Measuring Voters’ Multidimensional Policy Preferences with Conjoint Analysis: Application to Japan’s 2014 Election

Abstract

Representative democracy entails the aggregation of multiple policy issues by parties into competing bundles of policies, or “manifestos,” which are then evaluated holistically by voters in elections. This aggregation process obscures the multidimensional policy preferences underlying a voter’s single choice of party or candidate. We address this problem through a conjoint experiment based on the actual party manifestos in Japan’s 2014 House of Representatives election. By juxtaposing sets of issue positions as hypothetical manifestos and asking respondents to choose one, our study identifies the effects of specific positions on the overall assessment of manifestos, heterogeneity in preferences among subgroups of respondents, and the popularity ranking of manifestos. Our analysis uncovers important discrepancies between voter preferences and the portrayal of the election results by politicians and the media as providing a policy mandate to the Liberal Democratic Party, underscoring the potential danger of inferring public opinion from election outcomes alone.

The Politics of Trade-Offs: Studying the Dynamics of Welfare State Reform With Conjoint Experiments

“…However, in actual preferences over reforms, these elements are linked, and a voter may evaluate a measure differently depending on the other elements of the reform, for instance if a voter values the coherence of a reform package. Therefore, a standard random utility model that infers overall utility from several, separately measured components is unable to estimate overall utility correctly (Horiuchi, Smith & Yamamoto, 2015).…”

University Elections and Political Socialization in the Developing World

“…Conjoint experiments randomly assign attributes to pairs of candidate profiles presented to study participants. OLS estimates from a regression of respondents' binary candidate preference on a set of categorical variables representing each candidate attribute will produce an unbiased estimate of the Average Marginal Component Effect (AMCE) for each attribute-level (Hainmueller et al 2014;Horiuchi et al 2018). Conjoint analysis has been employed in a number of studies where the attribute of interest is difficult to understand a uni-dimensional survey question, usually because respondents hesitate to openly admit holding certain preferences regarding the religious, racial, ethnic, or political affiliation of the profile they are evaluating (Hainmueller et al 2014;Auerbach & Thachil 2018;Liu 2018).…”

“…The core of our analysis is derived from a series of choice-based conjoint experimental tasks administered at the conclusion of each wave of AUB-CONJOINT. Conjoint experiments typically present respondents with hypothetical profiles that differ on a set of attributes (Hainmueller et al 2014;Hainmueller & Hopkins 2015;Franchino & Zucchini 2015;Sen 2017;Horiuchi et al 2018). These attributes have two or more levels, each of which are randomly assigned to a profile.…”

Multi-dimensional policy preferences in the 2015 British general election: A conjoint analysis

“…Because of its intuitive design and ease of application, conjoint design has also been widely used in political science research to examine numerous topics, such as: support for immigrants' naturalization applications (Hainmueller and Hopkins, 2015;; evaluations of incumbent legislators (Vivyan and Wagner, 2016); preferences for presidential candidates (Hainmueller et al, 2014); and voters' evaluations of party policy platforms (Horiuchi et al, 2018).…”

“…Our conjoint design is analogous to the design implemented by Horiuchi, Smith and Yamamoto (2018) for the 2014 general election in Japan, in the sense that the policy components used to build up hypothetical policy packages in the experiment are derived from actual party manifestos. Their analyses concluded that policy preferences mattered little for voters' choice in the election: the Liberal Democratic Party of Japan, a governing party that enjoyed a decisive victory, had proposed a policy package that turned out to be one of the least popular among the general electorate.…”