Methodological Research for Longitudinal Surveys

Abstract

Longitudinal surveys face a number of unique methodological challenges in addition to having to deal with all the issues faced by other surveys. This presentation will focus on those aspects of methodology that are unique to longitudinal surveys or that have rather distinct properties in the case of longitudinal surveys. I will outline those aspects and, for each, will present a brief overview of methodological research in the area, focussing particularly on recent and current research and outlining a desirable future research agenda. Topics discussed will include sample design, between-wave intervals, keeping track of sample members and maintaining co-operation, adjusting for panel attrition, panel conditioning, and instrument design to minimise measurement error in measures of change.