Welfare Measurement Convergence Through Bias Adjustments in General Population and On-Site Surveys: An Application to Water-based Recreation at Lake Sevan, Armenia

Abstract

This paper compares household survey with on-site survey data for estimating the access value of a unique natural resource using a single-site travel cost model. The household survey model is adjusted for inflated zero observations for respondents who would not visit the site at any observable positive price. The on-site survey model is corrected for truncation and endogenous stratification, the latter being an adjustment for avidity bias. In an application to recreation at Lake Sevan (Armenia), consumer surplus estimates were not statistically different between the household model and the on-site model when zero-inflation and truncation and endogenous stratification are corrected In the respective models. This leads us to believe that either method can be used to derive a consistent welfare measure of access to a recreational site after appropriate adjustments and corrections are made. These results are somewhat reassuring as the choice between household and on-site surveys is often dictated by time and resource availability.

KEYWORDS:

• On and off-site sampling
• zero-inflation
• count data
• endogenous stratification
• Armenia