Open University Learning Analytics Dataset (OULAD)

Product or Programme · All products and programmes
Evidence · 8 claims (8 for) · 2 studies (1 causal, 1 design), q2 · 1 of 2 report an effect size · 8 claims rest on one study

Description

OULAD is an open-source dataset maintained for learning-analytics research, containing demographic, virtual learning-environment activity, and performance data from adult learners at The Open University.

Components

• Open University Learning Analytics Dataset (OULAD) used for fairness-audited student performance prediction: OULAD is an open-source benchmark dataset containing demographics, virtual learning activity, and performance outcomes for adult learners in online courses at The Open University. The article describes it as containing "demographics, virtual learning activity, and performance outcomes for over 35,000 adult learners," of which 25,690 students with nine features were used after aggregating Fail/Withdrawn into a Failed outcome and Distinction/Pass into a Passed outcome. The study used it to train and fairness-audit LR and XGBoost pass/fail predictors across gender, age band, disability, and IMD subgroups. (Raymond A. Opoku et al. (2025))

Claims

• Standard ML models predicting student pass/fail from VLE data exhibit biased true-positive rates across demographic subgroups [+W]
• Baseline LR and XGBoost models achieve comparable predictive performance on OULAD, with XGBoost slightly higher AUC-ROC [+W]
• Calibrated Equalized Odds Post-processing showed lower balanced accuracy and higher EOD than the preprocessing techniques DIR and RW in student performance prediction [+W]
• Preprocessing bias mitigation (DIR, RW, SUP) reduced subgroup disparities in TPR on OULAD, with effectiveness varying by demographic attribute [+W]
• Balancing the dataset significantly improved model predictive performance but increased bias under demographic parity [+W]
• Stage 1 (after the first unit review assignment) is the optimal point for identifying at-risk students, balancing timeliness, accuracy, and fairness [+W]
• Post-processing mitigation with ThresholdOptimizer reduced bias on most fairness metrics while maintaining accuracy reasonably well, with a slight performance drop [+W]
• Adding more learning activity data beyond Stage 1 did not significantly improve predictive performance of at-risk identification models [+W]

Related Products and Programmes

-

Key Sources

• Raymond A. Opoku, Bo Pei and Wanli Xing. (2025). Unveiling Accuracy-Fairness Trade-Offs: Investigating Machine Learning Models in Student Performance Prediction. Journal of Learning Analytics 12(2). https://doi.org/10.18608/jla.2025.8543

Page metadata

Field	Value
type	product
id	open-university-learning-analytics-dataset-oulad
title	Open University Learning Analytics Dataset (OULAD)
description	OULAD is an open-source dataset maintained for learning-analytics research, containing demographic, virtual learning-environment activity, and performance data from adult learners at The Open University.
status	review
moved_from	element/oulad-dataset-fairness-audit-element
product_kind	dataset
generated	by process:sweep-kinds on 2026-10-10
trust	unverified — no one has confirmed this page's content
sources	1 — raymond-a-opoku-2025

Frontmatter is the page's machine-readable record — see the Schema & Guide.