Abstract
Statistics on adult education and training (AET) are often considered as insufficient because they fail to deliver a comprehensive and consistent picture of this field of education. This study addresses a specific problem in AET statistics that is varying participation rates of adults in AET depending on underlying data sources. We elaborate potential causes for deviations in survey design and the measurement of participation in sample based AET statistics with reference to the Total Survey Error (TSE) approach. Our analysis compares AET participation rates and patterns from four representative German surveys and reveals substantial differences in participation rates and mixed results for patterns of participation in AET. We find similar relationships for the influence of employment and educational attainment. The relationship with region, gender, and age shows to some extent deviations that conclude in contradictory statements on probabilities of participation. The discussion addresses consequences for the interpretation of survey results on AET participations and draws conclusions for the further development of AET statistics.