An Assessment of the Current State of Dependent Interviewing in Household Surveys

Nancy A. Mathiowetz and Katherine A. McGonagle

Abstract:
Shifts over the past two decades in the method of data collection, from paper and pencil administration to computer assisted interviewing (CAI), have provided questionnaire designers with a large number of new design features, many of which can be used to reduce measurement error. These features include the expanded use of dependent interviewing. This article examines the historical roots of dependent interviewing and reviews current practices with respect to dependent interviewing within a computer assisted interviewing environment, examining the extent to which dependent interviewing is used in several longitudinal surveys and the empirical evidence with respect to the impact of dependent interviewing on data quality. Surprisingly, few studies have conducted empirical investigations related to either the design or the effects of dependent interviewing. The final sections of the article address the theoretical as well as practical issues related to the design and implementation of dependent interviewing and propose a research agenda to address gaps in our current knowledge of the effects of dependent interviewing.

Keywords:
Measurement error; human-computer interaction; seam effects.