Student dropout is a common problem at many tertiary institutions, with some reporting as high as 50% dropout among first years. The phenomenon affects the student and the student adversely in several ways. There are many factors contributing to student dropout such as socioeconomic position and family dynamics. This study explores the performance of machine learning methods of predicting student dropout. Five commonly used machine learning models, namely Gradient Boosting (GB), Support Vector Machine (SVM), Random Forest (RF), Logistic Regression (LR), and Artificial Neural Network (ANN) were considered. ANN exhibited superior performance in terms of Recall, whereas RF outperformed the other models in terms of accuracy and F1 score. ML can be applied to detect students at risk of dropping out. Once we detect student with a higher dropout predisposition, there is an opportunity to administer intervention measure to minimize the dropout rate.