VIEWER LEARNING PATTERNS AND ENGAGEMENT BEHAVIORS IN ONLINE EDUCATIONAL CONTENT CONSUMPTION: AN EMPIRICAL ANALYSIS WITH K-MEANS CLUSTER PROFILING AND SPEARMAN CORRELATION OF COMPLETION DETERMINANTS
DOI:
https://doi.org/10.70917/ijcisim-2026-2238Keywords:
online learning, course completion, K-Means clustering, Spearman correlation, learner personas, engagement behaviors, e-learning design, dropoutAbstract
This study analyzes the learning habits and engagement activities of users of online courses and examines the aspects of content design that are most influential in predicting the rate of course completion. A survey collected responses from 72 users and covered 58 variables related to demographics, content consumption, course participation, and learning satisfaction. K-Means clustering (k=3) was applied on the participants’ 28 behavior-related variables that pertained to course content, learning preferences, and course participation, and resulted in the identification of three empirical learner personas: Highly Engaged Achievers (Cluster A, n=10), Moderate Learners (Cluster B, n=59), and Disengaged Passive Learners (Cluster C, n=3). The separation of the clusters was confirmed by a Principal Component Analysis that described two dimensions that captured 38.2% of the variance. Spearman rank-order correlations confirmed that the strongest associations with learning satisfaction were completing practice exercises prior to taking the assessments (ρ=0.399, p<0.001), Reflection on Real-Life Application (ρ=0.385, p<0.001), and maintaining motivation through challenges (ρ=0.379, p<0.001). The weekly study time had the strongest overall association with learning satisfaction (ρ=0.436, p<0.001). A dropout rate of 61% was attributed mainly to online participants’ work/study conflicts (72.2%) and poor time management skills (61.1%). The study proposes, supported by the study findings, content design ideas for each identified learner cluster.