Systemic Reform Driven by Educational Big Data: A Mixed Study of Multimodal Learning Analytics and Evidence-Based Decision Making

With the rapid advancement of educational big data and learning analytics technologies, data-driven systemic transformation in education has become increasingly viable and impactful. As digital learning environments proliferate globally, the sheer volume of data generated offers unprecedented opportunities to understand and optimize the learning process. This study explores how multimodal learning analytics (MMLA) can support evidence-based decision making in educational transformation efforts. Using a comprehensive mixed-methods approach, we systematically analyze quantitative data extracted from digital learning platforms alongside in-depth qualitative insights gathered from educators, administrators, and policymakers. The integration of these diverse data sources allows for a holistic understanding of the complex educational ecosystem. The findings highlight how MMLA enables more accurate learner profiling, facilitates real-time feedback mechanisms, and promotes highly personalized instruction tailored to individual student needs. Furthermore, the empirical evidence generated through this approach plays a crucial role in informing strategic policy adjustments and driving systemic improvements across educational institutions. By bridging the gap between raw data and actionable pedagogical strategies, this research provides a robust, practical framework for integrating data-intensive tools into institutional decision-making processes. Ultimately, the proposed framework seeks to significantly enhance teaching effectiveness, boost learner engagement, and foster greater educational equity by ensuring that interventions are precisely targeted and empirically supported. This study underscores the transformative potential of combining advanced analytics with pedagogical expertise to shape the future of modern education.