Date
Publisher
arXiv
Student dropout is a global issue influenced by personal, familial, and academic factors, with varying rates across countries. This paper introduces an AI-driven predictive modeling approach to identify students at risk of dropping out using advanced machine learning techniques. The goal is to enable timely interventions and improve educational outcomes. Our methodology is adaptable across different educational systems and levels. By employing a rigorous evaluation framework, we assess model performance and use Shapley Additive exPlanations (SHAP) to identify key factors influencing predictions. The approach was tested on real data provided by the Moroccan Ministry of National Education, achieving 88% accuracy, 88% recall, 86% precision, and an AUC of 87%. These results highlight the effectiveness of the AI models in identifying at-risk students. The framework is adaptable, incorporating historical data for both short and long-term detection, offering a comprehensive solution to the persistent challenge of student dropout.
Study at a glance
- Question: Can machine learning models accurately identify students at risk of dropping out of school to enable early support?
- Setting and data: Fez-Meknes region, Morocco, 2015/2016 to 2020/2021 academic years, 1.4 million student profiles from primary through high school.
- Method: Machine learning classification models evaluated across multiple data-splitting strategies, imbalanced data techniques, and SHAP feature explanations.
Why it matters
Student dropout poses a significant barrier to personal growth and economic progress worldwide. Predictive modeling can uncover early warning signs to enable targeted, proactive interventions.
Key findings
- The best-performing model achieved 88% accuracy, 88% recall, 86% precision, and an AUC of 87% in predicting student dropout.
- Student dropout rates in Morocco consistently peaked during the final years of each educational cycle, particularly in middle and high school.
- Key factors influencing dropout predictions included overall grade averages, student age at current level, gender, and prior grade repetitions.
Summary drafted with AI from the full paper and may contain errors. See the full paper for details.
What is the application?
Who is the user?
Why use AI?
Study design
