Reliability and Validity Assessment: New Approaches

The quality of measurement is considered the sine qua non of research in all sciences. In the social sciences, the concern with measurement quality is critical because theoretical constructs are unobserved and often elusive. In order to evaluate the quality of measurement in these fields, it is necessary to, first, develop a model that specifies the linkage between observations and the latent variable(s) of interest; second, to develop a research design that permits the estimation of parameters of such a model; and third, to develop an interpretation of these parameters that is consistent with the concepts of measurement quality. Several traditions exist for the formulation of such models that relate observed to unobserved variables, and the concepts of reliability and validity are often used to interpret the quality of measurement from such designs. This article discusses the traditional techniques of reliability and validity assessment and briefly presents some new approaches to evaluating the properties of individual measures that are aimed at measuring theoretical constructs of interest.