Multidimensional Spatial Voting with Non-separable Preferences

Abstract

In most multidimensional spatial models, the systematic component of agent utility functions is specified as additive separable. We argue that this assumption is too restrictive, at least in the context of spatial voting in mass elections. Here, assuming separability would stipulate that voters do not care about how policy platforms combine positions on multiple policy dimensions. We present a statistical implementation of Davis, Hinich, and Ordeshook's (1970) Weighted Euclidean Distance model that allows for the estimation of the direction and magnitude of non-separability from vote choice data. We demonstrate in a Monte-Carlo experiment that conventional separable model specifications yield biased and/or unreliable estimates of the effect of policy distances on vote choice probabilities in the presence of non-separable preferences. In three empirical applications, we find voter preferences on economic and socio-cultural issues to be non-separable. If non-separability is unaccounted for, researchers run the risk of missing crucial parts of the story. The implications of our findings carry over to other fields of research: checking for non-separability is an essential part of robustness testing in empirical applications of multidimensional spatial models.

Modeling Preferences Using Roll Call Votes in Parliamentary Systems

“…Our study suggests that in this situation off-the-shelf estimates may be biased in various ways and that we should instead turn to more complex behavioral models to arrive at valid point estimates. Overall, our article contributes to recent innovations in political methodology that aim at increasing the validity of measures derived from roll call analysis (see Peress 2009;Aldrich, Montgomery, and Sparks 2014;Stoetzer and Zittlau 2015).…”

Identifying and Measuring Conditional Policy Preferences: The Case of Opening Schools During a Pandemic

“…Despite their relative prevalence in the mass public, conditional preferences are rarely studied as an outcome of interest in political science research. Indeed, since its publication, Lacy's theory of nonseparable preferences has been cited just over 100 times, 3 typically in the context of voting behavior (Stoetzer and Zittlau 2015;Franchino and Zucchini 2015) or for its optimistic interpretation of framing and question order effects (Druckman 2004;Gaines, Kuklinski, and Quirk 2007) rather than for measuring non-separable issue preferences directly. Moreover, when such preferences are the direct subject of empirical inquiry, they are typically measured using single-shot survey questions (Tate 2003;Neblo et al 2010), with the goal of estimating their prevalence in the aggregate.…”

A Matter of Representation: Spatial Voting and Inconsistent Policy Preferences

“…Based on this, 34 The exact wording of the question and descriptions are available in the Appendix. 35 Factor-analytic techniques to estimate policy platforms from these concrete issue item batteries have been used in various articles concerned with spatial voting decisions (see Schofield et al 1998;Schofield and Zakharov 2009;Stoetzer and Zittlau 2015;Quinn, Martin, and Whitford 1999). 36 A more detailed description of the results is provided in the Appendix.…”

“… 35 Factor-analytic techniques to estimate policy platforms from these concrete issue item batteries have been used in various articles concerned with spatial voting decisions (see Schofield et al 1998; Schofield and Zakharov 2009; Stoetzer and Zittlau 2015; Quinn, Martin, and Whitford 1999). …”