Determinants of teachers’ positive perception on their professional development experience: an application of LASSO-based machine learning approach

Abstract

ABSTRACTGiven the complex nature of teachers’ professional development (PD) processes, it is crucial to examine how various factors surrounding teachers are associated with the evaluation of their PD experience. By applying a machine-learning technique, least absolute shrinkage and selection operator (LASSO), we were able to include numerous factors in an integrated model to create a data-driven, parsimonious predictive model that is readily applicable. Using TALIS 2018 U.S. data (n = 2,418), we identified 16 important explanatory variables (out of 132 variables) in determining teachers’ positive perception on their PD. We found that teachers’ PD experience depends on multiple layers of factors such as features of PD activities (10 variables), teachers’ individual characteristics (four variables), and school organisational environments (two variables). Theoretical and practical implications are also discussed.KEYWORDS: Teacher professional developmentperceptions of teachersmachine learning techniqueLASSOTALIS Disclosure statementThere are no relevant financial or non-financial competing interests to report.