Conjoint Analysis Versus Contingent Valuation: Estimating Risk Values and Death Risk Equivalents in Road Traffic



Conjoint Analysis Versus Contingent Valuation: Estimating Risk Values and Death Risk Equivalents in Road Traffic

Authors

TRAWEN A and HJALTE K, Lund University of Technology and NORINDER A, The Swedish Institute for Health Economics, Sweden

Description

Eliciting preferences for non-market goods has traditionally been done by help of stated preference methods, which could be seen as a group of direct methods in comparison to those indirect methods usually denoted revealed preference methods. The direct v

Abstract

Eliciting preferences for non-market goods has traditionally been done by help of stated preference methods, which could be seen as a group of direct methods in comparison to those indirect methods usually denoted revealed preference methods. The direct valuation methods include variants like contingent valuation (CV) and conjoint analysis (CA). The CV method has since long time been the most used technique in environmental economics but is also used in health and transport economics, even though more sparsely. The dominant technique in transport economics for example in valuation of travel time savings has been the CA or closely related methods, in the transport economics literature usually named stated preference methods. An extension of the original CA is sometimes called choice experiments that involve more experimental and involved analysis of choice behaviour. The terminology as well as the advantages and drawbacks with these methods are not quite clear demanding additional research for example comparing different stated preference methods according the influence of the context of the questionnaire and according the psychological process of making choices in different formats. For instance, it may be quite different to state a maximum willingness to pay for a specified good than choosing among goods with assigned prices.

The purpose of this pilot study is to compare two direct valuation methods, the CV and CA, for estimating risk values for fatal and non-fatal road traffic accidents as well as death risk equivalents. The risk value for a fatal accident, the value of a statistical life, is defined as the average willingness to pay for a specified risk reduction divided by that risk reduction. Accordingly similar risk values are calculated for serious disabling injuries and slight injuries. With death risk equivalents, the health effects are valued in a common metric where the injury is presented in relation to death. A fatal injury has the value one and good health has the value zero. This implies that for example two injuries which both are valued to 0.5 death risk equivalent each, together correspond to one fatal injury.

Publisher

Association for European Transport