More complex IRT models

This chapter focuses on more complex Item Response Theory (IRT) models, relaxing some of the assumptions of models presented in Chapter 7. We discuss IRT models for polytomous data, including the Generalized Partial Credit Model and Graded Response Model for ordinal items, and Nominal Response Model (NRM) for multiple-choice items. We then focus on multidimensional IRT models extending the models for binary and polytomous items. Both the exploratory and the confirmatory models are discussed. Finally, we present estimation methods for the more complex models and algorithms which may speed up the computations.