What Do We Learn about Voter Preferences from Conjoint Experiments?

Political scientists frequently interpret the results of conjoint experiments as reflective of majority preferences. In this article, we show that the target estimand of conjoint experiments, the average marginal component effect (AMCE), is not well defined in these terms. Even with individually rat...

Full description

Saved in:
Bibliographic Details
Published inAmerican journal of political science Vol. 66; no. 4; pp. 1008 - 1020
Main Authors Abramson, Scott F., Kocak, Korhan, Magazinnik, Asya
Format Journal Article
LanguageEnglish
Published Oxford Blackwell Publishing Ltd 01.10.2022
Subjects
Candidates
Experiments
Majority rule
Political majority
Political scientists
Property
Research subjects
Voter behavior
Voters
Voting
Voting rules
Online AccessGet full text

Cover

More Information
Summary:Political scientists frequently interpret the results of conjoint experiments as reflective of majority preferences. In this article, we show that the target estimand of conjoint experiments, the average marginal component effect (AMCE), is not well defined in these terms. Even with individually rational experimental subjects, the AMCE can indicate the opposite of the true preference of the majority. To show this, we characterize the preference aggregation rule implied by the AMCE and demonstrate its several undesirable properties. With this result, we provide a method for placing bounds on the proportion of experimental subjects who prefer a given candidate feature. We describe conditions under which the AMCE corresponds in sign with the majority preference. Finally, we offer a structural interpretation of the AMCE and highlight that the problem we describe persists even when a model of voting is imposed.
Bibliography:Kocak gratefully acknowledges the support of the Research Program in Political Economy at Princeton University. The authors thank Naoki Egami, Matias Iaryczower, Kosuke Imai, John Londregan, Nolan McCarty, Teppei Yamamoto, and seminar audiences at Harvard, NYUAD, Princeton, Rochester, UCL, UdeM, the APSA Annual Meeting, the 2019 Conference of the Society for Political Methodology, and the 2019 Toronto Political Behavior Workshop for useful comments and encouragement.
ISSN:0092-5853
1540-5907
DOI:10.1111/ajps.12714