After the parties and candidates have presented their campaigns, the voters decide. Exactly how people make their voting decisions affects how well or how poorly elections contribute to the democratic control of government.
How Voters Decide
Years of scholarly research have made it clear that feelings about the parties, the candidates, and the issues have substantial effects on how people vote.1
Social Characteristics
People's socioeconomic status, religion, and ethnic background are significantly related to how they vote. Since the 1930s, for example, African Americans, Jews, and lower-income citizens have tended to vote heavily for Democrats, while white Protestants and upper-income Americans have voted mostly for Republicans. In 2004, 88 percent of blacks, but only 41 percent of whites, voted for John Kerry against George W. Bush; 55 percent of people with household incomes under $50,000, but only 41 percent of those with incomes over $100,000, voted for Kerry; and 74 percent of Jews, but only 29 percent of Protestant regular church-goers, voted for him. Recently, women have voted for Democrats in greater numbers than have men, though the differences were not as marked in 2004; 51 percent of women, but only 44 percent of men, voted for Kerry. Older people have traditionally supported Democratic candidates more often, and at higher rates, than other age groups, but this is no longer true. In 2004, 54 percent of those over 60 voted for Bush (see Figure 10.4). [Figure 10.4 about here; not yet available]
Party Loyalties
To some extent, these social patterns work through long-term attachments to, or identification with, political parties. As indicated earlier, a majority of Americans-between 60 and 65 percent--still say they consider themselves Republicans or Democrats. Party loyalties vary among different groups of the population, often because of past or present differences between the parties on policy issues, especially economic and social issues.2 For this reason, when people use their party identification as a short-cut for choosing a candidate, they are likely choosing a candidate who is close to them on the issues. The ability of party identification to serve as a useful tool for people to choose candidates that are close to them on the issues is further enhanced by the close linkages between the parties and ideologies, with Democrats generally more liberal (including party identifiers, activists, and candidates) and Republicans generally more conservative. 3
Party loyalties are very good predictors of how people will vote.4 Those who say they consider themselves Republicans tend to vote for Republican candidates in one election after another, and those who consider themselves Democrats vote for Democratic candidates. This is especially true in congressional elections and in state and local races, where most voters know little more about the candidates than their party labels, but the party loyalty factor is extremely important in presidential elections as well. Thus, in 2004, 89 percent of Democratic identifiers voted for Kerry, and 93 percent of Republican identifiers voted for Bush.
Candidates
Presidential election outcomes have not simply reflected the party balance in the country; if that were true, Democrats would have won most presidential elections during the post-World War II period. Voters also pay a lot of attention to their perceptions of the personal characteristics of candidates. They vote heavily for candidates who have experience, appear strong and decisive, and convey personal warmth. The Republican candidate in 1952 and 1956, Dwight D. Eisenhower, had a tremendous advantage in these respects over his Democratic opponent, Adlai Stevenson;5 so did Ronald Reagan over Walter Mondale in 1984, and George H. W. Bush over Michael Dukakis in 1988. Only in 1964 did the Democratic candidate (Lyndon Johnson) appeal to voters substantially more than the Republican candidate (Barry Goldwater). In elections between 1952 and 1972, the contrast between Republican and Democratic candidates typically gained the Republicans 4 or 5 percentage points-just enough to overcome the Democrats' advantage in what political scientists call the normal vote: how votes would be cast if only party identification determined voters' choices for president.
Issues
Voters also pay attention to issues. Sometimes this means choosing between different policy proposals for the future (as in the responsible party voting model), such as Reagan's 1980 promises to cut back federal government activity or Clinton's 1992 pledges of jobs and a middle-class tax cut. More often, however, issue voting has meant retrospective voting (the electoral reward and punishment model), making judgments about the past, especially on major questions about the state of the economy and war or peace. The voters tend to reward the incumbent party for what they see as good times and to punish it for what they see as bad times. In especially bad economic times, for example, Americans tend to vote the incumbent party out of office, as they did the Republicans during the Great Depression in 1932. In 1992, the electorate punished Republican George H. W. Bush for the poor state of the economy and, in 1996, it rewarded Bill Clinton for being president during good economic times. This did not happen in 2004, however, when protection against terrorism seemed to trump people's concerns about slow job growth in the economy. Several scholars and pollsters believe that cultural issues-such as gay and lesbian rights, abortion, civil rights and affirmative action, law and order, and the like-may have become more important than retrospective judgments about economic issues in determining voter choices. This may help explain why increasing numbers of affluent and educated Americans are voting Democratic, while lower income, less educated, and church-going whites are increasingly casting their ballots for Republicans.6
Foreign policy can be important as well, especially when war and peace are at issue. Bitter disillusionment over the Korean War hurt the Democrats in 1952, just as the Vietnam War cost them in 1968, and unhappiness about American hostages in Iran and the Soviet intervention in Afghanistan hurt Jimmy Carter in 1980. During nearly all of the past half-century, in fact, Republican candidates have been seen as better at providing foreign policy strength and at keeping us out of war. In most elections, however, foreign policy concerns take a back seat to domestic ones for most voters.
1Efforts to sort out their relative contributions include Benjamin I. Page and Calvin Jones, "Reciprocal Effects of Policy Preferences, Party Loyalties, and the Vote," American Political Science Review, 73 (1979), pp. 1071-1089; Gregory B. Markus and Philip E. Converse, "A Dynamic Simultaneous Equation Model of Public Choice," American Political Science Review, 73 (1979), pp. 1066-1070.
2Carlos Elordi, "Ideology: Assessing Its Impact on Political Choices," Public Perspective (March/April 2000), pp. 34-35.
3Erikson and Tedin, American Public Opinion, pp. 255-259.
4Larry M. Bartels, "Partisanship and Voting Behavior, 1952-1996," American Journal of Political Science, 44 (January 2000), pp. 35-50; and D. Sunshine Hillygus, and Simon Jackman, "Voter Decision Making in Election 2000: Campaign Effects, Partisan Activation, and the Clinton Legacy," American Journal of Political Science (2003), vol. 47, pp. 583-596.
5Donald E. Stokes, "Some Dynamic Elements of Contests for the Presidency," American Political Science Review, 60 (1966), pp. 19-28.
6Thomas B. Edsall, "The Shifting Sands of America's Political Parties," The Washington Post National Edition (April 9-15, 2001), p. 11.
