Cross-national surveys that consider mixed mode data collection face special problems. Because national “survey climates” differ according to respondent expectations, fieldwork expertise, and mode penetration, most cross-national surveys cannot implement the same mixed mode design in all participating countries. Yet differences in data collection designs pose a threat to measurement equivalence, potentially confounding country differences with errors introduced by country-specific survey practices.

In general, we can distinguish two kinds of cross-national mixed mode survey: in across-country mixed mode surveys, each country uses only a single mode of data collection, but not every country uses the same mode; in within-country mixed mode surveys, one or several countries use more than one mode within their own national survey. In either case, selection effects and measurement effects pose a threat to equivalent measurement. Several established cross-national surveys, such as the World Values Surveys and ISSP, are across-country mixed mode surveys. Yet the implications of this are rarely, if ever, considered or tested.

This presentation will outline the problems that cross-national mixed mode surveys face. We will summarize the evidence available, using results from the European Social Survey Mixed Mode Methodology Programme, which has been conducting cross-national research on mixed mode designs since 2003. We shall also identify gaps in the current scientific knowledge; in particular, the question whether mode effects might differ across countries has rarely been addressed. To conclude, we discuss the implications of our analysis for the design of cross-national mixed mode surveys.