Abstract
Four measurement models for use in performance evaluation (norm, criterion, domain, and objectives-referenced measurement) are presented and the contingent relationships among them are discussed. An argument is made for the use of criterion-referenced measurement in performance testing. Literature on two major problem areas in criterion-referenced measurement, reliability and validity, is briefly reviewed; and a recent example of criterion-referenced performance test development in an applied training context is described.
Get full access to this article
View all access options for this article.
