Reduced Reparameterized Unified Model Applied to Learning Spatial Rotation Skills

Abstract

There has been a growing interest in measuring students in a learning context. Cognitive diagnosis models (CDMs) are traditionally used to measure students’ skill mastery at a static time point, but recently, they have been combined with longitudinal models to track students’ changes in skill acquisition over time. In this chapter, we propose a longitudinal learning model with CDMs. We consider different kinds of measurement models, including the reduced-reparameterized unified model (r-RUM) and the noisy input, deterministic-“and”-gate (NIDA) model. We also consider the incorporation of theories on skill hierarchies. Different models are fitted to a data set collected from a computer-based spatial rotation learning program (Wang S, Yang Y, Culpepper SA, Douglas JA, J Educ Behav Stat, 2016. https://doi.org/10.3102.1076998617719727 ) and we evaluate and compare these models using several goodness-of-fit indices.