Computational Psychometrics: Analyzing Educational Behavior in Learners Using Machine Learning

Abstract

The intersection of psychometrics and machine learning offers new insights into student learning behaviors and cognitive patterns. This paper presents a computational psychometrics model that applies machine learning techniques to analyze students’ engagement, problem-solving approaches, and knowledge retention. Using real-world educational datasets, we develop predictive models that assess learning outcomes, detect at-risk students, and personalize learning paths. Our findings highlight the advantages of AI-driven adaptive learning systems and provide a comparative analysis with traditional psychometric approaches, demonstrating improved accuracy and interpretability in assessing student performance.