Elite Leadership of Opinion and the Public Polarization:

Elite Leadership of Opinion and the Public Polarization: The Same Sex Marriage Debate in the United States

Patricia Victorio

Key words, terms, names, concepts: same sex marriage, gay marriage; California proposition 8, Prop 8; public opinion; political ideology; elite leadership; religion, religiosity; polarization.

The California Supreme Court made a landmark decision with the court case In re Marriage Cases (2008), legalizing same sex marriage within the state, and overturning the California Defense of Marriage Act (Proposition 22). With a swift decision the supreme court put the controversial issue of same sex marriage back in the media spot light. Outside of California, states such as Arizona also reopened the debate of same sex marriage. The Arizona legislature put this issue up for a vote in the Fall 2008 election. The Arizona ballot measure, Proposition 102, wanted to define marriage between one man and one woman, and prevent gay marriage from being legal in the state of Arizona. The New York Times reported in October 2008 that Arizona wanted to stop politicans and judges from overturning the same sex marriage bans which happened in California and Massachusetts (McKinley2008). In 2008, with the issue in the media spolt light once again, same sex marriage became an important issue for the election.

With the debate in full gear, both sides became highly charged regarding the issue. Proposition 8 in California was recorded as one of the most expensive ballot propositions in the history of the United States with donanations coming in from all over the United States.Over $60 million was the total combined amount spent on both sides of the issues. The only more expensive campaign during the 2008 election was the presidential campaign (Ewers 2008). In the 2008 election, the public was exposed to opposing views in the media, on commericals, and from within various communities. If the media and elites are showing differing opinions from within the media and other elites, how then is the public to react to an issue such as gay marriage? This research seeks to find out how the public responded to the polarization of elite opinion on the issue of gay marriage.

Literature Review

Predispositions: Religiosity and Political Ideology

The issue of same sex marriage is a contentious issue because it crosses the political arena into the religious field and into the morals battlefield (Brewer, 2003). This investigation is important because it helps social scientists identify predispositions for support for gay marriage. In the article, "Conservative Protestantism and Tolerance toward Homosexuals: An Examination of Potential Mechanism" authors Burdette, Ellison, and Hill (2005) want to further explain why Conservative Protestants are less willing to support gay marriage and other gay rights. The authors argue that church attendance, beliefs about the bible, social contamination, moral privatism, and beliefs about the morality of GLBT people were characteristics that explained why conservative protestants were more unlikely to support same sex rights. Through the research, the authors found that church attendence and biblical literalism were very important characteristics of conservative protestantism that help indicate for support for same sex rights. More importantly from this research, the authors grounded the same sex marriage debate in religion and religiosity. Moreover, the authors of this article, while explaining why conservative protestantism produces less support for same sex rights, showed what characteristics of respondants were willing to support or not support same sex rights. These indicators were church attendance and biblical literalism.

In "Religion and Public Opinion" Olson, Cadge, and Harrison (2006) took the issue of gay rights further by specifically attempting to single out one GLBT political issue, same sex marriage, and how religion influences public opinion on this particular issue. Using the data from a telephone survey conducted by Greenberg Quinlan Rosner Research, Inc. in April 2004, the authors used a multiple logistical regression as their methodology. Unlike Burdette et al. (2005), the authors found variance within the myriad of religious affliliation. The authors found that being Jewish and mainstream Christianity were the greatest indicators affecting support for same sex marriage. Overall they concluded that religion is very powerful in influencing public support for same sex marriage, and that the opposition towards same sex marriage dissapproved of allowing government to recognize same sex marriages was because of moral reasons (Olson et al. 2006). This research also anchors the debate of same sex marriage in religion and morals. Although the authors uncovered religious variance on the topic of same sex marriage, the author do not dive into why certain religions seem to approve of same sex marriage, while others do not.

Paul Brewer (2008) found that partisanship and ideology might be a strong indicator for how people view same sex marriage. Using ANES data, he found that liberals and conservatives were divided on the issues of gay rights including nondiscrimination in the workplace, gays in the military, adoption rights, and same sex marriage. Conservative Republicans were more likely to oppose gay rights than liberal Democrats. The relationship between the support for same sex marriage and political idenfication and ideology can be seen as more complex. The Pew Reseach Center report from 2003 found that being Democrat or Republican was not a strong indicator for support for same sex marriage(Pew Research Center, 2003). Although Republicans are more strongly opposed in comparison to Democrats, the issue is that the Democratic party seems to be split on the matter. What is missing from this body of research is an explanation of why there are variations within political ideology and partisanship. The question we pursue becomes, how is the information being transmitted to the public that there is discrepancy within political ideology and partisanship? To answer this question, we need to look at how the public is filtering the information given to them by media and elites.

