With the support of the Gates Foundation, New York City created 150 small schools of choice between 2002 and 2008. Five previous rigorous studies of this program and other small school initiatives have demonstrated significant benefits for students. Now we have a sixth study from the School Effectiveness and Inequality Initiative at MIT.
The authors, Atila Abdulkadiroglu, Weiwei Hu, and Parag Pathak, are economists at Duke and MIT. They take advantage of lotteries to gain admission to these non-selective small schools of choice to conduct a random assignment experiment. The full study can be read here, but it does not allow me to cut and paste text to summarize the results. According to the press release:
The study follows cohorts of rising 9th graders for five application years from 2003-04 through 2007-08. For these students, small schools boost performance across all five major Regents exams: Math, English, Living Environment, Global History, and US History. Students randomly offered a seat at a small school accumulate 1.4 more credits per year, attend school for 4 more days each year, and are 9% more likely to receive a high school diploma.As the cohorts have aged, it is now possible to measure the effects of small schools on college enrollment and choice, outcomes that have never been examined before. Compared to the college enrollment rate of 37% for those not offered, students at small schools are 7% more likely to attend college and 6% more likely to attend a four-year college. Most of these gains come at four-year public institutions. There is a marked 7% increase in the fraction of students who enroll in the CUNY system. Small schools cause students to clear CUNY remediation requirements in writing or reading. The early evidence suggests that students are more likely to persist in college, as measured by attempting at least two academic semesters. Students in the lottery study are too young to say anything definitive about college graduation.A major innovation in the study is its use of information contained in NYC’s Learning Environment Surveys to characterize the small school environment for those in the experiment. Small schools are rated higher than fallback schools by student survey respondents on the overwhelming majority of questions on engagement, safety and respect, academic expectations, and communication. Surveys indicate that students feel safer and have closer interactions with their peers and teachers, despite reporting a smaller variety of course offerings and activities. Teachers indicate greater feedback, increased safety, and improved collaboration.This research was supported by the National Science Foundation. The research team includes Atila Abdulkadiroglu, Professor of Economics, Weiwei Hu, PhD Candidate at Duke University and Parag Pathak, Associate Professor of Economics at MIT and SEII Director. The study uses data provided by the New York City Department of Education. The findings are being released in the National Bureau of Economic Research working paper series this week.The study uses an innovative research design based on admissions lotteries contained in the high school match. The lottery-based research design relies on apples-to-apples comparisons: among those who apply to a given set of small schools, applicants who were randomly offered are compared with otherwise similar students who were not offered a seat. The study covers more than 108 oversubscribed high school programs with 9th grade entry, which represent 70% of unselective small high schools opened between 2002-2008.“These results indicate important possibilities for urban small schools reform,” said Pathak. “The collaboration partnership between key stakeholders in New York City shows that within-district reform strategies can substantially improve student achievement.”