2009; Nteta and Wallsten 2012). Instead, it is possible that the aspect of religiosity that truly matters in driving immigration policy attitudes is the socialization effect of face-to-face interaction, as described by Wald, Owen, and Hill (1988) and Fitzgerald (2012). Both sources attribute church-goers’ changes in political attitudes in part to these connections between fellow congregants (which often involve people of diverse backgrounds) rather than to interactions between congregants and clergy. This is an important shift in thinking that warrants further investigation.

Further research is warranted to isolate the alternate explanations described above and test them individually. For example, a future test could attempt to better simulate the nature of a direct elite cue by offering exact quotes from real religious leaders. Alternatively, another experiment could present multiple and obvious cues in an experimental survey context, with the possibility that this could allow a more thorough internalization of the message.

References

Bjarnason, Thoroddur, and Michael R. Welch. 2004. “Father Knows Best: Parishes, Priests, and American Catholic Parishioners’ Attitudes Toward Capital Punishment.” Journal for the Scientific Study of Religion 43(1): 103–18.

Brader, Ted, Nicholas A. Valentino, and Elizabeth Suhay. 2008. “What Triggers Public Opposition to Immigration? Anxiety, Group Cues, and Immigration Threat.” American Journal of Political Science 52(4): 959–78.

Brenneman, Robert. 2015. “Faith and the Foreigner: Exploring the Impact of Religion on Immigration Attitudes.” Notre Dame. http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.586.61&rep=rep1&type=pdf (June 15, 2016).

Brown, R. Khari. 2010. “Religion, Economic Concerns, and African American Immigration Attitudes.” Review of Religious Research 52(2): 146–58.

Burns, Peter, and James G. Gimpel. 2000. “Economic Insecurity, Prejudicial Stereotypes, and Public Opinion on Immigration Policy.” Political Science Quarterly 115(2): 201–25.

Citrin, Jack. 1990. “Language Politics and American Identity.” Public Interest 99: 96–109.

Citrin, Jack, Donald P. Green, Christopher Muste, and Cara Wong. 1997. “Public Opinion Toward Immigration Reform: The Role of Economic Motivations.” The Journal of Politics 59(3): 858–81.

Citrin, Jack, and Matthew Wright. 2009. “Defining the Circle of We: American Identity and Immigration Policy.” The Forum 7(3). http://www.bepress.com/forum/vol7/iss3/art6 (January 30, 2010).

Daniels, Roger. 2004. Guarding the Golden Door: American Immigration Policy and Immigrants since 1882. Fourth Printing edition. Hill and Wang.

Djupe, Paul A., and Christopher P. Gilbert. 2002. “The Political Voice of Clergy.” Journal of Politics 64(2): 596–609.

Djupe, Paul A., and Gregory W. Gwiasda. 2010. “Evangelizing the Environment: Decision Process Effects in Political Persuasion.” Journal for the Scientific Study of Religion 49(1): 73–86.

Dustmann, Christian, and Ian P Preston. 2007. “Racial and Economic Factors in Attitudes to Immigration.” The B.E. Journal of Economic Analysis & Policy 7(1). 

Espenshade, Thomas J., and Charles A. Calhoun. 1993. “An Analysis of Public Opinion toward Undocumented Immigration.” Population Research and Policy Review 12(3): 189–224.

Fetzer, Joel S. 1998. “Religious Minorities and Support for Immigrant Rights in the United States, France, and Germany.” Journal for the Scientific Study of Religion 37(1): 41–49.

Fischer, Sara. 2014. “U.S. Bishops ‘Applaud’ Obama on Immigration.” CNNPolitics.com. http://www.cnn.com/2014/11/30/politics/bishops-obama-immigration-pope-francis/ (June 15, 2016).

Fitzgerald, Jennifer. 2012. “Social Engagement and Immigration Attitudes: Panel Survey Evidence from Germany.” International Migration Review 46(4): 941–70.

Graham, Jesse, Jonathan Haidt, and Brian A. Nosek. 2009. “Liberals and Conservatives Rely on Different Sets of Moral Foundations.” Journal of Personality and Social Psychology 96(5): 1029–46.

