In support of turnarouds, Justin Cohen says, “we need an inside strategy that deals aggressively with failure where it exists.”

In support of restarts, Andy Smarick says, “My argument is that if a school is found to be failing kids, we need to try hard to fix it. If repeated efforts don’t work and it’s clear that we have a persistent failure, then you close it.”

I think Mr. Cohen and Mr. Smarick are both smart, experienced, and well-positioned in the field to know what courses of action work best and to be able to justify their opinions. But I can’t help feeling that there isn’t a “third way” solution, something that might triangulate the issue, unite the two positions, and produce a truly original approach to large-scale reform.

I call it the “philosophically constituted” school. (Not a great name, I know; I promise I’ll work on it with the folks from marketing.)

The "philosophically constituted school" is the optimal “inside strategy” for turnarounds but it would also be the best “outside” strategy for restarts. Best of all, it would be strongly resistant to turnbacks over long periods of time because philosophies, or what in this case might better be termed “belief systems”, are hard to get rid of once they’ve been acquired.

The solution I’m thinking about would involve a small but significant shift from a paradigm that says, “Education reform is all about creating effective schools.” to “Education reform is all about creating effective people who work in schools.”

This is also a different (and healthier) way of reframing the human capital problem. Instead of saying that “We need an army of well-educated and passionate young people to replace the teachers and principals we have now,” the approach I’m thinking about says “What we need is better teachING, not better teachERS, and better leadING, not better leadERS.”

I don’t mind at all, of course, if smart ambitious people want to devote their lives to education. After all, I’m one of them. But I don’t think there will ever be enough of us who will actually want to work in schools long enough to have any significant impact on our system. Nor do I think we’ll even produce a sparkling subset from this young, smart, ambitious group of educators to develop enough system-level leaders given the challenges of taking human capital to scale. No matter how much wonderful work TFA and KIPP do, (and they definitely do wonderful work) let’s all try to remember this adage: “Talent doesn’t scale; if it did, we wouldn’t call it talent.”

But something in schools does scale: beliefs.

And the proof is in the fact that so many schools are currently filled with people who hold virtually the same beliefs about education regardless of other issues like the socio-economic status of their students, the amount of money spent per student, the quality of their work environment, or whether they work in urban, suburban, or rural school.

Teachers and administrators are a fairly consistent lot when it comes to what might be called their “philosophy of school”. Obviously, if I can go into over 200 schools and have very different people in very different situations ask me the same questions over and over again, and experience the same shock and awe when I calmly describe the research base on things like grading or early literacy instruction, a common set of beliefs must exist even if the people who hold them aren’t aware of them as such. Whether we call this a “philosophy of education” or a “theory of action” or a “set of beliefs” or just common personality traits is irrelevant. At the level of belief, virtually all collections of educators are the same.

And the fact that most educators currently hold beliefs that are incompatible with effective teaching, learning, and school operation is the core problem we must address in reform.

So why not focus school improvement efforts at this foundational level? Why not start a turnaround or restart with a new set of beliefs? And why not target our work on individuals within a school rather than trying to change an entire school itself?

Virtually all reforms up to this point have been targeted at schools, not at the people inside of them. I believe this has been and will continue to be a losing proposition. It has been assumed all along that schools are the fundamental unit of reform. And, through NCLB, this idea has been codified into law. NCLB is really NSLB – No School Left Behind. Same goes for standards: they aren't for individual teachers or kids, they’re for making sure an entire state full of schools teaches the same thing. Charter school legislation is about creating new schools – not new school people. Vouchers are about sending kids to different schools – schools that may or may not be filled with different or better people.

Performance-based pay is the only reform focused directly on people – and it will be a very, very long time before we know which models are best and if any have sustainable positive effects. I predict that performance-based pay will gradually become part of school culture and that it will have virtually no system-wide effect whatsoever except for the increase of school budgets. People don’t go into education to make money, and people who are moved by money don’t go into education. Educators are, by and large, not an economically sensitive group of people; they're way more into authoritarian status levels and social capital. Don't forget that virtually all educators perform at or near 100% of their capacity (I didn’t say they performed well, just that most are doing pretty much everything they can do). Paying better teachers better money is certainly a fair and even righteous thing to do. But more money will not change the way they teach or inspire less-talented teachers to aspire to levels of practice they can’t or wouldn’t care to attain otherwise.

So what if we apply the notions of restart, turnaround, and turnback to people instead of to schools?

Why do consumers of education like Waldorf schools? Or Montessori schools? Or Friends schools? Or Core Knowledge schools? Or even military schools? Because these types of schools are grounded in an understandable philosophy and are happy to share their beliefs with everyone in their school community. But it is not the quality of their philosophies that makes them so attractive, merely the fact that they have them at all.

I have taught in over 200 schools. My consulting associates have probably taught in 100 additional schools. And, as a group, we’ve probably visited or read about another 200 schools. So from our, point of view even as a very small organization, we’ve had exposure to maybe 500 schools across the country. Here’s an interesting stat: We have never encountered a traditional public school that operated out of a well-articulated philosophy. I’m not talking about mission statements nobody follows or school improvement plans nobody reads. I’m talking about a coherent set of beliefs through which decisions are made and actions are taken.

This is why most schools don’t work. And why all schools are hard to change.

Most turnarounds fail for the simple reason that we fail to turn around the thinking of the individual people who make up the school community. Most restarts fail because we fail to start them off with a coherent set of beliefs in the first place. (“Work hard. Be nice.” is very good. But I’m thinking of something a bit more involved; and I think we all know that KIPP runs off a richer and more nuanced set of shared ideas.)

