ABSTRACT
One of the topics dealt with under generalizability theory (Cronbach, Gleser, Nanda, & Rajaratnam, 1972) has aroused relatively little attention among test developers and test users despite its importance to educational and psychological measurement. This topic pertains to the reliability or generalizability of a set of multiple scores used conjointly to assess certain attributes or traits of individuals. Such scores are usually obtained by one administration of a multiply-scored instrument such as the Wechsler Adult Intelligence Scale or the Minnesota Multiphasic Personality Inventory. They are interpretated by consideration of certain patterns or configurations or by the formation of weighted composites.
