Discrete Choice Experiments Are Not Conjoint Analysis

Abstract

We briefly review and discuss traditional conjoint analysis (CA) and discrete choice experiments (DCEs), widely used stated preference elicitation methods in several disciplines. We pay particular attention to the origins and basis of CA, and show that it is generally inconsistent with economic demand theory, and is subject to several logical inconsistencies that make it unsuitable for use in applied economics, particularly welfare and policy assessment. We contrast this with DCEs that have a long-standing, well-tested theoretical basis in random utility theory, and we show why and how DCEs are more general and consistent with economic demand theory. Perhaps the major message, though, is that many studies that claim to be doing conjoint analysis are really doing DCE.