|
Abstract:
|
With the increase in online and electronic learning, there is a need for tools to understand how students learn. One such interest is in classifying individuals into groups of similar learning trajectories. Cognitive diagnosis models (CDM) are used to classify students into groups based on latent skills estimated from item response data. However, CDMs traditionally have no capability for modelling dynamic, evolving skills. We propose a class of CDMs with an exploratory factor analysis (EFA) model to characterize latent learning trajectories over time, and to classify individuals into groups of similar learning trajectories. Parameter recovery of this model was evaluated through a Monte Carlo simulation study. The study results indicate that the proposed model provides good convergence rates and parameter recovery. The model was then fit on a working memory dataset, model fit was evaluated through posterior predictive model checking.
|