Developing a philosophy of education is much easier than one might think. We prove this in our consulting work with a simple activity we call “Core Beliefs”. We have even noticed that groups of people can work well together with different core beliefs as long as the following are all true:

1. They understand the beliefs of others.
2. They can see that everyone is trying to act with integrity out of their individual belief systems.
3. There is a small group of higher-level beliefs everyone CAN agree to.

Number three might seem to be the tough one, but it’s not.

For example, few teachers or principals or parents or district administrators have ever disagreed with my own personal core belief that “Literacy is the foundation of school success.” Or how about this one, "If you can't manage the classroom, you can't manage teaching and learning."

Or this one: after helping people realize that everyone passes their kids on to someone else in a school system, it’s easy to get people to see that, by virtue of simply working in a school, they all agree with the belief that “We’re all in this together.”

And, finally, in a paradigm-shifting tsunami of Homer Simpson-esque logic, I have also helped large groups of educators understand and reach agreement on the earth-shattering belief that “Teaching is for learning.”

This really IS a paradigm shift because our entire education system is founded on the notion that teaching is self-justifying and that learning is incidental; this is also why testing doesn’t work as a tool for accountability, but that’s another story for another time; consider Dr. Richard Elmore’s simple, insightful, and very correct idea that internal or personal accountability must precede external accountability.

As an outside change agent, as long as I can successfully inculcate these beliefs (or any other reasonable set), and then teach an administrator or two how to use them in school decision making, and how to monitor the ways people use them in their work, I can create what I like to call a “platform for progress”. If I also happen to be smart enough to feed the right teaching tools and curriculum and assessment tools into the school (those that match the school’s philosophy), progress will occur more quickly.

This is, of course, currently impossible given the slate of reforms we have chosen. And again, this is another reason why restarts and turnarounds are so hard. The system we’ve designed to force these things to happen, makes it’s highly unlikely that they will. (And, yes, I have an alternative set of reforms in my back pocket, but this post is way too long now as it is, so I’ll talk about these at another time, probably when the issue of NCLB reauth comes up.)

Here’s the big “Aha!” for me: I consider myself an extremely talented consultant. I have helped – literally – hundreds of thousands of teachers dramatically improve their practice, and even gotten a few principals to straighten up and fly right. But I have never turned a school around or contributed to a successful restart. Why? Because while I can help PEOPLE change their beliefs, I can’t do anything with a BUILDING. And, unfortunately, I work in a world where reform is building-wide and testing-bound.

I can get an entire elementary school full of teachers to buy into and apply a coherent and responsible theory of reading instruction. But I can’t get them to see that the success of this approach can be applied to any other subject. Why? Because currently we conceive of all change in education as school-wide and subject bound – instead of regarding it at as people-centered and belief-driven.

I would argue, then, that talking about turnarounds and restarts isn’t as important as talking about the people we’ll be restarting with or helping to turn around. And since we don’t have several million ambitious educational geniuses willing to spend 20-30 years working in schools, the most promising approach to reform that I see is helping people (in both turnarounds and restarts) derive practice from a shared philosophy and work together using that set of beliefs to drive their individual and collective decision-making.

The future of reform is the “philosophically constructed” public school. Shared beliefs form the foundation of long term success. They also give us what has been missing in so many failed restart and turnaround efforts to date, something Jim Collins identified in “Good to Great” as one of the key elements of enduring successful organizations. Collins notes in his research that the greatest organizations – the ones that perform the best and last the longest – find a way to “preserve the core and stimulate progress.” Without the core – or as I refer to it here, without a set of core beliefs – steady progress in any school is unlikely at best, and impossible at worst.

Though I like and respect Justin Cohen, his post below takes on a straw man, not me. I’ve never offered closures as a simple, easily implemented cure-all, and I’ve never said that all low-performing schools should be closed.

My argument is that if a school is found to be failing kids, we need to try hard to fix it. If repeated efforts don’t work and it’s clear that we have a persistent failure, then you close it. As the article put it, “When conscientiously applied strategies fail to drastically improve America’s lowest-performing schools, we need to close them.”

I’ve also argued that closures require thoughtful choreography and are just one part of a broader reform strategy. That is, no prudent system would shutter 30 schools at once and then let families fend for themselves. The system must carefully consider how you close a school (phasing it out or all at once), how to ensure kids have higher-performing schools to attend, how new starts will be integrated, etc. There are many difficult issues wrapped up in these decisions, such as geography (opening 10 new schools in lower Manhattan won’t really help the kids displaced by 10 closures in Harlem) and grade spans (shuttered elementary schools can’t be replaced by new middle schools).

Cohen makes the common argument that too many turnaround efforts have been soft interventions. But, as my article pointed out, we’ve tried a staggering array of strategies, including tough ones, and those too fail with remarkable consistency. The private sector tries tough interventions all the time, and those fail in the vast majority of instances. Steve Peha’s lesson about “turn backs” is instructive.

I wish I could share Nelson Smith’s optimism that the best charter operators will reject entreaties to get into the turnaround business sans a long list of guarantees. Unfortunately, it appears that lots of very good CMOs are getting pressured into this line of work. I worry about this greatly.

