Social media has become a useful tool for learning and teaching due to its functions for knowledge sharing, such as documents exchange, virtual communication, and knowledge formation. Higher education institutions (HEIs) that recognise the value of social media and the importance of individual motivation have sought to encourage its use to bolster learning performance. Using social cognitive theory and connectivism theory, we developed a model that theorises the influence of social media and individual motivation (reputation and altruism) on knowledge sharing and learning performance among students in higher education. To test the model, we employed covariance-based structural equation modelling (CB SEM) on survey data that was collected using questionnaires from 407 students enrolled in the top 10 private universities in West Malaysia. Congruently, we established the utility of the Heterotrait-Monotrait (HTMT) ratio (which was traditionally confined to variance-based or partial least squares SEM) as a discriminant validity measure for CB SEM. As a result, our study empirically evidences that social media functions (documents exchange, virtual communication, and knowledge formation) and individual motivation (reputation) are core factors that HEIs can leverage to encourage knowledge sharing and improve learning performance among tertiary students. The paper concludes with a discussion of its implications, limitations, and future research directions.