Analyzing Student Procrastination in MOOCs: A Multivariate Hawkes Approach [Konferenzbericht] Paper presented at the International Conference on Educational Data Mining (EDM) (13th, Online, Jul 10-13, 2020).

Autor/inn/en	Yao, Mengfan; Sahebi, Shaghayegh; Behnagh, Reza Feyzi
Titel	Analyzing Student Procrastination in MOOCs: A Multivariate Hawkes Approach [Konferenzbericht] Paper presented at the International Conference on Educational Data Mining (EDM) (13th, Online, Jul 10-13, 2020).
Quelle	(2020), (12 Seiten) PDF als Volltext kostenfreie Datei Verfügbarkeit
Sprache	englisch
Dokumenttyp	gedruckt; online; Monographie
Schlagwörter	Large Group Instruction; Online Courses; Student Behavior; Study Habits; Time Factors (Learning); Time Management; Interaction; Instructional Materials; Computer Science Education; Learning Activities; Assignments; Models; Grades (Scholastic) + Suchen Sie Ihr Suchwort? Online course; Online-Kurs; Student behaviour; Schülerverhalten; Study behavior; Study behaviour; Studienverhalten; Zeitmanagement; Interaktion; Lehrmaterial; Lehrmittel; Unterrichtsmedien; Computer science lessons; Informatikunterricht; Lernaktivität; Assignment; Auftrag; Zuweisung; Analogiemodell; Notenspiegel
Abstract	Student procrastination, as the voluntary delay of intended work despite expecting to be worse off for the delay, is an important factor with potentially negative consequences in student well-being and learning. In online educational settings such as Massive Open Online Courses (MOOCs), the effect of procrastination is considered to be even more prevalent and detrimental, as online courses are often self-paced and self-directed, where higher levels of self-regulated learning are expected from the students. Past research has mainly described students' procrastination by either static time-related measures (e.g. averaged starting time over all assignments per student), or by temporal models' parameters, under the assumptions that student activities take place at a constant rate (e.g. Homogeneous Poisson models), and that student interactions with one learning material are independent of interactions with another. In this work, we propose to consider the interdependence between the students' temporal activities while modeling their sequences in a continuous time scale. To this end, we propose to model the interaction sequence between each student and each course module, i.e. each module-student pair, as Multi-dimensional Hawkes processes, which not only capture the relationship between students' learning activities and their exogenous stimuli such as assignment deadlines, but also capture the endogenous responses within and between types of learning materials. Our experiments show that not only there exists dependencies between students' historical activities and the future ones when different types of learning materials are involved, such dependencies also provide meaningful interpretations in terms of students' procrastination behaviors. Furthermore, our findings show that in addition to association with delay, the parameters learned by multi-dimensional Hawkes processes provide more procrastination-related information and can improve our explanation of student grades. [For the full proceedings, see ED607784.] (As Provided).
Anmerkungen	International Educational Data Mining Society. e-mail: admin@educationaldatamining.org; Web site: http://www.educationaldatamining.org
Erfasst von	ERIC (Education Resources Information Center), Washington, DC
Update	2024/1/01