Polyhedral Methods for Adaptive Choice-Based Conjoint Analysis

• DOI: 10.1509/jmkr.41.1.116.25082
• Publication Date: 2004-01-14T16:45:39Z
• Authors: Olivier Toubia; John R. Hauser; Duncan I. Simester
• Keywords: Conjoint Analysis

The authors propose and test a new “polyhedral” choice-based conjoint analysis question-design method that adapts each respondent's choice sets on the basis of previous answers by respondent. Polyhedral “interior-point” algorithms design questions quickly reduce partworths are consistent with choices. To identify domains in which individual adaptation is promising (and it not), evaluate performance polyhedral methods Monte Carlo experiments. They vary magnitude (response accuracy), respondent heterogeneity estimation method, 4 × 2 3 experiment. hierarchical Bayes analytic center. latter individual-level procedure by-product question design. benchmarks for random designs, orthogonal aggregate customization. simulations suggest does well many domains, particularly those partworth magnitudes relatively large. feasibility, an important criterion (choice balance), obtain empirical data convergence describing application to executive education programs 354 Web-based respondents answered stated-choice tasks four service profiles each.