Rational herd behavior in online learning: Insights from MOOC

• Published in: Computers in human behavior Vol. 92; pp. 660 - 669
• Main Authors: Wang, Wei, Guo, Lihuan, Sun, Rui
• Format: Journal Article
• Language: English
• Published: Elmsford Elsevier Ltd 01.03.2019; Elsevier Science Ltd
• Subjects: CAI; Computer assisted instruction; Distance learning; Herding behavior; Internet; MOOC; Online instruction; Online learning; Rational herding; Websites; MOOC; Herding behavior; Rational herding; Online learning
• ISSN: 0747-5632; 1873-7692
• DOI: 10.1016/j.chb.2017.10.009

With the emergence of online learning, many similar courses are offered on the Internet. The Internet is an essential tool for improving learning efficiency by analyzing how learners select courses; it is also a convenient channel to extract learner preferences. We adopt dynamic learning data from a popular Massive Open Online Course (MOOC) website in China, imooc.com, to detect the online learning behavior pattern of Chinese online learners and estimate herding behavior. Considering the large number of similar online courses, this study determines how a suitable course can be selected. We first evince the existence of herding in online learning and then divide herding behavior into different groups: some learners passively mimic their peers (irrational herding), whereas some learners engage in active observational learning according to their needs (rational herding). The results reveal that rational herding dominates online learning. Subsequently, some variables amplify the herding effect, such as the difficulty level of the course and the experience of the learners. We conclude that the higher the difficulty level of a course, the more the rational herding of the learners and that the more experienced the learners, the higher the rational herding. This study provides implications for the way in course selection in MOOC, specifically in the learning analytics of the online learners. •Evidence the herding in online learning by learning data analysis empirically.•Rational herding dominates the learning process.•Instructor attributes and course attributes moderate the herding effect.•The higher the difficulty of a course, the more rational herding of the learners.•The more experienced the learners, the higher the rational herding.