Comparison of a full and partial choice set design in a labeled discrete choice experiment

Abstract

Labeled discrete choice experiments (DCEs) commonly present all alternatives using a full choice set design (FCSD), which could impose a high cognitive burden on respondents. In the setting of employment preferences, this study explored if a partial choice set design (PCSD) reduced cognitive burden whilst maintaining convergent validity compared with a FCSD. Respondents' preferences between the two designs were investigated. In the experimental design, labeled utility functions were rewritten into a single generic utility function using label dummy variables to generate an efficient PCSD with 3 alternatives shown in each choice task (out of 6). The DCE was embedded in a nationwide survey of 790 Australian pharmacy degree holders where respondents were presented with both a block of FCSD and PCSD tasks in random order. The PCSD's impact on error variances was investigated using a heteroscedastic conditional logit model. The convergent validity of PCSD was based on the equality of willingness‐to‐forgo‐expected‐salary estimates from Willingness‐to‐pay‐space mixed logit models. A nested logit model was used combined with respondents' qualitative responses to understand respondents' design preferences. We show a promising future use of PCSD by providing evidence that PCSD can reduce cognitive burden while satisfying convergent validity compared to FCSD.

Discrete Choice Experiment Consideration: A Framework for Mining Community Consultation with Case Studies

“…Also, Traets et al present a new R package, called idefix, which enables users to generate Bayesian adaptive designs which can be used to gather data for the mixed logit model [14]. Thai et al show a promising future use of partial choice set design (PCSD) by providing evidence that PCSD can reduce cognitive burden while satisfying convergent validity compared to full choice set design (FCSD) [15]. Mamine et al present a meta-analysis that explores the impact of methodological choices related to the timing of the DCE (design stage, implementation stage, and analysis stage) on farmers' contract preferences regarding the adoption of agrienvironmental practices [16].…”