Report , News of 20 December 2017

IAB-BAMF-SOEP Survey of Refugees in Germany: Revision of the Dataset

Johnny (Copyright)  Weltkarte Welt Weltkarten
Copyright: Johnny

The Institute of Employment Research (IAB), the Information Centre for Asylum and Migration of the Federal Office for Migration and Refugees (BAMF-FZ), and the Socio-Economic Panel (SOEP) at the German Institute for Economic Research (DIW Berlin) are currently working together on a longitudinal survey of refugees. The survey research institute Kantar Public (formerly TNS Infratest) has been commissioned to carry out the interviews. Kantar Public performs routine checks to ensure the quality of the survey. In preparation for the next round of interviews, Kantar Public found that one interviewer had not conducted interviews according to the established procedures. Six percent of the total household interviews were affected.

To address this issue, the research institutes involved in the study asked an expert in survey research methodology who is not part of the project, Prof. Dr. Jörg-Peter Schräpler of Ruhr University Bochum, to evaluate the dataset for statistical anomalies. In cases where additional statistical anomalies were identified, the respondents were contacted again. All of these respondents confirmed that their interviews had been conducted properly. These comprehensive checks provided no indication that further irregularities had occurred in the process of conducting the survey.

Kantar Public is currently assessing its quality control procedures and will take the necessary steps to improve quality control in its fieldwork processes and organization. Users of data from the IAB-BAMF-SOEP Survey of Refugees have been informed, and a revised version of the dataset will be provided to them soon.

After rechecking the results of studies published thus far based on the survey data, we have determined that the basic conclusions of these studies still hold after the data revision. The deviations in the revised dataset are either within the margin of statistical error, or they result in only slight differences of a few percentage points compared to the previously released version of the data. Details on the revised results can be found on the IAB website (in German).

The DIW publications affected by the data revision will be replaced with corrected versions in the course of 2018.