The factorial validity of the Maslach Burnout Inventory (MBI) was studied for 98 elementary school, 163 intermediate school, 162 secondary school, and 218 university teachers in Ottawa (Canada). The equivalence of factorial measurements and structure was tested across the groups. Exploratory factor analyses were used to examine the initial factor structure, and confirmatory factor analyses tested the hypothesized three-factor structure and its invariance across groups. In spite of the findings of a few aberrant items specific to each group, the MBI demonstrated remarkably sound psychometric properties with intermediate and secondary school teachers, less sound properties with university teachers, and poor psychometric properties with elementary school teachers. Tests of invariance across intermediate and secondary teachers revealed the equivalence of all item measurements for both groups. The results have important implications for substantive studies focusing on multigroup comparisons across teaching levels. Five tables summarize study findings. (SLD)