Polling and the Media |
Voters and Values in the 2004 Election
GARY LANGER is the director of polling at ABC News. JON COHEN is the assistant director of polling at ABC News.
Address correspondence to Gary Langer; e-mail: Gary.e.langer{at}abc.com.
 |
Abstract |
|---|
 TOP Abstract Appendix AAppendix BAcknowledgementsReferences |
|---|
A poorly devised exit poll question undermined meaningful analysis of voters’ concerns in the 2004 presidential election. Twenty-two percent of voters picked "moral values" from a list of "issues" describing what mattered most in their vote, more than selected any other item. Various commentators have misinterpreted this single data point to conclude that moral values are an ascendant political issue and to credit conservative Christian groups with turning George W. Bush’s popular vote defeat in 2000 into his three million–vote margin of victory in 2004. We suggest, rather, that while morals and values are critical in informing political judgments, they represent personal characteristics and ill-defined policy preferences far more than any discrete political issue. First by conflating morals and values and then by further conflating characteristics and issues, the exit poll’s "issues" list distorted our understanding of the 2004 election. In this article, we examine the flaws in the 2004 National Election Pool exit poll’s "most important issue" question and explore the presumed rising electoral importance of moral values and the conservative Christians who overwhelmingly selected this item. Using national exit poll data from 1980 through 2004 and other national surveys, we find that the moral values item on the issues list cannot properly be viewed as a discrete issue or set of closely related issues; that its importance to voters has not grown over time; and that when controlled for other variables, it ranks low on the issues list in predicting 2004 vote choices. The aggregated exit poll data also show that the voting behavior of conservative Christians is relatively stable over time, and these voters were not primarily responsible for Bush’s improvement in 2004 over 2000.
"Moral values" led the list of top issues cited by voters in the 2004 National Election Pool (NEP) exit poll, leading to widespread reports that a rise in moral concerns, particularly among conservative Christians, was the driving factor in George W. Bush’s reelection. This conclusion, the combined result of a poorly constructed survey question and incomplete data analysis, is misplaced.
Voting behavior depends on a panoply of influences—attitudes and emotions, issues and attributes alike (Miller and Shanks 1996
). Values—and for some voters, morals—are important elements of some such influences. But our analysis of the exit poll data shows that "moral values," when controlling for other variables, ranked only as high as fourth of seven competing items in predicting vote choices, behind terrorism, the economy, and Iraq and tied with health care.1 Nor were conservative Christians responsible for Bush’s improvement over the 2000 election; neither their share of the electorate nor their support for Bush increased.2
This was not the message delivered by many news outlets in their election coverage. "Voters who care about moral values delivered the election to President Bush," the Washington Times declared in an editorial. It was "an election that . . . amounted to a referendum on moral values," reported USA Today. On CNN’s Crossfire, cohost Tucker Carlson said, "Three days after the election, it is clear that it was not the war on terror, but the issue of what we’re calling moral values that drove President Bush and other Republicans to victory this week."3
These and other commentators were led astray by a seemingly simple and straightforward marginal result: Terrorism did not rank first among the issues presented to voters on the exit poll questionnaire as "most important" in their vote. As shown in table 1, 22 percent instead cited moral values; 20 percent, the economy and jobs; and 19 percent, terrorism.4
|
The inclusion of moral values distorted this list (Langer 2004). Compared with the other items, it is not commensurate, comparable, or a discrete political issue. Instead it served as an ill-defined grab bag, especially for Bush voters, who, compared with John Kerry’s voters, had fewer appealing options among the other items offered and among whom this phrase particularly resonates.5
The other terms on this list are not entirely precise: "Health care" may cover subtopics from insurance costs to prescription drug benefits; "economy/jobs" may boil down to interest rates or gasoline prices; and picking "the war in Iraq" does not tell us, in and of itself, the respondent’s position on the war. But these phrases have reasonably specific common meanings in a way that "moral values" does not.6 A Pew Research Center (2004) survey after the election asked respondents who cited moral values as an important issue in their vote what the phrase brought to mind. The open-ended question produced a broad range of responses: Twenty-nine percent said gay marriage; 28 percent, abortion; 18 percent, religious belief; 9 percent, honesty or integrity; and 7 percent, "other policy issues." Six percent cited Kerry as having or lacking such values; an equal 6 percent cited Bush as having or lacking such values. Pew produced 17 coded categories in all (multiple answers were accepted), including 2 percent who cited the media or "what’s on TV." The range of meanings underscores the point that moral values are not a discrete political issue. We suggest instead that, unlike the other six items on the list, moral values functions as an amalgam of personal attributes, policy-related predisposition, and preference concerning current conditions of immorality (Kinder and Sears 1985; Miller and Shanks 1996; Zaller 1992). It should not have been included with such unlike items in a single list question.
