Factors affecting teacher job satisfaction: a causal inference machine learning approach using data from TALIS 2018

Abstract

Teacher shortages and attrition are problems of international concern. One of the most frequent reasons for teachers leaving the profession is a lack of job satisfaction. Accordingly, in this study we have adopted a causal inference machine learning approach to identify practical interventions for improving overall levels of job satisfaction. We apply our methodology to the English subset of the data from TALIS 2018. Of the treatments we investigate, participation in continual professional development and induction activities are found to have the most positive effect. The negative impact of part-time contracts is also demonstrated.