An external validity test of intended behavior: Comparing revealed preference and intended visitation in response to climate change

We compare revealed preference and survey response estimates of changes in National Park visitation behavior from climate change. The revealed preference model is estimated from a time-series regression analysis of past visitation as a function of historic weather variability. We find no statistical difference between the revealed preference regression estimates and intended behavior estimates from a visitor survey for the total number of National Park visits. Confidence intervals for the two models overlap, and the mean estimates of the change in visitation are within 12% of each other. The paper concludes with the implications for natural resource planners and managers who must frequently evaluate the effects of alternative policies before the selection of a preferred alternative, and also must plan for accommodating changes in visitor use.