Modelling complex cognitive and psychological outcomes in, for example, educational assessment led to the development of generalized item response theory (IRT) models. A class of models was developed to solve practical and challenging educational problems by generalizing the basic IRT models. An IRT model can be used to define a relation between observed categorical responses and an underlying latent trait, such as, ability or attitude. Subsequently, the latent trait variable can be seen as the outcome in a regression analysis. That is, a regression model defines the relation between the latent trait and the set of predictors. The combination of both models, a regression model imposed on the ability parameter in an IRT model, can be viewed as an extension to the class of IRT models.
|Title of host publication||New developments in categorical data analysis for the social and behavioral sciences|
|Editors||L. Andries Ark, Marcel A. Croon|
|Place of Publication||Mahwah, NJ [etc.]|
|Publication status||Published - 2005|