Unrequited Promise

Illustration by Anita Kunz

In the summer of 1991, in response to President George H.W. Bush’s major education initiative, CEOs from a number of major corporations established the New American Schools Development Corporation (NASDC, later shortened to New American Schools, or NAS), a privately funded, nonprofit organization devoted to supporting the design and dissemination of “whole school reform” models. NASDC’s founders envisioned a complete overhaul of American education stimulated by the spread of these innovative designs. As one put it, school reformers who hoped to receive NASDC grants had to “cast aside their old notions about schooling–to start with a clean sheet of paper, and be bold and creative in their thinking, and to give us ideas that address comprehensive, systemic change for all students for whole schools.” To President Bush, NASDC represented a major step toward an “educational revolution” that would “seek nothing less than a new generation of schools.”

The idea was to apply a research -and -development model to a sector that too often fell for romantic–and untested–notions of how schools and learning should be structured. NASDC’s early leaders were determined to apply a no-nonsense business approach to their efforts, to create an organization that was as lean and agile as the corporations they led. New American Schools would be less bureaucratic and more aggressive in responding to new ideas than the typical government agency or major foundation. Moreover, NASDC would be a sort of venture capitalist for education, constantly evaluating its investments and continuing to fund only those designs that proved their effectiveness.

Initially, the plan for NASDC was to complement the creation of 535 federally funded “break the mold” schools by 1996. Each of the 435 congressional districts in the country would be home to one of these schools; two more schools per state would be funded as well. Each school would receive a one-time federal grant of $1 million in start-up funds. The privately funded NASDC would support R&D teams tasked with designing innovative reform models to be implemented in these schools.

NASDC’s original plan didn’t include supporting the implementation of these designs. But as the political environment soured early on, it became increasingly clear that NASDC wouldn’t have 535 federally funded laboratories in which to test its designs. In light of this, NASDC had to become more than just an R&D center; it would have to support not only the R&D efforts of the design teams, but also the dissemination and implementation of their ideas. Moreover, if it was to keep its promise of acting as a kind of venture capitalist for education reform, at some point these design teams would need to become self-sustaining, taking fees from school districts in return for their expertise.

Over a decade, this mandate would transform New American Schools from a privately funded R&D center that would support only effective whole-school reform models into a savvy service provider for research teams that needed to market and fund their designs. Their ability to sell those designs, in turn, was heavily dependent on the ability of New American Schools to lobby Congress for more federal funding of whole-school reform. In some ways, these changes strengthened NASDC’s claim of being a bold new venture in education reform, since few privately funded initiatives had ever committed to so ambitious a program. But along the way New American Schools lost sight of its mission and became more of a typical Washington insider than an education revolutionary.

Progressivism Rides Again

NASDC’s initial funding competition sought innovative ideas for creating schools in which all students would reach “world class standards in English, mathematics, science, history, and geography” and be “prepared for responsible citizenship, further learning, and productive employment.” The “request for proposals,” or RFP, process also required that the designs be replicable in other communities: “This is not a request to establish ‘model’ schools. The designs must be adaptable so that they can be used by many communities to create their own schools. The important thing is that long after NASDC has disappeared from the scene, its legacy for new designs will remain.” NASDC funds would support the start-up of these break-the-mold schools, but once they were running it was expected that they would operate on the same budgets as conventional schools.

Once NASDC chose the 11 winning proposals (see sidebar below), it established a three-phase process for refining, implementing, and marketing these designs. Phase one provided one year of funding, during which the teams fine-tuned their original proposals “into workable designs for school transformation.” At the end of phase one, NASDC dropped two of the designs, the Bensenville New American Schools and the Odyssey Project. Phase two provided an additional two years of NASDC funding for the design teams to implement (and further refine) their ideas and approaches. At the end of phase two, NASDC determined that two more designs, Community Learning Centers and Los Angeles Learning Centers, would not receive further support (although the Los Angeles group eventually rejoined New American Schools with a new name, Urban Learning Centers, and a more focused mission). Finally, phase three envisioned a two-year scaling-up process in which the remaining designs would be put into practice in a “large number of schools across the country.” NASDC stuck to its timetable for phases one and two, but the scale-up phase, which began in mid-1995, has taken longer and, in essence, continues today.

The popular and media response to the NASDC initiative was generally upbeat and positive, but a small, vocal group of critics repeatedly questioned NASDC and the designs. Initially the negative assessments came primarily from liberal critics of the Bush administration. Typical was a 1992 article in the Nation, in which the authors declared: “Most of the educational R&D teams endorsed by the corporation comprise an incestuous circle of right-wing ideologues and privatization advocates, teacher-hating technocrats and recession-rocked military contractors, their funding made palatable to the press by token support for established and respected liberal school-reform advocates.”

