Measurement error in earnings data in the health and retirement study

Article type: Research Article

Authors: Bricker, Jesse^{a; *} | Engelhardt, Gary V.^a

Affiliations: [a] Department of Economics and Center for Policy Research, Maxwell School of Citizenship and Public Affairs, 426 Eggers Hall, Syracuse University, Syracuse, NY 13244, USA

Correspondence: [*] Corresponding author. Tel.: +1 315 443 3114; Fax: +1 315 443 1081; E-mail: [email protected]

Abstract: This paper provides new evidence on the extent of measurement error in respondent-reported earnings data by exploiting detailed W-2 records matched to older workers in the Health and Retirement Study (HRS). There is a negative correlation between the measurement error and the true value of earnings as measured by the W-2 records, which indicates the presence of non-classical measurement error. Yet for men and women, this error shows little correlation with a standard set of cross-sectional earnings determinants. Quantitatively, the bias from using self-reported earnings, either as a dependent or explanatory variable, is larger than those from existing studies of the CPS and PSID.

Keywords: Measurement error, earnings

DOI: 10.3233/JEM-2008-0297

Journal: Journal of Economic and Social Measurement, vol. 33, no. 1, pp. 39-61, 2008