Moderation Effects in Structural Equation Modeling: A Comprehensive Approach to Complex Relationships

This study aims to explore the causal relationship between latent construct in a structural model by considering the role of moderating variables. Structural equation modeling (SEM) is the main method of its ability to test measurement and structural models simultaneously. Estimation is performed using the Generalized Least Squares (GLS) method, which is considered more efficient in conditions of non-normal data distribution. The study uses data from 332 managers from Muhammadiyah Hospitals in Indonesia. The research variable consist of servant leadership and emotional regulation as exogenous variables, leadership effectiveness as an endogenous variable, and managerial cognitive capability as a moderator variable. The study’s result indicate that servant leadership significantly impacts leadership effectiveness. Managerial cognitive capability acts as a moderator in the relationship between servant leadership and leadership effectiveness, which has a positive impact on leadership effectiveness. Similarly, for the exogenous variable of emotional regulation, the moderating variable strengthens the influence of emotional regulation on leadership effectiveness. Managerial cognitive capability as a moderator can either increase or decrease the influence of exogenous variables on leadership effectiveness.