A comparison of willingness to pay estimation techniques from referendum questions - Application to endangered fish

Referendum style willingness to pay questions have been used to estimate pa ssive use values. This referendum question format method may be problematic for many reasons, including the statistical techniques used to estimate wi llingness to pay from discrete responses. This paper compares a number of p arametric, semi-nonparametric and nonparametric estimation techniques using data collected from US households regarding Federal protection of endanger ed fish species.The advantages and disadvantages of the various statistical models used are explored. A hypothesis test for statistical equality among estimation techniques is performed using a jackknife bootstrapping method. When the equality test is applied, the modeling techniques do show signifi cant differences in some possible comparisons, but only those that are nonp aramentric. This can lead to conflicting interpretations of what the data s how. Resource managers and policy analysts need to use caution when interpr eting results until an industry standard can be developed for estimating wi llingness to pay from closed ended questions.