Decision models for use with criterion-referenced tests

The problem of mastery decisions and optimizingcutoff scores on criterion-referenced tests is considered.This problem can be formalized as an (empirical)Bayes problem with decisions rules of amonotone shape. Next, the derivation of optimalcutoff scores for threshold, linear, and normal ogiveloss functions is addressed, alternately using suchpsychometric models as the classical model, thebeta-binomial, and the bivariate normal model.One important distinction made is between decisionswith an internal and an external criterion. Anatural solution to the problem of reliability andvalidity analysis of mastery decisions is to analyzewith a standardization of the Bayes risk (coefficientdelta). It is indicated how this analysis proceedsand how, in a number of cases, it leads to coefficientsalready known from classical test theory. Finally,some new lines of research are suggestedalong with other aspects of criterion-referenced testingthat can be approached from a decision-theoreticpoint of view.