Description: Although Canadian students performed well on the 2000 Programme for International Student Assessment (PISA), there were considerable differences among provinces. Based on the Canadian sample of PISA 2000, our multilevel analysis examined the extent to which schools in different provinces "produced" differential impacts on reading achievement of their students. Our approach in this study involved an application of the input-process-output model of school effectiveness (see Willms, 1992). Reading achievement was the output. Data on family background and home environment were selected to represent inputs, data on school context and climate were selected to represent educational processes, and data on school policy and operation were selected to emphasize the policy dimension within educational processes. Methodology: The Canadian PISA sample included 29,687 students (at age of 15) from 1,117 schools across the ten provinces (Statistics Canada, 2001). PISA defined reading literacy as the ability to understand, use, and reflect on written texts in order to participate effectively in life (Organisation for Economic Cooperation and Development, 2001). Reading achievement was the dependent variable. The independent variables included student-level and school-level characteristics derived from student and school questionnaires and selected on the basis of the conceptual framework of school effectiveness (Willms, 1992). Multilevel analysis was chosen as the primary statistical technique for our data analysis (Raudenbush & Bryk, 2002). Conclusions: School characteristics that showed significant provincial effects on reading achievement included variables that described school context, school resources, and school climate, with the majority of these significant school-level variables being measures of school climate. This finding is encouraging in that it is school climate rather than school context or school resources that educators can reform in a meaningful way. Disciplinary climate and sense of belonging to school were the most important of significant school characteristics. They showed significant average province effects on reading achievement, and their effects varied significantly across provinces. The effect of academic pressure (to achieve) on reading achievement was negative in most provinces. Provinces with positive attitudes toward comparative evaluation (provincial assessments) showed a higher standing in reading performance. Provinces where school autonomy was strong delivered increased academic achievement. Results of our analysis suggest that provinces can use school effects to promote reading achievement of their students.