Web Survey Bibliography
Differences in answers to Internet and traditional surveys can be due to selection, mode, or context and question wording effects. We exploit unique experimental data to analyze mode and context effects controlling for arbitrary selection. The longitudinal Health and Retirement Study (HRS) surveys a random sample of the US 50+ population, with CAPI or CATI core interviews for the same respondents 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 regular and Internet survey answers of the same people, we analyze mode, context and wording 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. Further analysis suggests that these differences are the result of context and question wording effects rather than of a pure mode effect.
CentERdata (abstract)
Web survey bibliography - Kapteyn, A. (7)
- Use of Internet panels to conduct surveys; 2015; Kapteyn, A.; Liu, H., D.Hays, R. D.
- Recruiting an Internet Panel Using Respondent-Driven Sampling; 2014; Schonlau, M., Weidmer, B., Kapteyn, A.
- Methodology of the RAND Continuous 2012 Presidential Election Poll ; 2012; Kapteyn, A., Meijer, E., Weerman, B.
- Conducting Respondent Driven Sampling on the Web: An Experimental Approach to Recruiting Challenges; 2011; Kapteyn, A., Schonlau, M.
- Framing Effects and Expected Social Security Claiming Behavior; 2011; Brown, Je., Kapteyn, A., Mitchell, O. S.
- Mode and Context Effects in Measuring Household Assets; 2011; van Soest, A., Kapteyn, A.
- Beyond Demographics: Are ‘Webographic’ Questions Useful for Reducing the Selection Bias...; 2007; Schonlau, M., van Soest, A., Kapteyn, A.