When Choice Models Fail: Compensatory Models in Negatively-Correlated Environments.
Abstract
Linear compensatory models, which make tradeoffs between product attributes, provide reasonably good predictions of choices made by non-compensatory heuristics which do not make tradeoffs. This robustness to miss-specification of functional form, however, may fail when there are negative correlations among attributes in a choice set. A Monte-Carlo simulation first demonstrates that certain non-compensatory rules are poorly fit by linear models, even in orthogonal environments. Two laboratory experiments then assess the extend to which such model failure might arise in natural contests. The first, a process-tracing analysis, examines the decision strategies consumers use in non-orthogonal contexts. We conclude with a discussion of the work's implications for current research in applied choice modeling. (KR)
Document Details
- Document Type
- Technical Report
- Publication Date
- Feb 01, 1989
- Accession Number
- ADA205749
Entities
People
- Eric J. Johnson
- Robert J. Meyer
- Sanjoy Ghose
Organizations
- Duke University