At the request of the State of Pennsylvania, the School District of Philadelphia, in the summer of 2002, asked three for-profit firms to assume responsibility for 30 of its lowest-performing schools and it asked four nonprofit managers to assume the management of 16 other low-performing schools. A difference-in-differences analysis is used to estimate the impact of nonprofit and for-profit management on individual student achievement. Gains in test scores at the treated schools are estimated by comparing them with gains in other low-performing schools in the district. Students at schools under for-profit management outperformed those at schools under nonprofit management in all six years in both reading and math. Most estimations are statistically significant. Impacts of for-profit management relative to district management were positive in math, but no reading impacts could be detected. At nonprofits, students appear to have learned substantially less, especially in math, at nonprofit schools, than had their school remained under regular district management. However, impacts fell short of statistical significance. (Propensity Score Analysis is appended. Contains 21 endnotes and 9 tables.)