The authors examine the psychometric properties of the Collective Teacher Efficacy Scale (CTES), an instrument designed to measure collective teacher efficacy. Specifically, a multilevel confirmatory factor analysis is used to determine the factor structure of the CTES, comparing one- and two-factor models. The mediating role of the CTES factors in the relationship between student socioeconomic status and academic achievement at the school level are also examined. Implications for research and using the CTES in school reform agendas are discussed. (Contains 1 figure and 5 tables.)