Student Engagement and Academic Performance in Digital Learning Environments: A Learning Analytics Approach

Abstract

The increasing adoption of digital learning environments has transformed educational practices, raising important questions regarding the relationship between course design and learning outcomes. This study examines the influence of course-level characteristics on certification practices using a learning analytics approach. A structured dataset comprising 150 online courses was analysed to explore variations across subject areas, instructional formats, and engagement-related features. Descriptive analysis revealed moderate differences in course duration, instructor ratings, and certification rates across disciplines and formats. Interactive courses demonstrated relatively higher certification prevalence, while certain subject areas exhibited distinct patterns in course structure and credentialing. Inferential analysis, including chi-square tests and logistic regression modeling, indicated that none of the examined variables significantly predict certificate offering. These findings suggest that observable course design features, such as duration, instructional format, and assessment components, are not sufficient determinants of certification outcomes. The results highlight the complexity of digital learning environments, where outcomes are likely influenced by broader contextual and behavioral factors. The study contributes to the literature by emphasizing the limitations of relying solely on structural course characteristics and underscores the need for multi-dimensional analytical frameworks. Future research should incorporate learner-level engagement data and contextual variables to better understand the dynamics of digital education systems.