Others argued that NASDC had misplaced priorities and ignored such pressing issues as equalizing funding for poor districts and addressing the needs of minorities. Former U.S. commissioner of education Harold Howe II, for example, derided NASDC’s business-oriented approach, declaring, “You can bet that, when these parties get together, they will spend little time in making a diverse society work.”

Historians David Tyack and Larry Cuban wondered whether the New American Schools initiative was just the latest example of “ambitious reformers” promising “to create sleek, efficient school machines ‘light years’ ahead of the fusty schools of their times,” yet ultimately being overcome by the education system’s unreceptive environment. They argued that such efforts to reinvent schools have in practice “often resembled shooting stars that spurted across the pedagogical heavens, leaving a meteoric trail in the media but burning up and disappearing in the everyday atmosphere of the schools.”

These criticisms were provocative, but ultimately proved less than prophetic. Contrary to Howe’s prediction, the New American Schools designs have focused overwhelmingly on improving education for urban and minority students. And NASDC and its design teams, rather than “disappearing in the everyday atmosphere of the schools,” have survived for a decade, mainly due to their ability to adapt to the changing political and pedagogical environment.

Ironically enough, the attacks from conservatives have proven more prescient. They maintained that the RFP process was skewed to ensure the choice of safe rather than break-the-mold designs. The competition was more like those run by government agencies or major foundations than an agile start-up venture. This gave an advantage to experienced grant writers, and most of the winning proposals came from established education reformers and their organizations.

The competition rounded up not only many of the usual educational suspects but the usual educational ideas, too. Lynn Olson of Education Week pointed out that one “striking feature of the award-winning designs is how many ideas they have in common.” A more striking feature was how many of those ideas were deeply influenced by the questionable theories of the progressive education movement of the early 20th century, the original break-the-mold movement in American education. Indeed, most of the winning proposals read like echoes of progressive manifestos of the 1920s. Six of the eleven designs, for example, expressly declared that they were “learner-centered.” Eight sought to change the relationship between teachers and students by transforming the teacher into a “coach,” “facilitator,” or “guide.” All promised to meet “world class” curriculum standards, but only two (Expeditionary Learning and Modern Red Schoolhouse) focused their programs on students’ explicitly mastering academic disciplines. Nearly all the designs promoted interdisciplinary curricula to avoid what they saw as a key problem of traditional schools: teaching knowledge and skills in isolation.

Related to this interdisciplinary focus was the widely shared commitment to use the progressive-inspired “project method” to engage the interest of students. Eight NASDC designs proclaimed that project-based education would make learning exciting and relevant to students in ways that traditional textbook-centered education never could. Eight proposals envisioned using the community as a classroom to break down barriers between school and society, an idea championed by John Dewey, the great patriarch of progressive education. Like Dewey and other progressives, most of the proposals also argued that their design would not lead to the “piling up of facts,” but would produce critical thinkers.

The problem, as critics such as Robert Maynard Hutchins, E. D. Hirsch Jr., and Diane Ravitch have repeatedly shown, is that apart from a small number of schools serving highly motivated, well-to-do students, progressive education has rarely fulfilled its pedagogical promise. These critics have argued that progressive ideas often contribute to the evisceration of academic curricula and narrow the scope of student learning to entertaining but largely vapid topics. Especially relevant to the NASDC designs was the late Jeanne Chall’s finding, after surveying a host of research studies, that, on average, schools guided by progressive ideas have been less successful in raising academic achievement, especially among disadvantaged children.

It was, in fact, controversies over progressive ideas that were at least partly responsible for the Bensenville and Odyssey Project designs’ collapse in the face of angry community challenges. Both designs were grassroots efforts developed by local communities; neither was guided by or associated with the work of a national education leader. Still, among the most contentious elements of both designs was their progressive reorganization of the schools. A sociological study of the North Carolina-based Odyssey Project found that parents were bothered by the lack of course grades and college prep courses in the Odyssey curriculum. They saw Odyssey as an attempt to shift the curriculum to a more vocation-oriented track that would ensure a “compliant, trained workforce” for local industry.

These developments should have sent a warning to NASDC leaders about the politically charged and therefore tenuous nature of such broad efforts at education reform. These debacles showed that the success of NASDC-style reform–no matter how good the design nor how much money available–would depend ultimately on keeping all the community’s stakeholders committed to change over the long haul.

Regardless, phase one had gone generally well for the other nine design teams, and NASDC leaders plowed ahead undaunted. As businessmen, they recognized that any new venture has some failures and losses. More troubling for them at the time were two emerging problems–NASDC’s precarious financial situation and concerns about the organization’s survival following the defeat of George Bush by Bill Clinton in 1992.

