Impact of UK Postgraduate Student Experiences on Academic Performance in Blended Learning: A Data Analytics Approach

Blended learning has become a dominant educational model in higher education in the UK and worldwide, particularly after the COVID-19 pandemic. This is further enriched with accompanying pedagogical changes, such as strengthened asynchronous learning, and the use of AI (from ChatGPT and all other similar tools that followed) and other technologies to aid learning. While these educational transformations have enabled flexibility in learning and resource access, they have also exposed new challenges on how students can construct successful learning in hybrid learning environments. In this paper, we investigate the interaction between different dimensions of student learning experiences (ranging from perceived acceptance of teaching methods and staff support/feedback to learning pressure and student motivation) and academic achievement within the context of postgraduate blended learning in UK universities. To achieve this, we employed a combination of several data analytics techniques including visualization, statistical tests, regression analysis, and latent profile analysis. Our empirical results (based on a survey of 255 postgraduate students and holistically interpreted via the Community of Inquiry (CoI) framework) demonstrated that learning activities combining teaching and social presences, and tailored academic support through effective feedback are critical elements for successful postgraduate experience in blended learning contexts. Regarding contributions, this research advances the understanding of student success by identifying the various ways demographic, experiential, and psychological factors impact academic outcomes. And in theoretical terms, it contributes to the extension of the CoI framework by integrating the concept of learner heterogeneity and identifying four distinct student profiles based on how they engage in the different CoI presences.