Gerrymandering on Georgia's Mind: The Effects of Redistricting on Vote Choice in the 2006 Midterm Election*

Objective. We make use of individual‐level survey data from the 2006 midterm election in order to determine the degree to which redistricting affected the vote choice of whites residing in Georgia Congressional Districts 8 and 12. Methods. A multivariate probit model was used to assess the probability of voting for the GOP House candidate among voters represented by the same incumbent before and after redistricting versus voters who had been newly drawn into one of these districts. Results. Despite a national tide that favored the Democratic Party in the 2006 elections, redrawn whites were more likely to vote for the Republican challengers in the districts surveyed. Conclusions. Our findings indicate that redistricting can be used to dampen the incumbency advantage. In addition, the findings of this research also speak to the continuing Republican realignment of white voters in the Deep South and to the recognition that the effects of redistricting are dependent on political context.     

Structured Partisan Competition: A Vote Shares Model of Party Alignments and Realignments*

Objective. This study develops and tests an aggregate “vote shares” model of party alignments and realignments, building a theoretical framework around “structured political composition” (Rabinowitz, Gurian, and Macdonald, 1984, p. 6). The vote shares model conceptualizes party alignments as latent class constructs, or factors, and changes in these latent class constructs as party realignments. Methods. Confirmatory factor analysis (CFA) model with bias corrected bootstrap estimates and standard errors is used where the cases are counties and the variables are aggregate election outcomes. The data come from Kansas from 1900 to 2010. Results. Comparing the findings of the vote‐shares approach to those of Nardulli (2005) and Sundquist (1983) suggests that the vote shares model provides a greater depth of understanding and a more accurate portrait of the timing of realignments and the partisan bias of new alignments in Kansas. The vote shares model also overcomes the major problem of false positives associated with landslide elections, a problem that plagued most previous aggregate voting models. Conclusion. The vote‐shares model of party alignments and realignments provides a viable alternative for analyzing historic and current election returns where the votes are aggregated by a geographically defined government jurisdiction (parish, county, city, or district).     

Statewide Trial‐Heat Polls and the 2000 Presidential Election: A Forecast Model*

Objective. We provide an examination and update of a presidential election forecasting model that we have previously developed to predict state‐level presidential election outcomes. Method. Our model consists of September statewide trial‐heat polls in 1992, 1996, and 2000 along with a prior vote variable. We use this model to generate predictions for both state‐level and national‐level outcomes. Results. Although our model generated reasonably accurate point estimates of state‐level outcomes in 2000, it still incorrectly predicted a Gore victory in 2000. Conclusions. We discuss possible explanations for the 2000 misprediction and present updated coefficients to be used to generate a forecast for the 2004 election.     

The rising influence of evangelicalism in American political behavior, 1980–2004

Connections between evangelicalism and political behavior are explored using National Election Study (NES) data from 1980 to 2004. Our findings suggest that although white evangelical support for Republican presidential candidates has increased, the independent effect of evangelicalism on voting has weakened. We argue that this reflects an increasing convergence between evangelicalism and Republican Party identification. Beginning in the 1990s, as more and more evangelicals identified as Republicans, the independent effect of their religious commitment on voting weakened. We also find strengthening correlations between evangelicalism and policy preferences, especially on non-social issues, and enhanced Republican mobilization of evangelical voters in recent campaigns. We conclude with the suggestion that changes in evangelical political behavior over a 24-year period support the notion of an electoral realignment in American politics.     

Race–Based Redistricting, Core Constituencies, and Legislative Responsiveness to Constituency Change

Objective. Fenno (1978), Wright (1989), and other scholars suggest that legislators will be particularly responsive to various subgroups in their constituency, i.e., what might be termed the "core constituency." We explore the degree to which Republican and Democrat House incumbents respond differently to changes in the racial composition of their districts brought about by redistricting. We speculate that (1) Democrat House incumbents will be more responsive in their roll–call behavior to changes in African–American racial composition, since African–American voters are typically a major component of the Democrat core constituency, while (2) Republicans will be less responsive, since African–American voters are typically not part of the Republican coalition. Methods. We utilize data on the roll–call behavior, member characteristics, and constituency characteristics of House members who served during both the 102nd (1991–1992) and 103rd (1993–1994) Congresses. We model roll–call liberalism in 1993 as a function of levels of and changes in district racial composition, along with control variables. To capture the different effects of racial core constituencies, we estimate our models separately for Democratic and Republican House members. Results. Our findings provide strong support for our hypothesis: Democrat incumbents respond strongly both to levels of and changes in the African–American population in their districts, while Republican incumbents respond only modestly to changes in African–American population brought on by redistricting and negatively to African–American population levels. Conclusions. Democrats and Republicans appear to respond differently to constituent groups, depending on whether the groups are part of their core constituency. Our findings suggest the importance of considering core constituencies in studying roll–call behavior and legislative responsiveness.     

