If one is looking for a symbol of the rise and fall—and resurrection—of the American high school, one need but take the #1 IRT subway to 225th Street in the Bronx, then walk a few blocks up Marble Hill along the north shore of the Harlem River. The eight-story building covering some four city blocks looks as if it was lifted from the drafting table of a Soviet bloc architect.

Opened in 1972, John F. Kennedy High School housed in its heyday between 3,000 and 6,000 students—4,500 started that first September. According to Iris Zucker, who taught there in the 1990s, “It was as big as some towns. We had 350 teachers.” It had everything for everyone—except an education. By the end of the century, only one-third of its students graduated. Furthermore, it was dangerous, described in a 2004 New York Times story as a school that “has turned out more horrifying tales than success stories. There was the substitute teacher whose hair was set on fire, the assistant principal hospitalized after being knocked down by students, the assorted objects—trash cans, ceramics projects—hurled from windows, sometimes into teachers’ parked cars. In 2002, one summer school student fatally stabbed another outside the school. A few months later, things became so rowdy after a fire drill that the police officers on duty used Mace…”

Today, the building still holds some 3,000 students and 300 teachers, but a huge banner hanging from its towering façade announces a makeover: it lists five high schools. And even that is behind the times, since there are now seven: Marble Hill High School for International Studies (MHHS), the Bronx Engineering and Technology Academy (BETA), the Bronx School of Law and Finance (BSLF), the English Language Learners and International Support Preparatory Academy (ELLIS), New Visions Charter High School for Advanced Math and Science (NVAMS), the Bronx Theatre High School (BTHS), and New Visions Charter High School for the Humanities (NVH). Each school educates the same mostly poor, mostly black and Hispanic students as entered the building in 1972. Most now graduate at least two-thirds of their students and boast far more “success stories” than “horrifying tales.”

What happened to Kennedy happened all over New York City. Between 2002 and 2008, the number of high schools in New York increased from just over 250 to nearly 450, even as the number of high school students in the system remained the same. This resulted from closing 30 large schools, shrinking others (such as Kennedy), and creating dozens of small, themed high schools, with 100 students per grade instead of 1,000. At the same time, discovering that tens of thousands of high school students were hopelessly behind and on the fast track to dropping out, the district created a system of even smaller transfer schools.

As a series of studies began to emerge in 2012, it became clear that what Mayor Michael Bloomberg and Schools Chancellor Joel Klein had done in New York City over the preceding decade was real. While the nation seemed transfixed by No Child Left Behind, Race to the Top, and Common Core State Standards, “one of the most wide-ranging reforms in public education” during that time, according to a group of researchers from Duke and MIT, “was the reorganization of large comprehensive high schools into small schools” in New York City. Not only did the district, the largest in the country, take on a student population that had come to symbolize the impossibility of educating a certain kind of child—the urban poor who entered high school two and three grades behind—but it succeeded in getting those students to graduation. What worked in New York was a multifaceted, multibillion-dollar, multiyear overhaul of the city’s high schools. In an era when a high school diploma is the difference between a career and a lifetime on the dole, New York’s high-school reforms have increased the economic mobility of tens of thousands of students.

Characteristics of Success

Part of the reason that the small-schools effort was so remarkable is that it bucked the reform instinct to start when kids are young; it was also notable because it was so long in coming. The 1983 report A Nation at Risk called out comprehensive American high schools for their “smorgasbord” curricula that were “homogenized, diluted, and diffused to the point that they no longer have a central purpose.” Many urban educators complained that the large high school was simply too big to work and too impersonal to reach every child, much less hundreds of children. What had been conceived as an educational melting pot had for many become a cauldron of educational failure.

By the 1990s, alternatives were in the works. Through the $500 million Annenberg Challenge, Reader’s Digest founder Walter Annenberg provided matching grants ranging from $1 million to $53 million to 2,400 schools in 35 states, much of it to create small high schools. New Visions for Public Schools, a nonprofit established in 1989, launched a small-high-schools effort in New York City in 1993 with Annenberg’s help. By 2000, when New Visions created a separate organization, New Century High Schools, to run and expand its small-high-schools effort, the collaboration had created 40 such schools. Despite these efforts and the Annenberg philanthropy devoted to them, the academic performance needle for most urban students barely moved.