Including the word moral (as in "Moral Majority") compounded the distortion, making this not just any misplaced attribute in an issues list but also a religious attribute. Moral is a hot-button word (ill-advised in polling) that resonates with a particular, core Republican group, religious conservatives; its use makes it impossible to differentiate policy attitudes from a recitation of those characteristics. In a preelection ABC News poll, likely voters divided by 51 percent to 41 percent on which candidate best represented their "values." In the exit poll, by contrast, those who selected "moral" values favored Bush by 80 to 18 percent.
Indeed, the moral values item was overwhelmingly, almost irresistibly, appealing to religious conservatives. As shown in table 2, 72 percent of conservative religious evangelical white Christians selected moral values from the seven-item list, as did 61 percent of voters who said that the top attribute they sought in a candidate was "religious faith," 45 percent of conservative white Protestants or other non-Catholic Christians, 43 percent of weekly churchgoing white Protestants or other Christians, 42 percent of evangelical white Protestants or other Christians, and 37 percent of conservatives.7 It was chosen by far fewer Catholics (17 percent) and those who attend church infrequently if ever (15 percent), the groups that, as we explore below, shifted the most toward Bush in 2004 versus 2000.
|
In a multivariate model, presented in table 3, attending church weekly, self-identifying as a conservative, and self-identifying as an evangelical Christian are the strongest positive predictors of selecting moral values as the top election "issue." Each increases the likelihood of a voter selecting moral values by about 10 percent (being a Democrat has the largest negative impact, lowering the likelihood by 12 percent).
|
The moral values item on the exit poll issues list, then, created its own center of gravity. The word religion may just as well have appeared in its place—inarguably important to measure but simply not a comparable political issue. The question in effect asked voters: "What was the most important issue in your vote for president—the economy, terrorism, Iraq, health care, taxes, education—or, rather, are you a conservative, religious person?"
An argument might have been made for including values, moral or otherwise, on the issues list if it were volunteered in significant numbers in preelection polls that asked, open-ended, the most important issue in the election.8 But it was not; "moral values" as a phrase almost never came up in such surveys, and responses that could reasonably be coded as involving such values (from honesty, to abortion, to being a religious person) were not volunteered, in aggregate, beyond single digits.9
Regardless of who selected moral values, exit poll data do not support claims that such values were ascendant as a political concern. National exit polls by the Los Angeles Times included "moral values" as an option in 1992 and "moral/ethical" values as options in 1996, 2000, and 2004. In the 1992 L.A. Times poll, "moral values" was selected by 24 percent; in the 2004 NEP exit poll, as noted, by 22 percent. This is not a perfect comparison, because the questions include different items, the L.A. Times poll listed more items, and it allowed two responses. The 1996–2004 L.A. Times polls are more comparable; they listed different items but the same number of items and the same "top two" choice. In 1996 "moral/ethical" values was selected by 40 percent; in 2000, by 35 percent; and in 2004, again by 40 percent (see table 4).
|
While a set of ill-defined "moral values" was not ascendant, it is still possible that the voting patterns of the conservative Christian groups whose members disproportionately selected the item on the exit poll did in fact differ in 2004 relative to previous elections. If by topping the list moral values pointed to this group’s true impact, then, perhaps the poorly devised exit poll question would not necessarily have distorted election analysis. And in fact, political commentary on the election result reached beyond the selection of "moral values" as the top "issue" to a discussion of the voters who predominantly selected it. "The Republicans are in power because the evangelicals and social conservatives voted them into power," the evangelist Pat Robertson said in November 2004 (Dunham and Gleckman 2004). Some evangelicals used the moral values number not only to proclaim their impact but to demand political payback: "‘Values’ voters delivered for the president, and the president must now deliver for them—especially in the courts."10 But is the evangelical impact so clear?
