A General Form of Rasch's Extended Logistic Model for Partial Credit Scoring

Abstract

The purpose of this article is to recast the Rasch model in a generalizable form allowing for broader insights into the model's applicability beyond partial credit scoring. Applications of an extended logistic model to the study of dependencies within subtests of dichotomously scored items and the study of questionnaires in the Likert tradition are identified. One possible outcome of research using the extended logistic model is a better understanding of the cognitive processes in problem solving and other learning tasks.