Learning online: Do peers’ and teachers’ interactions influence dropout?

A Synapse Lab Seminar

• Date: Friday, May 24
• Time: 12.30 - 13.30
• Room: EOJ 277/276 and on Zoom

Abstract:

This study delves into the extent to which platform interactions of peers and teachers influence student dropout in online learning environments. We draw upon data from over a million users in various types of upper secondary schools engaged with an Italian Learning Management System between September 2019 and July 2021. Firstly, we identify the most effective machine learning technique for predicting educational platform dropout. Subsequently, we explore the distinct contributions of class elements, such as peers, teachers, and class size in elucidating this phenomenon. Our findings reveal that ensemble methods excel in predicting dropout incidents, probably given the complex relationship between diverse user interactions and student platform retention. Moreover, integrating both peer and teacher platform interactions significantly improves model performance compared to relying solely on student interactions. Notably, peers’ platform interactions alone are more effective in predicting individual student dropout than students’ own platform interactions.

Short Bio:

Melisa is a Post Doc Research Fellow at Politecnico di Milano, Department of Management Engineering, Economics and Industrial Engineering. Her research falls within the intersection of arts management, cultural economics and digital innovation. Exploring further the new trends and the determinants of cultural participation, intended as the engagement of people on cultural activities. She collaborates on projects developing performance measurement systems on the cultural sector in an institutional and a territorial level.