While I respect Governor Hunt’s work in education, the towering body of research and on-the-ground experience is stacked against his optimistic claims. More importantly, I need to take issue with his final statement that turnarounds “(have) to be done.” I believe that line of thinking, though rooted in the best of intentions, has led us down the wrong path for 40 years. In my opinion, what has to be done is our looking out for the best interests of kids. That’s quite different from (and is actually mutually exclusive from) endless efforts to fix all persistently failing schools.

In light of what has been said here in the last few days regarding the challenges of turnarounds, I wonder what everyone thinks of this:

"Ford Foundation plans to invest $100 million to school turnaround plans in seven cities, including Detroit."

http://bit.ly/4xXaZP

Having made three trips to the Motor City this summer to assist in the training of current and future entrepreneurs, I was struck by the "can-do" spirit that still exists in a town that has been hit harder than I could have imagined.

$100 million sounds like a lot of money, and if Detroit gets it's 1/7th share, that $14 million or so sounds at first like a good chunk of change. I want to believe -- especially in the case of Detroit -- that stunning successes will emerge, but so much of our discussion this week has pointed to the near-hopelessness of the turnaround proposition.

Is there a creative optimist in the house who can spin up an approach that makes turnarounds more likely to succeed? And since restarts are potentially so chaotic, and so challenging where the human capital issue is concerned, have we boxed ourselves in by favoring restarts so heavily over turnarounds when turnarounds will likely far outnumber restarts no matter what the probabilities are for success? Should ed think-tanks start thinking harder about ways to optimize turnarounds, especially in the inner city? With a huge base of well-educated but unemployed or under-employed people in Detroit, could a case be made for changing the Ford Foundation's approach to a restart model using some kind of alternative certification and superfast high quality training?

I have some thoughts on this but I'd love it if more senior voices chimed in first as I will have precious little data to support my ideas.

And then there's this question to consider: "if a turnaround costs $X but only succeeds Y% percent of the time, and a restart costs $W and succeeds Z% percent of the time, what's the actual cost-benefit of picking one strategy over another for a philanthropic organization like the Ford Foundation? And how does the size of school affect choice of strategy?"

Do hard numbers exist as yet for W, X, Y, and Z? Can the true scope of reform be quantified in any useful way? If we considered reform from a venture capitalists viewpoint, what would be the most responsible approach to hedging risk for a mixed portfolio of restarts and turnarounds?

PS Thanks, Eliza, for conferencing in Governor Hunt. As a fellow North Carolinian, I always enjoy getting his perspective, especially about where our state is headed.

James B. Hunt Jr., former Governor of North Carolina and founder of the Hunt Institute for Educational Leadership and Policy, on turnarounds. I interviewed him this week and this is what he had to say about turnarounds:

Turnarounds can work. The first thing you have to do is get new leadership. That certainly will mean a new principal and it probably means a lot of new teachers. But it also means upgrading and changing the ones that you have. In North Carolina we're trying to do it in a way that I think is going to be pretty successful. We've put all of our teachers through very rigorous training courses through the summertime, they have to do it, the principal has to do it. School board members have to be involved in it. And then of course we got to have better teachers in there, there's just no substitute. The thing that makes the most difference in a child's success in school is how good the teacher is. And the teachers won’t be good unless the principal supports them. And of course the family needs to be more involved, and they need to do a better job. But we can only do so much about the families. We ought to do what we can about the families. We can turn around schools. In my case in North Carolina, we sent that the state school chairman to turn around one high school in one school district. We've got a 100, probably 10 or 12 of those need to be turned around. So we are taking on one in one year. Now if we learn how to do it, and do it successfully, maybe next year we'll do two or three. But this is going to take a while to do, this turn around stuff. It's tough work but it has to be done.

Another reason why RTTT, in its present form, is more federal money “down the drain:” it will go into states with school finance systems that seriously under-fund their highest poverty, lowest wealth districts. And to states with chronically “underperforming” education departments, that is, state education agencies lacking the organizational structure to manage reform and the capacity to effectively deliver the expertise, technical assistance, rigorous evaluations and other supports necessary to improve high needs districts and schools.

We need a state, not just a school “turnaround strategy” to finally get the equitable funding and institutional changes required to ensure states meet their legal and moral obligation to provide a high quality education in all schools for all their children. And we don’t need a competition. Just pick a state, any state.

While we’re talking here about “turnarounds”, it occurred to me that there’s another phenomenon in the school improvement lifecycle that might be worth discussing. I call it the “turnback”.

When I look at the schools where I worked during the first five years of my practice, the graphs of their performance over time have a vaguely trapezoidal shape. That is to say: consultant arrives, school makes gains, consultant leaves, school flattens out, time passes, school declines – though not all the way back to where we started; that would just be too depressing for words.

I call this a “turnback”. And I think it’s a concept we need to add to the “turnaround” lexicon, and one we should be prepared to deal with especially if Ed Reform 2.0 ever launches. (I look at the reauth of NCLB as Ed Reform 1.1; it’s just gonna involve tinkering at the margins; not that we shouldn’t do it, of course, but, well, I guess we don’t really have a choice because we haven’t been very creative about the variety of approaches we've taken to reform.)

If there was a super hero known for his ability to retain his shape regardless of any force applied to his physique, his name would be “School”. One reason education doesn’t move forward is because the ability to morph back into its original state is written into the DNA of the school (and I literally mean a building, not an institution); rewriting that DNA is the solution, of course, but interestingly enough this is more than just a human capital problem. Something seems to happen to almost every group of well-intentioned adults when they spend too much time together in a school. To me it feels like entropy and then, later, atrophy.

