Purpose: To identify patterns of school environment and association with physical activity (PA) in different domainsMethods: Cross-sectional data from the National School Health Survey (2015) was used, and students enrolled in the ninth grade of elementary education were evaluated. Electronic questionnaires were answered by students and the school principal. The study outcomes were PA indicators (yes/no): commuting; Physical Education; leisure-time; reaching PA guidelines. The exposure was related to the school environment (physical, social and political factors). Sociodemographic characteristics were described using proportions and mean with standard deviations. Principal Components Analysis (PCA) was used to identify environmental school-related patterns. Multilevel binary logistic regression models were applied.Results: 81,154 students with a mean age of 14.3 ± 1.3 years old were investigated. Most of the students reported participating in Physical Education (80.8%), but only 8.6% of them met PA guidelines. The PCA revealed five patterns that explained 53% of the variance among the environmental school characteristics. Higher scores of both Sports Courts and Sports & Materials patterns were associated with a higher odds of participation in Physical Education (OR: 1.27 and 1.13, respectively). Higher scores for the Open Schools pattern increased the odds of PA in commuting to school (OR: 1.12), leisure-time PA (OR: 1.03), and met PA guidelines (OR:1.04); Higher scores for Track & Pool pattern increased the practice of Physical Education (OR: 1.05).Conclusion: The findings reinforce the role of the school environment within PA and reveal a complex association, as different patterns of the school environment may favor specific PA domains more than others. The promotion of PA must consider the multiple layers of the school environment to enhance PA practice across all domains.