Objective: This study explored how observed features of the school physical and social environment relate to students’ perceptions of school climate and how these in turn were associated with students’ involvement in violence. Method: Observational assessments were conducted of the environments (i.e., disorder, illumination, adult monitoring, proactive behavioral management, and negative student behaviors) of 58 high schools using a validated assessment (the School Assessment for Environmental Typology). Student perceptions of school climate (i.e., delinquency, rules and consequences, and physical comfort) as well as their perpetration of violence were collected from 28,592 adolescents in these same schools in the corresponding Spring. Multilevel structural equation models were used to test for indirect effects. Results: A good fit was found for all models. No direct effects of environmental observations on violence involvement were identified. However, significant indirect effects on violence were found, specifically for illumination through perceptions of disorder (estimate = −.01, p = .05), illumination through perceptions of rules and consequences (estimate = −.01, p = .03), and negative student behaviors through perceptions of rules and consequences (estimate = −.01, p = .01). Conclusion: Changes to the school environment may be associated with reduced violence involvement, but only insofar as they alter student perceptions of the environment.