Structural equation modeling of approval voting data.

Area of Science:

• Social Choice Theory
• Mathematical Psychology
• Statistical Modeling

Background:

• Approval voting allows voters to select multiple options.
• Previous models were computationally limited to three alternatives.
• Integrating normative theories with individual variability is key for social behavior modeling.

Purpose of the Study:

• To extend existing approval voting models to accommodate any number of alternatives.
• To overcome computational intractability in prior Thurstonian framework models.
• To enable more robust analysis of social behavior in voting.

Main Methods:

• Reparameterization of existing models within a structural equation modeling framework.
• Application of limited information methods for parameter estimation.
• Extension of the Thurstonian framework for approval voting.

Main Results:

• The proposed methods successfully extend approval voting analysis to an unlimited number of alternatives.
• Computational limitations of previous models are overcome.
• Demonstrated applicability through two real-world examples.

Conclusions:

• The developed approach provides a flexible and powerful tool for analyzing approval voting data.
• This research advances the statistical modeling of social behavior in elections.
• The findings facilitate a deeper understanding of voter preferences with diverse choice sets.