Running Head. Additive, Synergistic, and Global Effects Need-supportive teaching and engagement in the classroom: Comparing the additive, synergistic, and global contributions Elizabeth Olivier 1* , Benoit Galand 2 , Alexandre J.S. Morin 1 , & Virginie Hospel 2 1 Substantive-Methodological Synergy Research Laboratory, Department of Psychology, Concordia University, Canada 2 Psychological Sciences Research Institut, Faculté de psychologie et des sciences de l’éducation, Université catholique de Louvain, Belgium *Corresponding author Elizabeth Olivier Department of Psychology, Concordia University, Canada 7141 Sherbrooke W., Montreal, QC, Canada, H4B 1R6 (+1)514-848-2424 ext. 3633 Funding : This study was funded by a grant from the Social Sciences and Humanities Research Council (SSHRC) (#756-2017-0223), awarded to Elizabeth Olivier. This is the prepublication version of the following manuscript: Olivier, E., Galand, B., Morin, A.J.S., & Hospel, V. (2020). Need-supportive teaching and engagement in the classroom: Comparing the additive, synergistic, and global contributions. Learning and Instruction. Early view: https://doi.org/10.1016/j.learninstruc.2020.101389 ©2020. This paper is not the copy of record and may not exactly replicate the authoritative document published in Learning and Instruction. Abstract This study contrasts three hypotheses to determine the best configuration of teacher need-supporting practices (autonomy support, structure, and involvement) in terms of classroom-levels of behavioral, emotional, and cognitive engagement. Multilevel analyses were conducted among a sample of 1,193 8 th grade students nested in 57 math classrooms. Results failed to support the additive hypothesis (H1), which anticipated that all three practices would be associated with classroom-levels of engagement when jointly considered. Results also failed to support the synergistic hypothesis (H2), which predicted that the greatest benefits would emerge in classrooms characterized by a high level of two or three practices. Finally, results supported the global hypothesis (H3), which anticipated that the global level taken across the three practices—captured by a global factor—would provide optimal support to classroom-levels of engagement. Specific factors representing the imbalance in autonomy support, structure, and involvement also contributed to some aspects of classroom-levels of engagement. Keywords: student engagement; need-supportive teaching practices; bifactor-confirmatory factor analyses; multilevel analyses