Guth, James L., John C. Green, Corwin E. Smidt, and Lyman A. Kellstedt. 1997. The Bully Pulpit: The Politics of Protestant Clergy. Lawrence, Kan: University Press of Kansas.

Haidt, Jonathan. 2012. The Righteous Mind: Why Good People Are Divided by Politics and Religion. 1st ed. Pantheon.

Higham, John. 1955. Strangers in the Land : Patterns of American Nativism, 1860-1925. First Edition. Rutgers University Press.

Hood, M. V., and Irwin L. Morris. 1998. “Give Us Your Tired, Your Poor, … But Make Sure They Have a Green Card: The Effects of Documented and Undocumented Migrant Context on Anglo Opinion Toward Immigration.” Political Behavior 20(1): 1–15.

Hopkins, Daniel J. 2010. “Politicized Places: Explaining Where and When Immigrants Provoke Local Opposition.” American Political Science Review 104(1): 40–60.

Jackson, David. 2014. “Obama, Faith Leaders Talk Immigration.” USA Today. http://www.usatoday.com/story/theoval/2014/04/15/obama-faith-leaders-immigration-meeting/7741383/ (June 15, 2016).

Knoll, Benjamin R. 2009. “‘And Who Is My Neighbor?’ Religion and Immigration Policy Attitudes.” Journal for the Scientific Study of Religion 48(2): 313–31.

Layman, Geoffrey C. 1997. “Religion and Political Behavior in the United States: The Impact of Beliefs, Affiliations, and Commitment From 1980 to 1994.” The Public Opinion Quarterly 61(2): 288–316.

Lodge, Milton, Marco R. Steenbergen, and Shawn Brau. 1995. “The Responsive Voter: Campaign Information and the Dynamics of Candidate Evaluation.” The American Political Science Review 89(2): 309–26.

McDaniel, Eric Leon, Irfan Nooruddin, and Allyson Faith Shortle. 2011. “Divine Boundaries: How Religion Shapes Citizens’ Attitudes Toward Immigrants.” American Politics Research 39(1): 205–33.

Miller, Joanne M., and Jon A. Krosnick. 2000. “News Media Impact on the Ingredients of Presidential Evaluations: Politically Knowledgeable Citizens Are Guided by a Trusted Source.” American Journal of Political Science 44(2): 301–15.

Nteta, Tatishe M., and Kevin J. Wallsten. 2012. “Preaching to the Choir? Religious Leaders and American Opinion on Immigration Reform.” Social Science Quarterly 93(4): 891–910.

Obama, Barack. 2014. “Remarks by the President in Address to the Nation on Immigration.” Whitehouse.gov. https://www.whitehouse.gov/the-press-office/2014/11/20/remarks-president-address-nation-immigration (June 15, 2016).

Putnam, Robert D, and David E. Campbell. 2012. American Grace: How Religion Divides and Unites Us. New York: Simon & Schuster.

Rocha, Rene R., and Rodolfo Espino. 2009. “Racial Threat, Residential Segregation, and the Policy Attitudes of Anglos.” Political Research Quarterly 62(2): 415–26.

Smith, Gregory A. 2005. “The Influence of Priests on the Political Attitudes of Roman Catholics.” Journal for the Scientific Study of Religion 44(3): 291–306.

Smith, Gregory Allen. 2008. Politics in the Parish: The Political Influence of Catholic Priests. Washington, D.C: Georgetown University Press.

USCIS. 2015. “Executive Actions on Immigration.” USCIS. https://www.uscis.gov/immigrationaction (June 15, 2016).

Wald, Kenneth D., Dennis E. Owen, and Samuel S. Jr. Hill. 1988. “Churches as Political Communities.” American Political Science Review 82(2): 531–548.

Wilson, Thomas C. 2001. “Americans’ Views on Immigration Policy: Testing the Role of Threatened Group Interest.” Sociological Perspectives 44(4): 485–501.

Zaller, John R. 1990. “Political Awareness, Elite Opinion Leadership, and the Mass Survey Response.” Social Cognition 8(1): 125–53.