The educational data revolution has empowered universities and educational institutes with
rich data on their students, including information on their academic data (e.g., program
completion, course enrolment, grades), learning activities (e.g., learning materials reviewed,
discussion forum interactions, learning videos watched, projects conducted), learning process
(i.e., time, place, path or pace of learning activities), learning experience (e.g., reflections,
views, preferences) and assessment results.
In this paper, we apply clustering to profile
students from one of the largest Massive Open Online Courses (MOOCs) in the field of Second
Language Learning. We first analyse the profiles, revealing the diversity among students
taking the same course. We then, referring to the results of our analysis, discuss how profiling
as a tool can be utilised to identify at-risk students, improve course design and delivery, provide
targeted teaching practices, compare and contrast different offerings to evaluate interventions,
develop policy, and improve self-regulation in students. The findings have implications for the
fields of personalised learning and differentiated instruction