Skilled or unskilled - Improving the information on qualification for employee data in the IAB Employee Biography

Abstract

"This FDZ-Methodenreport describes the implementation of a deterministic imputation model for education in the Employee History (Beschäftigtenhistorik (BeH)) of the Institute for Employment Research (IAB). The imputation model was originally invented by Fitzenberger, Osikominu, and Völter, but now was adapted for the current Employee History. First the extent of the data quality problem is sketched out and different variants of the solving imputation model are evaluated.<br> Then the implementation of the selected variant in the Employee History is described in detail. Afterwards some descriptive graphs show the resulting imputed data. In order to give the user a good impression about the goodness of the selected imputation model the quality of the imputed information is tested with some regression models. The regression results indicate, which variables in the Employee History have an influence on the quality of the imputed values. This allows an assessment of its usability for certain subgroups in the data." (Author's abstract, IAB-Doku) ((en))

Cite article

Thomsen, U., Ludsteck, J. & Schmucker, A. (2018): Skilled or unskilled - Improving the information on qualification for employee data in the IAB Employee Biography. (FDZ-Methodenreport 09/2018 (en)), Nürnberg, 22 p. DOI:10.5164/IAB.FDZM.1809.en.v1