A PREDICTIVE PLS-SEM MODEL OF MATHEMATICS ACHIEVEMENT: IMPLICATIONS FOR EDUCATIONAL DATA ANALYTICS AND INTELLIGENT DECISION SUPPORT

Abstract

This study develops and validates a predictive PLS-SEM model that explains mathematics achievement through the interaction of students' math mindsets, grit, and learning disposition, providing an analytical framework for educational decision-making. A positive attitude toward math is crucial as it develops students' perseverance, influences their approach to learning, and greatly improves their performance in the subject. This research ultimately investigates the indirect effects of students' perseverance and attitudes on the link between their mathematical mindsets and their performance. The participants in this research comprised 360 high school students selected from ten different schools within the Pampanga Division, representing various types of schools. These students were chosen using stratified proportional sampling to ensure a fair and accurate representation across different school categories. The results showed that students fixed and growth mindsets related to mathematics significantly impact their levels of perseverance, attitude, and overall achievement in the subject. Furthermore, the findings revealed that students' perseverance and attitudes towards mathematics have both direct and indirect effects on their performance in math. In light of these results, it can be concluded that mathematical perseverance and attitude play an important mediating role in the relationship between students' mindsets and their academic success in mathematics. A growth mindset by itself is insufficient, perseverance and a positive attitude are vital for transforming beliefs into consistent academic effort and achievement. Therefore, educators should promote perseverance and attitude alongside mindset to enhance long-term achievements in mathematics. The findings extend beyond mathematics education by contributing to the fields of computer information systems, predictive analytics, and intelligent decision support. The validated PLS-SEM model demonstrates how advanced computational modeling techniques can transform educational data into predictive insights that support evidence-based decision-making and performance optimization. As a data-driven analytical framework, the model has potential applications in educational information systems, learning analytics platforms, and academic management systems for identifying students at risk, guiding targeted interventions, and enhancing institutional decision-making. These contributions highlight the growing role of computational intelligence and predictive analytics in addressing complex educational challenges through information systems and analytical modeling.