Prediction of social media effects on students’ academic performance using Machine Learning Algorithms (MLAs)

The global increase in social media (or social networking site) usage among college and university students, both outside and inside lecture halls, has open discussions on its effects on their academic performance. However, existing research probing its effects on students’ academic performance is not settled. Thus, a section of studies on social networking site usage and multitasking strongly supports a negative consequence on students’ academic performance. At the same time, other scholars have found little to no effect of social networking site usage on academic performance. The current study examined the association between the usage rate, the nature of use, usage in class, and level of exposure to a social networking site by students and their academic performance. Five-Hundred and Fifty (550) were sampled using a convenient sampling technique; an effort was made to address the divergent views on social media use and academic performance in the literature. Also, we predicted the student’s GPA using a predictive framework based on Decision Tree (DT) and Random Forest (RF) machine learning algorithms. The outcome of the current study shows that the rate of social networking site usage and its use in class partially affects students’ academic performance. © 2021, Beijing Normal University.