Abstract

Increasing cost pressure has led to population surveys being conducted less frequently with interviewer-administered survey modes (CAPI, CATI) and increasingly in more cost-effective self-administered survey modes (PAPI/CAWI). But changes in the coverage of individual survey modes can also suggest a mode change.However, a mode change is potentially problematic, particularly with existing panels, since mode-specific effects on response behavior and data quality are possible. In order to be able to identify and quantify the effects of a mode change, however, an experimental design is necessary which, among other things, allows the type and extent of error sources (nonresponse, measurement error) to be analyzed. Such an experimental design is planned for the sixth and seventh waves of the Linked Personnel Panel (LPP).This study could identify savings potentials if the findings should show that a mode change from the interviewer-administered mode CATI to a more cost-effective self-administered mode (CAWI) can be carried out without significant losses in quality.