Application of Learning Analytics in Higher Education: Datasets, Methods and Tools

The accumulation of big educational data on the platforms of universities and social media leads to the need to develop tools for extracting regularities from educational data, which can be used for understanding the behavioral patterns of students and teachers, improve teaching methods and the quality of the educational process, as well as form sound strategies and policies for universities development. This article provides an analysis and systematization of datasets on available repositories, taking into account the learning analytics problems solved on their basis. In particular, the article notes the predominance of datasets aimed at solving analytical problems at the level of student’s behavior understanding, Datasets aimed at solving analytical problems at the level of understanding the needs of teachers and administrative and managerial staff of universities are practically absent. Meanwhile, the full potential of learning analytics tools can only be revealed by introducing an integrated approach to the analysis of educational data, taking into account the needs of all participants and organizers of the educational process. This review article discusses learning analytics methods related to the study of social interaction patterns between students and teachers, and learning analytics tools from the implementation of simple dashboards to complex frameworks that explore various levels of learning analytics. The problems and limitations that prevent learning analytics from realizing its potential in universities are considered. It is noted that universities are generally interested in introducing learning analytics tools that can improve the quality of the educational process by developing strategies for targeted support for individual groups of students, however, teachers treat such initiatives with caution due to a lack of data analysis skills and correct interpretation of analysis results. The novelty of this analytical review is associated with the consideration of learning analytics at different levels of its implementation in the context of approaches to openness, processing and analysis of educational data. This article will be of interest to developers of learning analytics tools, scientific and pedagogical workers, and administrative and managerial staff of universities from the point of view of forming an idea of the integrity of the university analytics process, taking into account various levels of analytics implementation aimed at understanding the needs and requirements of all participants in the educational process.