View Proposal

---

Proposer

Ian Tan

Title

Predictive Analytics and Causal Inference for Personalised Remediation in Education

Goal

To develop a causal-driven prescriptive analytics framework that generates personalised learning pathways

Description

This project explores the development of data-driven personalised learning interventions within the Kayam School App ecosystem, an ed-tech platform deployed in collaboration with Chumbaka Asia. Building on the 2025 longitudinal dataset from Program Anak Kita (PAK) Sarawak, which served over 8,114 pupils across 204 primary schools, the project aims to move beyond retrospective analysis toward actionable, real-time student support. The data contains real-world variations in app usage intensity, school-level fidelity, and demographic covariates.

Resources

Background

Some background on causality studies, you can review foundational frameworks from Joshua Angrist on instrumental variables, natural experiments, and technology evaluations in education * Angrist, J. D., & Lavy, V. (2002). "New Evidence on Classroom Computers and Pupil Learning." The Economic Journal, 112(482), 735-765. How is this relevant? A seminal study examining the direct causal impact of educational technology (ICT) tools on student learning outcomes using exogenous variation to control for selection bias. * Angrist, J. D., Gao, C., Hull, P., & Yeh, R. W. (2025). "Instrumental Variables in Randomized Trials." NEJM Evidence, 4(4). How is this relevant? Clarifies how to use instrumental variables methods to measure per-protocol causal effects (Local Average Treatment Effects) when real-world participants show irregular compliance or varying dosage—directly mapping to real-world variations in app usage intensity.

Url

Difficulty Level

High

Ethical Approval

Full

Number Of Students

1

Supervisor

Ian Tan

Keywords

causality, educational pathway recommendation, statistics, econometrics

Degrees

Bachelor of Science in Computing Science
Bachelor of Science in Statistical Data Science