Surveys from inside: An assessment of unit nonresponse bias with internal criteria

Abstract

The article uses the so called “internal criteria of representativeness” to assess the unit nonresponse bias in five European comparative survey projects. It then goes on investigating several ideas why unit nonresponse bias might vary between surveys and countries. It is proposed that unit nonresponse bias is either caused by country characteristics or survey methodology. The empirical evidence presented speaks more in favour of the latter than of the former. Among the survey characteristics the features that strengthen the leverage to control interviewers’ behaviour have top priority