As difficult as we all acknowledge turnarounds to be, I think it’s important that we also acknowledge that turnbacks are likely, especially over periods of 5, 10, or 15 years. Just as we have magically resegregated our schools circa 1968, most schools we turn around are going to try to turn back.

In my practice, I have noted the following reasons why the schools I work with turn back. And strangely enough, every single one of these trumps even my return. So even if I do get hired back at some point, we’re starting over from scratch, and I feel like I’m caught in one of those temporal causality loops that I see on re-runs of Star Trek: The Next Generation.

Why do schools turn back?

1. The best teachers in a successful school are promoted by the district, (to become literacy coaches, TOSAs, curriculum directors, etc.); the best teachers are "encouraged" to transfer to other schools so one school isn't better than another (an ironic attachment to the misplaced ideal of fairness that permeates school culture); the best teachers leave the district altogether to explore different career paths (consulting, principalship, curriculum director, MA, Ph.D., EMO, NPO, parenthood, tai chi, gardening, etc).

2. The principal leaves (often by transfer because talented principals are often shuffled with less talented principals to put out fires in less successful schools). Not to mention retirements and the same career explorations as sighted in #1 above.

3. No succession planning. Few exiting principals I’ve known have known in advance who their replacement would be. Thus, new principal = new school. And since principals are often quite skittish their first couple of years in a new building, schools that might still be on the path of improvement tend to lose the “Big Mo”.

4. The normal course of teacher attrition and transfers – another form of poor succession planning, I suppose.

5. Loss of institutional knowledge. I can get around this to some extent by using some very "compact" custom materials I've made for clients. But it still happens. I think of it as "leakage" more than total loss; but eventually all the air leaks out anyway, so it really is a total loss at some point. (I just don't like to face it all at once.) I’ve revisited some of my early schools only to find them carping about the problems we solved 5-7 years previous. Mid-carp I often mention that everyone still has the same shelf full of tools they used way back when, but somehow folks can’t remember that they used to use them and that they worked pretty well. So, loss of institutional knowledge might simply be a form of selective amnesia – a wonderful strategy for the change-resistant educator.

6. (Most fascinating and frustrating of all, in my opinion) “The 7-year Itch." Seven years is roughly how often a district will do a new curriculum adoption in a given subject. Weird thing is, districts will blithely swap something old for something new, even if they have seven years of solid data saying that the old thing works! It's just a relic of the pre-accountability age, I guess, when schools were in the business of selling hope and hype, (or nothing at all) and each new curriculum adoption got everyone excited because at least the boxes were bigger and the colors were brighter and nobody worried if anything was getting better or not.

All my clients LOVE going through adoption cycles, even after I've trained them in a literacy method that doesn't require that they buy anything to continue using it! My company’s biggest client of all time spent almost $2 million one year, while we were still working in the district, purchasing materials that were almost all incompatible with the model they were paying us to help them implement. We use our own version of the "workshop" method for literacy and provide all the materials teachers need – other than paper, pencil, and books – for free. But even our "Yes, this really IS a free lunch!" model hasn't discouraged anyone from spending big bucks on adoptions that don’t support their own curricular and instructional goals. In the case I mention here, purchasing new materials actually discouraged teachers from using the method the district wanted them to learn, and lured them back to the textbook-based approach the district hired us to eradicate. (Oh, the irony!)

Yet another reason things that were good go bad is the odd discomfort many teachers seem to have about using the same method year after year. (Of course, the opposite is common as well: some teachers who would gladly teach the same thing 30 years in a row, but I’m assuming that since a school has been through an initial change process, those people and that tendency may have been weeded out.)

I've tried to combat this “gotta try something new for no good reason” attitude in my clients by sending them improved materials from time to time, but it doesn't seem to help them improve no matter how much better the materials get. The basic "workshop" method is our foundation, so that stays the same, and it seems very hard for many teachers to "grow within a method" as their careers develop. Normally, teachers grow, if they grow at all, by changing methods. One mark of a great teacher, however, is someone who grows "in-method" based typically on a strong and early commitment to an effective philosophy of teaching and learning. But this is rare in my experience; maybe 2% of all teachers.

Mind you, restarts may be just as susceptible to turnback as turnarounds. We don’t know yet because there hasn’t been enough time. At what point does the “selfish gene” with its new DNA evolve a new organism that is completely superior to all previous iterations such that there is literally nothing to turn back into? The timeline probably has something to do with teacher and principal career cycles. One of the questions I have about many of the “heroic” models of school-wide change in use today (KIPP would be a good example) is how long they can keep people pumped up to put in the 10-hour teaching days and take the dinner-interrupting cell phone calls. Might “heroic” leadership in education accelerate burnout? Might "solving" the human capital problem produce a generation of talented educators so ambitious and intelligent that attrition actually increases as a result? I’m not saying that KIPP isn’t doing some of the best work in the country; they are. I’m just curious about how schools like these will retain their shape or, better yet, evolve into organisms that are more successful by virtue of becoming more efficient, more distributive, less heroic, and in the end, more sustainable.

Proud as I am that both ends of this discussion are anchored by Alliance alumni (Smarick the Critic and Cohen the Turnaround Czar), it's getting needlessly muddled."Turnaround" for starters, is a catchall phrase that focuses on buildings rather than kids. Andy's right that it has a dismal record, not just in education but in other sectors, and that disruptive innovators are a better bet. I don't care much whether we hand the building over to the innovators or let them do new starts in the adjoining neighborhoods - -what matters is that the kids get a new deal.

