Researchers at MIT and Duke University just released a study entitled Small Schools and Student Achievement: Lottery-Based Evidence from New York City. The findings add to the available evidence about New York City's success. They show that students who attend small, nonselective high schools earn more credits, score higher on Regents exams, and have a significantly better chance of graduating and attending college than comparable students in schools that have not been explicitly designed to give them the supports they need. The results are striking for several reasons.
First, the researchers used gold-standard research methods that relied on New York's lottery system for assigning students to schools that are over-subscribed. This random selection allowed the researchers to overcome a common objection raised by small school skeptics: that even among the predominantly poor, minority, under-prepared students who attend these high schools, those who opt for and are accepted by small schools are inherently different from those who do not. Matching students precisely by key characteristics, including academic performance, demographics, and a host of other factors -- for example, whether or not they attended a school open house -- the researchers achieve a true "apples-to-apples" comparison among applicants.
Second, the new study extends earlier work by researchers at MDRC, whose 2010 and 2012 reports documented the positive impact of "small schools of choice" on high school graduation rates. For this new study, funded by the National Science Foundation, MIT and Duke economists followed five cohorts of students through high school and into the years beyond. Their findings show clearly that students who attended these small schools were not only more likely to graduate but also substantially more likely to go to college, place out of remedial courses once they got there, and stay in college for at least two semesters. (Students in the study are not yet old enough to have completed college.) The new study also used a slightly different randomization methodology, strengthening the authority of the earlier results. This is precisely the sort of exhaustive, long-term, cross-validating research that school leaders need in order to justify important policy decisions regarding the design of schools.
Third and most important, Small Schools and Student Achievement proves that it is indeed possible to provide adolescents -- even those who enter high school substantially behind -- with accelerated learning of an academically challenging curriculum that lets them catch up, get on track, and graduate ready for college. A crucial key, as revealed by the researchers' review of standardized school survey results, is a school environment where students feel safe, respected, and engaged with adults, and in which adults hold students to high expectations. The implications of these findings are immense as our education system prepares for the challenge, now acute in the face of Common Core implementation, of educating all students -- including those who are already struggling and already in high school -- to much higher standards.
On a personal note, I acknowledge that I did not read this new study, or the earlier ones by MDRC, with an unbiased eye. I was an architect of New York City's initiative to replace its lowest performing large high schools in high poverty neighborhoods with new small high schools, first at Carnegie Corporation in 2001 when I joined with the Bill & Melinda Gates Foundation, the Open Society Institute and New Visions for Public Schools to launch the New Century High School initiative, and then from 2002-2007 during my tenure as senior counselor to Chancellor Joel Klein. During this period, the NYC Department of Education undertook a full-scale secondary school reform, phasing out long-failing, large high schools where only one in three students typically graduated. They were replaced with hundreds of new, academically-themed small high schools educating no more than 450 students.
It's gratifying to see these affirmative results and to know they were produced through such rigorous research; honestly, it's also a relief. The truth is, when we started in 2002, we hoped, like so many school reformers before us, that small schools would have even greater impact on student achievement than they had so far achieved. We were working largely on the basis of our experience, assembling proven practices into an unproven whole. Fortunately, that's no longer necessary.
Today, thanks in part to research like this latest study, we know much more about what it really takes to redesign schools, and systems of schools, to serve all students. At Carnegie Corporation, we are building on the lessons from New York City through our Opportunity by Design initiative, funding four urban districts -- Cleveland, Denver, Philadelphia and New York City -- to advance new high school designs and system reforms. And we are looking for more districts to add to this initiative. In light of the new findings, an analysis of New York City's small-high schools initiative is instructive. Robert Hughes, President of New Visions, and I have distilled some key lessons:
•Be Bold: From the start, the small-schools effort was designed to improve the student experience with strategies that could be replicated system wide. We targeted the lowest-performing high schools and mobilized stakeholders in a commitment to own past failures and adopt meaningful changes.
•Adopt research-based design principles: Proposals for new schools needed to demonstrate the capacity to put in place many of the elements of successful change. If not, the plans would be rejected. The criteria include strong and accountable leadership, high-quality teaching, a curriculum leading to a Regents diploma, and engagement with students and the community.
• Redefine community schools: Community groups and service providers were invited to contribute to the school experience in new ways. Partnerships focused on coordinating resources to better support students and provide for an extended school day.
•Measure success with student outcomes: Rather than set a goal of mere improvement over the abject failure of the schools they replaced, New York City set a goal of 80 percent graduation and 92 percent attendance. The idea was to spur accelerated change with very high expectations.
•Change the district: The small-schools initiative was supported by New York City's systemic reform plan that targeted improvement in school leadership and teacher recruitment, deployed resources to high-poverty schools, redesigned student admissions, introduced school-based budgeting, and greatly improved use of data for planning. Engagement by union leadership and senior district staff in meeting with teacher groups from all the affected schools enabled the schools to attract good and committed teachers with high expectations for the students.
Perhaps the most important lesson is that the reforms worked in concert with one another, not in isolation, and the improved results can be replicated elsewhere. In the coming years, we need to push further and reach more students. The new MIT/Duke report will strengthen the knowledge base that undergirds the challenging work ahead.