This longitudinal study explores the application of Machine Learning (ML) to predict academic success based on non-intellective factors. The study involved 1,439 students pursuing National Diplomas in engineering at a University of Technology (UoT) in South Africa. At the start of their first academic year, students completed the Emotional Skills Assessment Process (ESAP) and the Learning and Study Strategies Inventory (LASSI) to gather data on non-intellective factors. We demonstrate that an ensemble of baseline machine learning algorithms can predict a student’s Credit Accumulation Rate (CAR) using solely non-intellective data without considering any intellective (i.e., cognitive) pre- or post-admission variables, such as National Senior Certificate (NSC) results or Mathematics and Science scores. The CAR reflects academic progress by integrating both the duration of enrolment in higher education and the total credits earned. These variables were sourced from institutional student records. Using non-intellective factors to predict student success opens new avenues for bridging the gap between secondary and tertiary education. Based on our results, we recommend ways in which our framework could identify and support students at risk of academic failure and propose directions for further research.