What can education leaders, advocates, and policy makers learn from the failures of recent high-profile school reform initiatives?
Ever tried. Ever failed. No matter. Try again. Fail again. Fail better.
— Samuel Beckett
American education is littered with failed reforms. Across the country, we see charter schools that have been shuttered, federal funding streams that have run dry, philanthropic initiatives that never panned out, and brand-new teacher evaluation systems that have already been marked for the junkyard.
Of course, failure isn’t necessarily a bad thing — when pursuing a goal as urgent and complex as school improvement, some amount of failure will be inevitable. The problem, though, is that policy makers, foundation officials, and pundits have strong incentives to deny that their favored initiatives have gone badly, and they rarely acknowledge and learn from those failures before moving on to the next reform. As a result, they tend to repeat their mistakes and make much less progress than they should.
Nor have past failings caused education reformers to slow down. The last few years have seen major new philanthropic organizations, such as the Chan Zuckerberg Initiative and the Emerson Collective, enter the fray and make splashy investments in personalized learning, school redesign, and a host of other strategies. Meanwhile, with every new legislative session, a raft of new policies, regulations, and programs emerges from state capitals and trickles through public agencies, local school districts, and eventually into schools. Further, because the Every Student Succeeds Act gives states more policy-making discretion than they have had in years, the pace of experimentation will likely continue to increase.
What can we do to make the next generation of reforms more successful than the last one? As the old proverb puts it, the best time to plant a tree is 20 years ago, but the second-best time is today. We cannot go back and undo the mistakes that education reformers and policy makers have made over the last few decades — having to do with standards and accountability, teacher evaluation, and more — but we can learn from them.
To acknowledge mistakes is not to malign those who committed them. Some critics may choose to portray the contemporary school reform movement as a nefarious conspiracy on the part of evil plutocrats hoping to destroy America’s schools, but in fact, the vast majority of reformers (ourselves included) are well-meaning people who are acting in good faith and trying to do right by kids. No doubt, we’ve done many things wrong, but in addressing those mistakes, it’s not particularly useful to focus on whose side we’re on or what motives drive us. Rather, the question is, what lessons can we learn, so that we can “fail better” in the future?
With that question in mind, we convened a conference in May 2017 in Kansas City, featuring a number of the country’s leading education scholars who represent a broad range of perspectives on educational improvement: Larry Cuban (from Stanford University), Matthew DiCarlo (the Shanker Institute), Anna Egalite (North Carolina State University), Rick Hess and Paige Wiley (the American Enterprise Institute), Ashley Jochim (the Center for Reinventing Public Education), Matthew Ladner (the Charles Koch Institute), Megan Tompkins-Stange (the University of Michigan), Martin West (Harvard University), and Daniel Willingham (the University of Virginia).
What lessons can we learn, so that we can “fail better” in the future?
We asked each to write and deliver a paper about an education reform that failed, with attention to the lessons learned from that experience. We didn’t tell the participants what to write about, but we did encourage them to focus on a reform strategy with which they were generally sympathetic. (It’s tempting to ask scholars to criticize others’ mistakes, but we reasoned that it would be more productive to ask them to reflect on failures that challenged their own beliefs — most of them did so.) The result was an edited volume, Failure Up Close, released in January 2018.
While the themes of these papers varied widely, their arguments tended to overlap in interesting ways, suggesting that for all of the sharp disagreements that have divided the field of K-12 education in recent years, it may be possible to build a strong consensus in some areas. As we describe below, many of our authors and conference participants agreed that school improvement cannot succeed unless policy makers, advocates, and educators are willing to make a few difficult trade-offs and observe a few critical lessons.
Most important, our contributors voiced a strong consensus about the nature of education policy: After years of treating school reform as a competition among theoretical perspectives and technical strategies, the time has come for all of us to recognize that education is an inherently political enterprise, in which we all constantly struggle over how best to spend the public’s money and serve the nation’s children. The sooner we stop trying to “get politics out of education” — as though that were possible — the sooner we can engage in the real work of educational improvement. Politics is about persuasion. Politics is about interests. Politics is about consequences. Education policy cannot escape those dynamics, nor should it, because politics is the best tool we have as a society to adjudicate competing values. It is also the means by which we must navigate the following trade-offs and learn the following lessons.
Those hoping to improve the education system must be honest about the difficult trade-offs inherent in education reform. Very rarely are decision makers presented with one option that’s clearly better than all the others. Rather, they have to choose from among any number of competing philosophies, values, and beliefs about teaching, learning, and school improvement. And when they make a specific choice, it’s often because somebody — e.g., a state official, the federal government, the head of a large philanthropic organization — has put their thumb on the scale, using their influence to favor one approach at the expense of another. Such trade-offs are inevitable, and we ignore them at our peril.
#1: Urgency vs. prudence
Sometimes, two conflicting arguments are both true at the same time. For example, it’s true that millions of American children are stuck in schools that are not meeting their needs, and if we care about those kids, then we should fix their schools immediately. On the other hand, it’s also true that school improvement happens slowly, in fits and starts, and if we care about those kids, then we shouldn’t rush into anything. Often, when education reformers try to make quick, sweeping changes, they just make things worse.