Fast-forward 15 years, and the story is radically changed. When Bloomberg and Klein took the reins, in 2002, the stage was set for top-to-bottom transformation. West Coast billionaire Bill Gates had begun steering his Bill and Melinda Gates Foundation to education, and would eventually devote a billion dollars to the small-high-schools effort, spreading its largesse to some 300 school districts across the United States, including New York City.

According to Becky Smerdon and Kathryn Borman, who led the Gates-sponsored research team that evaluated the initiative, by the late 1990s some consensus had emerged among reformers about what made schools successful: “a shared vision focused on student learning, common strategies for engendering that learning, a culture of professional collaboration and collective responsibility, high-quality curriculum, systematic monitoring of student learning, strong instructional leadership (usually from the principal), and adequate resources.”

A growing body of research supported the idea that these characteristics were more easily achieved in smaller schools than in larger ones. The Gates grantmakers created seven “attributes of high-performing schools” that would guide its giving to those who wanted to create small high schools: a common focus; high expectations; personalization; respect and responsibility; time to collaborate; performance-based instruction; and using technology as a tool. Properly implemented, the foundation believed, these attributes would “lead not only to better outcomes for students attending the schools, but to increased demand for such schools.”

Unfortunately, in the first five years of the initiative, according to the Evaluation of the Bill & Melinda Gates Foundation’s High School Grants Initiative: 2001–2005 Final Report, released in August 2006, there were mixed results at best. The researchers who studied the project, from the American Institutes for Research and SRI International, analyzed the grants to a sample of 17 school districts in 11 states. Their report found that most districts in the sample registered positive results with personalization and collaboration, but struggled with efforts to raise the expectations bar and implement performance-based instruction. For a program that was supposed to improve the educational outcomes of low-income high-school students, this was not good news.

In a book that Smerdon and Borman would curate for the Urban Institute in 2009, Saving America’s High Schools, many of the members of the research team expanded on the findings from the Gates report, offering a wealth of specific findings for many of the larger districts receiving Gates funds. The conclusions were the same: the major problem was implementation, especially with academics.

In the end, according to the final report of the Gates Foundation, “both new and redesigned schools needed more help with issues of curriculum and instruction.” As Smerdon and Borman would conclude in their subsequent book, “there is good reason to expect that the success of this ‘raise-the-bar’ approach to school improvement will depend on stakeholders’ abilities to provide the academic supports that students, particularly struggling students, need to be effective learners. Without these supports, the benefits of entering a ‘rigorous’ high school with more course requirements or a college-preparatory mandate may not be realized….”

Indeed, though the Gates Foundation would move on to other things, Smerdon and Borman had, in effect, suggested why New York City’s small-schools program has worked: academic instruction.

Sovereignty, Not Johnny Appleseed

Robert Hughes recalls his first meeting with Joel Klein, in 2002, at an opening-day ceremony at South Bronx High School, the newest of New Century’s small high schools. “And it’s a beautiful day and he sees what we’re doing,” recalls Hughes, who had taken the top job at New Visions in 2000, “and he turns to me and he says, ‘Can you create 200 more of these?’ I said, ‘Sure,’ because you always say ‘Yes’ to the new chancellor.”

But Hughes recognized immediately what that question meant. One of the lessons from the failed Annenberg Challenge, he would later explain, was that “You have to have superintendency”—by which he means authority—“so you start to change the system itself… You want to find new ways of supporting education improvement as a matter of routine.”

To make improvement a matter of routine may have been Bloomberg’s and Klein’s greatest contribution to New York’s public school ethos.

“Another critique of Annenberg,” says Hughes, “was that its theory of change was a little bit like Johnny Appleseed. You sprinkle good schools throughout a system and they’ll start to grow and sprout and other people will replicate them.”

Greg Duncan and Richard Murnane echo Hughes’s Johnny Appleseed observation in their book Restoring Opportunity: The Crisis of Inequality and the Challenge for American Education. Previous efforts, they said, were characterized by “a lack of a cogent framework for structuring these schools.” School administrators, they continued, often “viewed [seed schools] as exceptions in a system of centralized control, tolerating them only because they pacified innovative educators who would otherwise have been more vocal critics of the system.”

Until Bloomberg and Klein, the system tolerated the new small-school “seeds,” but didn’t fertilize them. By backing up the reform efforts, Bloomberg and Klein provided the “cogent framework”—the fertilizer, the water, the sun. And that was just the beginning.