Leadership and Polarization in Public Opinion

Looking at one of the very salient issues of the GLBT movement, gay marriage, authors Becker and Scheufele want to break down public opinion to the individual level in their article "Moral Politicking: Public Attitudes towards Gay Marriage in an Election Context." They answer the question "Why do people choose to support or disfavor gay rights?" The authors analyzed data from the Cornell Media Attitudes Survey, a nationwide from 2003. The dependent variable is the support for gay marriage and the independent variables are ideology, religious values, political knowledge, political tolerance, and the media. Becker and Scheufele used hierarchical ordinary squares (OLS) regression to evaluate the data. The authors, Becker and Scheufele wanted to examine predispositions, media use, and political inputs influence public support of gay marriage. They examined a variety of factors that can influence the opinions of individuals towards support of same sex marriage. The authors found that religiosity and ideology act as a filter and moderate the effect of attention towards entertainment and news campaigns in the issue of gay marriage. Therefore, according to the authors of the article, the media had little impact on actually influencing the public opinions towards gay rights. Political knowledge and political toleration was weaker in explaining the support or anti-support for gay marriage. By their research, outside influence had little effect on the individual.

The authors looked at different relationships such as media influences, political knowledge and political tolerance as other factors that could explain the person's ideology. They were able to rule out some other factors such as the media, and asserted that ideology and religiosity have a greater influence in support for gay marriage. This was important because both act as filters when evaluating the media and other outside forces. Moreover, the media, according to this article, was less of an influence than religion and party identification. The authors interpreted their data to mean that media had little effect since the information was filtered by existing predispositions by the individual, but media do have an effect on individuals by further grounding their support either for or against same sex marriage. Their premise about the media should have stated that exposure to media does not act as a variable that changes support for same sex marriage. The authors do not dive in further to see if the media actually reaffirms their beliefs.

To help unravel the role of the media, Brewer (2008) in his article "From the Podium to the Pulpit: Opinion Leadership and Gay Rights" in the book Value War: Public Opinion and the Politics of Gay Rightstested for polarization effects on the issue of same sex marriage. His goal was to show that leaders polarized on the subject extend that conflict to the public. Leaders or elites could be media, government and religious elites. By testing for public polarization on the issue of the gay marriage, Brewer can indirectly show a relationship between media and the public that Becker and Schefuele (2009) were unable to do. Moveover, Brewer can show the individual receives the flow of information about same sex marriage policy from the top elites in the media, government and from church.

Controlling for political ideology, partisanship, presidential approval for George W. Bush and religious doctrine, Brewer found that the polarization did happen within the public amongst the most politically attentive citizens and with the people with higher church attendence or levels of religiosity. Therefore, his initial hypothesis about the public taking cues from the elite about gay marriage is proven correct. Brewer used the Pew Research Center data from August 2003. To measure for political attentiveness, Brewer used levels of education as an indicator. To test for polarization within the religious department, Brewer used the Cooperative Clergy Study Project and the Pew Research August 2003 data. He came to the same conclusion regarding polarization of the religious congregation from the clergy. Although Brewer's study shows the polarization of opinion in the political arena, he is unable to rigoursously show the same for the effect of elite clergy signals to the congregation. Moreover, Brewer chooses levels of education as an indicator for political attentiveness which can be used but a more specific variable might yield more concrete results.

Definitions and Measures

In order to find out if there is elite leadership in public opinion regarding the debate of same sex marriage, first we need to understand how the public opinion is formed and how to define and measure polarization in public opinion. The formation of public opinion on an issue consist of two things: first the information given to the person and, second, the predispositions he or she holds (Zaller 1992). The information gives the person an image of what is going on. The predispositons help that person interpret and analyze the information, and then come to certain conclusions about the information. But the information has to go through elites, and the elites transfer that information to the public. The public never receives the true neutral form, so to speak, because the elites that carry the information change the information by framing it in certain ways (Zaller 1992). Moreover, elites are defined as persons who have given full time and resources to a certain aspect of politics or public affairs; this could include but is not limited to politicians, government officials, journalist, activist, and experts in policy.

