status quo, described the average visibility level at the site during the summer months.
Results
WTP. Results from the analysis of the WTP survey data were consistent with prior
expectations. Higher income and larger improvements in visibility increased the probability that
the respondent would accept the bid offered, while higher bids in turn lowered this probability.
Thirty three percent of respondents in the sample accepted the offered bid. The model correctly
predicted 70.3 percent of responses.
In a linear model, estimated mean and median will be the same since both logistic and
normal distributions are symmetrical. The general formula to compute the mean WTP/WTA
estimate (and the median for the log-linear model) is β'XZγ. Estimates of the median economic
value of visibility derived from the semi-log WTP logit models were calculated by the following
formula:
Pr(Accept) = 1/(1 + e - (a + β Ln Compensation))
where α and β are estimated parameters.
WTA. Results from the WTA survey analysis are also shown in Table 2. Relatively few
respondents were willing to make a tradeoff between electricity bills and reduced visibility (20.1
percent). The model correctly predicted 80.6 percent of responses. The probability of future
visits (FVISIT) and the amount of visibility loss (VISLOSS) both reduced the probability of a
“yes” response. A surprising result was that the amount of compensation offered did not have a
statistically significant effect on respondents. This seemingly counterintuitive result will be
discussed later.
That relatively few respondents were willing to accept a tradeoff between visibility and
10