ENVIRONMENTAL INTEREST GROUP-BEHAVIOR AND SELF-SELECTION BIAS IN CONTINGENT VALUATION MAIL SURVEYS

This paper analyzes the behavior of a general sample and an environmental interest group sample in a contingent market for wetlands preservation. Mail survey response rates and environmental values for wetlands preservation are significantly greater in the environmental interest group sample than in the general population sample. An estimate of the potential self-selection bias in the benefits of wetlands preservation is made. These results suggest that self-selection bias in contingent valuation mail surveys could upwardly bias aggregate benefit estimates as much as 50 percent. Potential, but costly, solutions to the problem of self-selection bias are suggested.