School-wide positive behavior interventions and supports (SWPBIS) is widely implemented across the United States. Evidence suggests that SWPBIS, when implemented with fidelity, has a significant and positive impact on schools, including lower suspensions and higher reading and mathematics achievement. A critical component of this research is measuring fidelity of SWPBIS implementation in schools. In this study, we re-evaluated the reliability and validity of a widely used fidelity tool, the Tiered Fidelity Inventory (TFI), within a single-state sample. Using data from 693 schools in the U.S. state of California, we found a hierarchical factor structure based on subscales and tiers fit the data best when controlling for the multilevel nature of the data; however, there was no relation between the latent fidelity variables and changes in suspensions. Implications of these results are discussed, including different approaches to scoring and conceptualizing fidelity of implementation for SWPBIS, along with considerations for future research using the TFI.