Conferences

Among the range of approaches proposed to estimate nonresponse bias and correct for it (Groves 2006) several of these are applied in past rounds of ESS. These approaches are: comparing sample statistics with population statistics and poststratification weighting; using information from reluctant respondents based on converted refusals; asking a small set of crucial questions at occasion of first contact (and refusal) or in a period after the main survey; and collecting information of the live environment of the sampling units by means of observation; a follow up survey among nonrespondents. Other approaches may be used in other cross-nation surveys. The benefits, costs, challenges, and weaknesses of these approaches in single country surveys or in surveys with a small number are known. What are the specificities, the requirements, and weaknesses in cross-nation surveys covering a large number of countries and aimed to produce data that are useful for cross-nation comparisons and models that include all of the participating countries. We are now in a stage where the theoretical discussions about opportunities and challenges can be completed by real cross-nation data. The contributions of this section are expected to provide critical reflection of approaches to nonresponse bias used in cross-nation surveys and to find solutions for the problems that are met.

References

Billiet, J., Phillipens, M., Fitzgerald, R. & Stoop, I. (2007). Estimation of Nonresponse Bias in the European Social Survey: using Information from Reluctant Respondents. Journal of Official Statistics, vol. 23 (2), pp. 135-162.
Groves, R.M. (2006). Nonresponse Rates and Nonresponse Bias in Household Surveys. Public Opinion Quarterly, 70, 646-675.