SAMPLE NONRESPONSE IN A MAIL CONTINGENT VALUATION SURVEY - AN EMPIRICAL-TEST OF THE EFFECT ON VALUE INFERENCE

A mail contingent valuation survey was used to assess the environmental values related to northern Swedish forests. A follow-up survey, focused on attitudes toward the forest environment and toward forest environmental research, was also conducted. The follow-up survey was sent to respondents and nonrespondents of the contingent valuation survey. Data from these two surveys formed the basis for a test of sample nonresponse effect on value inferences. Attitudes toward the forest environment significantly affect willingness-to-pay though attitudes toward research do not. Respondents and nonrespondents for the contingent valuation survey do not differ significantly on attitudes toward the forest environment or toward research. Value inference from the initial sample (the respondents to the contingent valuation survey) can therefore be considered as representative for the population.