Conservative Christians are a core Republican voting bloc (Guth and Green 1991; Layman 1997; Pew Research Center 2005). But simple mathematics requires that for this group to be responsible for the change in popular vote margins from 2000 to 2004, its members must have made up a greater share of the electorate and/or supported Bush in disproportionately greater numbers. Neither is substantively so.
Exit poll trend data are not available for evangelicals; the 2000 survey did not include the question.11 But various closely related groups, including conservative white Protestants, weekly churchgoing white Protestants, and all weekly churchgoers, did not make up a greater share of the 2004 electorate than they did in 2000, and their vote patterns were largely similar.12 For example, table 5 shows that weekly churchgoing white Protestants were no more likely to vote for Bush in 2004 than they were in 2000. For other categorizations, the changes in vote choice were relatively small; indeed, a bigger change is among Catholics. Nor was there any significant change in the share of the electorate made up by conservative white Protestants or similar groups: Weekly churchgoing white Protestants if anything made up a somewhat smaller share of the 2004 electorate than the 2000 electorate.
|
As shown in table 6, the percentage of Republican voters who are white Protestants or other Christians was essentially the same in 2004 as it was in 1980. In 2004, white Protestants and other Christians cast 55 percent of all votes for Bush; in 1980, 56 percent of all of Ronald Reagan’s votes came from this group. Similarly, 34 percent of Bush’s 2004 votes came from conservative white Protestants and other Christians; in 1980 this group accounted for 35 percent of Reagan voters.13
|
While the relative size of these voting blocs has been stable, conservative white Protestants and other Christians have been increasingly likely to self-identify as Republicans (see table 6). The changing partisan identification of conservative white Protestants and other Christians has electoral implications, but they are not chiefly responsible for Bush’s improvement from 2000 to 2004. Indeed, Bush got more additional votes from Catholics than from white Protestants in 2004 versus 2000.
A related measure, church attendance, tells a similar story. Frequent church attendees did not make up a larger share of the electorate in 2004 than they did in 2000; nor, again, was there any significant difference in their likelihood of voting for Bush. Weekly church attendees—particularly white Protestant and other Christian weekly churchgoers—were much more likely to vote for Bush than for Al Gore in 2000 or John Kerry in 2004 (see table 7). However, table 8 shows that this relationship is severely attenuated when controlled for political party identification and ideology and that the estimated effect of frequent churchgoing on a Bush vote did not increase from 2000 to 2004.
|
|
Indeed, contrary to the notion that frequent church attendees delivered the 2004 election to Bush, his bigger gains over 2000 were in the opposite groups—those who attend church infrequently or not at all. Table 7 shows that voters who reported never attending church were four points more likely to support Bush in 2004 than they were in 2000; among those who attend church once a week or more, the increase was not significant—a single point. (Specifically among white Protestants and other Christians, Bush’s support improved by six points among those who attend church less than weekly, versus two points among those who attend weekly or more often.) As noted above, 32 percent of weekly churchgoers picked moral values as the top issue in their vote, compared with 15 percent of less-frequent churchgoers.
