ABSTRACT
A generalized partial credit model (GPCM) was formulated byMuraki (1992). Based on Masters’ (Volume One,Chapter 7) partial credit model (PCM), the GPCM relaxes the assumption of the uniform discriminating power of the test items. However, the difference between these two types of item response models is not only the parameterization of item characteristics but also the basic assumption about the latent variable. An item response model is viewed here also as a member of a family of latent variable models, which includes such models as the linear or nonlinear factor analysis model, the latent class model, and the latent profile model as well (Bartholomew, 1987).
