Search

Following the 2004 election, exit polls and pundits suggested that moral values were a determining factor in voting for president. Subsequent analyses have suggested that this initial interpretation may have been overstated and that moral issues-such as gay marriage and abortion in particular-were less consequential for presidential voting than the economy, terrorism, and the Iraq war. Using data from the 2004 NES, I confirm these findings that moral issues were generally not consequential but I show that moral values not only mattered, they were among the most important predictors of presidential voting in 2004 save party identification. The results of this study confirm the initial interpretation that the 2004 presidential election hinged, in part, on moral values voting. ..PAT.-Unpublished Manuscript [ABSTRACT FROM AUTHOR]

Given the focus of media and scholarly attention on the increasingly ideological nature of American politics, it is somewhat surprising that so little is known about the structure and meaning of political ideology in the electorate. In this paper, we identify a number of methodological and conceptual limitations to existing measures of political ideology, and then offer an improved measure of ideology, an explicitly issue-based latent measure estimated using an item response model (IRT). We compare our latent measure of ideology with ideological self-placement using the 2000 National Election Study. We find that our latent issue-based measure of ideology is a better predictor of presidential vote choice than a traditional self-placement measure and that both economic and social ideological dimensions were critical to voter decision making in the 2000 election. ..PAT.-Conference Proceeding [ABSTRACT FROM AUTHOR]

Young Americans' non-engagement in politics has been raised as a critical issue for the last 10 years. People claim that young people today are more ignorant, apathetic, and cynical about politics than at any other time in the past. A recent report from the Washington Post quoted a young voter from Arkansas as saying, "The young people don't feel connected to the candidates, because they don't campaign to them. And the candidates don't feel connected to the young people, because they think they don't vote." Using the National Election Study (NES) data, this study seeks to discover the relationships between various factors that influence voting behavior and voter turnout among young people aged 18 to 35 in the 2000 presidential election. Using the hierarchical regression, this study tried to predict how much of variance in voter turnout among young Americans was explained by certain factors. This study further develops a profile of the young people who are likely to turn out to vote on an election day using the discriminant analysis. [ABSTRACT FROM AUTHOR]

Every four years presidential candidates devote months and hundreds of millions of dollars campaigning for the highest office in the land. Some question whether these presidential campaigns actually make a difference. In 2000, the Bush and Gore campaigns provided a golden opportunity to empirically assess campaign effects because both candidates focused their campaigns on a set of "battleground" states. 2000 NES data indicates that those who lived in "battleground" states had significantly more issue knowledge and issue salience than citizens from other states. Because this kind of targeted campaigning is encouraged by the Electoral College, with its winner-take-all rules, the data show that the Electoral College has a demonstrably negative effect on representative democracy. [ABSTRACT FROM AUTHOR]

Extant research has typically relied on the use of political knowledge items orinterviewer evaluations to measure the political sophistication of survey respondents.In this paper, I further develop Luskin's approach to measuring sophistication usingknowledge items, items based on recognition and understanding of ideological terms,and items based on correct candidate and party placements on issue scales. I alsopresent the advantages of developing a measure based on an item-response theorymodel rather than the typical approach of using an additive index. ..PAT.-Unpublished Manuscript [ABSTRACT FROM AUTHOR]