While Bush’s support among conservative white Protestants changed little from 2000 to 2004, his support among Catholics, quintessential swing voters, increased more significantly. (White Catholics, in particular, have gone with presidential election winners since at least 1980, counting Bush in 2000 as the winner. Bush won white Catholics by a 13-point margin in 2004, compared with a seven-point margin in 2000.) But moral values was comparatively less important to the Catholic vote: As noted, 17 percent of Catholics (and 19 percent of white Catholics) selected moral values as the most important issue in their vote, half the number of evangelical white Protestants who did so.14 If anything, then, the effect of religion in 2004 was more about Bush’s gains among infrequent churchgoers and Catholics than his support among the evangelical white Protestants who have been the primary focus of postelection analyses.15
Momentarily setting aside conceptual problems with the moral values item, a closer analysis of the data shows that it was far from the "most important issue" in Bush’s electoral victory. A logit analysis predicting Bush votes, presented in table 9, shows that moral values ranks much lower than other issues in our model. Controlling for partisan self-identification, race, ideology, religion, and church attendance, moral values has less predictive power than terrorism, the economy, and Iraq; it is tied with health care for fourth in terms of predictive probability. In the Bush model, moral values has less than half the predictive power of the terrorism item.16 (The single biggest predictor of Bush votes, not surprisingly, is self-identifying as a Republican.)
|
These data leave open the question of what moral values actually means, beyond invoking conservatism and religiosity. The Pew results cited above suggest that, albeit only for some respondents, the question may have served as a stand-in for two implied but unspecified issues, same-sex marriage and abortion. Unfortunately, the NEP exit poll asked the issues list in one version of its questionnaire and questions on abortion and same-sex couples on a different version, making cross-tabulation of these variables impossible. (In an ABC News/Washington Post poll in April 2005, opposition to abortion and to same-sex marriage, religious conservatism, and the view that elected leaders should rely on their religious beliefs in making policy decisions all related positively with one another—with correlations between .23 and .36.)17
In the exit poll, 26 percent of voters said both that abortion should be illegal in all or most cases and that there should be no recognition of same-sex marriages—roughly equivalent to the 22 percent who selected moral values as the most important "issue" in their vote. Bush’s support is as highly correlated with opposition to abortion and to same-sex unions (phi = .345; p < .001) as it is with the selection of moral values (phi = .311; p < .001). Moreover, controlling for party ID, ideology, religion, and church attendance, attitudes toward abortion and same-sex unions predict Bush votes as well as—indeed slightly more strongly than—moral values (see table 10).
|
For some voters "moral values" may have meant abortion and/or same-sex marriage; for others these two issues and "moral values" may be related but not identical; for yet others, moral values may entirely reflect personal attributes or predispositions related to morality or any of a range of other issues, sentiments, or beliefs. It is this lack of definition that renders the phrase "moral values" so inapt as a political "issue."
The conflation of political, ideological, and religious views represents a critical aspect of election politics. Measuring and understanding voters’ values and religious beliefs, and the extent to which these inform vote choices, together and separately, is crucial to election analysis. Indeed these influences are too important to be mismeasured with an ill-defined phrase in an ill-constructed list, as occurred in the 2004 National Election Pool exit poll. The inclusion of "moral values" on the exit poll as part of a seven-item issues list undermined rather than elucidated our understanding of the 2004 election. The list conflated voter attitudes on issues and underlying predispositions. Moreover, even given this faulty list, "moral values" was not, in fact, the most important "issue" in the election; nor were the conservative Christian groups who disproportionately selected it the main movers in turning Bush’s popular vote defeat in 2000 into his 2004 victory. Future exit polls should measure issues and underlying values discretely; exit polls, after all—like all other survey research—require balanced questions and cautious analysis alike.
|
Appendix A |
|---|
|
TOPAbstractAppendix AAppendix BAcknowledgementsReferences |
|---|
ABC News, Good Morning America, November 4, 2004: Well, the campaign is over but one of the big surprises still has everyone talking. Voters saying that their top issue in choosing a candidate, not the economy, not terrorism, not Iraq, but moral values.
AP, November 3, 2004: Moral values—heavily emphasized by the president—edged terrorism and the economy as the top issue.
Atlanta Journal-Constitution editorial, November 10, 2004: . . . the message was clear. Faith and family values matter; they are not just relevant but dominant. Last week’s overwhelming 3.5 million-vote presidential margin has given us "values voters" an important reassurance that will resonate in politics for years to come. . . .
Chicago Tribune, November 4, 2004: There is little doubt that faith fueled the re-election of President Bush, which some described as electoral evidence of the growing political proximity of "God and country." Asked which single issue mattered most in their choice for president, 22 percent of voters chose moral values above everything else. . . .
