The electoral choice context and support for democratic norms

• 成果类型: Article
• 署名作者: Huang, Kang; Lovett, Mitchell; Helmke, Gretchen
• 署名单位: University of Rochester; University of Rochester
• 刊物名称: PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
• ISSN/ISSBN: 0027-8424; 1091-6490
• DOI: 10.1073/pnas.2521380123
• 发表日期: 2026-05-12
• 页码: e2521380123
• 关键词: democratic backsliding; ELECTIONS; voting; DEMOCRACY; polarization

摘要

In this paper, we explore whether different choice environments affect voters' willingness to put democracy over party and policy. Using a series of candidate choice experiments in which we manipulate not only candidate attributes, but also the number of races on the ballot, we find that when voters are able to vote in multiple races on a ballot they are substantially more likely to punish antidemocratic candidates than when they are only given the opportunity to cast one vote. We refer to this effect as democratic balancing. Substantively, our experiments show that the magnitude of the total punishment effect increases from 11% in single-race ballot settings to 17.9% in multirace ballot settings. Such effects remain largely consistent across different candidate platforms, separate waves of the survey, different types of norm violations, and are robust to different electoral race combinations and different information environments. Observational data from the 2022 U.S. midterm elections further corroborates our experimental findings. For researchers, our results underscore the importance of taking into account the institutional context in which voters make decisions. For citizens concerned with democratic backsliding, our results offer some reassurance: Moving to a more realistic electoral context in which multiple races appear on the ballot tempers the willingness of respondents to trade off democracy for their preferred policies and party.