Weights and Imputations in SHARE Wave 8

Abstract

In this chapter, we first use the different patterns of participation to define three subsamples of primary interest for the analysis of the SHARE data collected in Wave 8: CAPI, CATI and CAPI & CATI. We then describe the procedure used to construct calibrated cross-sectional and longitudinal weights for handling, respectively, problems of unit non-response and attrition in the CAPI subsample. Afterwards, we describe the model used to obtain multiple imputations of the missing values due to item non-response in the CAPI data.