Using SAS PROC IRT for Multidimensional Item Response Theory Analysis

Abstract

ABSTRACT Statistical Analysis Software (SAS) is a widely used tool for data management analysis across a variety of fields. The procedure for item response theory (PROC IRT) is one to perform unidimensional and multidimensional item response theory (IRT) analysis for dichotomous and polytomous data. This review provides a summary of the features of PROC IRT specifically for multidimensional data with examples provided for simple structure data, complex structure data, and bifactor data. Instructive examples for dichotomous data (using the Rasch and 2-parameter logistic models) and polytomous data (using the graded response model) are given. Explanations of the syntax are also presented.