Teacher Attrition Prediction Model and Analysis of the Associated Factors

Teachers are the foundation of a well-educated society and play a pivotal role in guiding the coming generations to success. Untimely departure of teachers hampers the education of students and makes it difficult for the school authorities to find a suitable replacement within a reasonable time-frame. The study proposes to build a supervised machine learning model using Decision Tree coupled with ADABoost to predict the likelihood of a teacher to leave the school with around 90% accuracy, taking into account the various personal and professional factors, so that pre-emptive actions can be taken. The supremacy of this model is verified by comparing its accuracy with other machine learning algorithms. A detailed analysis of the various factors and their intensity in contributing to a teachers unforeseen departure is also presented. This study will benefit the school administrations to develop retention and replacement strategies by observing the behaviour of their faculty members.