Cognitive diagnosis modelling incorporating item response times

Area of Science:

• Educational Measurement
• Psychometrics
• Cognitive Psychology

Background:

• Traditional cognitive diagnosis models often overlook response time data.
• Integrating response times (RTs) can provide richer diagnostic information.
• Current methods may lack precision in attribute profiling.

Purpose of the Study:

• To propose a joint modeling approach for cognitive diagnosis that simultaneously uses item responses and response times (RTs).
• To extend the deterministic input, noisy 'and' gate (DINA) model to incorporate RTs for enhanced diagnostic feedback.
• To evaluate the performance of the proposed joint model in attribute and profile classification.

Main Methods:

• Developed an extended DINA model integrating item responses and RTs.
• Employed the Bayesian Markov chain Monte Carlo (MCMC) method for parameter estimation.
• Utilized PISA 2012 mathematics data for initial analysis and simulation studies.

Main Results:

• The proposed joint model demonstrated good parameter recovery using the MCMC approach.
• Incorporating RTs into the DINA model significantly improved attribute and profile correct classification rates.
• The joint model yielded more accurate and precise estimations of model parameters compared to traditional methods.

Conclusions:

• Joint modeling of responses and RTs offers a more refined approach to cognitive diagnosis.
• The extended DINA model provides a viable framework for integrating response speed information.
• This methodology enhances the accuracy and precision of cognitive attribute assessment.