Dynamic Generalizations of the Rasch Model

Abstract

An overview is given of the different approaches to modeling dynamic phenomena within the framework of the Rasch model. Two main classes of dynamic phenomena are discerned: in the first class, change of the latent ability between test sessions is considered; in the second class, change within teat sessions is studied. The models of the first class are mainly covered by the detailed study given in Chapter 9, although some other models have been discussed in the literature. The second class comprises two parts: models which violate the assumption of local stochastic independence, and models which do not. This latter class is shown to be a generalization of mathematical learning models which were popular in the fifties and sixties.