At a time when reformers and philanthropists have largely turned their back on the 鈥渟mall schools鈥� movement, a of New York City high schools has found that students are more academically successful in smaller, more personal high schools that they choose for themselves than they are in larger, more traditional schools.
The report released last night by MDRC, a New York-based research group, focuses on the 1.1-million-student school system鈥檚 effort from 2002 to 2008 to shut down 20 large, failing high schools and replace them with 200 smaller schools where students might be more likely to receive the attention they need. Researchers found that students who ended up in the small schools were more likely than peers in other kinds of city high schools to be on track by 9th grade to graduate in four years, to stay on track for three more years, and graduate from high school on time.
鈥淭his is the first convincing evidence that the 鈥榮mall schools鈥� model can be effective at improving student outcomes,鈥� said Brian Jacob, a professor of education policy and economics at the University of Michigan in Ann Arbor. 鈥淚t鈥檚 incredibly important.鈥� Mr. Jacob advised the researchers on an early version of the study.
The new study was financed by the Seattle-based Bill & Melinda Gates Foundation, which was an enthusiastic early supporter of New York City鈥檚 small-schools initiative and others around the country.
Sparse Evidence
Like New York, many districts in the late 1990s and the early 2000s began to replace large, comprehensive high schools with smaller schools.
Many such efforts were spurred by the Gates Foundation, which poured nearly $1 billion into small-schools programs before deciding in 2008 to move in a new direction after some of those efforts proved disappointing. (The Gates Foundation provides funding to Editorial Projects in Education, the publisher of Education Week.)
One such disappointment was the high-profile closing of Denver鈥檚 Manual Arts High School, which closed in 2006, after persistent, poor academic performance.
And the handful of studies up until now on the topic of small schools were mostly descriptive or anecdotal.
鈥淣obody really had strong systematic evidence about the effects of those schools,鈥� said Gordon L. Berlin, MDRC鈥檚 president. Mr. Berlin and others said the new study is important because it tracks a large-scale improvement effort, involves mostly disadvantaged students, and uses a more rigorous research methodology to compare students who were randomly assigned to their top-choice schools with those placed elsewhere or in regular city high schools.
鈥淚 hope this will lead to a conversation which involves the re-evaluation of small schools,鈥� said Charles Payne, a social work professor at the University of Chicago. 鈥淚n the 鈥�80s, the talk about small schools focused on relationships, and that seems to have taken a back seat now to things like turnarounds.鈥�
Use of various 鈥渢urnaround鈥� strategies to improve troubled schools is a policy prescription heavily promoted by the Obama administration.
But other experts expressed more skepticism about the results of the MDRC study.
鈥淚 don鈥檛 know how much of this is about choosing and how much is about small schools,鈥� said Amy Ellen Schwartz, a professor of public policy, education, and economics at New York University and the director of its Institute for Education and Social Policy. 鈥淲hat would have happened if students were choosing large schools of choice?鈥�
Home-Grown 69传媒
The study zeroes in on the 123 small schools that opened in New York City after 2002. Dubbed 鈥渟mall schools of choice鈥� by the researchers, those nonselective public schools all enrolled 550 or fewer students in grades 9-12. Drawing mainly from disadvantaged student populations, most of the schools were located in Brooklyn and the Bronx.
While the schools incorporated a variety of themes, such as coastal studies, sports management, and media studies, all were required to offer features common to the small-schools movement. Such features include an 鈥渁dvisory鈥� period to provide for closer attention by a teacher to a small group of students, partnerships with the local community, and common planning times for teachers.
From these and other schools, 80,000 rising 9th graders each year are required to pick 12 as their top choices. Based on those student lists and schools鈥� geographic locations, students are then assigned, via a centralized-placement system, to a high school.
When a school is oversubscribed, the system randomly assigns students. The researchers used those 鈥渕ini-lotteries鈥� to assign students to either control or experimental groups. Randomized studies are considered a powerful means of determining whether an intervention causes an effect because students are not handpicked or self-selected.
Since not every school was oversubscribed, though, the researchers based their findings on a much smaller final sample of 21,000 students, roughly half of whom landed in a top-choice school. The rest attended a mix of other kinds of city schools, some of which were lesser-ranked smaller schools.
Overall, however, the control-group schools were much older and larger than the 鈥渟mall schools of choice.鈥� More than half the students in the comparison group attended schools with more than 550 students, according to the report.
Perspective on Numbers
While the students entering small schools were academically and demographically similar to their control-group peers upon entering high school, they were more likely, by the end of 9th grade, to be on track to graduate, the study found. In the small schools, 58.5 percent of students had passed enough courses to be on the four-year graduation track, compared with 48.5 percent of the students in the control group.
By the end of the fourth year of high school, the graduation rate for small-schools enrollees was 68.7 percent, compared with 61.9 percent for the control-group students. A larger proportion of the small-schools graduates had also earned the state鈥檚 Regents diploma, which requires students to earn a higher passing score on state exams.
The researchers also found that the positive effects were similar for students from a variety of subgroups, including minority students, males, and those with a history of poor achievement.
Some researchers disagree, however, on how to think about the 6.8-percentage-point difference in graduation rates between small-schools graduates and their counterparts elsewhere in the city.
Mr. Berlin said the difference is large, given that the results come from randomized studies, which rarely show large effects in education. It鈥檚 one-third the size of the graduation-rate gap that separates black and Hispanic students across the city from their higher-achieving white peers.
But David C. Bloomfield, the program head for educational leadership at Brooklyn College and a professor of educational leadership at City University of New York, said the magnitude of the impact was 鈥渞elatively small for all the attention that has been lavished on these schools.鈥� The small schools all received $400,000 in startup grants over four or five years, as well as technical assistance from intermediary organizations such as New Visions for Public 69传媒鈥攈elp that many experts credit as a key to New York City鈥檚 good results.
The Gates Foundation, which ultimately invested $150 million in the New York small-schools effort, issued a prepared statement yesterday calling the findings 鈥済reat news for New York City students and for all school reform efforts dedicated to increasing college-readiness rates and grounded in improving the quality of the interaction between students and teachers inside the classroom.鈥�
