The Psychometrics of Order Effects

Abstract

The chapters in this volume and its predecessor (Hippler, Schwarz, & Sudman, 1987) attest to a broad interest in item-order effects. A better understanding of their causes and implications has come from perspectives in cognitive psychology, social cognition, and survey methodology. The importance of contemporary psychometric theory for evaluating order effects has been overlooked. We believe that recognition of psychometric contributions can help researchers better understand item-order effects. One obstacle faced thus far has been that many of the latest psychometric developments have not been readily accessible (or comprehensible) to researchers in this area. The goal of this chapter is to show how such contributions logically relate to testing and interpreting item-order effects. In doing this, we define some necessary preconditions that should exist for the effects obtained to date, and for those obtained in future research, to be interpreted unambiguously.