Predicting Academic Success Based on Learning Material Usage

Abstract

In this work, we explore students' usage of online learning material as a predictor of academic success. In the context of an introductory programming course, we recorded the amount of time that each element such as a text paragraph or an image was visible on the students' screen. Then, we applied machine learning methods to study to what extent material usage predicts course outcomes. Our results show that the time spent with each paragraph of the online learning material is a moderate predictor of student success even when corrected for student time-on-task, and that the information can be used to identify at-risk students. The predictive performance of the models is dependent on the quantity of data, and the predictions become more accurate as the course progresses. In a broader context, our results indicate that course material usage can be used to predict academic success, and that such data can be collected in-situ with minimal interference to the students' learning process.

Introductory programming: a systematic literature review

“…Although such resources are widely used and studied across disciplines, there is a growing body of work specifically focused on their use in introductory programming. Some researchers have applied techniques for machine learning [323,370] and data mining [36] to study interaction by introductory programming students with an LMS in order to identify behaviour patterns and investigate whether these can predict academic success. Other papers focus on introductory programming students' use of specific types of resource provided within a learning environment, such as video lectures [149,428] and interactive reading and homework materials [183] designed to enable a flipped classroom approach.…”

Predicting Students Success in Blended Learning—Evaluating Different Interactions Inside Learning Management Systems

“…Results show that the students who followed the personalized guidance and recommendations performed better in exams. The usage of online learning material (in an introductory programming course) was used to predict academic success [27]. The results obtained have shown that the time spent with the material is a moderate predictor of student success.…”

Early Identification of Student Struggles at the Topic Level Using Context-Agnostic Features

“…Other studies track interactions with online learning materials offered as a part of the course. Leppänen et al monitored how long each element, be it a paragraph or an image, of the online textbook was visible on each student's screen [11]. Similarly, Yang et al collected clickstream data from students watching lecture videos in a massive online open course (MOOC) and used time series machine learning techniques to develop a predictor to much success [24].…”

“…The nature and source of such data vary greatly. Examples include student programming behavior, assignment performance, clicker data, and question-and-answer interactions on online forums [10][11][12]. These recent studies improved our understanding of the characteristics of low-or high-performing students, as well as the extent to which machine learning can predict struggling students.…”