Large scale study of primary school student performance relative to their LMS activity and socioeconomic demographics using a Bayesian Additive Regression Trees containing random effects

TL;DR

This large-scale study uses Bayesian Additive Regression Trees with random effects to analyze how LMS activity and socioeconomic factors relate to primary school student performance in Uruguay, enabling early risk detection.

Contribution

It introduces a novel application of BART with random effects to educational data, revealing insights into LMS impact and socioeconomic influences on student performance.

Findings

High LMS usage benefits low socioeconomic students more

Model identifies at-risk students early

Highlights schools needing intervention

Abstract

Using data collected on almost every 9-12 years old student in Uruguay, we show how to apply Bayesian Additive Regression Trees (BART) with random effects to study performance association with Learning Managment System (LMS) activity and socioeconomic status. Performance data is joined with LMS activity pattern data. BART is chosen because it is possible to include school-level random effects. The model can be used for early identification of at-risk students, and highlights schools that are successful or need intervention. An interesting finding is that high levels of LMS usage show larger positive effects on performance in low socioeconomic status.