Unstructured Direct Elicitation of Decision Rules

Abstract

The authors investigate the feasibility of unstructured direct elicitation (UDE) of decision rules consumers use to form consideration sets. They incorporate incentives into the tested formats that prompt respondents to state noncompensatory, compensatory, or mixed rules for agents who will select a product for the respondents. In a mobile phone study, two validation tasks prompt respondents to indicate which of 32 mobile phones they would consider from a fractional design of features and levels. The authors find that UDE predicts consideration sets better, across both profiles and respondents, than a structured direct-elicitation method. It predicts comparably to established incentive-aligned compensatory, noncompensatory, and mixed decompositional methods. In a more complex automotive study, noncompensatory decomposition is not feasible and additive-utility decomposition is strained, but UDE scales well. The authors align incentives for all methods using prize indemnity insurance to award a chance at $40,000 for an automobile plus cash. They conclude that UDE predicts consideration sets better than either an additive decomposition or an established structured direct-elicitation method (CASEMAP).