LevelDataStudentSuccess

The Level Data Student Success Insight project, developed during the Fall 2024 AI Studio, focuses on identifying key factors affecting student success in K-12 school districts using data science and machine learning techniques. The goal was to build a predictive model enabling school districts to intervene and support at-risk students using data-driven insights. The dataset includes anonymized 2023 data from District 18, covering 11,633 students with test scores (ACT, SAT, state exams), demographic attributes, district benchmarks, and vendor usage data from platforms like IXL Math and Accelerated Reader. After thorough data cleaning, z-score standardization, and benchmark score calculations, the data was split by school level and gender for targeted analysis. We built and evaluated Logistic Regression and Decision Tree models, focusing on metrics such as accuracy, precision, recall, F1-score, and ROC-AUC. Key insights included positive correlations with specific educational vendors and varying impacts based on student engagement. Despite challenges like missing demographic data and limited vendor usage granularity, the project highlighted critical academic success factors, enabling school districts to implement data-driven interventions.