Scale-Up

In its first two years of operation, NASDC raised only about $50 million of a planned $200 million. Concerns about fund-raising deepened with fears that NASDC was too closely tied to the Bush administration to win the new president’s support.

In May 1993, however, both President Clinton and Secretary of Education Richard Riley stated that the NASDC’s reform efforts dovetailed nicely with the new administration’s education agenda. In fact, this was no real stretch for Clinton. He had represented the nation’s governors at the 1989 meeting at which Bush and the governors had agreed on this agenda, and his administration embraced many of the same goals.

Yet Clinton’s support had little immediate effect on fund-raising, and NASDC watchers began asking whether the organization could remain in business much longer. In response, the NASDC board elected David Kearns CEO. Kearns, who had previously served as deputy secretary of education in the Bush administration and as CEO of Xerox, was widely respected in political, education, and business circles. The board hoped he would find ways to shore up the organization’s precarious financial situation. In January 1994, these hopes were borne out when philanthropist Walter Annenberg donated $50 million to NASDC–a life-saving development for the organization. Without these funds it was unlikely that NASDC could have made the crucial transition from serving a modest number of schools to its broader scale-up to the state and district level.

Further enhancing NASDC’s national profile was the fact that its key scale-up idea–moving from reforming individual schools to transforming entire systems–was quite similar to the idea of systemic reform that was one of the cornerstones of the Clinton administration’s education policy. NASDC was now freed from the allegation that it was little more than a partisan GOP initiative. In the process, however, NASDC also appeared to be distancing itself from its “revolutionary” free-market approach to education reform and was increasingly looking like a respectable if rather corporate member of the education establishment.

With its status bolstered, NASDC turned to its biggest challenge: scale-up. Between 1992 and 1995, nine NASDC-supported design teams had implemented their plans in 148 schools. Phase three called for a dramatic increase, with projections reaching into the thousands. At the time, however, none of the designs had demonstrated that it would lead to dramatic improvements in student achievement (although Robert Slavin’s Success for All program, the cornerstone of the Roots and Wings initiative, had shown promising results).

Nevertheless, NASDC moved forward with its scale-up plan. Agreements were ultimately reached with the states of Kentucky and Maryland and nine large urban districts: Cincinnati, Broward County (Fort Lauderdale), Dade County (Miami), Memphis, Philadelphia, Pittsburgh, San Antonio, San Diego, and Seattle (and a group of contiguous school systems). By mid-1997, two years into phase three, the number of participating schools had climbed to 553.

Memphis would become NASDC’s signature district. In 1997 its student population was more than 80 percent African-American, and more than 60 percent of all students qualified for free or reduced-price lunch. Only 40 percent of its 9th graders were passing the state’s minimum-competency test. However, Superintendent Naomi “Gerry” House was determined to use whole-school reform to try to solve Memphis’s education ills. By 1997, Memphis had almost reached the NASDC goal for jurisdictions participating in the scale-up of having 30 percent of its schools adopting some form of whole-school design. Encouraging news about student achievement soon followed. A study comparing 25 NASDC school sites in 1995-96 with a control group of non-NASDC schools found “significantly greater gains” in the redesigned schools.

Meanwhile, less encouraging news was emanating from the RAND Corporation’s studies of NASDC’s phases one and two. RAND, with which NASDC had been working from the beginning, found that one of the main problems in the early stages was the grating between these innovative designs and the superstructure of state and district policies. This spelled trouble for the scale-up phase, when entire systems would be expected to convert to these new designs while keeping such major influences as union-negotiated contracts and state policies largely intact. For instance, in a RAND survey of principals in NASDC-supported schools, a large majority reported that “standardized, multiple-choice tests are misaligned with the classroom practices of reforming schools.” This is a common criticism among all educators, but here the NASDC principals were voicing a more pointed concern, namely that there was a mismatch between the student-centered approaches of the design teams and the more traditional evaluation practices used by states and districts.

Moreover, a 1995 survey of parents in several NASDC jurisdictions found that they were deeply divided about reform, with over a third of the respondents wanting “the best of new and innovative methods”; just under a third wanting “tried and true” basics; and the remaining third wanting “a combination of old and new methods.” Getting such communities to commit to long-range plans for whole-school transformations would be a formidable challenge. This indicated a process of change that was more incremental and political than the NASDC teams had expected, as they would soon learn in Memphis. The early phase indicated that NASDC’s original perspective was somewhat simplistic, even naive. NASDC could not simply apply the business metaphor–“develop and test a new product and then go sell it”–to educational change.

