Investigating How the Perceived Utility of Artificial Intelligence and Social Media Influence Students’ Academic Performance: The Mediating Roles of Student Engagement and Mental Health

This study examines how artificial intelligence (AI) tools (e.g., ChatGPT, tutoring systems) and social media platforms (e.g., LinkedIn, YouTube, Coursera) influence students’ academic performance, focusing on the mediating roles of mental health and student engagement. Drawing on Social Learning Theory, we conceptualize AI and social media not merely as resources but as socio-technical learning environments that shape academic outcomes through psychological and behavioral mechanisms. Using survey data from 440 students across five public universities in Bangladesh, structural equation modelling (PLS-SEM) was employed to test the hypothesized relationships. The findings reveal that both the perceived utility of AI and social media positively affect academic performance, but their impact is indirect: improved outcomes depend on enhanced engagement and preserved mental health. Mediation analyses confirm that engagement and mental health jointly explain how digital tools foster or hinder learning success. These results reconcile conflicting findings in prior research by demonstrating that AI and social media are neither inherently beneficial nor detrimental; instead, their effects depend on student well-being and active participation. The study advances theory by extending Social Learning Theory into digitally mediated contexts. It contributes practical insights for educators and policymakers to design balanced digital ecosystems that promote academic achievement and student well-being.