Structural equation modeling of approval voting data

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

The method of approval voting is a commonly used voting procedure in which each judge selects a subset of the alternatives. By postulating that the random utilities associated with the choice options in approval voting elections follow a multivariate normal distribution under the Thurstonian framework, Regenwetter, Ho, and Tsetlin (2007) attempted to integrate the normative theories and individual variabilities in modeling social behavior. However, their approach is limited to only three alternatives, due to computational intractability as the number of alternatives increases. In this article, we reparameterize extensions of their models under the structural equation modeling framework and propose the use of limited information methods for estimating model parameters. As a result, we are able to extend their previous approach to the analysis of approval voting data with any number of alternatives. Two applications are presented to illustrate the usefulness of such an approach.

Original languageEnglish
Pages (from-to)798-808
Number of pages11
JournalBehavior Research Methods
Volume42
Issue number3
DOIs
Publication statusPublished - 2010 Aug

Fingerprint

Politics
Social Behavior
Normal Distribution
Structural Models
Structural Equation Modeling
Voting

ASJC Scopus subject areas

  • Psychology(all)
  • Psychology (miscellaneous)
  • Experimental and Cognitive Psychology

Cite this

Structural equation modeling of approval voting data. / Tsai, Rung Ching.

In: Behavior Research Methods, Vol. 42, No. 3, 08.2010, p. 798-808.

Research output: Contribution to journalArticle

@article{c5616bba5a3e466397635709195a6d73,
title = "Structural equation modeling of approval voting data",
abstract = "The method of approval voting is a commonly used voting procedure in which each judge selects a subset of the alternatives. By postulating that the random utilities associated with the choice options in approval voting elections follow a multivariate normal distribution under the Thurstonian framework, Regenwetter, Ho, and Tsetlin (2007) attempted to integrate the normative theories and individual variabilities in modeling social behavior. However, their approach is limited to only three alternatives, due to computational intractability as the number of alternatives increases. In this article, we reparameterize extensions of their models under the structural equation modeling framework and propose the use of limited information methods for estimating model parameters. As a result, we are able to extend their previous approach to the analysis of approval voting data with any number of alternatives. Two applications are presented to illustrate the usefulness of such an approach.",
author = "Tsai, {Rung Ching}",
year = "2010",
month = "8",
doi = "10.3758/BRM.42.3.798",
language = "English",
volume = "42",
pages = "798--808",
journal = "Behavior Research Methods",
issn = "1069-9384",
publisher = "Springer New York",
number = "3",

}

TY - JOUR

T1 - Structural equation modeling of approval voting data

AU - Tsai, Rung Ching

PY - 2010/8

Y1 - 2010/8

N2 - The method of approval voting is a commonly used voting procedure in which each judge selects a subset of the alternatives. By postulating that the random utilities associated with the choice options in approval voting elections follow a multivariate normal distribution under the Thurstonian framework, Regenwetter, Ho, and Tsetlin (2007) attempted to integrate the normative theories and individual variabilities in modeling social behavior. However, their approach is limited to only three alternatives, due to computational intractability as the number of alternatives increases. In this article, we reparameterize extensions of their models under the structural equation modeling framework and propose the use of limited information methods for estimating model parameters. As a result, we are able to extend their previous approach to the analysis of approval voting data with any number of alternatives. Two applications are presented to illustrate the usefulness of such an approach.

AB - The method of approval voting is a commonly used voting procedure in which each judge selects a subset of the alternatives. By postulating that the random utilities associated with the choice options in approval voting elections follow a multivariate normal distribution under the Thurstonian framework, Regenwetter, Ho, and Tsetlin (2007) attempted to integrate the normative theories and individual variabilities in modeling social behavior. However, their approach is limited to only three alternatives, due to computational intractability as the number of alternatives increases. In this article, we reparameterize extensions of their models under the structural equation modeling framework and propose the use of limited information methods for estimating model parameters. As a result, we are able to extend their previous approach to the analysis of approval voting data with any number of alternatives. Two applications are presented to illustrate the usefulness of such an approach.

UR - http://www.scopus.com/inward/record.url?scp=77957708854&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77957708854&partnerID=8YFLogxK

U2 - 10.3758/BRM.42.3.798

DO - 10.3758/BRM.42.3.798

M3 - Article

C2 - 20805603

AN - SCOPUS:77957708854

VL - 42

SP - 798

EP - 808

JO - Behavior Research Methods

JF - Behavior Research Methods

SN - 1069-9384

IS - 3

ER -