ability to pay since the amount they said they would pay exceeds 

 their income. 



The other group of respondents excluded from the analysis were 

 those hunters who were protesting the simulated market. The 

 U.S. Water Resource Council recommends that a followup question 

 be asked to determine the reasons for a negative response. 

 Mountain lion hunters who indicated they would not pay the bid 

 amount but had a valid reason were left in the sample. Hunters 

 were excluded from the sample as protest bids if they indicated 

 they did not understand the question or were opposed to any 

 increases in fees or taxes, etc. 



Model Specification 



Lion hunters' net willingness to pay (net economic value) was 

 estimated by analyzing their responses to the CVM question using 

 logistic regression. A comprehensive discussion of this theory 

 and methods is provided in Duf field and Patterson (1991) . 



Economic theory suggests that particular variables will influence 

 an individual's response to a CVM question. This study used a 

 bivariate model that regressed "yes and no" responses upon the 

 bid amounts asked. It is assumed that as the bid amount 

 increases, the probability of a "yes" response will decrease. 



The specification of the logit equation is shown in Equation 1. 



ln(P/l-P) = BO + Bl In(Bid) 



where: P = probability of a "yes" response 



Bid = dollar amount of increased trip costs respondent 

 was asked to pay. 



The estimated equations are shown in Appendix B. The 

 coefficients for the independent variable In (Bid) had the 

 expected sign (negative) and were significant at the 95% level. 

 Based on the results of the models, the responses are logical and 

 consistent with economic theory. 



Benefit Estimates 



The measure of economic benefits (net economic value) used in 

 this study is the truncated mean. The bivariate form of the 

 model can be graphed with the probability of a "yes" response on 

 the vertical axis and the bid amount on the horizontal axis. 

 Graphing this relationship shows a high probability of acceptance 

 at low bid amounts with the probability decreasing and 

 approaching zero at high bid levels. Integrating the logit 

 function from zero to an upper limit, generally the highest bid 



15 



