This study investigates the effects of WhatsApp social media learning group in the teaching and learning of mathematics with first year underprepared university students. The study was undertaken using a quantitative research method. It employed a randomised post-test-only with a non-equivalent groups design to investigate if there is a statistically significant difference between university students who studied through the traditional face-to-face lecturing approach and students who utilised a blended approach to learning mathematics. The study population comprised first-year university students enrolled for the university extended programme offered by all universities in South Africa. The sample consisted of 192 experimental groups and 341 control group students conveniently sampled from a university in South Africa. The main instruments used in this study were two standardised semester exams. These tests were checked and moderated by senior mathematics lecturers to ensure they conformed with the module content and satisfied all assessment policies of the university. The Cronbach alpha coefficient was used to measure the consistencies in these two-semester exams. The study's main finding showed no statistically significant difference in results between students who studied mathematics through a face-to-face lecturing approach and students who studied mathematics through a blended learning approach. The study concludes that the latter performed slightly better than students in the former, confirming that a WhatsApp learning group can be a viable alternative to the teaching and learning at the university when face-to-face learning is not possible, as for an example in the COVID-19 era. The study recommends that more profound research be conducted to identify and analyse positive indicators when learning is done through social media interaction.