A Comparison of Joint and Stepwise Dynamic Cognitive Diagnostic Models

2026-04-17

AI Summary

This study addresses the parameter bias commonly introduced by stepwise estimation in longitudinal cognitive diagnostic models. To overcome this limitation, the authors propose a joint Bayesian dynamic cognitive diagnosis model that simultaneously estimates both measurement and state transition components within a unified framework. By integrating Bayesian inference, latent variable transition modeling, and Monte Carlo simulation, the proposed approach achieves more accurate recovery of state transition parameters, particularly under conditions of limited test length and small sample sizes. Empirical results demonstrate that, compared to bias-corrected stepwise methods, the joint modeling strategy yields significantly superior parameter estimation accuracy, especially in short-test and small-sample scenarios. This work thus offers a more robust modeling paradigm for longitudinal cognitive diagnosis.

Technology Category

• Cognitive Modeling & Cognitive Systems: Adaptive Behavior
• Intelligent Robots: State Estimation
• Reasoning under Uncertainty: Relational Probabilistic Models

Application Category

• User Modeling, Personalization and Recommendation: Studies of user behavior, including longitudinal effects of personalized systems
• Web Mining and Content Analysis: Models for Web evolution
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

To extend cognitive diagnostic models (CDMs) to longitudinal settings, stepwise approaches that integrate a CDM model with a latent transition model and covariates are widely used due to their flexibility. Previous research has shown that stepwise estimation can yield biased results, motivating classification-error correction as a means of improving inference over uncorrected stepwise procedures. In this study, we evaluate a unified Bayesian dynamic cognitive diagnostic model that jointly estimates measurement (item parameters, latent attribute profiles) and transition components (transition parameters) in longitudinal settings with covariates. We compare this joint approach with the bias-corrected stepwise latent transition CDM through a Monte Carlo study. Results demonstrate that joint modeling provides more accurate recovery of transition parameters, particularly under limited test length and sample size, underscoring its advantages for longitudinal diagnostic analysis and offering practical guidance for applied researchers.

Problem

• cognitive diagnostic models
• longitudinal analysis
• latent transition model
• parameter bias
• dynamic modeling

Innovation

• joint modeling
• dynamic cognitive diagnostic model
• Bayesian estimation
• latent transition model
• longitudinal CDM