The Bayesian Hierarchical Approach with Adaptive Lasso Technique to Identify Factors Affecting Student Dropout Rates in Secondary Education in AL Najaf

Abstract

 The problem of student dropout is one of the most significant challenges facing the educational process today. Furthermore, choosing an accurate statistical method to reduce this phenomenon is one of the most essential and necessary measures for decision-makers in mitigating it. This research aims to apply the bayesian hierarchical method with adaptive Lasso to identify the most significant factors influencing student dropout rates in secondery  schools. Adaptive Lasso is crucial for model interpretation and improving prediction accuracy. Furthermore, the Bayesian approach combined with this technique facilitates accurate estimations by incorporating raw data into the study. The Gibbs Sampler algorithm was developed and used in the R program for the estimation process, utilizing subsequent functions from both the Bayesian and adaptive Lasso methods. The results indicate that family encouragement, family income, and motivation are among the most significant factors contributing to increased student dropout rates, with p-values ​​of 0.003, 0.01, and 0.001, respectively. The findings also demonstrate the accuracy and importance of using the Bayesian hierarchical approach with adaptive Lasso compared to traditional methods.