Propensity Score Adjustment as a Weighting Scheme for Volunteer Panel Web Surveys

Abstract

Propensity score adjustment (PSA) has been suggested as an approach to adjustment for volunteer panel web survey data. PSA attempts to decrease, if not remove, the biases arising from noncoverage, nonprobability sampling, and nonresponse in volunteer panel web surveys. Although PSA is an appealing method, its application in web survey practice is not well documented, and its effectiveness is not well understood. This study attempts to provide an overview of the PSA application by demystifying its performance for web surveys. Findings are three-fold: (a) PSA decreases bias but increases variance, (b) it is critical to include covariates that are highly related to the study outcomes, and (c) the role of nondemographic variables does not seem critical to improving PSA.