The Measurement and Control of Beta Change

Applied researchers have long been confronted with the complex statistical problems associated with the measurement of change. A method frequently used in such analyses is the pretest-posttest research design. This approach incorporates the often overlooked assumption that the pretest and posttest scores are comparable. However, for the scores to be comparable, a common metric must exist between them. Distinguishing between three types of change in test scores (alpha, beta, and gamma), we present an original statistical procedure for measuring and controlling the confounding influence of beta change — i.e., the problem of scale recalibration in the minds of respondents.