9th International Conference on ICT for Sustainable Development

Authors - Vinayak Hegde, Manoj S, Eshwar H S, Pallavi M S
Abstract - The research desire to better comprehend the complex interplay that leads to student success in college and universities is the main motive behind this research where three hypotheses on psychological well-being, social interaction, and academic performance were tested. It was observed, based on Pearson correlation results that there is a positive correlation coefficient of: r = 0.79, p < 0.001 between happiness, as measured by the Oxford Happiness Questionnaire Short Form, and student’s performance. As a result, the null hypothesis was rejected, showing that increased happiness is related positively to higher academic performance. Multiple regression also demonstrates that social support and family time share a positive effect on students’ happiness, as can be seen with factors and their positive coefficients. This confirms our conjecture that social support and family time hint towards better or worse psychological wellness. Finally, using Multi-Layer Perceptron neural network classifier, the model is 79% accurate in discerning academic successes or failures by evaluating past trauma and substance addiction of students. As evident in the confusion matrix, the model can accurately categorize 54 out of 68 observations, with 10 false negatives and 4 false positives. These mathematical representations suggest that the model can be trusted in marking the association between past trauma, substance addiction and academic performance. Overall, the results indicate that student wellbeing and social factors are central to academic outcomes. They also provide a measure of student indicators that can inform the creation of intervention programs to enhance student wellness and academic progress.