Aggregation Bias in Recreation Site Choice Models: Resolving the Resolution Problem

Abstract

This paper examines the effect of differing levels of spatial resolution on recreation site choice models and welfare resulting from changes in site attributes. These issues are important where the spatial scale at which recreationists make choices is unknown, but information exists on choice attributes at larger spatial scales. We estimate choice models at various scales of spatial resolution and incorporate the size of the aggregate sites and heterogeneity parameters in the model. Accounting for the size of the aggregates in estimation improved model fit and alleviated aggregate parameter bias. We provide advice for applied modeling based on these results. (JEL Q26)