Student Attendance and Early Warning System Dataset: A Case Study of Pelita Kasih Christian School (K-12)

This dataset was curated and utilized for the research project titled: "Early Warning Framework for Student Attendance Using Interpretable Machine Learning for K-12 Students: A Case Study at Pelita Kasih Christian School." The primary objective of this dataset is to support the development of predictive models that identify students at risk of chronic absenteeism. By employing Interpretable Machine Learning, the associated research aims to provide educators not only with predictions but also with the underlying factors (features) contributing to a student’s attendance patterns. Context & Data Collection The data was collected from Pelita Kasih Christian School, covering K-12 educational levels. It encompasses various parameters that influence student behavior and attendance, providing a localized context for educational data mining in Indonesia. Dataset Structure The dataset contains 7000+ records. Data Privacy & Ethics To comply with ethical standards and protect student privacy: All Personally Identifiable Information (PII) such as Student Names, ID Numbers, and specific addresses has been removed or anonymized. The data is intended for academic and research purposes only. Potential Use Cases Training and benchmarking Early Warning Systems (EWS) in education. Researching Model Explainability (using SHAP, LIME, etc.) in the K-12 context. Comparative studies on student attendance factors in private religious schools.