Imagine a situation where we could alter the outcome of an election simply by changing the number of offices on the ballot, the availability of candidates' political party affiliations, and the amount of dialogue between the candidates. We would like to think that voters would reach the same decision regardless of the campaign environment and the outcome of the election would not change, but this may not be the case. While most scholars have discarded the notion of a perfectly rational voter, one who is fully informed about the candidates and who compares the costs and benefits associated with each candidate, scholars still hold hope that voters can overcome their short-comings through decision strategies, such as heuristics (party labels) or confirming their notions about candidates enough to satisfy some minimal criteria. However, all elections are not equal. For exmaple, the November 2008 ballot in Boulder, Colorado included over 40 items on the ballot for which voters had to make a decision. In contrast, voters in Brunswick, Maine only had to deal with 7 items on the ballot. Likewise, some states hold non-partisan elections for positions such as the state supreme court, meaning that voters cannot rely on the candidates' party affiliations. In other elections, there is a lack of dialogue between candidates where a candidate will avoid issues that he or she is not comfortable with, making it more difficult for voters to compare the candidates' policy preferences. Yet, we still do not know what ramifications this has for decisions voters make.
To fill this gap, the investigators examine how changes in the campaign environment affect how voters search for information. For the purposes of this project, three specific changes are explored: the number of offices on the ballot, the availability of candidates' political affiliations, and the amount of dialogue between candidates.
This project examines three different campaign environments in which voters learn about candidates and make decisions: long versus short ballots, partisan versus non-partisan campaigns, and campaigns with and without dialogue between the candidates. To test these findings I developed and programmed a dynamic information board that is used to dynamically present campaign information to subjects in a way that mimics the characteristics of a real campaign. The overarching finding is the way in which we design our electoral institutions has implications for how voter acquire information and make decisions. While voters faced with longer ballots appear to gather more information and compare more information, they appear to spend significantly less time actually learning about the candidates. On average, the length of the ballot does not appear to affect whom a person will vote for. However, when broken down by demographic, there are some differences. For example, there are differences between the probability of a women voting for a Republican under the shorter ballot and a women voting for a Republican under a longer ballot, even when the subject’s party affiliation is taken into account. Likewise there are differences in how voters search for information in partisan and nonpartisan campaigns. First, it is important to note that there are important differences in who subjects vote for in the two conditions. In the non-partisan election, votes were split almost evenly between the two candidates. In the partisan condition, there were significant differences in who subjects voted for. Moreover, there was a different winner for the two elections, even though the only difference between the two elections was the availability of the candidates’ party labels. One important finding is that voters compare more information between candidates in non-partisan elections. Other aspects of how voters gather information are driven by their strength of partisanship. Strong partisans gather less information in non-partisan campaigns and weak partisans and independents search for less information about the in-party candidate in partisan elections. Endorsements, an important piece of information in the absence of partisan cues, are used by the most knowledgeable, especially in non-partisan campaigns. Finally, there were also differences in how subjects gather information in campaigns in which there was dialogue versus no dialogue. In a campaign in which there is dialogue, candidates address the same set of issues. In a campaign without dialogue, candidate address a different set of issues. In this study, subjects searched for and compared more information in the campaign with dialogue between the candidates. Likewise their search appears to be more systematic, signaling that they were better able to handle the environment when dialogue between the candidates was present. Like in the experiment that examined the length of the ballot, the campaign context on average did not affect the outcome of the election. The results show that the information environment in which voters makes decisions affects how they gather information and make decisions. These findings highlight a discrepancy between our democratic ideals of the informed and rational voter and how voters actually behave. However, based on this project, this may not entirely be due to the voters and may have to do with the design of our electoral institutions. The results from this project indicate the amount of information that voters compare between candidates, the amount of information that voters search for, and their ability to sift through information in an organized fashion is driven by the campaign environment. This also means that the campaign environment drives how voters learn about candidates, which potentially affects who they will vote for.
