The Effects of Dependent Interviewing on Responses to Survey Questions

Abstract

Dependent interviewing techniques are becoming widely used on longitudinal surveys (Mathiowetz and McGonagle, 2000). However, there have been few systematic studies of the effects on survey data and estimates based upon them. This presentation will describe an empirical study designed to enable comparisons of two alternative approaches to dependent interviewing (proactive and reactive) with traditional independent interviewing. This is the first large-scale experimental comparison of proactive and reactive dependent interviewing. Preliminary results will be presented, relating to modules of questions about sources of income and occupation. The study design also enables identification of the characteristics of respondents whose responses are sensitive to the mode of interviewing.