CNN, Crossfire, November 5, 2004 (Tucker Carlson, cohost): Three days after the presidential election, it is clear that it was not the war on terror, but the issue of what we’re calling moral values that drove President Bush and other Republicans to victory this week.
Cox News Service, November 3, 2004: Headline: Moral values a deciding election issue
Dallas Morning News, November 4, 2004: Their seeds planted 20-plus years ago by the Moral Majority, America’s values voters blossomed this year into a political force that could portend a lasting Republican majority.
Deseret Morning News/Religion News Service, November 4, 2004: Forget Iraq. Forget terrorism. Forget the economy. The biggest factor shaping people’s votes Tuesday was the mother of all sleeper issues—"moral values."
Hartford Courant, November 4, 2004: Headline: An unexpected "moral" victory; Bush benefits as voters put heavy emphasis on values
Minneapolis Star Tribune, November 4, 2004: More than any other concern—not job creation, not Iraq—voters in exit polls declared "moral values" their top priority, and 80 percent of them backed the president. Those 20 million voters were at the heart of Bush’s reelection which gave him new clout in pursuing his conservative agenda.
N.Y. Times, November 4, 2004: . . . Americans said they were motivated to vote for President Bush on Tuesday by moral values as much as anything else. . . . In the survey, a striking portrait of one influential group emerged—that of a traditional, church-going electorate that leans conservative on social issues and strongly backed Mr. Bush. . . .
Scripps Howard News Service, November 4, 2004: . . . one statistic might help to put into perspective the verdict from the rest of America. . . .
USA Today, November 4, 2004: In an election that . . . amounted to a referendum on moral values. . . .
Washington Times editorial, November 4, 2004: Voters who care about moral values delivered the election to President Bush. . . . The exit polls couldn’t have been clearer. They showed that more voters think moral values—that is, the vaunted "God, guns and gays" questions—are the most important question facing the nation than think the same about the state of the economy, the terrorist threat or the Iraq war.
|
Appendix B |
|---|
|
TOPAbstractAppendix AAppendix BAcknowledgementsReferences |
|---|
This analysis used the following questions from the 2004 National Election Pool exit poll:
Are you:
- White
- Black
- Hispanic/Latino
- Asian
- Other
In today’s election for president, did you just vote for:
- John Kerry (Dem)
- George W. Bush (Rep)
- Ralph Nader (Ind)
- Other: Who? ___________
- Did not vote for president
Which ONE issue mattered most in deciding how you voted for president? (Check only one)
- Taxes
- Education
- Iraq
- Terrorism
- Economy/Jobs
- Moral values
- Health care
Are you:
- Protestant
- Catholic
- Mormon/LDS
- Other Christian
- Jewish
- Muslim
- Something else
- None
Would you describe yourself as a born-again or evangelical Christian?
- Yes
- No
How often do you attend religious services?
- More than once a week
- Once a week
- A few times a month
- A few times a year
- Never
No matter how you voted today, do you usually think of yourself as a:
- Democrat
- Republican
- Independent
- Something else
On most political matters, do you consider yourself:
- Liberal
- Moderate
- Conservative
Which comes closest to your position? Abortion should be:
- Legal in all cases
- Legal in most cases
- Illegal in most cases
- Illegal in all cases
Which comes closest to your view of gay and lesbian couples:
- They should be allowed to legally marry
- They should be allowed to legally form civil unions, but not marry
- There should be no legal recognition of their relationships
We also used the following questions from the Los Angeles Times exit polls (multiple responses recorded):
- 2004: Which issues, if any, were most important to you in deciding how you would vote for president today? moral/ethical values, jobs/economy, terrorism/homeland security, situation in Iraq, education, social issues such as abortion and gay marriage, taxes, health care, foreign affairs, social security, Medicare/prescription, none of the above
- 2000: Which issues, if any, were most important to you in deciding how you would vote for president today? moral/ethical values, education, jobs/the economy, the environment, taxes, abortion, health care, Social Security, budget surplus, Medicare/prescription drugs, foreign affairs, none of the above
- 1996: Which issues, if any, were most important to you in deciding how you would vote for president today? moral and ethical values, education, jobs/the economy, the environment, taxes, abortion, health care, poverty, federal budget deficit, crime/drugs, foreign affairs, none of the above
- 1992: Which issues—if any—were most important to you when deciding how you would vote for president today? moral values, education, jobs/the economy, the environment, taxes, abortion, health care, poverty, federal budget deficit, crime/drugs, foreign affairs, none of the above
- 2000: Which issues, if any, were most important to you in deciding how you would vote for president today? moral/ethical values, education, jobs/the economy, the environment, taxes, abortion, health care, Social Security, budget surplus, Medicare/prescription drugs, foreign affairs, none of the above
|
Acknowledgements |
|---|
|
TOPAbstractAppendix AAppendix BAcknowledgementsReferences |
|---|
The authors would like to thank Dalia Sussman, Daniel M. Merkle, John Sides, Benjamin Highton, Felicia Cote, Howard Schuman, two anonymous POQ reviewers, and the editors of this special edition for their helpful suggestions.
