The mental health and well-being of students and teachers during the COVID-19 pandemic: combining classical statistics and machine learning approaches

Abstract

Abstract The aims of this study were to (1) to explore the state of students’ and teachers’ well-being and (2) examine the factors that predict their well-being during the pandemic-related school closures in the Philippines. Our sample comprised 733 students and 1168 teachers. During the height of the pandemic, 22.10% of the students and 13.44% of teachers met the cut-off for depression; 13.91% of the students and 15.92% of the teachers met the cut-off for anxiety. Both classical statistics and machine learning approaches were used to identify the roles of demographic, psychological, and socio-contextual factors that statistically predicted well-being outcomes. Results highlighted that family support was the strongest predictor of students’ and teachers’ positive well-being. For mental health outcomes, the strongest predictors of depression were anxiety and stress, while the strongest predictors of anxiety were depression, stress, and fear of COVID. Implications for students’ and teachers’ well-being amidst COVID are discussed.