This article presents quantitative results from the first three cohorts of a school leadership residency program in an urban district. The two-year leadership residencies were selective, multifaceted, and intensive, combining at least 370 hours of professional development with on-the-job training, in which aspiring school principals held either assistant-level administrative or teacher leadership roles. Using a difference-in-differences framework with school fixed-effects for all schools in the state, we present estimated effects of school staffing by leadership program residents and alumni on schools' academic achievement, behavior, graduation rates, and school culture as described by teachers. The effect of each leadership resident-by-year in an administrative role is estimated as an additional 14% of a standard deviation in math scale scores, 10% of a standard deviation in ELA scale scores, and 2.5 percentage points in four-year graduation rates. However, estimated effects on student attendance and suspensions move in the opposite of the desired directions, and there is little evidence of a relationship between leadership residency placements and teachers' reports of school culture. Robustness checks limited to the treatment city indicate some sensitivity to the definition of the counterfactual sample. Three years after program inception, due to the contracting nature of the district, one of 30 entering residents had moved into a position as a school principal. Consequently, these estimates shed new light on the effects of instructional leadership by educators in supporting roles.