Loi Zedda, MaudeMaudeLoi ZeddaBouchard, LaureLaureBouchardPidoux, MarcMarcPidouxOppliger, SabineSabineOppligerCourt Urzelai Intza, LaurenceLaurenceCourt Urzelai IntzaMartin, BorisBorisMartinPanayiotou, GeorgiaGreiff, Samuel2025-10-032025-10-032025-06https://hdl.handle.net/20.500.12162/15376In Switzerland, around 20% of teachers leave the profession within the first five years. One of the causes of this problem is the difficulty teachers have in dealing with difficult student behavior. Thus, the development of good classroom management skills, including the ability to deal with difficult student behavior, is a key issue in teacher training. Our research focuses on two concepts that have an impact on teachers' professional development. The first is the teachers' self-efficacy in classroom management, which influences competence in classroom management. The second is teacher empowerment, which positively influences teachers' decision-making ability, professional learning, motivation and impact on student achievement. Since self-efficacy is a factor in teacher empowerment, there is a link between these two concepts. However, to the authors' knowledge, no studies appear to investigate the links between these two variables for teacher training in classroom management. Thus, the aims of this paper are to determine and verify the existence of relationships between self-efficacy for classroom management and teacher empowerment, and to compare the results between two research studies conducted in 2023 and 2024. The methodology used for both studies is predictive correlational. This choice makes it possible to determine whether there are relationships between the variables. 245 Bachelor and Master of Teaching students (research conducted in 2023) and 144 students from the same programs (research conducted in 2024) participated in this study. The results reveal, namely for the 2023 study, some positive relationships between classroom management self-efficacy and empowerment, with correlations ranging from 0.18 to 0.32. However, these results are hampered by certain limitations. For example, the absence of previous studies linking the two variables studied in the same population is one of them. Similarly, the methodology used does not allow causal relationships to be established.enType de référence::Communications::Communication scientifique publiée10.1024/2673-8627