Cluster analysis for tailored tutoring system


Abstract

In online courses, the tutoring activities are relevant for the learners’ training; they affect the course quality and imply creating dynamic and active online learning environments. The research aims to define a model based on a structured and data-driven tutoring system to identify homogeneous groups of students attending a blended degree course and, then, set tailored interventions as support. We chose a multivariate data analysis technique, cluster analysis, and used personal data and academic achievements of first-year students (n=110) in Degree Course in Digital Education at the University of Modena and Reggio Emilia to create groups of similar learners, identify common characteristics of students in each group and define personalized tutoring strategies for each cluster. The analysis allows identifying six homogeneous clusters of students and defining activities to design that concern the fields of content, motivation and metacognition, involve degree course tutors and teachers, and be carried out individually or in small groups of students.