Abstract
This article discussed the concept of validity from both historical and psychometric perspectives. We started by describing the so-called “trinitarian view of validity” in a detailed historical context. This view presumes that criterion-related validity, construct validity, and content validity constitute the “different types” of validity. We delineated several problems with this view and introduced the more recent “unitary concept of validity,” in which the construct validity is viewed as the central theme and subsumes all types of validation process. We next discussed several topics on practical measurement validation from psychometric perspectives. These topics include 1) incompatibility between broad-band scales and traditional psychometric measurement models, 2) distinctions and relations between within-person and between-person correlations, and 3) the impact of metric arbitrariness on the interpretation of data.
