Assessing the Effect of Data Collection Mode on Measurement

This paper discusses the challenges in assessing the effects of mixed modes on measurement. We briefly review theories about why different modes of data collection can lead to differences in survey responses and statistical methods typically used to assess mode effects. We then discuss the challenges, including which mode effects are identified, how to test for mode effects, and whether these would affect substantive conclusions. The issues raised are illustrated with examples from the European Social Survey, which is conducting a programme of experimental research to inform decisions about whether to use mixed modes of data collection. The paper concludes with general implications for mixed modes research.