I want to make two points specific to charter involvement in this debate:

First, you'll find very few charter operators who would accept the deal Checker Finn describes, just being handed a lousy school and told to fix it under a new 'charter" label. The only way their involvement in "turnarounds" makes sense is if they can actually create a real charter school with real autonomy over budget, personnel, curriculum, and time. "The movement" has made this clear to Arne and co., and I actually think he gets it.

Second, don't assume (as Diane Ravitch does) that any random charter school would be chosen to occupy or replace a failed district school. The strategy is to replicate the best, the proven models -- and even they will need to demonstrate the extra capacity needed in these tough circumstances.

Justin C. Cohen, President of The School Turnaround Strategy Group at Mass Insight Education & Research Institute submitted the following in response to this week's question:

“Because the problem of fixing failing schools is so vast and complex, the only “fundamental flaw” in a strategy to fix failing schools would be to search for and implement a magic bullet cure, which is what Andy Smarick suggests when he proposes that we close all low performing schools. I am huge supporter of starting new schools and chartering, but we need an inside strategy that deals aggressively with failure where it exists. Districts and states pursuing turnaround should adopt a portfolio of strategies.

There needs to be more consensus on definitions and terminology. As Checker Finn suggests below, when turning around a failing school, the only thing that we should hold constant are the students and the neighborhood (i.e. location). Everything else – from the adults in the building to the instructional program – should be subject to change. Moving teachers and administrators out of a building should fit within any reasonable definition of turnaround. Unfortunately, the word “turnaround” traditionally has been used to define not only this type of dramatic change, but also to define half-measures. Sending a coach to a school once or twice a week is not turnaround. Changing the reading program and leaving everything else the same is not turnaround. When private equity firms pursue turnaround in the private sector, they don’t assume everyone in the company is fit to get the job done, and neither should we when pursuing aggressive school turnaround strategies.

Also, while I partially agree with Richard Rothstein and Ted Herschberg about the need to have better identification metrics, the problem is more pronounced on the back end than on the front end. I think we have a pretty darned good idea which schools are our failing schools, especially at the high school level. And at the elementary school level, it’s even more important to rely on good standardized and formative assessments, because the age-appropriate misbehaviors are less “in your face” than those of high school students. The bigger problem is defining when a school has “turned around.” Is it an average of 25+ point gains on high stakes assessments after just a few years? Is it some predetermined proficiency bar? We need some shared understanding of this, because it would be a huge mistake to declare victory once we make hard changes on the inputs; results for children are the only thing that really matters.

Finally, as others have suggested, turning around failing schools is a zero-sum exercise unless we are simultaneously making progress on other critical factors that drive reform, namely: human capital, appropriate use of data, and systems reform. At The School Turnaround Strategy Group, we view failing schools as the entry point for broader reform. If we’re not simultaneously thinking of how to consolidate change and scale the strategies we learn from working in turnaround, school change will not be sustainable.”

These responses are somewhat surprising and quite encouraging. There’s more agreement than I expected about the lack of promise of turnarounds. I just wish more folks would be vocal about this; the Race to the Top is $4.35 billion and the School Improvement Grant program currently has $3 billion. Lots of money, time, and human resources are about to be invested in a venture likely to bear very little fruit. Add to that the opportunity costs of not pursuing the closure-new start option and you have a major problem.

Some respondents below lamented the lack of research on successful turnarounds and suggest that we need to keep trying and keep studying. My argument is that turnarounds, as a field, seldom work. Regardless of the industry, the evidence shows that the vast majority of persistently broken institutions are, for whatever reason, unable to become consistent high performers.

Another point raised several times is that we can’t be sure that new schools will be excellent. I concede that readily. But that’s not the end of the conversation by any means. First, that shouldn’t surprise us. No industry guarantees the success of every new entity. But that doesn’t stop those industries from constantly welcoming new starts.

Second, if you look at America’s best urban schools—with the levels of achievement we should be aspiring to—they are almost always new starts or long-excellent schools. They are not turnarounds.

Finally, closures and new starts are an indispensible part of every high-performing industry out there.

Two quick final points: Steve Peha raises the issue of the definition of turnarounds. He’s right that this is a massive area of confusion. During my research, I found six different definitions.

And, as Chad Wick notes closures in rural areas would be considerably more difficult.

The consensus among education experts is that RTTT is not backed by research. It is based, it would seem, on the same test-and-punish ideology that produced NCLB. If anything, RTTT is NLCB on steroids. It extends to teachers the punitive attacks based on limited and flawed standardized tests that have been waged on schools, and it intensifies NCLB's inadequate requirements for "transforming" schools, which are either unproven or aleady proven to fail. Einstein’s definition of insanity was "doing the same thing over and over again and expecting different results" The new definition should be doing the same thing ever more intensely.

The real question is whether, in response to the detailed, research-based criticisms that have been levied at RTTT, Secretary Duncan and his staff will fundamentally overhaul their proposed requirements, including those for "turnarounds." If the Department does, it will show that research, evidence and reason matter. If not…

Richard Rothstein explained clearly why test scores are not a good means of identifying schools for anything we might label "turnaround." Several other bloggers have pointed out, in different ways, why the proposed interventions are not well-grounded. Among them, they explain why RTTT is very unlikely to improve learning for poor kids in most under-funded schools in low-income communities. I would note further that the National Research Council's Board on Testing and Assessment said that high-stakes decisions should not be made on the basis of a set of test scores (which is how schools are to be identified) and that "value-added" models are not ready for high-stakes uses (contrary to Ted Herschberg's call for their use).

