responses. Interestingly, time spent in the Park was found to have a
statistically significant and negative effect on the probability that a
respondent would be willing to pay the specified bid amount. Individuals
who saw a bear on their most recent trip to Yellowstone
were more likely to be willing to pay than those who did not, all
else constant, and this variable is significant at the 1% level. Finally,
the probability that a respondent is willing to pay the specified bid
amount decreases with age and increases with the number of years
of education they have. Variables controlling for respondents who
participated in bear viewing or other wildlife viewing on their most
recent trip to Yellowstone compared to those who did not were not
found to have a statistically significant effect on willingness-to-pay.
Similarly, expecting to see a bear, as well as whether the respondent
had seen a bear in Yellowstone on past trips, did not have a
statistically significant effect on willingness-to-pay.
While the results of this model could be used to estimate a
measure of central tendency for willingness-to-pay, as can be seen
in Table 7, the majority of survey respondents were willing to pay at
even the highest bid amounts offered. The range of the bid amounts
used was simply too narrow; respondents were never presented
with a bid amount high enough that a majority would not be
willing to pay that amount. The result is a lack of data to characterize
the full response distribution, meaning the use of a parametric
model to estimate a measure of central tendency for
willingness-to-pay may not be appropriate. A reasonable strategy
when presented with this situation is to employ a conservative
approach, such as estimating the lower bound of expected