Engaging Educators

Chancellor Klein hit the ground running, talking to Michele Cahill, a senior program officer at the Carnegie Corporation with vast experience, the day after his July 2002 appointment—and asking her to lead the high-school reform efforts. This sent a signal to the bureaucracy that change was coming. The following October, just after posing the question to Hughes, Klein announced the district’s intent to open 200 new small high schools.

While at Carnegie, Cahill had worked closely with New Visions and helped, in the spring of 2001, to secure an additional $30 million for New Century High Schools from the Gates Foundation, the Open Society Institute, and Carnegie, each interested in a different piece of the pie.

“Carnegie was interested in systems, how do we systematically think about reform, and wanted us to look at both small schools and large school transformations,” recalls Hughes about the $30 million small-schools grant he received in 2000. “Gates was about small, so small was an option that we put on the table. And Open Society was about highest need.”

Balancing those funder desires, New Visions created a Request for Proposals to all community school districts and high school superintendents in the city, inviting any group of educators to propose a small high school—limited to some 100 students per grade—with a focus on the Bronx, which had the highest concentration of low-performing schools.

“Annenberg was all about outside groups coming in,” explains Hughes, “so we wanted to use community-based organizations to drive change and ensure that there was a sense of urgency from the community, a kind of youth development perspective or civic perspective, that could be incorporated in what was going on in education.”

The Request for Proposals that had gone out was still reverberating when Bloomberg became mayor on January 1, 2002. The mayor’s reform efforts were aided by the fact that veteran teachers and administrators were excited again about school because of the small-schools efforts spearheaded by New Visions. Dozens of new school-team hopefuls had responded. “We all had a passion for this,” recalls Kirsten Larson, an English as a Second Language teacher at Morris High School in the Bronx. Larson was one of four teachers, an assistant principal, and guidance counselor from Morris determined to take the plunge. “But the writing was also on the wall,” she recalls. “They were going to close Morris.” Morris, like Kennedy, had become a dropout factory. Its last principal had described it, in 2001, as “a place out of control.”

New Visions and the New York City Department of Education provided technical assistance to the 75 applicants, convening workshops and advising the teams about curriculum, parent engagement, student engagement, teacher recruitment, the grading system, the floor plan, administrative priorities, and New Century’s 10 principles. Number one was “a rigorous instructional program.” That meant a Regents diploma curriculum. And a Regents diploma meant earning 22 credits of core subject courses and passing five different (and rigorous) domain-specific tests (in English, math, science, U.S. history, and global history).

In the end, Larson and her colleagues were 1 of only 15 of the 75 applicants that made that first cut. They moved to the eighth floor of Kennedy High School, then in the process of being remade, and opened Marble Hill School for International Studies in September 2003. Today, with 440 students, Marble Hill has a four-year graduation rate of 89.7 percent.

Over the next six years, the small-high-schools team succeeded in creating the 200 schools that Klein had imagined. All were mission-driven, most with a specific theme or subject, including college prep and career and technical specialties. And these were the schools that would prove so successful: raising graduation rates of previously underperforming students by some 10 percentage points.

“We never lost track of the fact that it was about graduating more kids career- and college-ready,” says Hughes. “But I think equally important was the fact that you had everybody at the table, and so you could learn and make mistakes together and build a sense of collective trust as you went forward.”

Building a System That Works for Kids

Eventually, Cahill and her colleagues would draft a Secondary Education Reform Plan that included literacy programs, introduced “small learning communities” in to larger schools, and provided the administrative support necessary to ensuring success.

Cahill had realized early on that there were several tracks to the high-school turnaround gauntlet, and that she didn’t have enough data to be sure exactly what kind of system to build.

“We knew what made effective schools,” she recalls. “Leadership, high-quality teaching, coherence, mission, youth development…. But we didn’t know how many of what kind of kid was actually in the system.”

Cahill coaxed her longtime collaborator JoEllen Lynch into joining the effort. Lynch had worked in the trenches of inner-city education for nearly 20 years, helping a nonprofit organization called Good Shepherd Services create education alternatives for the city’s most disenfranchised children. Cahill and Lynch reached out to the Parthenon Group, a data analysis and research firm in Boston, to find out how many of which kind of student was out there, which students fell behind, how they progressed through the system, what the outcomes were, and how those outcomes differed by program.

Parthenon began gathering data on every student who entered New York City’s high schools in 1999, nearly a quarter million of them, and by 2005, as education journalist Sarah Garland reported in a 2010 Washington Monthly story, had accumulated data that were “shocking”: “Nearly 140,000 high-school-age youth in the city were at least two years behind where they needed to be to graduate on time. They had failed one or more grades in elementary or middle school and were way behind in accumulating the 44 high school credits they needed to graduate.”

