Predicting Individual Learning Growth through Learning Diagnosis: Assessing with Random Effect Diagnostic Classification Multilevel Growth Curve Model

Learning diagnosis is essential in effective education, with formative assessments shown to significantly enhance academic performance. Diagnostic classification models have been developed to assess the learning status of students and provide remedial instructions. However, the impact of mastery or non-mastery of specific attributes on long-term learning development remains uncertain. If certain non-mastered attributes hinder the growth of mathematical ability, early intervention becomes essential. In this study, we developed a random effect diagnostic classification for a multilevel growth curve (RDC–MGC) model to identify the specific effects of attribute mastery on the individual-level mathematics ability growth. The model was applied to arithmetic test data from second- to sixth-grade elementary students. Diagnosis was conducted in the second grade, and the effects of mastery on mathematics ability growth from the third to sixth grades were assessed. The results showed that attribute mastery in the second grade influenced both the intercept and slope of individual ability growth, highlighting the importance of early-stage diagnosis in supporting mathematical development. Potential extensions of the proposed RDC–MGC model are also discussed.