Research on the Group Behavioral Characteristics of Learners in Blended Courses Based on Cluster Analysis

This study selected 247 learners who participated in the blended learning of Probability Theory and Mathematical Statistics at a certain university. Six variables, namely, duration of video viewing, interaction score rate, online test score rate, regular assignment score rate, in-class answering score rate, and final exam score rate, were used to reflect the learners' behavioral characteristics in online and offline learning. Agglomerative hierarchical clustering analysis, K-Means clustering algorithm, correlation analysis, and logistic regression analysis were employed to process the learning data. The research results indicate that the learner groups of this course can be divided into four categories: active learners, offline-preferred learners, average learners, and passive learners. There are differences in the six variables among the four learner groups. The interaction score rate, in-class answering score rate, and final exam score rate have a significant impact on the learners' learning outcomes.