The influence of socio-economic characteristics on the political attitudes of American Indians

We examine the political attitudes of American Indians in eastern Oklahoma where they make up almost 20% of the population. We argue that American Indians in the region play the same role that other minorities in the southern states—notably blacks and Hispanics—do in resisting the region's realignment from the Democratic to the Republican Party. American Indians in the region are populists in that they are economically liberal and religiously conservative. The results of our analysis suggest that the socio-economic characteristics of American Indians play a greater role in explaining American Indians’ political attitudes than their ethnic identification.     

Institutional and District‐Level Sources of Competition in State Legislative Elections*

Objectives. What are the sources of partisan competition in state legislative elections? Specifically, what impact do institutional features have relative to district‐level conditions on competition between party nominees? Methods. Using data from 30 states in 1994 and 1996, a range of factors are examined to determine their influence on both the likelihood that a race is contested and the degree of competition that results. Results. Multiple regression analyses indicate that a district's characteristics, measured as social and partisan diversity, have a strong and durable influence on elections. Contested elections are more likely and competition levels are higher in districts with heterogeneous populations. Institutional characteristics such as legislative professionalism also have a large influence, although the direction of their impact varies by stage of the electoral process examined. For example, professionalism increases the likelihood that an election is contested, but leads to a lower level of competition between opposing candidates. Conclusions. District‐level conditions have a large influence on competition; however, incentives created by institutional features are also critical for understanding the competitiveness of state legislative elections.     

Candidates and Competition: Variability in Ideological Voting in U.S. Senate Elections*

Objectives. In choosing candidates to support in congressional elections, voters consider both policy and nonpolicy factors. However, the relative importance of incumbency or presidential approval versus candidates' ideological platforms likely varies across elections. Specifically, stiffer electoral competition should encourage ideology‐based voting because candidate information is more plentiful. In contrast, incumbents' ability to garner votes simply by virtue of already holding office should depress proximity voting in elections with incumbents. Methods. Using data from the 1988–1992 Pooled Senate Election Study, I estimate logistic regression models of individual vote choice. Results. I find that open‐seat elections do promote the use of candidate ideological proximity in the voting calculus but that the effects of election competitiveness are less clear. Conclusions. The findings have important implications for normative democratic theory, for our constitutional framework, and for elite behavior and aggregate‐level electoral outcomes.     

Presidential voting in the west

Voters in the western United States are becoming more Republican than the rest of the country in their presidential choices. The Solid West has replaced the Solid South as a reality of presidential campaigns. This is in spite of a long term national trend away from strong party loyalties. This article examines the ten states of the Interior West at the county level during the eight presidential elections from 1956 through 1984. All parts of the region did not simultaneously embrace the Republican Party from the onset of the New Western Normal Vote episode. Those areas that lagged were characterized by sizable employment in forestry and mining, large Hispanic or Native American populations, or were among the region's few large urban concentrations. Some of these areas still remain outside the Republican fold in presidential elections. The overall magnitude of change in Republicanism in the 549 counties was not uniform, but rather impacted some areas such as Nevada and Idaho to a much larger degree than others such as Montana and the Dakotas. This article generally confirms Archer and Taylor's assertion that the Western Periphery constitutes a new Republican heartland. But some Indian and Hispanic populations continue to vote for Democratic presidential candidates. Rural areas with strong organized labor such as the mining and forestry areas of Montana, and big cities like Denver are only weakly Republican. Hence, the conversion to strong and unwaivering Republicanism is not total and may never occur.     

How Do Voters Form Expectations about the Parties' Chances of Winning the Election?*

Objective. This article examines the factors that form voters' perceptions of the parties' chances of winning at both the national and the local levels. Method. We make use of the 1988 Canadian Election Study and we employ a HLM model to estimate the effect of individual‐level and contextual‐level variables. Results. It is shown that voters' expectations are affected by a combination of “objective” contextual information and personal preferences (projection effects). Conclusion. The basic contextual information that is utilized to ascertain local chances is the outcome of the previous election in the local constituency, whereas polls are crucial in the case of perceived national chances. We also find that the most politically aware are more strongly influenced by “objective” indicators.     

Voting for Catholic Candidates: The Evolution of a Stereotype*

Objectives. This article looks at the effects of candidate Catholicism on individual voters, turning the traditional inquiry into voters' religion on its head. Specifically, it hypothesizes that individuals stereotype Catholic candidates based on the voting behavior of Catholics in general, and that these stereotypes help voters make a decision in elections with Catholic candidates. Methods. Using data from the American National Election Studies (ANES), this article argues that citizens hold stereotypes of Catholics based on actual Catholic political behavior—solidly Democratic from the 1950s to the 1970s, but trending Republican starting in the 1980s. It also tests these stereotypes with Gallup data on hypothetical Catholic candidates and analyzes, through election‐day exit polls, the impact on voters of the Catholic conversion of a sitting U.S. Senator. Results. The data analyses strongly support the article's hypothesis, demonstrating that partisan attitudes toward hypothetical Catholic candidates shifted direction as the Catholic electorate shifted. In addition, Senator Sam Brownback's (R‐KS) conversion to Catholicism increased partisan polarization in his support—indicating a conservative shift in perceptions. Conclusions. Candidate religion plays a role in elections. Specifically, voters stereotype candidates based on candidate religion and use this stereotyped information to help them make an electoral decision. This article demonstrates this effect for Catholicism, but other religions should evidence similar impacts.     

