What Learning Analytics Tells Us: Group Behavior Analysis and Individual Learning Diagnosis based on Long-Term and Large-Scale Data

• Published in: Educational Technology ＆ Society 2018-04, Vol.21 (2), p.245-258
• Main Authors: Zhang, Jia-Hua, Zhang, Ye-Xing, Zou, Qin, Huang, Sen
• Format: Article
• Language: English
• Subjects: Academic Achievement; Academic learning; Access control; Analysis; Analytics; Behavior; Behavior Patterns; Big data; CAI; Collaborative learning; Computer assisted instruction; Correlation; Correlation analysis; Cybersecurity; Data Analysis; Data Collection; Data mining; Diagnosis; Diaries; Distance learning; Educational activities; Educational aspects; Educational environment; Educational Research; Educational Researchers; Educational Resources; Educational Technology; Electronic Learning; Feedback (Response); Foreign Countries; Group Behavior; Group dynamics; Individualized instruction; Integrated Learning Systems; Interaction; Internet; Learning; Learning Analytics; Learning Processes; Network Analysis; Online Courses; Online education; Online learning; Process management; Resource management; Social factors; Social interaction; Social network analysis; Social Networks; Special Issue Articles; Standardization; Statistical Analysis; Statistical methods; Students; Teachers; Teaching; Teaching Methods; Technology; Technology assessment; Technology Uses in Education; Tests; Time Management; Use Studies; Visualization
• ISSN: 1176-3647; 1436-4522; 1436-4522

The practice and application of education data mining and learning analytics has become the focus of educational researchers. However, it is still a difficult task to explore the law of group learning and the characteristics of individual learning. In this study, the online learning logs of 1,088 students from 22 classes were analyzed from the aspects of their login behaviors, resource utilization, quizzes, interactive behaviors, and academic achievement. To address these issues, multiple methods, including statistical analysis, visualization social network analysis and correlation analysis, were used to analyze the process and results of online learning. The results reveal the characteristics of group behavior of online learners and highlight the key factors that influence the learning process and outcomes of individual learners. From the view of students, these factors include the length and allocation of online time, the effective utilization of resources, social interaction, online learning support and services, etc. From the perspective of teachers, the factors include the management of online teaching, the appropriateness of learning resources, the effectiveness of online intervention strategies, the accurate feedback for online learners, etc. Therefore, learning analysis technology can not only standardize the assessment of learning outcomes, but can also focus more attention on the standardization of learning process assessment. It also identifies the main factors that affect the online learning outcomes and the group characteristics of online learners. At the same time, it provides the learners with personalized learning diagnosis reports which can both help learners understand their own learning status and promote instructors' accurate teaching and reasonable evaluation.