A reference model for learning analytics

TLDR

Learning analytics is a growing, multidisciplinary field that uses educational data to support learning, integrating machine learning, AI, statistics, and visualization, and converging with related areas such as academic analytics, action analytics, and educational data mining. The study investigates the connections between learning analytics and these related fields. The authors propose a reference model for learning analytics comprising four dimensions—data and environments, stakeholders, objectives, and methods—and map recent publications onto this framework. They identify challenges and research opportunities for each dimension of the model.

Abstract

Recently, there is an increasing interest in learning analytics in Technology-Enhanced Learning TEL. Generally, learning analytics deals with the development of methods that harness educational datasets to support the learning process. Learning analytics LA is a multi-disciplinary field involving machine learning, artificial intelligence, information retrieval, statistics and visualisation. LA is also a field in which several related areas of research in TEL converge. These include academic analytics, action analytics and educational data mining. In this paper, we investigate the connections between LA and these related fields. We describe a reference model for LA based on four dimensions, namely data and environments what?, stakeholders who?, objectives why? and methods how? We then review recent publications on LA and its related fields and map them to the four dimensions of the reference model. Furthermore, we identify various challenges and research opportunities in the area of LA in relation to each dimension.