The desire to go big and go fast has its advantages. Ambitious school reform movements create a sense of urgency, optimism, and emotional commitment, and they tend to win the backing of powerful political leaders and the financial support of major philanthropic organizations (who always make those movements look great in their glossy annual reports, attracting even more supporters).
The sooner we stop trying to “get politics out of education,” the sooner we can engage in the real work of educational improvement.
But going small and slow has its virtues as well. It is far easier to identify problems and make course corrections to a program that serves 5,000 students than one that reaches 5 million. Smaller, more cautious initiatives are far less likely to burn out their own leadership or exhaust their teachers and staff. They are less likely to promise the moon and then, once their lofty ambitions fall short, alienate the families and communities that were counting on them. And when small initiatives fail, the effects are felt by a relatively modest number of students.
To be sure, the ability to act on a large scale is one of the chief virtues of public policy, but it is also a danger. Whether it is “no-excuses” schools sucking up most of the charter school movement’s oxygen (and funding, and cap space) or the rush to adopt new classroom technologies before teachers are ready (or want) to use them, reformers often get ahead of themselves in their quest for scale. If they hope to arrive at effective, widely adopted policies and practices, they must figure out how to balance their ambitions with a sense of prudence and a willingness to tinker and make adjustments along the way.
#2: Top down vs. bottom up
Nearly 40 years ago, Richard Elmore (1979) pointed to what he called the “noble lie” of public administration: Some organizations — schools, for instance — feature a large amount of ground-level decision making that central planners simply cannot control, despite claiming to do so. A decade later, in his seminal work Bureaucracy, James Q. Wilson (1989) tugged on this thread a bit harder, arguing that central offices often struggle even to keep track of what goes on in such “coping organizations” on any given day. As a result, trying to impose top-down policies on schools is fraught with challenges.
This hasn’t stopped policy makers from trying, though. Over the last decade, for example, many states and districts rushed to implement centralized teacher evaluation systems, even though critics warned, for good reason, that they wouldn’t provide valid or reliable measures of classroom practice. Similarly, the federal government’s School Improvement Grant program required applicants to choose from a list of highly prescriptive school turnaround models. Not surprisingly, neither of these reforms achieved their intended results.
Don’t get us wrong — “bottom-up” reform has its challenges, too. Working to improve practice at the school or district level is a messy and uncertain process, requiring coordinated efforts by disparate groups of people. The old adage “too many cooks spoil the broth” remains apt. Local politics is often dysfunctional, and relying on a grassroots political process can seem like a painfully slow and inadequate way to address students’ urgent needs. On the other hand, without the buy-in, input, support, and elbow grease of the people actually doing the work at the ground level, no school improvement strategy will succeed, no matter how elegant its design.
#3: Politics as the cause of failure vs. politics as the solution
It was no less a philosopher than Homer Simpson who came up with that memorable toast to alcohol: “The cause of, and solution to, all of life’s problems.” If he were involved in school reform, he might just as well have addressed his toast to politics.
Politics certainly does cause many of the problems we face in K-12 education. Thanks to political pressure, for instance, state and district leaders often cut promising initiatives and programs prematurely, before they’ve gone through the multiple years of implementation, assessment, redesign, and improvement that might have enabled them to succeed. If one political party backs a new program, then the other party will try to kill it off at the first sign of weakness. In lean times especially, partisans reach quickly for the budget axe.
But at the same time, politics is the lifeblood of school improvement. No new educational program can flourish without political support, and, in turn, no program can maintain that support unless it delivers on at least some of its promises. To borrow a line often attributed to Harry Truman, good policy is good politics.
Perhaps even more important, politics is the final arbiter of the success or failure of any given policy, teaching strategy, or school improvement model. Research findings, anecdotes, and test scores can inform the debate about whether a particular school reform has worked or not worked, or whether it is a “good” approach or a “bad” one, but the decision to continue or abandon that approach will always be political in nature.
So how can policy makers, philanthropists, politicians, school leaders, parents, and taxpayers navigate the contradictory world of education policy, balance these trade-offs, and maximize their chances of doing right by kids? There are no easy answers. If we’ve learned anything from the last two decades of school reform, it is that education reform is hard, uncertain, and often frustrating work. But, there are some lessons that we can glean from past efforts.
Lesson #1: Be humble
Whether it is the 100% proficiency goal of No Child Left Behind or the School Improvement Grant program’s promise to turn around the bottom 5% of schools, education reformers have wildly exaggerated the potential effects of their policy proposals. Worse yet, supporters of these programs knew full well, from the beginning, that these promises were ridiculous, but in hopes of scoring quick political wins, they were all too willing to say things that they knew to be false.
In the short term, this kind of bravado may have helped reformers win support, but it also guaranteed that their initiatives would be viewed as failures over the long term. When told to pursue unrealistic goals, schools and communities eventually lose faith, and programs die on the vine. Modest goals might not be as enticing, but they’re much more easily reached.
