Self-Efficacy Prediction Model Using Bayesian Networks

Abstract

This study introduces a Bayesian network model designed to predict teacher self-efficacy, a critical factor influencing pedagogical effectiveness. The model integrates a range of teacher characteristics, including educational background, professional experience, training, attitudes, and burnout indicators, to provide a probabilistic estimate of self-efficacy. A survey of 44 secondary school teachers in N'Djamena, Chad, was conducted to gather data for model validation. The results demonstrate the model's ability to effectively predict both individual and collective self-efficacy, highlighting the significant impact of factors such as emotional exhaustion and attitude on teachers' perceived competence. The findings suggest that this model can serve as a valuable tool for educational administrators in developing targeted interventions and support programs to enhance teacher well-being and improve classroom effectiveness. Future research should explore the integration of machine learning techniques to refine predictive accuracy and expand the model's applicability across diverse educational contexts.