Travel cost demand model based river recreation benefit estimates with on-site and household surveys:: Comparative results and a correction procedure -: Reevaluation

[ 1] Loomis ( 2003) reported to have measured the effect of endogenous stratification on benefit estimates resulting from using on-site data. He compared corrected and uncorrected benefit estimates of water-based recreation using on-site sample data and estimates based on data from a household survey. We clarify his analysis and show that his comparison was actually between uncorrected estimates and estimates that corrected for both endogenous stratification and zero truncation. Our reanalysis of Loomis's ( 2003) on-site sample data confirms that zero truncation accounts for most of the on-site sample bias. The properly isolated effect of correcting for endogenous stratification only is much less substantial than reported by Loomis.