Conjoint respondents as adaptive decision makers

One implicit assumption in conjoint measurement is that respondents solve the conjoint tasks by using some form of weighted additive rule for preference judgements. The weighted additive rule is assumed to be associated with a high level of accuracy, but at the same time to be among the most cognitively demanding processes for arriving at preference judgements. Research from other domains, including consumer behaviour, indicates that people often use highly simplified rules to arrive at preference judgements. This suggests that the weighted additive assumption might be unrealistic. In this paper we report on an empirical study that was designed to investigate the decision rules used by respondents in making judgements in full profile conjoint analysis. Results from self-reports based on a set of judgement rule descriptions show that only 40% of the subjects reported using a weighted additive rule. A set of hypotheses derived from an adaptive decision-making perspective is also developed and tested.