Estimating Measurement Error in SIPP Annual Job Earnings: A Comparison of Census Bureau Survey and SSA Administrative Data

Abstract

The research program of the Center for Economic Studies (CES) produces a wide range of economic analyses to improve the statistical programs of the U.S. Census Bureau. Many of these analyses take the form of CES research papers. The papers have not undergone the review accorded Census Bureau publications and no endorsement should be inferred. Any opinions and conclusions expressed herein are those of the author(s) and do not necessarily represent the views of the U.S. Census Bureau. All results have been reviewed to ensure that no confidential information is disclosed. Republication in whole or part must be cleared with the authors.To obtain information about the series, see www.census.gov/ces or contact Cheryl Grim, Editor,

A Method of Correcting for Misreporting Applied to the Food Stamp Program

“…This assumption may not be warranted with other linked records, for example if the administrative data are also based on reports, as with tax records. Nonetheless, such records may still be a better measure than survey reports and both measures can be combined to improve inference as in Abowd and Stinson (2011) or Kapteyn and Ypma (2007).…”

“…A more complex model may be required if the linked data come from a stratified sample or suffer from problems such as sample selection. In other cases, the assumption that the administrative records represent "truth" may be too strong, as in Abowd and Stinson (2011). Nonetheless, the assumptions they make in their analyses identify the conditional density of income conditional on survey reports, which could be used in the es-timators discussed above.…”

Econometric Analyses of Linked Employer–Employee Data

“…7 Further details of identi…cation requirements are discussed in section 5.1. 8 The zero conditional mean assumption (8) has been interpreted as an assumption of "exogenous mobility," since it precludes any relationship between an individuals employment location (measured by F i ) and the errors " i : See AKM for further discussion, and section 8 below for recent work that accomodates endogenous mobility.…”

“…We focus on the cases treated by Woodcock (2003) and Abowd and Stinson (2003), where the parameters on observable characteristics are treated as …xed, and where the pure person and …rm e¤ects and are random. 17 This speci…cation corresponds closely to the hierarchical models that are common in some other applied settings, for instance in the education literature.…”

“…Woodcock (2003) considers several in detail. Abowd and Stinson (2003) consider two more in the context of speci…cations that allow for multiple jobs in the same (i; t) pair and multiple measures of the dependent variable. The simplest parameterization is R = 2 " I N : This speci…cation is useful for making comparisons with the …xed-e¤ect estimation procedure.…”