Late-deciding voters in presidential elections

This article examines the behavior of voters who decide on their presidential candidate preferences within the final two weeks of the presidential campaigns. Relying on data from American National Election Studies (ANES) for elections, between 1972 and 1988, we demonstrate that late deciders are different from other voters in a variety of respects. They are less involved politically and far less predictable in their behavior than other voters. More significantly, we find that the candidate preferences of late deciders are not determined by the conventional political forces that motivate other voters. Rather, the choices that late deciders make appear to be nearly random in character. The implications of these findings for the analysis of electorates are discussed.     

Turnout and presidential coattails in congressional elections

Estimating the impact of turnout on House election results is problematic because of endogeneity and omitted variable bias. The following study proposes an instrumental approach to correct for these problems by using a series of fixed effects two-stage least squares panel-data regression models covering three congressional apportionment cycles (1972–1980; 1982–1990; 1992–2000). The analysis tests whether voter participation decreases the House incumbent’s electoral support, regardless of the level of competition in the district. The study also aims to determine if an increase in participation benefits Democratic candidates and whether this effect is constant across apportionment cycles. The results show that the influence of turnout on incumbency vote share is conditional on the level of presidential support in the district. This finding is explained by the surge and decline thesis of Campbell (1960).     

Ideological representation and competitive congressional elections

Competitive elections do not produce representation. We demonstrate that elections in which incumbents win by landslides yield Representatives who are ideologically closer to their voters than elections with narrow margins. Furthermore, we demonstrate that ideological proximity to one's Representative creates feelings of trust and efficacy, but that competitive elections do not. In fact, since competitive elections produce ideological distance between voters and their Representatives, and that distance produces dissatisfaction, competitive elections indirectly reduce voters' feelings of trust and efficacy. Thus, competitive elections are paradoxically harmful to representation.     

Competitive primaries and party division in congressional elections

We examine the effects of competitiveness and divisiveness in the 2010 congressional election, as the emergence of the Tea Party resulted in many competitive Republican primaries and highlighted significant divisions within the party. We find that competitive Republican primaries and competitive Democratic primaries increased turnout in the general election. The presence of divisiveness in the Republican Party has no discernible effect on turnout on its own, but actually increases the vote share captured by the Republican Party in November. Additionally, we find that while either a competitive primary or the presence of a Tea Party candidate was advantageous to the Republican Party, the presence of both in the same election was no more beneficial than if there was either a competitive primary or a division in the party.     

Candidate ideology and electoral success in congressional elections

In examining the factors that contribute to electoral success in congressional elections, legislative scholars often consider the actions of elected representatives; however, other research suggests that one must consider what challengers are (or are not) doing as well. For instance, inexperience and poor funding can significantly inhibit challenger success. We expand this list of potential shortcomings by arguing that ideological congruence with a constituency may be another factor in explaining challenger defeat. Using ideology measures derived from campaign contributions, we find that unsuccessful challengers in the U.S. House are generally more extreme than those who win, but ideological extremity is not a disadvantage to those seeking to represent an extreme constituency. More importantly, our existing political institutions may actually serve to mitigate the already high levels of partisan polarization in Congress.     

A test of the revealed-preference phenomenon in congressional elections

This study tests for the existence of a revealed-preference phenomenon in recent congressional elections. A revealed-preference analysis predicts that if candidate one defeats incumbent candidate two, then candidate one has been revealed preferred to candidate two, and can expect to secure higher pluralities in subsequent reelection contests than candidate two would have expected. I collect data on incumbents in 1946–1980 congressional elections to examine this prediction. The data provide qualified support for the presence of a revealed-preference phenomenon. The essay closes with a discussion of the possible connection between the revealed-preference phenomenon and the recent electoral experience of congressional incumbents.     

Explaining the unpopularity of public funding for congressional elections

This article uses data from the 2008 Cooperative Congressional Election Study to explain weak support for public financing of congressional campaigns. Previous studies lack theory to explain variation in support and use a flawed measure of the dependent variable. We argue that low support reflects a failure resulting from a collective action dilemma. Citizens desire a campaign finance system that weans politicians from private donors, but are unwilling to pay a small amount in taxes to support public financing. In contrast to conventional wisdom, we show that support for public financing is highest among those perceived to benefit the most from the current system. Our results suggest that most Americans would rather not pay for politics, and that reform proposals must avoid incurring transparent costs on individual citizens to pay for reform.     

Incumbency advantage in multi-member districts: Evidence from congressional elections in Chile

Do parties enjoy an advantage to incumbency in multi-member districts? In this paper we answer this question by adapting a regression-discontinuity design to multi-member districts in congressional elections in Chile. The electoral system in place generates discontinuities in the number of elected representatives from each coalition at the 1/3 and 2/3 thresholds of the two-party vote share. Regression-discontinuity estimates indicate that, by holding two seats as opposed to holding only one, the left-leaning coalition obtains an extra 4.5% vote share in the next election and increases by 28 percentage points the probability of electing two candidates again. These results are in line with those obtained in previous studies in the U.S. but contrast with results obtained in developing countries which find a negative advantage to incumbency.     

Voters and roll call voting: The effect on congressional elections

The subject of this research is whether ideological preferences play a major role in explaining voters' refusal to reelect some members of the U.S. House of Representatives. If ideological control is important, one would expect to find a large difference between the voting record of a rejected incumbent and his or her replacement. In distinction, whenever voters must replace a congressman or congresswoman because that person had died in office or chose to run for higher office, the hypothesis of ideological tracking implies that the newly elected member of Congress will resemble his or her predecessor. The data confirm these hypotheses and show, as well, that ideological control exists within parties and not only between them; that the degree of voters' ideological control is as great over senior congressmen and congresswomen as over junior ones; and that voters' concern about ideology has increased over the last two decades.     

Spatial elections with endorsements and uninformed voters: Some laboratory experiments

This paper presents a laboratory experiment designed to examine voting behavior, within a spatial election framework, when subjects are only endowed with endorsement information. In the experiment, subjects are assigned ideal points on a single-issue dimension. With the aid of an endorsement, their task is to vote for the candidate whose position is closer to them. The results show that under these conditions, subjects are in fact playing a lottery, where the odds of being correct is a function of spatial location. The study points out that researchers should be aware of an “information bias” in models that use this particular information source.