Article contents
Voting in a Multi-dimensional Space: A Conjoint Analysis Employing Valence and Ideology Attributes of Candidates
Published online by Cambridge University Press: 28 October 2014
Abstract
Most formal models of valence competition add a single, separable and unweighted component to the standard one-dimensional utility function of voters. This article presents the results of a conjoint analysis experiment in which respondents were asked to choose between two candidates whose profiles vary along five attributes. Four of these traits behave like valence or policy issues as expected, but one, which has been employed in recent formal and empirical works, does not. Moreover, policy and valence are not separable. They interact at least in some cases, taking a competency form whereby the marginal impact of valence on voters’ choice is conditional on candidates’ policies. This result lends support to recent studies that have found more extensive valence voting under ideological convergence. Finally, policy trumps valence in awkward choices. Respondents even prefer corrupt candidates with similar policy views to honest ones with different opinions, despite integrity being declared the most important attribute.
- Type
- Original Articles
- Information
- Copyright
- Copyright © The European Political Science Association 2014
Footnotes
Fabio Franchino and Francesco Zucchini are Associate Professors of Political Science, Department of Social and Political Sciences, Università degli Studi di Milano, Via Conservatorio 7, Milan, 20122, Italy (fabio.franchino@unimi.it, francesco.zucchini@unimi.it). This article has greatly benefitted from presentations at the 71st Annual Conference of the Midwest Political Science Association (Chicago, April 2013), the 3rd Annual Conference of the European Political Science Association (Barcelona, June 2013) and at the seminar series of the Department of Social and Political Sciences, Università degli Studi di Milano. The authors would like to thank Luigi Curini, Walter R. Mebane, Guy D. Whitten and the reviewers of PSRM for their useful comments, and Alessandra Caserini of the Opinion Polls Laboratory (Laboratorio Indagini Demoscopiche, Università degli Studi di Milano), for research assistance. Supplementary material is available at http://dx.doi.org/10.1017/psrm.2014.24.
References
- 68
- Cited by