ABSTRACT
This chapter focuses on the passive type of interviewer effects and discusses the basic statistical model used in the research tradition of analyzing interviewer effects. It argues that this basic model is not suitable for comparing interviewer effects across different respondent groups. The chapter shows how one can extend the basic model to allow for such investigations. It demonstrates that lower-educated respondents elicit larger interviewer effects. The interviewer effects express the variability between interviewers after adjusting for these respondent characteristics. The chapter focuses on two procedures based on the basic model that allow for a direct investigation of the relationships between respondent characteristics and the occurrence of interviewer effects as measured by the intra-interviewer correlations. The relationship between intra-interviewer correlations and education level was formally evaluated by the specification of a cross-classified multilevel model with the intra-interviewer correlations (expressed as percentage of explained variance) as the dependent variable.
