Institute for Education and Social Policy (IESP) Study Funded by New Visions for Public Schools Analyzed 67 Small and Medium Schools in New York City

Student outcomes at small and medium sized high schools in New York City are better compared to the citywide average, and show no difference between small and medium size schools, according to two new reports by the New York University Institute for Education and Social Policy (IESP). The reports are significant in that they are the first long-term studies of the first wave of small school development in New York City from 1994-95 through 1997-98. The studies suggest that optimal school high size may be larger than previously thought.

A joint initiative of the Steinhardt School of Culture, Education, and Human Development and the Robert F. Wagner School of Public Service, IESP has explored the development of small high schools in New York City for a 10-year time period. Funded by New Visions for Public Schools, the studies examine the effectiveness of small high schools during the period between 1994-95 and 2003-04 as well as the stability of student and teacher characteristics.

“While the findings reinforce what prior research has shown regarding the positive influence of small schools on student outcomes, these new data demonstrate that the previous definition of small (under 500 students) might, in fact, be too strict. Our findings support other research that suggests that the optimal high school size may be 600 to 900 students,” said Amy Schwartz, director of IESP.

Among the reports’ findings:

  • Small (under 500 students) and medium (between 501 and 1500 students) schools in the study had significantly better four-year graduation outcomes and lower dropout rates than the citywide average.
  • The characteristics of entering 9th and 10th grade student populations (e.g. previous academic performance, eligibility for free lunch, special education and language status) are critical factors in influencing subsequent performance of schools. Higher performing schools are more likely to have a greater percentage of entering students who have met standards on both 8th grade ELA and mathematics and lower percentages of English language learners.
  • High-performing small high schools exhibited cohesion of instructional practices, learning routines, and rules across classrooms that lower-performing small high schools lacked.
  • Small high schools produce better outcomes even though they have higher percentages of harder-to-educate student populations and less qualified teachers compared to high schools citywide.
  • New small schools are less stable in terms of student population and teacher experience over time compared to the citywide average, potentially making it difficult to sustain the instructional practices that support a small school’s specialization. One reason for this may be that teachers at small schools play multiple roles and over time find it difficult to sustain the amount of energy needed to work in these schools.

Only small schools that opened during the period were studied, in order to control for some of the unobservable factors that may influence school outcomes in newer schools, such as incomplete faculty, lack of supplies, or limited curricula. The findings demonstrate that regardless of size (from under 500 to 1500 students), the new schools performed better than the citywide average for the years studied.

“These findings provide important insights on how new small urban high schools are established and supported, and will help in the development of mechanisms to assist new small high schools after their initial development period,” said Meryle G. Weinstein, co-project director of the study.

The redesign of American high schools, including an increased focus on creating smaller schools and small learning communities, has been a key strategy for reforming public schools in urban districts around the country. Because most studies on small high schools are able to use only one or two years of data, it is particularly important to assess how these schools do over time, and whether initial achievement gains are maintained as the school develops.

The studies can be found at or by calling Tim Farrell in the Office of Public Affairs, 212.998.6797 or emailing

Press Contact