Multidimensional Item Response Theory
Description
Analysis of dichotomous and polytomous response data using latent
trait models under the Item Response Theory paradigm. Exploratory models
can be estimated via quadrature or stochastic methods, a generalized
confirmatory bi-factor analysis is included, and confirmatory models can be
fit with a Metropolis-Hastings Robbins-Monro algorithm which may include
polynomial or product constructed latent traits. Multiple group analysis
and mixed effects designs may be performed for unidimensional or
multidimensional item response models for detecting differential item
functioning and modelling item and person covariates.