Answers are grounded in this paper and cite their sources.
Survey and register data each have their advantages and disadvantages. Record linkage of these different data sources can offset many of the disadvantages, while capitalizing on their combined strengths can unlock new analytical potential. This paper presents SOEP-CMI-ADIAB, a novel dataset based on a linkage of the Socio-Economic Panel (SOEP) with administrative data from the Institute for Employment Research (IAB). We describe the linkage process and examine selectivity of the data. Cross-validation analyses reveal that survey and administrative data largely correspond regarding individuals’ employment status and exhibit a high correlation in wages while discrepancies in occupational classifications are more frequent. We discuss how reporting practice and survey method may explain the detected differences. Lastly, we use the linked data to replicate an exemplary SOEP study and showcase the enhanced research potential of the linked dataset.
No takes yet. Share an insight, caveat, or question.
Keita et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: