Response biases following a change of the survey data collection method shown on the example of the Austrian and German segment of the European Social Survey
Sprache des Vortragstitels:
Englisch
Original Tagungtitel:
5th International European Social Survey (ESS) Conference
Sprache des Tagungstitel:
Englisch
Original Kurzfassung:
As the European Social Survey European Research Infrastructure Consortium (ESS ERIC) considered to put its established face-to-face approach to survey data collection to rest by 2027, the Covid-19 pandemic offered an opportunity for a practical test of the self-completion mode that has been seen as a fitting replacement. While the majority of participating countries kept the face-to-face approach for the tenth wave of the ?European Social Survey? (ESS), several countries turned to a web- and paper-based self-completion mode with a marginally deviating questionnaire instead. Two of the nine countries switching modes for the tenth wave of the ESS have been Austria and Germany. The data collection started in Austria in August 2021 with Germany following suit in October 2021, already well into the pandemic. As such, the data may be regarded as a potential spotlight on the biases that may occur with the full transition to self-completion.
Alike to the population samples already underrepresented subsamples within the ESS (such as immigrants) show significant discrepancies in relation to the data collection modes. Online respondents both with and without a history of migration were younger, more educated, more frequently part of the active workforce and had better financial resources than paper questionnaire participants. Although balancing effects can be attested for both immigrant and population samples, biases are still noticeable in comparison to earlier waves of the ESS collected face-to-face as well to the Austrian ?Sozialer Survey Österreich? (SSÖ) and the German ?Allgemeine Bevölkerungsumfrage der Sozialwissenschaften? (ALLBUS) undergoing similar pandemic-related developments as the ESS. Aside from being younger and more educated in average, the participants of the tenth wave were also more often areligious and politically left-oriented. Specific to Germany a systematic withdrawal of respondents from the East has been observed at last.
Forschungs-