Explanatory Cognitive Diagnostic Modeling Incorporating Response Times

This study proposes explanatory cognitive diagnostic model (CDM) jointly incorporating responses and response times (RTs) with the inclusion of item covariates related to both RTs. The joint modeling RTs intends provide more information for diagnosis while can be used predict parameters when calibration is not feasible in assessments or parameter estimation errors could too large due small sample sizes calibration. In addition, allows evaluation theories underlying test design development. Model explored using Bayesian Markov chain Monte Carlo (MCMC) method. A simulation conducted examine recovery proposed under different simulated conditions comparison alternative competing models. Further, application illustrated Programme International Student Assessment (PISA) 2012 problem-solving items RT data. results indicate that well recovered MCMC algorithm CDM holds promising applications digital-based assessments.