Using proxy measures of survey outcomes in post-survey adjustments

Abstract

"Sample survey nonresponse may introduce bias if non- respondents are not missing at random. Post-survey weighting adjustments are one strategy for addressing survey nonresponse. To be e®ective, the variables used in the adjustment need to be highly correlated with both response propensity and with the survey variable of inter- est. Interviewer observations made during the ¯eld pe- riod are increasingly available for respondents and non- respondents. Using data from the ESS this paper will examine the extent to which these observational data are correlated with response propensity and a selected set of survey variables." (Author's abstract, IAB-Doku) ((en))

Cite article

Kreuter, F., Lemay, M. & Casas-Cordero, C. (2007): Using proxy measures of survey outcomes in post-survey adjustments. Examples from the European Social Survey (ESS). In: American Statistical Association (Hrsg.) (2007): Proceedings of the Survey Research Methods Section, American Statistical Association (2007) : session 414: use of measures related to the survey variables in post survey adjustment for nonresponse, p. 3142-3149.