Confirmatory Factor Analysis of the Maslach Burnout Inventory A Bayesian Structural Equation Modeling Approach

In the Maslach Burnout Inventory (MBI), burnout is conceptualized as a combination of emotional exhaustion, depersonalization, and personal accomplishment. However, the factorial structure for the MBI remains controversial. We examined the factorial structure of the MBI, relying on Bayesian structural equation modeling (BSEM). BSEM allows the investigator to take into account sources of influence such as approximately-zero factor cross-loadings and between-item residual covariances. Data from a convenience sample of 5,575 French teachers were mobilized. One-, two-, three-, and bi-factor models were tested using BSEM. Maximum likelihood (ML) estimation was also implemented to examine the robustness of the BSEM analysis. The BSEM analysis showed that a two-factor model, consisting of a burnout factor combining emotional exhaustion and depersonalization and a separate personal accomplishment factor, best fitted the data. ML estimation showed a substantial enhancement of model fit when information derived from the BSEM analysis was incorporated. The MBI appears to be underlain by a unified factor involving the emotional exhaustion and depersonalization items and a separate factor involving the personal accomplishment items. These findings are compatible with a research practice that involves combining the emotional exhaustion and depersonalization dimensions of burnout and considering personal accomplishment individually.