This study tested a model in which perceived self-efficacy in classroom management explained the influence of student disruptive behavior on teacher burnout. Dutch secondary teachers completed the Maslach Burnout Inventory, Self-Efficacy Scale for Classroom Management and Discipline, and Order and Organization subscale of the Classroom Environment Scale. Before analysis, completed questionnaires were divided into two groups. Utilizing a structural equation modeling procedure with maximum likelihood estimation and a two-step modeling approach, researchers first tested the measurement model using half of the sample. Pseudo chi-square was significant, indicating the specification of the measurement model needed improvement. They related depersonalization and emotional exhaustion to one construct (the core of burnout). The measurement model was adequate since pseudo chi-square was not significant. Tests performed on the structural model indicated that it could be improved by adding a direct effect of personal accomplishment upon perceived self-efficacy. After this modification, Normed Comparative Fit Index (CFI) and Tucker Lewis Index (TLI) exceeded .90, a recommended criterion. The modified model was tested using the rest of the sample. CFI and TLI exceeded .90. After modification of the measurement model and structural model, researchers concluded that the hypothetical model was empirically sound. (Contains 75 references.) (SM)