Evaluation of teaching effectiveness in data analysis courses using a behavioural big data model

Abstract

This study proposes a behavioural big data model for evaluating teaching effectiveness in data analysis courses across primary, secondary, and higher education levels. The framework integrates learning management system data, classroom engagement indicators, and student interaction behaviours to provide a comprehensive understanding of how teaching strategies influence learning outcomes. At the primary level, the model captures early learning patterns such as task attention, problem-solving attempts, and basic data reasoning through gamified digital activities. For secondary students, behavioural indicators – including learning persistence, collaboration, and response accuracy –are used to assess the development of analytical thinking and computational skills. Machine learning and statistical techniques, such as clustering, regression, and correlation analysis, identify patterns linking teaching approaches with student performance and motivation. With a predictive accuracy of 89%, the model demonstrates strong adaptability across age groups. Findings show that interactive, feedback-rich, and project-based learning environments significantly enhance students’ comprehension and retention of data analysis concepts.