Browsing by Author "Johnston, Christopher"

A commonly accepted model of public attitudes toward election rules assumes that citizens follow the cues of their preferred party’s elites and support rules that would benefit that party in elections. However, a separate literature on procedural fairness suggests that the public places a high priority on the fairness of democratic institutions. This dissertation tests which model predominates in the public’s decisions on election rules across a variety of policies and political contexts. It finds that most citizens prefer fair electoral institutions at the expense of partisan interest when that choice is made explicit, and a minority of committed partisans are driven by partisanship. While most partisans are unwilling to manipulate election rules to benefit their own party, they react negatively to attempts at manipulation by the other party. Citizens are susceptible to influence from elite messaging on election law issues but are resistant to attempts to influence their core democratic principles.

Researchers have dedicated substantial effort to investigating important non-material motivations for people to get involved in politics, such as duty, emotions, and identities. Less attention, however, has been paid to the expectations people develop for what governments and politicians will deliver. This dissertation is about what people think elections will do for them, where those expectations come from, and their political consequences.

The first substantive chapter explores the policy changes people expect from elections, and how those expectations influence the decision to vote. There I study voters' beliefs about what candidates would actually do if given political power. I first find that public respondents likely underestimate the impediments that the separation of power poses to policy change. Just before the 2020 election, these general population respondents expected much more legislation than political scientists completing an identical survey. Second, among the general public, there was a 16 percentage point difference between voters and non-voters in expectations for policy change resulting from the election. Most importantly, these high expectations predicted validated voter turnout better than education, identifying as a Democrat or as a Republican (as well as partisan strength and ideology), having voted in 2016, and political interest. These results support explanations for the decision to turnout which center on the benefits, whether individual or social, that people believe their preferred candidate will deliver.

Next, Chapter 3 argues that a psychological bias called focalism contributes to an overestimation of the differences between political candidates, which in turn increases participation and polarization. Focalism causes people to confuse the allocation of attention to things with the importance of those things. Because attention to politics typically centers on conflict, the result is an exaggeration of differences across the partisan divide. I test this intuition using an experimental design that provides all respondents with all of the information they need to estimate how much Joe Biden and Donald Trump objectively disagreed on policy positions just before the 2020 election. I find that shifting attention – towards either those positions the candidates agreed or disagreed with each other on – influences beliefs about the differences between candidates. The effect exceeds that of identifying as a Democrat or as a Republican. Beyond those perceptions, focalism increases turnout intentions, perceptions of election importance, negative feelings towards the out-candidate, and affective polarization.

Finally, Chapter 4 attempts to moderate people's expectations using a series of real-world experiments. That final essay asks: would learning about coverage biases as people learn about the news soften people's beliefs about how different Democrats and Republicans are? To test this question, I use two experiments, one of which recruited participants to consume news covering the full population of partisan and non-partisan sources and the second of which randomized coverage among a sample predisposed to change their minds. I find that giving people the tools to understand media bias does give people the opportunity to choose to consume centrist news. Exploring app-use data, I show that people who explicitly choose to engage with stories favored by these moderate sources stories while avoiding stories favored by partisan sources feel less polarized.