|
Footnotes |
|---|
1. Moral values tied with health care as the fourth biggest predictor of voting for Bush (in terms of predicted probability). See table 9 below. See also Hillygus and Shields (2005)
for an evaluation of vote preference data from an Internet panel. 
2. Bush lost the popular vote in 2000 by 537,179 votes (per America Votes 24 [Scammon, McGillivray, and Cook 2001]) and won it in 2004 by 3,012,171 votes (per America Votes 26 [Scammon, McGillivray, and Cook 2006]).
3. See appendix A.
4. The 2004 exit poll was conducted by Edison/Mitofsky for the National Election Pool (ABC, AP, CBS, CNN, Fox News, and NBC). The 1996 and 2000 exit polls were conducted by the Voter News Service, a consortium of these same news organizations, absent Fox in 1996. The 1980, 1984, 1988, and 1992 exit poll data used in this article were gathered by ABC News. For a list of questions used in the analysis, see appendix B.
5. Kerry overwhelmingly won voters who picked four issues—the economy (80 percent to 18 percent), Iraq (73 to 26 percent), health care (77 to 23 percent), and education (73 to 26 percent). Bush won voters who picked moral values (80 percent to 18 percent), terrorism (86 to 14 percent), and taxes (by a much narrower 57 to 43 percent).
6. Gary Bauer, former presidential candidate and president of a group called American Values, told ABC News, "Well, obviously the phrase moral values or moral issues is a little vague, and so it leaves it open to a lot of interpretation" (ABC News transcript, This Week with George Stephanopoulos, November 28, 2004).
7. From this point forward, "other Christians" refers to other non-Catholic Christians.
8. Again, our argument is not that moral values are irrelevant (see Abramowitz 1995); to the contrary, they are too important to mismeasure (Stoker 1987).
9. See, e.g., Gallup’s "most important problem" questions from July 8–11 and August 8–11, 2004. In the July and August Gallup surveys, just 7 percent cited dishonesty, lack of integrity, abortion, homosexuality, gay issues, or a decline in ethics, morals, religion, or family as the country’s most important problem (Roper iPoll: USGALLUP.04JULY8.R07, USGALLUP. 04AUST09.R07).
10. Executive Director Gary Cass, Center for Reclaiming America, in ABC News report, "Evangelicals to Bush: Payback Time," November 28, 2004 (http://abcnews.go.com/WNT/story?id=280881&page=1).
11. We limit this trend analysis to comparable exit poll items.
12. In the 2004 exit poll, evangelical white Protestants and other Christians correlated highly with each of these groups, e.g., .56 with weekly churchgoing white Protestants and .47 with conservative white Protestants.
13. The data in table 6 show a small decline in the percent of Republican voters who are white Protestants and other Christians in 2004 relative to the 1988–2000 elections, which suggests growth in the Republican coalition outside the groups now widely supposed to have produced Bush’s 2004 popular vote victory.
