What question did this study set out to answer?

The aim is to develop reliable, interpretable predictive models for classifying student stress levels.

July 21, 2026Open Access

Predicting Student Stress Using Machine Learning Ensemble Models: A Multi-Criteria Comparison with Explainable Artificial Intelligence Analysis

DADaniel Cristóbal Andrade-GirónUniversidad Nacional José Faustino Sánchez Carrión WRWilliam Joel Marín RodriguezUniversidad César Vallejo MZMarcelo Gumercindo Zúñiga-RojasUniversidad Nacional José Faustino Sánchez Carrión

Key Points

The aim is to develop reliable, interpretable predictive models for classifying student stress levels.
Analyzed public Student Stress Factors dataset with 1100 records and 20 predictors.
Employed supervised machine learning with stratified data partitioning and nested cross-validation.
Compared nine algorithms, including Gradient Boosting, XGBoost, and LightGBM, assessing performance with discrimination and calibration metrics.
Gradient Boosting achieved an accuracy of 89.55% and ROC-AUC of 98.59% during cross-validation.
Final calibrated model exhibited accuracy of 88.18% and ROC-AUC of 98.61% on independent dataset.
No statistically significant performance differences were found among algorithms (χ2 = 10.953, p = 0.204).

Abstract

Student stress is a significant mental health issue in educational settings; therefore, developing reliable, calibrated, and interpretable predictive models can support the classification of observed stress levels. This study analyzed the public Student Stress Factors dataset, comprising 1100 records, 20 predictors, and one target variable, using a supervised machine learning pipeline designed to reduce information leakage. The pipeline included stratified data partitioning, encapsulated preprocessing, nested cross-validation restricted to the training data, and independent holdout evaluation. Nine ensemble and boosting algorithms for tabular data were compared: AdaBoost, Gradient Boosting, Random Forest, Extra Trees, Bagging, Voting, Stacking, XGBoost, and LightGBM. Model performance was assessed using key discrimination and calibration metrics, together with the nonparametric Friedman test for statistical comparison. Gradient Boosting achieved the best average performance in nested cross-validation, with an accuracy of 89.55 ± 3.16%, F1-weighted of 89.54 ± 3.17%, MCC of 0.845 ± 0.047, and ROC-AUC weighted of 98.59 ± 0.92%. XGBoost and LightGBM showed comparable performance. In the independent holdout set, the final calibrated model maintained robust predictive performance, achieving an accuracy of 0.8818, F1-weighted of 0.8818, MCC of 0.8237, and ROC-AUC weighted of 0.9861. Although the overall results indicate stable and high predictive performance, the Friedman test did not identify statistically significant differences among the algorithms, χ2 = 10.953, p = 0.204. Therefore, model selection should consider not only predictive accuracy but also computational efficiency, calibration, interpretability, and implementation feasibility. Despite the internal stability of the pipeline and satisfactory holdout performance, the public and cross-sectional nature of the dataset limits causal inference and model transferability. Consequently, external and prospective validation is required before integration into institutional early warning systems.

AIに質問

Bookmark

View Full Paper