The nation faces a situation in which hopelessly inadequate tools (large-scale tests) are used to identify schools for drastic action. The track records of the proposed actions reveal they are unlikely to succeed: charters seem to do slightly less well than regular public schools, forcing schools to become charters usually does not work, replacing the staff rarely succeeds unless the students also are replaced, etc. Recent research from Chicago shows the various options under Arne Duncan’s "Renaissance 2010" scheme have failed, yet Chicago is the core model for RTTT.

Let's learn from the NCLB educational disaster and the failures of assorted "turnaround" and privatization efforts. That means, first of all, not intensifying these failures, which is what RTTT will do unless the Department thoroughly overhauls the guidelines in response to the excellent advice provided in many of these comments and submitted in responses to the RTTT draft guidelines.

Andy Smarick is right, unfortunately. School "turnarounds" rarely succeed, and they're least apt to succeed when, as the Education Department (and NCLB) seem to expect, they're undertaken by the very school systems that allowed these same schools to fester to the point that they need a radical makeover. Starting from scratch is a lot more promising. (That can include closing a bad school and starting from scratch within the same building--but with everything else different, especially the instructional team.) But of course that's what school systems are least likely to do. Even more troubling is the ED assumption that turning a bad school into a charter school will cure what ails it. What's far more likely to result is another bad charter school, which is not good for kids, not good for the "turnaround" strategy and not good for school choice or the charter movement. Unfortunately, that movement, because it doesn't want to give any grief to its friends Messrs. Duncan and Obama, is biting its tongue on this particular topic. Truth to tell, the proposition that a bad school can be "turned around" by chartering it is among the worst policy ideas I've encounted in many years.

Any fund used for the improvement of our schools is money well spent. My hope is that the RTTP funds have a significant impact on persistently low performing schools and succeed beyond our wildest of dreams. Will we then have the courage and the political will to admit that public education needs more money to get the job done nationwide and replicate these successful models IN ALL OF OUR SCHOOLS?

Chad P. Wick, President & CEO, KnowledgeWorks, submitted the following in response to this week's question:

There is little doubt that taking on large, chronically low-performing high schools is difficult, messy work. But we must do it.

In many already devastated communities, closure is not a viable option. We must remember there are contextual differences to the communities where schools are located. Closing a school in a large district and replacing or reconstituting it may be possible. However, how do you close a low- performing school in a rural area or a small town where only one school serves the entire community?

For the past several years, KnowledgeWorks has been leading high school transformation in some of Ohio’s lowest-performing communities, launching 79 high schools, and many of them have made dramatic improvements in graduation rates, standardized test scores, and narrowing the racial achievement gap.

We can positively say that we have been most successful in turning around those schools where the adults recognize that they must radically change the way the school is structured, the way they teach and lead, how they support classroom teachers, and their expectations for all students. They determined that failure is not an option, nor is dropping out a viable alternative.

If turnaround strategies are going to be successful, these schools need new motivation, new structures, and a new commitment to excellence.

I agree the Race to the Top Fund is a likely blueprint for the Administration’s plans for reauthorization of the Elementary and Secondary Education Act, which is one reason I’ll be watching closely when the final regulations are released this month. I’m struck by the notion that states – including my own – are relying on grants from nonprofit foundations to navigate the complex RttT grant application process. I’m reserving judgment until the final regulations are released, but the complexity and prescriptive detail of the draft regulations certainly gave me pause.

Because of RttT’s implications for ESEA reauthorization, it seems to me this question doesn’t quite get at the right issue. The ‘turnaround strategy’ for chronically underperforming schools is certainly a major focus for the Administration, but RttT appears far more expansive in scope. Looking at the preliminary guidelines, RttT could have consequences for every single school in a state receiving a grant – not just those schools that are persistently underperforming.

For example, the preliminary guidelines create new categories for measuring teacher and principal effectiveness – on top of the “highly qualified teacher” requirements that have often proven troublesome under the No Child Left Behind Act. The draft regulations also require states (and their schools) to dramatically improve reading and math scores measured against NAEP and increase graduation rates at a time when schools and school districts are currently struggling to meet the accountability requirements in the law.

I have been vocal in my support for some of the concepts being championed by the Administration, including efforts to improve teacher quality through performance pay and expand access to charter schools. Nonetheless, to me, the real question we ought to be asking about RttT is whether being too prescriptive will stifle the innovation we’re trying to foster.

Returning to the original question, I agree that turning around chronically underperforming schools has been one of the truly intractable challenges facing our education system for years. I’m not convinced the federal government has all the answers for transforming these schools, but I don’t question Secretary Duncan’s intentions for attempting to focus on the schools with the greatest challenges. I would simply suggest that the focus of our policy should be to help spur local innovation, which in turn can help improve education for these and all schools. I do not believe a heavy-handed federal program – whether through RttT or a broader reauthorization – will get us there.

Richard Rothstein is correct. For starters the feds should change the means by which “persistently low-performing” schools are identified. Growth measures should be used to distinguish “high growth” from “low growth” schools among those not meeting proficiency targets under current law. Those in the latter group deserve the label.

Though evidence is limited, some charter networks, such as Mastery Charter Schools in Philadelphia, demonstrate that these schools can in fact be turned around. There is no secret to their success: students learn when they have high-quality instruction. Turn-around plans should be judged by how well they address this central issue.

