Using individual student test-score data from the School District of Philadelphia, we estimated the impact of for-profit and non-profit school management on student achievement by tracking the performance of students in math and reading from 2001 to 2006. After four years, the average student at schools managed by for-profit firms learned roughly two-thirds of a year more in math than would be expected had the schools remained under district management. However, the positive impact of for-profit management on average reading gains was smaller and not statistically significant. For non-profits, we found mainly negative impacts on student performance in both math and reading, but none were statistically significant. Treatment effects were identified using a quasi-experimental research design known as “difference-in-differences” analysis. We used as a control group the 71 schools under regular district management at which students were performing below the district median.
Peterson, Paul E., and Matthew M. Chingos. "Impact of For-Profit and Non-Profit Management on Student Achievement: The Philadelphia Experiment." KSG Faculty Research Working Paper Series RWP07-055, November 2007.