Identifying at-risk students in online learning by analysing learning behaviour: A systematic review

Abstract

As the development of online learning is growing, a large amount of log data on student activity is available and accumulated in Learning Management Systems (LMS). This massive volume of data can be analysed in order to discover hidden information about students' learning behaviours and, in particular, to identify at-risk students and to advise instructors to provide additional assistance and interventions to help those who are at risk, to increase retention rates. As a result, Learning Analytics (LA) has emerged, which aims to measure, analyse and report students' learning behaviours using data analytics techniques. In this regard, this paper aims to provide an overview of how at-risk students are identified or predicted by analysing their learning behaviours in online learning, including the types of data and tools or analytics methods used. The findings indicate that most of the data analytics techniques have successfully identified and predicted at-risk students and various types of data are determined as attributes in predicting at-risk students. Finally, further research should be conducted to identify and predict at-risk students, since there are diverse students in online learning who have different demands, and interventions should be developed to tackle the problems faced by at-risk students in order to enhance their learning performances.