As a result, by 1995, NASDC leaders rarely spoke about creating break-the-mold schools; the words “education revolution” had disappeared entirely. The realities of education politics were forcing NASDC to adapt and adjust its approach and expectations.

Congress to the Rescue

By mid-1997, NASDC, now named New American Schools, or NAS, was well into becoming more of a clearinghouse than an initiator of reform and was changing the design teams into market-oriented purveyors of education improvement. NAS increasingly focused on helping the teams become self-sufficient organizations capable of marketing and supporting their products. For some teams, this change was relatively painless. Success for All, the parent organization of Roots and Wings, was already a fee-for-service institution with substantial experience in this area. But others, such as the Modern Red Schoolhouse, had virtually no business or marketing experience. Thus, during phase three, NAS began sending financial and marketing experts to the design teams to help them develop business plans and marketing strategies. At the same time, NAS began addressing the problem of creating markets for the designs.

The most immediate problem for NAS and the teams was finding funds that districts could use to pay for their services, since design-team fees for the first year of implementation alone could range from $50,000 to more than $500,000 per school. This effort was especially crucial given that so many of the clients targeted by NAS were low-performing, financially strapped urban systems.

Help came in the form of the Comprehensive School Reform Demonstration Program, better known as the Obey-Porter Amendment. Passed by Congress in 1997, it provided about $150 million for competitive grants to aid schools in adopting “proven” whole-school reform models. Supporters of the amendment routinely cited the NAS designs as representative of the kinds of programs they hoped school districts would adopt. In fact, NAS marketing efforts and Obey-Porter funds have brought a five-fold increase since 1997 in the number of schools using NAS designs, to some 3,000 schools in 2001. Not surprisingly, NAS had lobbied Congress on behalf of the amendment.

One could view Obey-Porter as helping NAS fulfill its original role, namely acting as a private R&D agency whose designs would be implemented in schools that received federal start-up funds. Yet if Obey-Porter aided NAS in fulfilling an important goal, it also indicated that NAS was changing from an agency committed to operating outside the constraints of Washington into one that depended on Washington for the success of its venture.

However, the most troubling aspect of Obey-Porter and, by implication, of the new role played by NAS design teams in education reform, was not the dependency on federal funding but the claims of “proven effectiveness” that the law explicitly made on behalf of the designs. With the possible exception of Roots and Wings, at the time that Obey-Porter was enacted the NAS designs still couldn’t provide substantial evidence of effectiveness.

Disappointing Results

Indeed, NAS officials’ reaction to the 1999 American Institutes of Research (AIR) study An Educator’s Guide to Schoolwide Reform would reveal an organization that had not come to terms with the importance of evaluating student achievement in the reform process. Of 24 whole-school reform designs, AIR found that only three had “strong” positive effects on student achievement: Direct Instruction, High Schools That Work, and Success for All, none of which were NAS designs. Of the eight NAS programs, AIR rated Expeditionary Learning “promising” and Roots and Wings (the NAS version of Success for All) “marginal” in boosting achievement. All the others received no rating because there weren’t enough data.

New American Schools president John Anderson’s reaction, in a 1999 column for Education Week, was to maintain that the AIR review was flawed in its approach and conclusions. “Not surprisingly,” he argued, “AIR took a traditional approach, gathering the relatively small body of published research on the programs as its evidence of effectiveness. But this sort of reliance on limited and often time-consuming research as a way of assessing comprehensive designs may be outmoded.” Anderson held that “traditional” evaluation took too long and ignored data on progress that schools were gathering on an almost daily basis. In addition, he wrote, “Even the most careful research on a handful of campuses and a handful of matched schools could not give us a picture of what happens when models are used in hundreds of schools.” Anderson seemed troubled by the attention AIR gave to student achievement in making its assessments. “Beyond academic results,” he asserted, “design teams should be judged on other practices involving their relationship with schools and school districts. Practices involving implementation, communication, and data collection are all vital to viable campus change.”

Was Anderson implying that the NAS designs were so novel that standard approaches to research could not capture what they were doing? Was he questioning all traditional external reviews of the NAS programs? If so, these concerns didn’t stop him from citing data from a traditionally designed study on student achievement in Memphis as evidence that the designs in that city were working.

