A computational investigation of learning behaviors in MOOCs. Issue 5 (17th May 2017)

• Record Type: Journal Article
• Authors: Zhong, Sheng‐Hua; Li, Yanhong; Liu, Yan; Wang, Zhiqiang
• Is Part Of: Computer applications in engineering education. Volume 25:Issue 5(2017)
• Page Start: 693
• Page End: 705
• Publication Date: 2017-05-17
• Subjects: behavior study -- computer science education -- learning style -- massive open online courses; Engineering -- Study and teaching (Higher) -- Periodicals; Engineering -- Computer-assisted instruction -- Periodicals; 620
• DOI: 10.1002/cae.21830
• Languages: English
• ISSNs: 1061-3773

Abstract

Abstract: Massive open online courses (MOOCs) are the latest e‐learning initiative to attain widespread popularity in the world. Thus, it is highly required to have a throughout analysis of learning in MOOCs, from theoretical to practical. Our primary goal is to take a detailed and comprehensive investigation into the learning behaviors in MOOCs, as well as to identify issues that have not yet to be adequately resolved. We employed commonly used educational data mining methodologies to analyze and interpret the behaviors in a computer science course based on the questionnaire survey data and daily activity data. We find most of the students could be divided into several groups that are coincident with their learning styles. Moreover, we can easily predict students' learning styles based on their learning behaviors. This finding means the learning style could be a factor to indicate students' learning behaviors, or even measure whether a student is appropriate to learn via MOOCs.