Cahill asked Parthenon to find out the exact role played by school size in student outcomes. “So many people were saying to me,” she recalls, “‘If size is the problem, why isn’t it the problem for Stuyvesant?’” One of eight specialized public “exam” schools in New York, Stuyvesant had 3,200 students and a 98.4 percent four-year graduation rate.

Parthenon discovered that school size mattered much less (it explained 9 percent of the variation in outcomes) than did concentrations of low performers in the schools (which explained 22 percent of the variation). And with another statistical flourish, Parthenon determined that, together, school size and concentrations of low performers explained 41 percent of the variation in the outcomes.

Just those two variables, concluded the Parthenon researchers, were a “a powerful predictor of an individual school’s ability to prevent Level 1 and Low Level 2 students from falling behind.” (Level 1 and Level 2 were New York State score categories on standardized state math and ELA tests, where Level 1 was not proficient and Level 2 was below proficient. Thus Level 1 [L1] and Low Level 2 [LL2] scores on 8th-grade exams, though not a perfect metric, suggested that a student was one to three grade levels behind when entering high school.) Together with the significance of school size, the predictive power of the concentrations of L1 and LL2 represented something like the keys to the kingdom. The researchers could then measure a high school’s “preventive power”—its capacity to prevent students from becoming over-age and under-credited.

The Parthenon report put 14 sample high schools on a chart to illustrate the point. The Manhattan Village Academy, with just 359 students—52 percent of them L1/LL2—had a preventive power score of 86. This meant just 14 percent of its low-performing students would become over-age and under-credited with a high probability of dropping out before graduating. At the other end of the chart was Richmond Hill High School with 3,696 students, 58 percent of whom were L1/LL2. Parthenon determined that Richmond Hill had a preventive power score of just 55, that is, 45 percent of its students would end up over-age and under-credited—in other words, it was a dropout factory.

In sum, the report provided Cahill and her team with powerful evidence that they were on the right track in their pursuit of a small-schools strategy. But now they knew that not only would they need to create what Parthenon called “beat-the-odds” small schools, but they also had to dilute the concentrations of low performance in those schools. And so in 2004, a citywide system of choice for middle school students going to high school was born. With the new open-enrollment system, educators believed they could capitalize on the Small Schools of Choice reform.

Conclusion

After jump-starting small school creation in New York City and in districts throughout the country, the Gates Foundation has since turned its attention away from small schools. “Foundation president Bill Gates concluded that small schools did not have the effect on college readiness and graduation rates that he expected,” explained researchers from Duke and MIT.

New York City, however, showed how important all the other “attributes” are. Academic rigor and personalization are critical, and the layers of implementation require administrative expertise, management finesse, and political savvy. “Personalization,” for instance, doesn’t just mean making eye contact. It means giving teachers and students a focus to their school mission and a personal stake in the school, and creating a system that ensures accountability for results.

“The new small schools actually only worked because we were making systemic changes,” says Cahill, to ensure “that teaching and learning, human resources, finance, facilities, accountability, procurement, partnerships would be coordinated and problems solved rather than going into the black hole of bureaucracy.”

“In summary,” concluded a 2012 MDRC report that first gave evidence to the stunning success of New York’s small-high-schools program, “the present findings provide highly credible evidence that in a relatively short period of time, with sufficient organization and resources, an existing school district can implement a complex high-school reform that markedly improves graduation rates for a large population of low-income, disadvantaged students of color.”

MDRC issued a 2014 follow-up report, noting that “these graduation benefits do not come at the cost of higher expenditures per graduate.” Why? Because Cahill and her team worked smarter and, by getting so many more kids to graduation a year earlier, cheaper.

New York has proved that high school reform is possible; that boosting graduation rates of the poor and unprepared, even if the effort is begun in high school, is possible; that small alone is not enough; that choice alone is not enough. The package of elements that make for successful schools, identified by educators for several decades, is what is needed. And by following the money and making sure that it is targeted toward student achievement, it is a package that is affordable.

Adapted from Peter Meyer’s “Breaking the Mold: New York City’s Small High Schools Give All Students a Chance to Graduate,” prepared for the Thomas B. Fordham Institute’s Education for Upward Mobility Conference, December 2, 2014.