Convergent Validity of Contingent Behavior Responses in Models of Recreation Demand

Recreation demand modeling efforts are often limited by the range of variation in observed environmental quality. To address this limitation, the practitioners increasingly makes use of contingent behavior (CB) data; i.e., asking survey respondents to forecast their trip patterns under hypothetical quality conditions. However, relatively little is know as to whether these stated responses are consistent with how households response to actual quality variation. The purpose of this paper is to investigate the convergent validity of CB data with observed trip patterns. Toward this end, we jointly model the recreation lake usage for in Iowa using observed and CB trip data collected from the 2004 Iowa Lakes Survey. The Iowa lakes survey collected three sets of trip data for 131 lakes in the state: (a) actual trips in 2004, (b) anticipated trips in 2005 to the same lakes given current lake conditions and (c) anticipated trips in 2005 given hypothetical improvements to a subset of the lakes. The three types of recreation demand data provide a unique opportunity to investigate the convergent validity of individual responses to actual versus hypothetical environmental conditions.