Red states,blue states: How well do the recent national election labels capture state political and policy differences?

The alleged polarization between the so-called red (Republican) and blue (Democratic) states during the presidential elections has been examined using only voter surveys. Focusing on the recent thirteen national elections from 1964 to 2012, we examine social, political, institutional, and policy indicators of the 50 American states to (1) gauge the extent to which national election results reflect significant policy and political differences between the red and blue states and (2) to assess the explanatory power of the dichotomous red–blue label relative to a continuous variable of “redness” or “blueness” by the percentage of votes received. We find substantial political and some moderate social differences between red and blue states but fewer institutional and policy differences than one would expect if there were actually deep divisions between the states. We find that the red–blue state distinction performs well when compared to the explanatory power of the more precise redness or blueness of a state.     

Seeing Red (and Blue): Effects of Electoral College Depictions on Political Group Perception

Colored maps depicting electoral results may exacerbate perceptions of polarization, rather than merely reflecting them. Participants viewed maps of state-by-state Presidential election results that were either Electoral (red/Republican or blue/Democrat) or Proportional (purples that proportionally reflected each group's support). Half of the maps also displayed state-level numeric electoral results. Participants viewing Electoral maps perceived the nation as more politically divided, stereotyped the political beliefs of residents of various states more, and saw people holding views in the political minority as less agentic and less likely to vote. These differences occurred even in the presence of numeric data. Implications of these findings for intergroup perception in several domains are discussed, including the impact of electoral depictions on political campaigns and elections .     

Ideological polarization in a decentralized party system: Explaining interstate differences

During the past decade scholars have noted growing ideological polarization between Democratic and Republican Party elites in the United States. This trend has occurred in a party system traditionally characterized as decentralized. This paper examines whether the trend towards partisan polarization noted by scholars at the national level has affected state party systems in similar ways. Are some state party systems more polarized than others? The paper uses a classification scheme of state party systems developed by David Mayhew to try to explain interstate differences in partisan polarization. The paper concludes that states with political environments that supported pragmatic and non-ideological traditional party organizations are less polarized in the modern era than states without such environments.     

Why do American Indians vote Democratic?

We examine the relationship between a relatively large population of American Indians in eastern Oklahoma and the slow pace of that region's partisan realignment. With a 2012 exit poll, we empirically examine whether American Indians’ cultural ties and their socio-economic characteristics affected their vote choices. Because of their greater poverty, American Indians are more economically progressive and vote more Democratic than their white counterparts.     

Third‐Party Candidates and the 2000 Presidential Election*

Objective. This article examines how third‐party candidates influenced the 2000 presidential election. Methods. Two surveys provide information on a hypothetical election between only George Bush and Al Gore. The determinants of voting behavior in this election are then used to estimate how votes cast for third‐party candidates would have been partitioned between Bush, Gore, and abstentions had the other candidates dropped out of the race. Results. The estimates suggest that the Ralph Nader votes would have gone primarily to Gore (giving him the win in Florida) while Bush would have received more of the Pat Buchanan votes. The results also show that Nader's presence in the race gave Gore an incentive to position himself farther to the left ideologically. Bush's ideological position was not affected by Buchanan's participation. Conclusion. The third‐party candidates' participation was a critical factor in George Bush's Electoral College victory over Al Gore.     

Ideological Realignment in the Contemporary U.S. Electorate Revisited*

Objective. In this study, we revisit the ideological‐realignment theory proposed by Abramowitz and Saunders (1998) by assessing the varying impact a person's ideology had on his or her partisan identification for individuals in different regions and between men and women. Method. Through an examination of the NES cumulative data file and the 1992–1994 NES Panel Study, we present a series of tabular findings, an OLS regression model, and partial correlation coefficients to assess these changing subgroup dynamics. Results. Between 1978 and 1994, the impact of a person's ideology on his or her party identification grew in magnitude, while the impact of respondents' parental party identification on their own party identification waned. However, these changes were not felt uniformly throughout the electorate, with Southern white males and Southern white females exhibiting the greatest changes. In the case of Southern white males, racial issues now appear to be related to their party identification. Conclusion. We find support for the ideological‐realignment theory, but find evidence that such changes occurred rather unevenly throughout the electorate, suggesting that different dynamics of realignment may be at work simultaneously.