Schräpler, Jörg-Peter Schupp, Jürgen Wagner, Gert G.
Year of Publication:
[Journal:] Journal of Official Statistics [ISSN:] 0282-423x [Volume:] 26 [Issue:] 2 [Pages:] 239-269
This article examines the implications of moving to Computer-Assisted Personal Interviewing (CAPI) for data quality by analyzing the transition from Paper-and-Pencil (PAPI) to Computer-Assisted Personal Interviewing (CAPI) on a subsample of the German Socio-Economic Panel (SOEP) conducted using an "experimental design" in Wave 1. The 2,000 addresses for the sample E of SOEP were split into two subsamples with the same structure, E1 and E2, using twin-sample points. Each of the 125 sample points contained 16 addresses (8 for E1 and 8 for E2) and with each interviewer alternating in the first wave between PAPI and CAPI mode. In the subsequent waves, the PAPI mode was partly replaced by CAPI. With this design we are able to control for possible interviewer effects in the analysis of mode effects in Wave 1. The article assesses whether any mode effects or mode-related compound effects are apparent for the response rate. Within the data, we examine monetary dimensions such as gross income and item and unit nonresponse rates.
CAPI mode effects data quality interviewer effects