A Comparative Analysis of Student Performance Predictions in Online Courses using Heterogeneous Knowledge Graphs

TL;DR

This paper compares student performance prediction models in online and on-campus courses using heterogeneous knowledge graphs and Graph Convolutional Networks, highlighting differences in engagement and success prediction accuracy.

Contribution

It introduces a novel approach using heterogeneous knowledge graphs and GCNs to predict student performance across different course modalities.

Findings

Prediction accuracy of 70-90% for at-risk students.

Models reveal engagement pattern differences between online and on-campus students.

Heterogeneous knowledge graphs effectively model student interactions.

Abstract

As online courses become the norm in the higher-education landscape, investigations into student performance between students who take online vs on-campus versions of classes become necessary. While attention has been given to looking at differences in learning outcomes through comparisons of students' end performance, less attention has been given in comparing students' engagement patterns between different modalities. In this study, we analyze a heterogeneous knowledge graph consisting of students, course videos, formative assessments and their interactions to predict student performance via a Graph Convolutional Network (GCN). Using students' performance on the assessments, we attempt to determine a useful model for identifying at-risk students. We then compare the models generated between 5 on-campus and 2 fully-online MOOC-style instances of the same course. The model developed achieved a 70-90\% accuracy of predicting whether a student would pass a particular problem set based on content consumed, course instance, and modality.