Beliefs, Biases and Ballots: A Bayesian Exploration of Mismatch Between Community Preference and Voting Behavior

Voting is a mechanism to aggregate individual preferences into a collective decision. How individuals vote is informed both by their own preferences and by their beliefs about the preferences of others. However, evidence from across climate policy, civil rights, and electoral politics suggest that people consistently misjudge community preferences, a phenomenon called pluralistic ignorance. Pluralistic ignorance can lead individuals to act against their preferences by modifying their behavior to either align with a presumed majority or strategically avoid outcomes that bring negative utility. How do systematic biases in people’s beliefs about the preferences of others affect voting outcomes? And how does the structure of the information environment influence this process? To address these questions, we will develop a Bayesian framework for modeling opinion spread and voting decisions in a social network. We will use this framework to study mechanisms that give rise to discrepancies between true preferences and voting outcomes, with a focus on the role of individual incentives, prior assumptions, and network structure.