Adaptive Learning Analytics and Personalized Education Management: A Decision Support Framework for Academic Performance Optimization
Abstract
In the context of adaptive learning environments, researchers have argued that for education to evolve, that is, to personalize learning pathways and optimize outcomes, it must integrate the analytical foundations and decision structures in the educational management system. Our objective was to design an analytical framework to aid personalized management of academic performance. The proposed framework aims to point out that the integration of a multi-criteria hierarchy and a predictive modeling approach is a key innovation due to the fact that it has the most robust explanatory power in comparison to the use of a linear, univariate regression, or the most commonly used descriptive statistical analysis. The results indicated that the decision hierarchy and predictive modeling when the system began to process learner data and classify the patterns of behavioral and performance indicators, and respond to the contextual and cognitive states of the students were validated as a functional model. Correlation analysis was used to ensure that learners in the treatment group sample were very similar to those in the control group sample in terms of prior achievement and engagement levels. Regression results clearly confirmed this, with learner's personal engagement having three times the predictive power of baseline aptitude. A hierarchical integration of analytic hierarchy process, regression modeling, and correlation mapping concluded that only by aligning the adaptive mechanisms and decision processes in a way that monitors, supports, predicts, and enhances the learning experiences of students from all kinds of backgrounds can the personalized education system meet its goals for inclusivity and academic excellence, and sustainable performance growth.