The need for humility extends beyond goal setting, though. Every day, learning goes on in millions of classrooms at more than 100,000 schools across the U.S., and presiding over those classrooms are millions of human beings, each with their own thoughts, strengths, talents, and desires. Policy can only do so much to compel those individual teachers to act in particular ways. Thus, as the economist Friedrich von Hayek (1975) put it in his acceptance speech for the Nobel Prize, governments (and one might say the same of philanthropists, leaders of charter school management organizations, technology entrepreneurs, and others) should behave not like craftsmen but like gardeners — “not to shape the results as the craftsman shapes his handiwork, but rather to cultivate a growth by providing the appropriate environment.”
We hope that readers will examine policies with humility, recognizing that even though policy has its limits, it can positively affect outcomes.
State legislators do not operate schools, nor do foundation executives, nor even school board members. At best, they can create the conditions whereby thoughtful and talented educators can teach children the things they need to be successful in life. Focusing on how to create those conditions, rather than on trying to engineer a particular outcome, will maximize the likelihood of long-term success.
Lesson #2: You can’t make an end run around democracy
If it’s true (as we describe above) that politics is both the cause of and the solution to the problems we face in K-12 education, then reformers have no choice but to embrace the tensions of political life. Ultimately, no policy can succeed without securing a strong base of political support. And we’re not talking about some “grass tops” agreement amongst elites or a law enacted hastily by a state legislature without much public input. Given American educators’ proven ability to resist top-down control, it’s pointless to try to impose decisions on schools from above. Ideas need to be debated in the open, the proper channels need to be respected, and the hard work of convincing people has to be done in earnest. There is simply no other way.
The Common Core State Standards offer a prime example of how this can go awry. Supporters rushed the standards’ adoption, used whatever political channels were most convenient, eschewed public debate and discussion, and in doing so triggered a massive backlash that either caused states to reject the standards outright or led them to water down the initiative to such an extent that there’s little left of the original vision, which called for the careful alignment of standards, tests, curricular resources, and teaching strategies, all to be shared across state borders. To stand any real chance of success over the long term, that vision, and its tangible benefits, would have had to be embraced by a real constituency.
Lesson #3: You can’t hide behind technocracy
If policy advocates cannot make an end run around democracy, nor can they hide behind technocracy (loosely defined as “rule by experts”). When trying to build support for a policy proposal, it’s always tempting to claim “This is strongly supported by the research” or “We’re seeking to implement evidence-based best practices.” The problem, though, is that educating children is not like solving an engineering problem, and the research rarely generates clear evidence that a particular teaching strategy or educational model works. In fact, we have no common agreed-upon definition of what it even means to say something “works” in the classroom, or that it might be replicable. More often than not, research can only help us understand the various trade-offs at play in any decision we make about policy or practice — it might tell us, for example, that a particular approach has worked for some students and not others, or that it has led to student improvements on a certain measure but only when implemented in a very expensive way. That is, the evidence provided by educational research doesn’t tell us which course of action we ought to take; at best, it helps inform our all-too-human process of choosing one option over another.
It is also worth noting that the “expertise” that policies rest on can be an awfully thin reed. For example, researchers with deep expertise in one area are frequently asked to weigh in on another. And studies of “best practices” may simply call attention to a smattering of things observed at successful schools, without asking whether the same things might also be going on in less successful schools. In short, it’s always wise to be skeptical of education experts (present company included!) and to ask whether the subject they’re opining about is one that they actually understand.
The world of education reform needs to get more comfortable with complexity. More often than not, contextual factors affect the implementation of policies and even the definition of success and failure. This is not a bad thing. We live in a big, diverse, pluralistic nation that draws tremendous strength from the wide spectrum of ideas and opinions that our citizens possess. Our political process, frustrating though it may be, winnows and refines ideas, and our decentralized, federal system allows for experimentation and pressure-testing. All of this informs and shapes policy, and, when channeled to productive ends, makes policies better.
We do not want people to walk away from this article as nihilists. The purpose here is not to argue that nothing will ever work or that all sins can be washed away by saying that success is context-dependent. Rather, we hope that readers will examine policies with humility, recognizing that even though policy has its limits, it can positively affect outcomes. We hope they will also have more faith, and invest more time and energy, in our political process to put policies through the necessary paces and get them enacted with real support. We hope that they won’t pooh-pooh small innovations like a one-off charter school with a unique design, a teacher evaluation tool that a school district has worked hard to create and implement, or a set of standards that a community has created to reflect its values. Rather, we hope that they will agree that small improvements can and should be woven into our broader understanding of educational improvement and can help our school system move forward.
Elmore, R. (1979). Backward mapping: Implementation research and policy decisions. Political Science Quarterly, 94 (4), 601-616.
Von Hayek, F.A. (1975). The pretence of knowledge. The Swedish Journal of Economics, 77 (4), 433-442.
Wilson, J.Q. (1989). Bureaucracy. New York, NY: Basic Books.
Citation: Greene, J.P. & McShane, M.Q. (2018). Learning from failure. Phi Delta Kappan 99 (8), 46-50.