SOEP-Core - 2018: Sampling, nonresponse, and weighting in Sample O

This paper provides details on the sampling design, field results and nonresponse, as well as population adjustments for the 2018 Sample O of the Socio-Economic Panel (SOEP). Sample O refreshes the SOEPCore sample by adding the residents of households in neighborhoods that receive government funding as part of the Soziale Stadt (Socially Integrative City) urban development and planning program. To provide a sample of households within these neighborhoods, the SOEP implemented a novel sampling approach integrating grid level census counts on the population and georeferenced building addresses. The approach is both cost efficient and lends itself as an alternative to the often criticized random route techniques. Obtaining nearly 1,000 interviews and panel consent of households in the neighborhoods was a demanding task in the fieldwork period. Nevertheless, nonresponse on the household level was driven primarily by neighborhood characteristics such as social composition and by whether survey staff were able to contact the household.