Explainable Artificial Intelligence Framework for Predicting Educational Inequality Using Random Forest and XGBoost Across Multiple Educational Levels Globally

Educational inequality remains a major global challenge influenced by socioeconomic, regional, and developmental disparities across educational systems. However, accurately predicting educational inequality while maintaining model interpretability remains difficult due to the complex and nonlinear nature of educational data. This study proposes an explainable ensemble machine learning framework for predicting educational inequality across Primary, Lower Secondary, and Upper Secondary education levels using Random Forest and XGBoost models. The study employed regression evaluation metrics, including Mean Absolute Error (MAE), Root Mean Square Error (RMSE), coefficient of determination (R²), and five-fold cross-validation to evaluate model performance. The experimental results demonstrated that XGBoost achieved the best predictive performance with an R² value of 0.712 and a cross-validation mean score of 0.584, indicating strong predictive capability and stable generalization performance. To enhance model transparency, SHAP (SHapley Additive Explanations) analysis was applied to identify the most influential predictors contributing to educational inequality. The explainability results revealed that Education_Level_Upper Secondary, development-related variables, regional characteristics such as Sub-Saharan Africa, and temporal factors were the most significant contributors affecting prediction outcomes. In addition, the findings showed that educational inequality becomes increasingly distinguishable at higher educational levels, particularly within Upper Secondary education systems. The integration of Explainable Artificial Intelligence successfully improved model interpretability by providing transparent explanations regarding feature contributions and prediction behavior. Overall, this study demonstrates that explainable ensemble machine learning offers substantial potential for supporting evidence-based educational policy analysis, improving educational monitoring systems, and assisting policymakers in identifying critical factors contributing to educational inequality across diverse regional and educational contexts.

Authors

Institutions

Publication Details

Journal
Artificial Intelligence in Learning
Published
2026-11-17
DOI
https://doi.org/10.63913/ail.v2i4.74
Primary Topic
Explainable Artificial Intelligence (XAI)
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Explainable Artificial Intelligence Framework for Predicting Educational Inequality Using Random Forest and XGBoost Across Multiple Educational Levels Globally

Aulia Abikhair
Artificial Intelligence in Learning
Explainable Artificial Intelligence (XAI)
article

Explainable Artificial Intelligence Framework for Predicting Educational Inequality Using Random Forest and XGBoost Across Multiple Educational Levels Globally

Aulia Abikhair
article en

Abstract

Educational inequality remains a major global challenge influenced by socioeconomic, regional, and developmental disparities across educational systems. However, accurately predicting educational inequality while maintaining model interpretability remains difficult due to the complex and nonlinear nature of educational data. This study proposes an explainable ensemble machine learning framework for predicting educational inequality across Primary, Lower Secondary, and Upper Secondary education levels using Random Forest and XGBoost models. The study employed regression evaluation metrics, including Mean Absolute Error (MAE), Root Mean Square Error (RMSE), coefficient of determination (R²), and five-fold cross-validation to evaluate model performance. The experimental results demonstrated that XGBoost achieved the best predictive performance with an R² value of 0.712 and a cross-validation mean score of 0.584, indicating strong predictive capability and stable generalization performance. To enhance model transparency, SHAP (SHapley Additive Explanations) analysis was applied to identify the most influential predictors contributing to educational inequality. The explainability results revealed that Education_Level_Upper Secondary, development-related variables, regional characteristics such as Sub-Saharan Africa, and temporal factors were the most significant contributors affecting prediction outcomes. In addition, the findings showed that educational inequality becomes increasingly distinguishable at higher educational levels, particularly within Upper Secondary education systems. The integration of Explainable Artificial Intelligence successfully improved model interpretability by providing transparent explanations regarding feature contributions and prediction behavior. Overall, this study demonstrates that explainable ensemble machine learning offers substantial potential for supporting evidence-based educational policy analysis, improving educational monitoring systems, and assisting policymakers in identifying critical factors contributing to educational inequality across diverse regional and educational contexts.

Artificial Intelligence in LearningVol. 2(4)
Indonesia University of Education (ID)
Explainable Artificial Intelligence (XAI)
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon