摘要：Purpose – The purpose of this research is to formulate a measurement model of Emotional Exhaustion for non-permanent lecturers with SEM-PLS at Private Islamic Higher Education (PTAIS) in North Sumatra.Methodology – The research population is all non-permanent lecturers of Private Islamic Higher Education (PTAIS) in North Sumatra. To determine the sample size, Maximum Likelihood estimation is used in the Structural Equation Model (SEM). Therefore, based on the need for researchers to use data analysis techniques using the Structural Equation Model and also because later the maximum likelihood estimation technique will be used which requires a sample size of 100-200 samples, the researcher will take a sample of 100-200 non-permanent lecturers. The analytical method used is SEM-PLS analysis.Findings – The results showed that there was a significant effect of Role Conflict on Emotional Exhaustion. There is a significant effect of Self Efficacy on Emotional Exhaustion. There is a significant influence of Leadership on Emotional Exhaustion. There is a significant effect of Emotional Exhaustion on Job Satisfaction. There is a significant effect of Role Conflict on Job Satisfaction. There is a significant effect of Self Efficacy on Job Satisfaction. There is a significant influence of Leadership on Job Satisfaction. There is no significant effect of Leadership on Job Satisfaction through Emotional Exhaustion. There is no significant effect of Role Conflict on Job Satisfaction through Emotional Exhaustion. And there can be no significant effect of Self efficacy on Job Satisfaction through Emotional ExhaustionOriginality/Novelty – The research model uses five (5) variables and applies the concept of mediation, where each relationship in the research model is analyzed by SEM-PLS.Implications – This study has shown that Emotional Exhaustion does not have a mediating effect on the influence of leadership, self-efficacy and role conflict on job satisfaction. Starting from the results of this study, the recommendation for further research is to try to analyze or use the Emotional Exhaustion variable as an independent variable, not as a mediating variable.
