Report

Identifying and Evaluating Sample Selection Bias in Consumer Payment Surveys

Abstract: This paper develops a two-stage statistical analysis to identify and assess the effect of a sample selection bias associated with an individual’s household role. The methodology is applied to the 2012 Survey of Consumer Payment Choice. Survey responses to a series of questions about the respondent’s role in household finances are combined and adjusted for response error to estimate a latent variable that represents each individual’s share of household responsibility. The distribution of this variable among survey respondents suggests that the sampling procedure favors household members with higher levels of financial responsibility. We also find that the estimated household role relates to the number of noncash payments made. As a result, estimators that do not account for the selection bias overestimate population means for these payment instruments by around 10 percent. We discuss ways to account for this selection bias in future surveys.

Keywords: survey error; household economics; Dirichlet regression; Survey of Consumer Payment Choice;

JEL Classification: C11; D12; D13;

Bibliographic Information

Provider: Federal Reserve Bank of Atlanta

Part of Series: Consumer Payments Research Data Reports

Publication Date: 2015-11-23

Number: 2015-07