Title: Measurement errors: Evidence from reports of program participation from multiple surveys
Authors: Pablo Celhay - University of Chicago (United States)
Bruce D Meyer - University of Chicago and NBER (United States)
Nikolas Mittag - CERGE-EI (Czech Republic) [presenting]
Abstract: Measurement error is often the largest source of bias in survey data. Little is known about the determinants of such errors, making it difficult for data producers to reduce the extent of errors and for data users to assess the validity of analyses using the data. We study different causes of survey error using high quality validation data from three major surveys in the U.S. that are linked to administrative data on government transfers. The differences between survey and administrative records show that up to six out of ten cash welfare recipients are missed by surveys. We find that survey design and post-processing as well as misreporting by respondents affect survey errors systematically. Imputation for missing data induces substantial error. Our results on respondent behavior confirm several theories of misreporting, e.g. that errors are related to salience of receipt, respondents degree of cooperation, forward and backward telescoping, event recall, and the stigma of reporting participation in social programs. Our results provide guidance on the conditions under which survey data are likely to be accurate and suggest different ways to control for survey errors.