Arie Kapteyn (RAND)
Differences in answers in Internet and traditional surveys can be due to selection, mode, or context effects. We exploit unique experimental data to analyze mode and context effects controlling for arbitrary selection. The Health and Retirement Study (HRS) surveys a random sample of the US 50+ population, with CAPI or CATI core interviews once every two years. In 2003 and 2005, random samples were drawn from HRS respondents in 2002 and 2004 willing and able to participate in an Internet interview. Comparing core and Internet survey answers of the same people, we analyze mode and context effects, controlling for selection. We focus on household assets, for which mode effects in Internet surveys have rarely been studied. We find some large differences between the first Internet survey and the other three surveys which we interpret as a context and question wording effect rather than a pure mode effect.