With various digital technologies increasingly integrated into higher education, understanding how students engage with such technologies has become vital. There are different ways to measure student engagement; however, self-reported measures such as questionnaires are predominantly used to understand student engagement. In contrast, this study utilises an Educational Data Mining (EDM) technique to discover students' engagement patterns (N = 54) in a Blended Learning (BL) environment. SimpleKmeans clustering technique is applied to students' learning data obtained from a BL environment and patterns of student engagement are identified. Findings suggest students engage differently with learning resources as students have different engagement patterns based on low, moderate and high engagement levels. The analysis of student generated data can help provide timely interventions that enhance student engagement. Furthermore, educators should concentrate on best practices in order to engage students bearing in mind that students engage differently with learning resources.