Linking survey data with administrative employment data: The case of the German ALWA survey

This study evaluates the linkage of the ALWA survey with administrative labor market data from the German Federal Employment Agency. Exact, probabilistic or manual matches could be found for 86% of those respondents who consented to the linkage. Both the consent bias and the selectivity in linkage success are comprehensively examined based on survey and paradata. The results are informative for potential data users, for survey practice as well as for practitioners of data linkage. The highest consent rates are achieved by older, female and more experienced interviewers. The implementation of probabilistic record linkage after exact matching substantially increases the number of observations without introducing additional selectivity to the linked sample. Manual matching further increases the number of observation at the cost of more pronounced selectivity of the resulting sample. Selectivity of the linked data is mainly driven by the age, the immigrant background and the employment status of the respondents. JEL classification: C81