An experimental investigation of preference elicitation methods in policy decision-making

Preference elicitation methods support computational and behavioural aspects of preference assessment and resolution of inconsistencies. The purpose of this research was to investigate efficient and accurate ways for analytically disinclined policy decision makers to assess priorities and preference intensities. A total of 89 subjects participated in a laboratory experiment performing preference assessments. The results suggest that policy decision makers should first use a behavioural non-compensatory method to derive priorities from which numerical intensities can be computed. The numerical values should then be presented to the policy makers for them to make changes if desired. The results also indicate that many of the theoretical assumptions of computational compensatory methods fall short, indicating that a sensitive balance between computational and behavioural aspects is crucial for the design of preference elicitation systems.