Abstract
PURPOSE: To address the limitations of static and subjective methods, this study aims to examine the dynamic interplay among factors influencing student engagement in physical education across different developmental stages. METHOD: This study integrates an enhanced Delphi method with a Time-Varying Influence Network model and dynamic MICMAC analysis to compute time-varying driver power and dependency from the evolving total influence matrix. RESULTS: The dynamic system evolution was characterized by three key patterns: convergence toward equilibrium, transient dominance, and structural stability. DISCUSSION/CONCLUSION: The dynamic system evolution is characterized by three key patterns: convergence toward equilibrium, transient dominance, and structural stability. These results underscore the importance of timing and targeting in intervention strategies and the need to reinforce stable core elements.