Most Americans are "rationally ignorant" about politics, so the majority look towards the elites for guidance on the issues (Zaller 1992). So when elites decide on an a certain policy issue, the public will mostly likely adopt the same policy issue. There are differing levels of ignorance. When elites disagree along party/partisan and there is an even flow of information, what does the public do? Zaller (1992) argues in his book The Nature and Origins of Public Opinion that the public follows the elites' cues. Taking the signals from the elites, the more politically aware a person is the more likely he or she will be persuaded by the information favored by the same party he or she aligns. Thus political attentiveness is defined through interest to the government and politics. If there is a wider gap between favor and opposing to the most attentive citizens than the least attentive citizens, then it shows that citizens do take signals from elites on the issue of gay marriage.


By confirming whether or not he public follows the elite cues on the issue of gay marriage we can better understand how opinion is formed on such an important issue. The role of the elite leaders through the media can also be comprehened in the debate in same sex marriage. This is very significant especially to demonstrate the tension between the media, political and religious elites, and policy making process in a democratic society. This research can help citizens in the United States understand the importance of elite leadership in politics and in the media. Further, it draws further attention to the value of agenda setting within the media and government, and it can show the importance of how a policy is framed can shift public opinion.

Premise for Investigation

Ample research has been done to show that religion is an important variable in influencing the attitude and support for same sex marriage. Therefore I hypothesize that religiosity is an important predictor towards support for same sex marriage. Brewer (2003) and The Pew Research Center (2003) found through their data that political ideology and partisanship does not yield a strong relationship in influencing the public support for same sex marriage. Therefore I hypothesize that political ideology is not important predictor towards support for same sex marriage. Looking at polarization of public opinion in the support for same sex marriage could potentially shed light on how citizens form their opinion on same sex marriage and the influences on their decision.

Under the rubric and guidance of Brewer's research in his article From the Podium to the Pulpit: Opinion Leadership and Gay Rights" (2008) and Zaller's (1992) research on elite leadership of public opinion, when the public are exposed to equal but opposing information about same sex marriage that they will rely on elite signals or cues to come to their decision about support for same sex marriage. Finding polarization within the citizens with the highest attention to government and politics shows if the public does take cues from elites on the issue of same sex marriage. Therefore, opinion polarization on gay marriage is wider among the politically attentive than the less politically attentive.

H1: Religiosity is an important predictor towards support for same sex marriage.
H2: Political Ideology is not important predictor towards support for same sex marriage.
H3: Opinion polarization on gay marriage is wider among the politically attentive than the less politically attentive.

Tools and Data


The data used in this research was the American National Election Studies 2008 Time Series data. A linear regression was preformed to illustrate the relationship between religiosity and partisanship to same sex marriage. To measure religiosity, I used the variable of church attendance. To measure ideology, I used a variable asking the respondent their ideology was either conservative, moderate, or liberal. The variable to determine political attentiveness was asking how interested the respondant was in government and politics. The answers were grouped and recoded into three answers extremely interested, interested, not interested. Finally, to measure the support for same sex marriage, I used the variable that asked the respondants if gay marriage should be allowed, should not be allowed, should not be allowed but civil unions, or other. Since this paper looks at specifically supporting same sex marriage not unions, the same sex variable was recoded to 'should not be allowed to include the allowing civil unions.'


Table 1: Religiosity and Political Ideology as Influences for Support for Same Sex Marriage



Unstandardized Coefficients

Standardized Coefficients




Std. Error









Church Attendance






Religion provides some guidance in day-to-day living






Political Ideology






a. Dependent Variable: X16. R position on gay marriage

Table 1 shows the strength of the variables political ideology and religiosity to support for same sex marriage. The p value is significant if it is less than or equal to 0.05. Political ideology has a p value of .283 while religiosity measured in terms of church attendance has a p value of 0.014. But, interestingly, religion measured by guidance in every day life shows a weak relationship to support for same sex marriage. Thus church attendance seems to play an imporant role in influencing the public's opinion on same sex marriage. Political ideology is not a strong indicator when determining the approval of same sex marriage.

Tables 2a, 2b, and 2c show three graphs involving the variables interest in government, support for same sex marriage, and church attendance. Controlling for religiosity with the variable of church attendance helps uncover the relationship between political attentiveness and support for same sex marriage. By conducting a cross tabulature and graphing the results, the widest gap between the opposing opinions happens with the citizens with the most political attentiviness.With the graph for the highest religiosity, church attendance is at a frequency of every week, the gap between the opinion for those who have the highest political attentiveness. The graph with religiosity was church attendance of a few times a year, followed the same trend as the graph before. The widest gap occurred with the highest political attentiveness. It is important to note that with the least political attentiveness (not interested in government and politics) the number of respondants opposed for same sex marriage and the number of respondents in favor of same sex marriage were equal. Therefore the gap between the opinions in the lowest political attentiveness is zero. The graph with the lowest religiosity, repsondents attending church never, did not follow the trend from the previous gaps. But this graph is not a significant outlier and can be thrown out because of the number of respondants in the category of never attending church. In comparison to the other tables, the number of respondents in the lowest religiosity graph is 6. While the number of the other graphs were 384 and 388 respondents. Therefore the last graph is not an accurate representation of the sample population.

