Contributions to Learning Analytics Focused on Assessment and Self-Regulated Learning

Abstract

Learning analytics is a fairly recent discipline of increasing relevance, as educational environments as a whole are undergoing a technological transformation which, as a side effect, causes student data to be more easily accessible than ever. This thesis explores ways in which these data can be used to assess how well students regulate their own learning, a factor that has a direct impact on their academic performance. Ultimately, the goal is to build an early warning system able to identify students’ negative attitudes related to self-regulation at early stages of a course, enabling the possibility of providing personalized help to each student depending on their profile. The study focuses on the educational context of higher education, particularly, first-year engineering students, who typically struggle to adapt to the vastly different academic demands of college compared to high school. Among the types of data that are used in this project, information related to assessment and exams stands out as one of the most complete and relevant.