Meanwhile, the data finally produced by RAND on the performance of NAS schools in scale-up jurisdictions relative to other schools in the same district showed mixed results. About half of the NAS schools in Cincinnati, Memphis, Miami, and Kentucky (the cities where RAND studied math performance) outperformed the district average in mathematics, while 50 percent or more of the schools in Cincinnati, Philadelphia, San Antonio, and Washington (the cities where RAND studied reading performance) improved in reading. In other words, about half of the NAS schools in the cities studied by RAND only met the district average or underperformed relative to other district schools. RAND identified Roots and Wings as the most consistent design, with 10 of 21 schools making progress in both reading and mathematics relative to the district. Modern Red Schoolhouse did even better, with 7 of 11 of its schools improving in math and 8 in reading. Other designs showed considerable variation. Of the Audrey Cohen schools, for example, 5 of 8 schools improved in math, but only 2 in reading. Similarly, 9 of 24 ATLAS schools and only 4 of 16 Expeditionary Learning schools improved in math.

The inescapable finding from RAND was that, despite millions of dollars and enormous effort, over the course of a decade what NAS had brought about was at best incremental change and modest achievement gains. RAND’s mixed assessment was: “Dramatic achievement gains were not made, although some designs fared better than others.”

Another blow to NAS’s performance record was dealt in June 2001, when the Memphis school district announced that it was abandoning the district’s six-year experiment in whole-school reform. In the spring of 2000, when Johnnie Watson took over as superintendent from Gerry House, the great promoter of whole-school reform, he found a deeply troubled district. In the late 1990s, House had mandated that all of Memphis’s more than 160 schools adopt a reform model, a policy that angered and alienated many teachers. Worse, an internal district study found that after six years of reform and the expenditure of some $12 million, Memphis students had exhibited virtually no gains on state tests. Moreover, the study revealed that only 3 of 18 reform models implemented in the city had boosted student achievement; none of the three was a NAS design. Indeed, one NAS design, ATLAS, had “the most negative impact” on student achievement of any of the reform models. The report found that even Success for All, which was being used in almost 25 percent of the Memphis schools, performed poorly. There have been questions about the quality of the study, but this merely underscores one of NAS’s fundamental problems: the lack of independent, reliable research on the effectiveness of its designs, a deficiency that allows decisions like Memphis’s to be made more on the basis of politics than sound evidence about what works.

These events should signal to NAS that the future of whole-school reform rests as much on the fluid and unpredictable nature of educational politics, changes in leadership, and the sufferance of parents and teachers as it does on successful marketing, buy-in from key administrators, and substantial federal funds. Moreover, the Memphis experience highlights the importance of the bottom line–student achievement–for all such reform efforts.

How will Memphis’s rejection of whole-school reform affect NAS? While it is certainly too soon to offer a definitive answer, it seems likely that NAS will weather this storm, although its momentum may be somewhat slowed. The flow of Obey-Porter and, increasingly, Title I funds to support whole-school reform will sustain NAS and its design teams for many years to come.

Lost Chances

At its inception New American Schools held the promise of being an extremely exciting research-and-development initiative in education. It would sponsor the creation of innovative designs, pilot test them in a select group of schools, and decide whether they were effective enough to warrant wide dissemination. Its evolution into a financier, lobbyist, and marketing shop for a variety of as-yet-unproven whole-school designs must be considered at best a modest contribution to the cause of education improvement. In its current incarnation, NAS is almost indistinguishable from the foundations and government agencies that have dominated school reform efforts for decades. It has become part of the very education establishment its leaders once sought to circumvent.

More important than the changed status of NAS is the increasing prominence of NAS-style whole-school reform as one of the most popular strategies for education change. NAS has played a key role in that development. Yet its own experience in Memphis raises questions about whether the commitment to whole-school reform is the best strategy for improving educational outcomes in troubled urban schools. With so many unmanageable and unpredictable variables, whole-school reform will always be tenuous, because it attempts to simply glide by many of the institutional constraints–the variety of programs operating within a school, all with different goals; the requirements of the administrators’ and teacher unions’ contracts–that make large-scale change so tortuous. University of Michigan historian Maris Vinovskis and other scholars have suggested that less ambitious and less costly efforts such as more learning time during the summer might yield gains as great as or greater than comprehensive school reform.

One of Johnnie Watson’s first moves after ending Memphis’s experiment with whole-school reform was to order a new district-wide curriculum. A primary reason was the problem the district had experienced in adopting so many different designs. One teacher noted: “Since Memphis has such a mobile student population students fell even further behind because nothing was the same from school to school.” High rates of student mobility are an ever-present feature of all urban school districts. Yet the NAS design teams did not appear to have taken that vital fact of urban life and education into consideration when they developed their programs. The seeds for significant gains may lie in such small insights about student mobility rates and district-wide curricula.

-Jeffrey Mirel is a professor of educational studies and history at the University of Michigan and the author of The Rise and Fall of an Urban School System: Detroit, 1907-81 (University of Michigan Press, 1993).