Identifying and analysing dominant preferences in disrete choice experiments: an application in health care

The assumption that goods are traded-off to maximise utility underpins many choice-based empirical methods used to elicit stated preferences. The aim of this paper is to test and examine the implications of this assumption in the context of a discrete choice experiment that examined stated preferences for different models of out of hours care provided by General Practitioners. The results indicated that 45% of individuals exhibited evidence of dominant preferences, a special case of a lexicographic ordering. Factors influencing the existence of dominant preferences included the complexity of the choices, but also individuals' past experiences. The analysis of dominant preferences provided important information about individuals with particularly strong preferences, and suggested different policy conclusions for this group of respondents. (C) 2002 Elsevier Science B.V. All rights reserved.