Data Quality of Cross-National Surveys

Cross-national surveys provide the opportunity for research using secondary data. One issue encountered by researchers is the identification of data quality of survey research in order to obtain reliable findings. The issue is even more critical when using data from cross-national or cross-cultural surveys, as each of the cross-national survey projects has specific research purposes and approaches. When examining data quality from the perspective of total survey errors, survey aspects, such as sampling design, questionnaire design, mode of data collection and survey agency, may have impact on data quality. It is common that probability sampling is required for most of cross-national surveys. A standard source questionnaire is often used, although some cross-national surveys allow different rating scales or response labels in corresponding to the preference of local researchers. Mode of data collection, other other hand, may be more flexible depending on the social, political, and economic context of the participating countries. It can also be found that some countries adopted a mixed-mode design in cross-national surveys, either for experiment or simply for reducing the cost. This session aims to provide and foster the exchange among provider and researchers using cross-national surveys. The focus of this session is the methodological challenges in relation to data quality of cross-national surveys. The use of total survey error approach is especially welcome.