Detection and classification of dropout behavior

Timothée Duron (1) , Laurent Gallon (1) , Philippe Aniorte (1)

1 LIUPPA - Laboratoire Informatique de l'Université de Pau et des Pays de l'Adour

This article presents the first results of our work aimed at detecting students' risk of dropping out and dropping out during block-based computer programming lessons in the classroom. We use information related to the use of the mouse and the interface of the programming software to detect behaviors characteristic of dropout. The paper describes the methodology chosen to build the rules for detecting these behaviors. The results obtained on a first case study are also given. These results were obtained within the framework of the PERSEVERONS project 1 , which aims to study the perseverance of students induced by the use of a digital object.

Mots clés: code.org; block programming; dropout; perseverance; MHI; HCI; Flow theory

Domaines: Environnements Informatiques pour l'Apprentissage Humain

Timothée Duron, Laurent Gallon, Philippe Aniorte. Detection and classification of dropout behavior. FECS'22 - The 18th Int'l Conf on Frontiers in Education: Computer Science and Computer Engineering, Jul 2022, Las Vegas, United States. ⟨hal-03834009⟩