This thesis contributes to the limited body of empirical evidence on nonresponse in establishment surveys by examining unit and item nonresponse, their consequences, and the methods used to address nonresponse before, during, and after data collection. The main objectives are to investigate the prevalence of nonresponse, nonresponse bias, and subgroup participation patterns. Furthermore, the thesis explores the potential of different mode esigns and question clarification information to questions in mitigating unit respectively item nonresponse and nonresponse bias. Further, it evaluates the effectiveness of nonresponse adjustment procedures based on innovative big data and machine learning techniques in reducing nonresponse bias.
Dieser Eintrag ist Teil der Universitätsbibliographie.
Das Dokument wird vom Publikationsserver der Universitätsbibliothek Mannheim bereitgestellt.