Assessing learners’ satisfaction in collaborative online courses through a big data approach

• Published in: Computers in human behavior 2019-03, Vol.92, p.589-599
• Main Authors: Elia, Gianluca, Solazzo, Gianluca, Lorenzo, Gianluca, Passiante, Giuseppina
• Format: Article
• Language: English
• Subjects: Analytics; Big Data; CAI; Clustering; Collaboration; Collaborative learning; Computer assisted instruction; Data management; Distance learning; Learning analytics; Learning satisfaction; Online instruction; Real time; Sentiment analysis; Students
• ISSN: 0747-5632; 1873-7692
• DOI: 10.1016/j.chb.2018.04.033

Monitoring learners' satisfaction (LS) is a vital action for collecting precious information and design valuable online collaborative learning (CL) experiences. Today's CL platforms allow students for performing many online activities, thus generating a huge mass of data that can be processed to provide insights about the level of satisfaction on contents, services, community interactions, and effort. Big Data is a suitable paradigm for real-time processing of large data sets concerning the LS, in the final aim to provide valuable information that may improve the CL experience. Besides, the adoption of Big Data offers the opportunity to implement a non-intrusive and in-process evaluation strategy of online courses that complements the traditional and time-consuming ways to collect feedback (e.g. questionnaires or surveys). Although the application of Big Data in the CL domain is a recent explored research area with limited applications, it may have an important role in the future of online education. By adopting the design science research methodology, this article describes a novel method and approach to analyse individual students' contributions in online learning activities and assess the level of their satisfaction towards the course. A software artefact is also presented, which leverages Learning Analytics in a Big Data context, with the goal to provide in real-time valuable insights that people and systems can use to intervene properly in the program. The contribution of this paper can be of value for both researchers and practitioners: the former can be interested in the approach and method used for LS assessment; the latter can find of interest the system implemented and how it has been tested in a real online course. •Real-time monitoring of LS is crucial for CL processes.•Real time monitoring of LS may support course delivery and reduce early drop-outs.•Learning Analytics provide valuable insights about the level of LS.•Big Data may support learning managers to enhance decision making processes.