The results illustrate polarization of public opinion on the issue of gay marriage at the highest level of political attentiveness. This translates to the elite leadership of opinion on the issue of same sex marriage (Zaller 1992, Brewer 2008). The people with the highest level of political attentiveness have the highest exposure to signals or cues. Since the elite signals to the public are opposing, the individual use predispositions such as religion to filter out the opposing information, but the views which aligns with the predispositions further ground the opinion in the individual. What we see and hear is elite leadership in opinion for same sex marriage.

Further Discussion

As the Pew Research Center, discussed earlier, ideology and partisanship did play a role but there is not a strong relationship. The findings show that possibly the reasons why it is not a strong indicator for this issue is that Democrats seem to be divided on the issue. This is most likely because the debate on gay marriage is drawn on moral and religious lines. The regression results indicating that church attendance has a more significant relationship to same sex marriage in comparison to religion as guidance for everyday life. This could potentially mean that chuch goers take the cues from religious elites. I do not want to overstate my case, this study was not looking for religious elites and signals to their congregation.

The significance of elite leadership of public opinion is that the conflict within the realm of the elites are being extended to the public. Authors Layman and Carsey (2002) in their article "Party Polarization and "Conflict Extention" in the American Electorate" define conflict extension as the spreading of the conflict from the elite level to the electorate level. This is an important issue to contend with because it raises the issue of the lack possibility of bipartisanship with the issue of same sex marriage. If Americans, as Brewer (2008) argues, are voting based on their morals then how the media and elites in general form those issues becomes critical. To people on both sides of the issue of gay marriage could attempt to create a dialoge that could garner public support and bi-partisanship. Further, polarization of public opinion on the issue of same sex marriage is not a healthy sign of our liberal democratic system.

Missing from this body of research is a comphrensive study on how religious elites and the religious congregation interact in forming the opinion of same sex marriage. What we do know is that the gay marriage debate is drawn on moral and religious lines and that certain characteristics of religion provide a better relationship between support for same sex marriage and religion. Are there any regional differences within religion that affect the formation of opinion on the subject? Does the congregation signal the religious elites or vice-versa? The next step in the research in same sex marriage in the United States could be attempting to tease out how religion, religious elites, and the congregation exchange information that lead to religiosity being an important influence in the same sex marriage debate.

Works Cited

Becker, Amy B and Dietram A Scheufele. "Moral Politicking: Public Attidues towards Gay Marriage in an Election Context." The International Journal of Press/Politics 14.186 (2009): 186-211.

Brewer, Paul R. "The Shifting Foundation of Public Opinion about Gay Rights." The Journal of Politics 65.4 (2003): 1208-1220.

Brewer, Paul R.. Value War: Public Opinion and the Politics of Gay Rights. Boulder: Rowman & Littlefield Publishers, INC. , 2008

Burdette, Amy M, Christopher G Ellison and Terrence D Hill. "Conservative Protestantism and Toelrance toward Homosexuals: An examination of Potential Mechanism." Sociological Inquiry 75 (2005): 177-196.

Ewers, Justin. (2008, October 29). California same-sex marriage initiative campaigns shatters spending records. U.S. News & World Report, Retrieved from http://politics.usnews.com/news/national/articles/2008/10/29/california-same-sex-marriage-initiative-campaigns-shatter-spending-records.html.

Layman, Geoffrey C and Thomas M. Carsey. "Party Polarization and "Conflict Extension" in the American Electorate." American Journal of Political Science 46.4(2002): 786-802.

McKinley, Jesse. "Same-Sex Marriage on the Ballot in Arizona, a Second Time." New York Times. October 29, 2008.

Olson, Laura R, Wendy Cadge and James T. Harrison. "Religion and Public Opinion about Same Sex Marriage." Social Science Quarterly 87.2 (2006): 340-360.

The Pew Research Center For the People & The Press. "The Pew Research Center." 18 November 2003. people-press.org. 22 April 2010 <http://people-press.org/>.

Zaller, John R. The Nature and Origin of Mass Opinion. New York City: Cambridge University Press, 1992.