Leveraging Student Feedback and Data-Driven Modeling to Enhance Academic Outcomes: A Correlational Study in Computing Education

This study examines student course feedback within the College of Computing Studies at EARIST, focusing on the interconnection between course difficulty, delivery method, and grade performance. Using a quantitative correlational research design, the study employs sentiment analysis, Pearson correlation, multiple linear regression, and mediation analysis to identify patterns in student responses. Significant findings show that students with high satisfaction and engagement tend to achieve better academic results. Conversely, those who perceive courses as burdensome face challenges that hinder learning. Notably, students enrolled in online classes consistently outperform those in traditional face-to-face settings. It can be attributed to online learning environments\' flexibility and self-paced nature. The study also emphasizes that student feedback, which encompasses satisfaction, engagement, and perception, does more than merely reflect course difficulty; it actively influences academic outcomes. A positive feedback loop is observed, where bigher grades lead to more favorable evaluations, reinforcing the importance of student experience. Data-driven models, particularly neural networks, were used to predict student sentiment, offering valuable insights for refining teaching strategies. These findings underscore the importance of developing balanced instructional design that incorporates adaptive learning and targeted support for students who struggle. The research provides actionable recommendations for educators and institutions to improve course structures and teaching practices. By aligning educational approaches with diverse student needs, schools can enhance satisfaction and academic success in today\'s evolving higher education Landscape.