If we are to succeed in improving persistently low-performing schools, we’ll have to align the interests of their in­di­vi­dual educators with the goal of sig­nif­i­cant increases in student achieve­ment. This means putting in place a different human capital development system with new ways to evaluate, compensate, remediate and provide pro­fes­sional de­vel­op­ment for teach­ers and ad­min­i­stra­tors alike.

“Race to the Top” guidelines were designed to generate a broad range of experiments based on these changes which can then be measured for effec­tiveness.

Increased compensation for top performing teach­ers will address both the long-term, by attracting more of our best and brightest college graduates, and the short-term by creating incentives to bring talented teachers in sig­nif­i­cant numbers into low performing schools. But pay-for-per­form­ance programs will only succeed if there are negative as well as positive consequences associated with instruction. Systems must be put in place that provide struggling educators with the time and resources to improve, but that also can lead to fair and timely dismissal should these efforts fail.

Eliza - I think there is no "yes or no" answer to your questions.

So, let me answer your question with a question. Will Race to the Top funding be grounded in strong and enforceable expectations that states have and implement turnaround strategies based on the wisdom shared here by Rick, Sherman, and Diane, among others?

If so, it could be effective. If, on the other hand, it's more like "revenue sharing" for high hopes and promises in states that have merely used the right words in their proposals, it won't.

Everyone seems to be making strong and varied cases for why Secretary Duncan’s Race to the Top approach is likely to fail when it comes to helping states turn around low-performing schools. It would be wonderful if the Secretary would join us in our discussion this week and give us his side of the story. Specifically, since most of seem to have a negative view of RttT’s turnaround potential, I would love to know why he believes otherwise.

Much as I enjoy playing the affable contrarian here, I’m afraid I must agree with just about everyone so far. What strikes me, though, is not the incidence of agreement but the fact that each of us has so many reasonable reasons to support the notion that Race to the Top will have little impact in turning around low-performing schools.

In addition to my endorsements of the ideas offered by Mr. Hess, Mr. Dorn, Mr. Rothstein, and probably many others yet to post, I have a few reasons of my own to add to the critical onslaught. (Mind you, I have applied to be a reviewer for RttT applications, so I’m hoping my thoughts here won’t hurt my chances of being chosen as I am very much looking forward to the experience.)

First of all, does the Secretary have a practical definition of what a turnaround is? Many of my lowest performing clients are thrilled if I can help them raise their scores to state averages – and quite happy to let them languish there indefinitely. But when states are in the business of consistently making tests easier to pass, staying where you are could actually mean you’re losing ground. Without at least some vague definition of what a turnaround would be, we could end up with thousands of them or none at all using the exact same data to justify whichever side of the argument we prefer to be on. (Note that this exactly how NCLB has turned out and exactly how we argue about it – to no particularly constructive end, I might add. For all the pomp and circumstance we make in this country about creating world-class schools, we have no idea what world-class schools are and no reforms targeted at creating them.)

Second, in order to count as a “turnaround” how long do change effects have to stick? If you look at some of my clients over ten years, they tend to begin improving when we begin working together, peak two or three years after I leave, and then drop slightly. As I revisit many of oldest clients, I see more “turn backs” than I do “turn arounds”. This is usually due to the fact that most of the people who were in the building for the turnaround are gone. It is also due in part to the “Groundhog Day” phenomenon which I discuss below.

Next, are we focusing turnaround efforts at the true fulcrum point? That is to say, if we’re thinking of turning around schools, are schools the place to start? I would say no, and that the district office is the place to start. Until certain “rules of engagement” have been agreed to (and probably ratified by the school board, too), any turnaround action at the school level can easily be undone by almost any district administrator at any time: shuffle a dozen teachers, transfer an effective principal, adopt curriculum that contradicts the turnaround effort and enforce district rules to prevent the turnaround school from what it may have just recently succeeded with. I’ve seen all of these happen and they are all disastrous for continuous improvement and long-term change. It just doesn’t take very much to tip a school off its tipping point.

I pose this perhaps unusual notion based on my own experience. Of the 200 or so client schools my company has worked with, most were relatively low-performing, most made immediate gains to one degree or another, and then, in many cases, something happened to short-circuit further growth. Tops on the list has been negative intervention from the district office.

There’s a huge cultural problem in most districts in the United States that never gets talked about: few districts want any particular school to be doing much better than any other school (although it's just fine for a few schools, you know, the ones with "those" kids, to be doing worse). The problem here is one that many administrators I have spoken with refer to as “camps”. “We can’t have any camps,” districts folks have often said to me. This is a euphemism for groups of people who don’t agree on something, a quintessential element of the healthy democracy our founders envisioned but not something we can tolerate in schools.

For example, when I helped an elementary school turn low and mediocre scores into district-leading scores by way of a particular method of literacy instruction, this method was suddenly outlawed within the district and replaced with a method that had a better chance of achieving poorer but more uniform results. In a similar situation, my company helped one urban high school move from somewhere near the bottom of their district to somewhere near the top. This wasn’t looked upon at the district office as a positive contribution either, again because the instructional methods that contributed to the change were not shared by many other buildings, and were, in fact, strenuously opposed by many traditional teachers and principals simply because they were afraid they’d be made to use it, too -- even though methodology was a building-level choice and no teacher had to use any particular methodology even if his or her building chose it.

