Testing the consistency between standard contingent valuation, repeated contingent valuation, and choice experiments

Choice experiments are a relatively new approach to valuing environmental resources. Initial tests of the validity of the approach have either compared benefit estimates generated using choice experiments with those estimated using contingent valuation or used more sophisticated hypothesis tests of parameter equality. Although useful, existing tests have been restricted to testing consistency based on a single policy scenario (standard contingent valuation). We argue that, though these tests are informative, they fail to take full advantage of the richness of choice experiment data. In particular, choice experiment data allow for the calculation of benefit estimates over a range of policy scenarios (i.e. attribute combinations). A similar range of benefit estimates may be generated by pooling scenarios in a repeated contingent valuation study. In this paper we explore this relationship between contingent valuation and choice experiments by conducting validity tests between a choice experiment model and a repeated contingent valuation model over a range of three levels of improved water quality at Clear Lake, Iowa. Evidence from this test suggests that the choice experiment and contingent valuation data are consistent.