To investigate claims that New York City charter schools push out low-performing students in order to inflate academic achievement scores, this report uses six years of student-level data that allows for tracking student movements over time in both district and charter schools. It relies on a regression equation that includes “low scorers” (defined in several ways), whether a school is a charter or a traditional public school (TPS), and an interaction between those two characteristics in order to estimate the probability of a student exiting. The report concludes that charter schools have exit rates similar to TPSs, and there is no significant interaction between low-performing students and charter schools; and thus, there is no charter push-out effect for low-achieving students. The main problem is that the research design does not address its primary push-out question. The brief report has little detail and does not examine a host of other relevant factors. Dichotomous test scores are a proxy for low-achievement, reasons for disenrollment are not addressed, mid-year vs. end-of-year mobility is not parsed, cumulative rates of attrition are not examined, a possible data discrepancy between the two sectors in grades 5 and 6 is not considered, and 5% of the student population is missing. These unresolved issues are particularly regrettable in light of the rich dataset. While the report’s central question is important, this report fails to provide policymakers with new or definitive guidance.
Manhattan Institute for Policy Research