Learning Analytics Intervention: A Review of Case Studies

Intervention has been claimed to be the greatest challenge in learning analytics. As the provision of just-in-time and personalised support for learners, intervention has yet to be widely implemented in learning analytics practices. This paper reviews intervention practices in higher education in 23 case studies. The cases were categorised into four types - direct message, actionable feedback, categorisation of students, and course redesign - according to the nature of the methods of intervention; and the intervention methods were summarised. Most of the intervention cases belonged to the first two types. Direct message involves contacting at-risk students via channels such as emails or phone calls to encourage their participation, provide additional learning resources, or remind them of deadlines. Actionable feedback involves the provision of suggestions or information for students to help them understand their performance and possible ways of improving it. Only a few cases were identified for the other two types of intervention, which involve categorisation of students into different groups based on their risk levels for taking specific remedial actions for each group, and redesigning the course structure or contents based on the analysis of data. The results of this review serve to facilitate the formulation of intervention strategies for higher education institutions which practise learning analytics.