Estimating and Using Individual Marginal Component Effects from Conjoint Experiments

Abstract

Conjoint experiments are quickly gaining popularity as a vehicle for studying multidimensional political preferences. A common way to explore heterogeneity of preferences estimated with conjoint experiments is by estimating average marginal component effects across subgroups. However, this method does not give the researcher the full access to the variation of preferences in the studied populations, as that would require estimating effects on the individual level. Currently, there is no accepted technique to obtain estimates of individual-level preferences from conjoint experiments. The present paper addresses this gap by proposing a procedure to estimate individual preferences as respondent-specific marginal component effects. The proposed strategy does not require any additional assumptions compared to the standard conjoint analysis, although some changes to the task design are recommended. Methods to account for uncertainty in resulting estimates are also discussed. Using the proposed procedure, I partially replicate a conjoint experiment on immigrant admission with recommended design adjustments. Then, I demonstrate how individual marginal component effects can be used to explore distributions of preferences, intercorrelations between different preference dimensions, and relationships of preferences to other variables of interest.

The Demand Side of Democratic Backsliding: How Divergent Understandings of Democracy Shape Political Choice

“…According to our theoretical argument, divergent understandings of democracy affect to what extent citizens are likely to overlook democratic violations when selecting among competing candidates. We use our conjoint experiment to measure what we term revealed democratic attitudes by assessing the weight of candidates' democratic attributes in individual respondents' candidate ratings by computing Individual Marginal Component Effects (IMCEs) (Zhirkov 2021). Respondents' IMCEs then serve as a measure of our dependent variable.…”

“…IMCEs rely on the same set of assumptions as AMCEs. That is, only when the assumptions of (1) stability and no carryover effects, (2) no profile-order effects, and (3) completely independent randomization of the profiles in a conjoint experiment hold, IMCEs can be estimated independently for each respondent (Hainmueller, Hopkins, and Yamamoto 2014;Zhirkov 2021). For our candidate choice experiment, we verified assumption 1 (see Appendix B.2.2) and assumptions 2 and 3 are true by design, allowing us to proceed with the estimation of IMCEs.…”

“…To assess the linkages between understandings of democracy and vote choice, we implement a preregistered, well-powered candidate choice conjoint experiment among a representative sample of citizens in Poland. We measure revealed understandings of democracy and their impact on candidate choice by leveraging the novel approach of Individual Marginal Component Effects (IMCEs) (Zhirkov 2021) that allows us to estimate how individual-level citizen commitment to democracy affects vote choice for candidates expressing differing democratic views.…”

“…Zhirkov 2021). Yet it is precisely such divergent understandings of democracy at the individual level that underpin the causal mechanism we seek to probe in our study, leading us to adopt a recently proposed alternative approach to conjoint analysis via Individual Marginal Component Effects.…”

“…Individual Marginal Component Effects (IMCEs) as Measures of Revealed Democratic Attitudes Individual Marginal Component Effects (IMCEs) have been recently introduced byZhirkov (2021) …”