Among the various kinds of learning analytics emerging especially in the latest decade, clicking patterns cover a prominent role, fostered by their success in analyzing several types of data concerning activity on the web. They can be defined as sets of clicks performed by users, in which every set is treated as the basic unit. Few research has been performed on clicking patterns in educational contexts. In this paper, we perform analysis regarding clicks to an online course in Mathematics, aimed at allowing students to follow courses at a distance, both before and after enrolling at University. We used clustering techniques on students learning behavior, which have been defined for this research as visualizations of activities and resources of the course, to detect differences on students’ grade according to their online learning behavior. Our results show that students tend to proceed on the course in both activities and resources. There is no correlation between participation and course grades, even if the most active students show higher scores. Moreover, patterns differ significantly according to the degree program of each student, showing the importance of tailored path.
Clustering Techniques to investigate Engagement and Performance in Online Mathematics Courses
Francesco Floris
;Marina Marchisio
;Sergio Rabellino
;Fabio Roman
;Matteo Sacchet
2022-01-01
Abstract
Among the various kinds of learning analytics emerging especially in the latest decade, clicking patterns cover a prominent role, fostered by their success in analyzing several types of data concerning activity on the web. They can be defined as sets of clicks performed by users, in which every set is treated as the basic unit. Few research has been performed on clicking patterns in educational contexts. In this paper, we perform analysis regarding clicks to an online course in Mathematics, aimed at allowing students to follow courses at a distance, both before and after enrolling at University. We used clustering techniques on students learning behavior, which have been defined for this research as visualizations of activities and resources of the course, to detect differences on students’ grade according to their online learning behavior. Our results show that students tend to proceed on the course in both activities and resources. There is no correlation between participation and course grades, even if the most active students show higher scores. Moreover, patterns differ significantly according to the degree program of each student, showing the importance of tailored path.File | Dimensione | Formato | |
---|---|---|---|
2_CELDA2022_F_197.pdf
Accesso aperto
Tipo di file:
POSTPRINT (VERSIONE FINALE DELL’AUTORE)
Dimensione
356.17 kB
Formato
Adobe PDF
|
356.17 kB | Adobe PDF | Visualizza/Apri |
Celda2022_Clustering.pdf
Accesso riservato
Tipo di file:
PDF EDITORIALE
Dimensione
2.09 MB
Formato
Adobe PDF
|
2.09 MB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.