A New IRT (Item Response Theory)-Based Continuous Conjunctive Diagnostic Modeling Approach

• In Event: Roundtable Session 13
  In Roundtable Session: Cognition in Assessment and Cognitive Diagnostic Modeling

Abstract

This study proposes a new modeling approach for
implementing cognitive diagnosis in large scale testing, namely, the IRT-based continuous conjunctive latent trait model, which combines
the features of multidimensional IRT (MIRT) model and diagnostic classification model (DCM) into a single model. Two models are proposed within this framework, one is the continuous extension of
the NIDA model, and the other can be viewed as an extension of the Embreston's (1987) multicomponent latent trait model (MLTM;
Embreston, 1985,1997; Whitely, 1980). Model estimation methods for both new models will be developed and their performance will be
evaluated via extensive simulation studies.

Authors

• Chun Wang, University of Minnesota

• Hua-Hua Chang, University of Illinois at Urbana-Champaign

• Jeffrey Douglas, University of Illinois at Urbana-Champaign

• Keith A. Boughton, CTB/McGraw-Hill LLC