What did you really earn last year?: explaining measurement error in survey income data

Stefan Angel, Franziska Disslbacher, Stefan Humer, Matthias Schnetzer

Publication: Working/Discussion PaperWU Working Paper

28 Downloads (Pure)


This paper analyses the sources of income measurement error in surveys with a unique dataset.We use the Austrian 2008-2011 waves of EU-SILC which provide individual information on wages,pensions and unemployment benefits from survey interviews and officially linked administrativerecords. Thus, we do not have to fall back on complex two-sample matching procedures likerelated studies. We empirically investigate four sources of measurement error, namely (i) socialdesirability, (ii) socio-demographic characteristics of the respondent, (iii) the survey design, and(iv) the presence of learning effects. We find strong evidence for a social desirability bias inincome reporting, while the presence of learning effects is mixed and depends on the income typeunder consideration. An Owen value decomposition reveals that social desirability is a majorexplanation of misreporting in wages and pensions, whereas socio-demographic characteristicsare most relevant for mismatches in unemployment benefits.
Original languageEnglish
Place of PublicationVienna
PublisherWU Vienna University of Economics and Business
Publication statusPublished - 2019

Publication series

SeriesINEQ Working Paper Series

WU Working Paper Series

  • INEQ Working Paper Series

Cite this