Using Elicited Choice Probabilities in Hypothetical Elections to Study Decisions to Vote

Area of Science:

• Political Science
• Survey Methodology
• Behavioral Economics

Background:

• Traditional voting studies rely on actual election data, limiting the scope of analysis.
• Survey research offers a flexible method to explore diverse voting scenarios.

Purpose of the Study:

• To demonstrate the feasibility and utility of using hypothetical election scenarios in survey research.
• To enrich data for studying the determinants of voting decisions.

Main Methods:

• Presented over 4,000 participants in the American Life Panel (ALP) with hypothetical presidential election scenarios.
• Scenarios varied candidate preference, election closeness, and time cost of voting.
• Assessed respondents' self-reported probability of voting and analyzed data using direct variation studies and a random utility model.

Main Results:

• Voting time and election closeness were significant predictors of voting probability.
• Candidate preference did not notably influence the decision to vote in hypothetical scenarios.
• Model findings were corroborated by analyzing actual voting behavior data from the 2012 election.

Conclusions:

• Survey research using hypothetical election scenarios is a valuable tool for understanding voter behavior.
• Voting costs and perceived election competitiveness are key factors in the decision-making process.
• The methodology provides insights applicable to real-world election dynamics.