14. This finding is confirmed by a logit analysis predicting Bush vote that shows that moral values has a greater estimated effect in predicted probability of voting for Bush among evangelical Protestants than it does among Catholics.
15. These are numerically significant groups: Infrequent churchgoers accounted for 57 percent of voters in the 2004 election, and Catholics accounted for 27 percent, compared with evangelical white Protestants/other Christians at 23 percent.
16. This article deals with the direct effects of the most important issues on vote choice, as respondents’ choices on that question are strongly correlated with their votes. We also estimated models with interactions, not shown here, that indicate that while the top issue highlighted in this analysis, terrorism, had a direct effect on respondents’ likelihood of voting for Bush regardless of religion, moral values influenced presidential vote choice only for people with particular religious backgrounds, particularly evangelical Protestants.
17. This ABC News/Washington Post poll was conducted by telephone, April 21–24, 2005, among a random national sample of 1,007 adults. The fieldwork was done by TNS of Horsham, PA.
|
References |
|---|
|
TOPAbstractAppendix AAppendix BAcknowledgementsReferences |
|---|
Abramowitz, Alan I. 1995. "It’s Abortion, Stupid: Policy Voting in the 1992 Presidential Election." Journal of Politics 57:176–86.[CrossRef]
Dunham, Richard S., and Howard Gleckman. 2004. "Can the GOP Get Down to Business?" Business Week, November 22, p. 43.
Guth, James L., and John C. Green, eds. 1991. The Bible and the Ballot Box: Religion and Politics in the 1988 Election. Boulder: Westview.
Hillygus, D. Sunshine, and Todd Shields. 2005. "Moral Issues and Voter Decision Making in the 2004 Presidential Election." PS: Political Science and Politics 38:201–10.[CrossRef]
Kinder, Donald, and David Sears. 1985. "Public Opinion and Political Action." In Handbook of Social Psychology, 3rd ed., ed. G. Lindzey and E. Aronson, pp. 714–26. New York: Random House.
Langer, Gary E. 2004. "A Question of Values." New York Times, op-ed, November 6, p. 19.
Layman, Geoffrey C. 1997. "Religion and Political Behavior in the United States: The Impact of Beliefs, Affiliations and Commitment from 1980 to 1994." Public Opinion Quarterly 61:288–316.[CrossRef][ISI]
Miller, Warren E., and J. Merrill Shanks. 1996. The New American Voter. Cambridge: Harvard University Press.
Pew Research Center. 2004. "Voters Liked Campaign 2004, but Too Much ‘Mud-Slinging’; Moral Values: How Important?" November 11. Available online at http://people-press.org/reports/display.php3?ReportID=233 (accessed November 7, 2005).
Pew Research Center. 2005. "Religion and Public Life: A Faith-Based Partisan Divide." Washington, DC: Pew Research Center. Available online at http://pewforum.org/publications/reports/religion-and-politics-report.pdf (accessed November 7, 2005).
Scammon, Richard M., Alice V. McGillivray, and Rhodes Cook. 2001. America Votes 24: A Handbook of Contemporary American Election Statistics. Washington, DC: CQ Press.
Scammon, Richard M., Alice V. McGillivray, and Rhodes Cook. 2006. America Votes 26: A Handbook of Contemporary Election Statistics. Washington, DC: CQ Press.
Stoker, Laura L. 1987. "Morality and Politics: Conduct and Control." Report to the NES Board of Overseers based on the 1987 National Election Pilot Study. Available online at ftp://ftp.nes.isr.umich.edu/ftp/nes/bibliography/documents/nes002273.pdf (accessed November 7, 2005).
Zaller, John R. 1992. The Nature and Origins of Mass Opinion. New York: Cambridge University Press.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  I. H. Indridason Does Terrorism Influence Domestic Politics? Coalition Formation and Terrorist Incidents Journal of Peace Research, March 1, 2008; 45(2): 241 - 259. [Abstract] [PDF]
|
|
|
|
 |
|
 S. McCloud Putting Some Class into Religious Studies: Resurrecting an Important Concept J Am Acad Relig, December 1, 2007; 75(4): 840 - 862. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||