An Empirical Bayesian Approach to Item Banking

Abstract

A procedure for the sequential optimization of the calibration of an item bank is given. The procedure is based on an empirical Bayesian approach to a refor mulation of the Rasch model as a model for paired comparisons between the difficulties of test items in which ties are allowed to occur. First, it is shown how a paired-comparisons design deals with the usual in completeness of calibration data and how the item pa rameters can be estimated using this design. Next, the procedure for a sequential optimization of the item pa rameter estimators is given, both for individuals re sponding to pairs of items and for item and examinee groups of any size. The paper concludes with a dis cussion of the choice of the first priors in the proce dure and the problems involved in its generalization to other item response models.