Doctoral Dissertation Research in Political Science: Finding their Way: Rational Voters in an Uncertain World

University Of California-Santa Barbara, Santa Barbara CA

Investigators

Abstract

Primary Investigators: Garrett Glasgow and Stuart Kasdin Title: Doctoral Dissertation Research: Choosing the Right Tool for the Job: Voter Decisionmaking and Uncertainty in Non-Partisan Elections 0718716 Abstract In representative democracy people do not vote directly on policy concerns. Voters must select a representative to make those choices for them. This paper starts with the premise that voters are instrumentally rational; they select the candidate who, if elected, is most likely to accomplish the goals they prefer. Thus they must predict what each candidate would choose to do in office and how well they would do it. However, the public has only imperfect knowledge of their representative's preferences and actions, as well as uncertainty about the types of issues and circumstances that are likely to arise in the future. Thus, to predict how candidates will behave in office, given different possible future challenges or issues environments, voters rely on different estimators, such as the candidate's political platforms, prior performance, as well as personal background, experiences, and group loyalties or identity. These measures offer varying degrees of accuracy in estimating candidates' likely performance in office, if elected, depending upon the information available for each of the metrics. Less information makes a metric less efficient; hence less accuracy. In addition, voters may change their preferences for the type of representative, based on political circumstances. This change depends upon the likely issue agenda, in which voters may vary their preferences as the expected political economic environment changes. I test this theory experimentally in an internet survey using a random selection of students. Each student will participate in a series of mock elections to see how they alter their vote as the candidate combinations are systematically changed (candidates' platforms, ethnicity, and reported competence). Furthermore, different elections include an additional altered element, such as varying the issue environment and the issue content of the candidates' platforms. Thus, since voters are concerned with outcomes, they are expected to change their vote depending upon the context, even for the same set of candidates. Some of the implications of this decision process investigated in this paper include: Do voters rely on the same basic framework for decisionmaking in which they evaluate each candidate across several dimensions, considering dimension uncertainty and salience? When are voters more likely to cast aside an ineffective incumbent? What factors condition political accountability, making it less likely that voters do not reject the poorly performing incumbent? Are voters acting in choosing a candidate from the same ethnic or racial background as themselves? Why are issues in campaigns often vague and symbolic? Does a politician adding new, popular issues to his platform during a campaign increase the votes proportionally to how strongly people feel about an issue? Does a candidate receive a policy mandate from an election victory?