This is a variation on the Jonah Complex, or the fear of one’s own greatness. It lurks in the psyche of many educators and creates a dysfunctional district culture where no one is allowed to succeed unless everyone else succeeds in exactly the same way at exactly the same time -- so everyone can be equally great which, ironically, means no one is great at all. Pair this with the preternatural aversion to conflict that is almost part of the job description for school administrators and you have a perfect storm of conformity.

Now, about this "Groundhog Day" thing. I had a chance recently to swap school stories with a prominent national education researcher whose work focuses on classroom practice and school-wide change. This gentleman works in dozens of schools each year and lately describes his visits as “Groundhog Day”, a reference to the overly sentimental but oddly moving Bill Murray film about a TV weatherman who finds himself living the same day over and over again. The analogy to turnarounds relates as follows: It is not terribly difficult to get most schools to clean themselves up a bit, improve school climate, increase order, and even make modest test score gains. But once schools reach this point of initial improvement, they flatline. Why? Nobody wants to try anything new because they’re afraid they’ll lose the small gains they’ve just achieved. Much like the publicly traded company that forgoes long-range planning and R&D investment in order to meet analysts’ quarterly targets, these schools feel they cannot risk another multi-year initiative to make the significant gains that might qualify them as a legitimate turnaround. No risk, no reward. It’s as simple as that.

In this context, NCLB has had a chilling effect on the tolerance schools have for innovation. For example, my company embodies this almost perfectly as our existence spans a 7-year period before NCLB and the current 7-year period during NCLB. In our pre-NCLB days, we got paid much less money, and had many fewer days in schools to make improvements. However, we produced better results per school than in the current period. Now, during the NCLB era, we have made an order of magnitude more money, had many more days to help many more schools, and yet have had only a modest impact in most situations. The culprit, of course, is low implementation rates of contemporary research-based practice. During the Age of Accountability, few schools want to risk using even well-researched ideas if they differ in any significant way from what they are doing already. Instead, they prefer to hunker down with mindless test preperation and the teaching with which they are most comfortable – that being the traditional teacher-centered paradigm of transmission model delivery combined with textbook-driven programmed instruction. Everybody does the same thing, the same way, at the same time, on the same day. Consistency equals order, order equals predictability, and we can all then predict that scores will stay right where they are.

Finally, with specific reference to RttT, the significant nature of the cash infusion, and the broad spectrum of its possible uses, lead me to believe the following:

1. Big Money Causes Big Problems. Much of the money will be misspent, not spent, or unlawfully spent. I don’t mean to imply that states are scheming to defraud the government, but large cash infusions into systems that have not had much success with applying money toward large-scale reform is a recipe for fiscal chaos. The same thing happens in young public companies. I’ve worked with a few post-IPO firms and I can tell you that the first six months are absolute bedlam. The money that flies in seems to fly out just as quickly, and few people in the organization can explain where it went or even why it was spent in the first place. (Remind me to tell you the story about the company that purchased 90,000 square feet of office space for about 40 employees. Each of us could have built a house for our office.)

2. Funding Needs to Fit the Facts on the Ground. States are going to get one or two cash infusions within one or two years. But the problems they are trying to solve will take many more years than that. This means that serious cash flow management will be required. States proposing long term initiatives – and all of them will because the problems we want to solve are all long term – will have to hold in reserve enough money to see their projects through to completion. This is very hard for states to do, especially in education, and especially in a down economy. My hunch is that some RttT funds will get redirected much as some ARRA funds have. It would have been better to parcel out the cash more gradually as states hit reasonable milestones over the course of completing their projects. Hit mileston #1, fund milestone #2, and so on.

3. The “Mice and Men” Problem. As Steinbeck reminds us, “the best laid plans of mice men often go awry.” To win big money, states are going to propose big initiatives, many of which may not have reached even the proof-of-concept stage. Forget the fact that no sane venture capitalist would ever invest such nascent schemes, how do the states know that these are the right things to do and that they can actually do them? Since virtually all large-scale efforts in reform to date have failed or produced no clear results, it seems to me that many of these ambitious proposals will face the same fate.

We all remember studying realpolitik from our history and poli-sci classes, perhaps it’s time to promote realpolicy. RttT is truly quite exciting. And I’m sure Secretary Duncan feels particularly honored to be the steward of so much money. But the plan simply doesn’t match the problem. If Mr. Duncan’s top priority is turnarounds of failed schools, the money needs to be targeted at solving human capital problems, along with teaching and administrative quality issues. If his priority is quantification and measurement, the money needs to focus on data systems. If his priority is something like scaling high-quality instructional techniques to hundreds of thousands of teachers, the money needs to go to identifying those techniques, documenting them, distributing them via technology, and monitoring their use. But RttT isn’t set up that way. Instead of encouraging states to pick something smart from a small menu, it invites them to belly up to the buffet.

Realpolitik is based on power and pragmatism. Realpolicy could be, too. What is the true fulcrum point within the system? (It’s in the principal’s office, by the way.) Where is the leverage? (Look to Asst. Superintendents on this one.) Who pulls the strings in large urban districts – or works like mad to keep those strings from being pulled in order to maintain the status quo? (Often it’s whoever is in charge of Title I money plus a rag-tag band of veteran naysayers, along with certain department chairs at the high school level.) If we’re going to fix big problems in education, and that’s clearly what RttT was created to support, we need precision strikes not carpet bombing. How likely is it that Secretary Duncan will get to play with another 4.35 billion dollars during his tenure? As my grandpa always said when he gave me a buck for my birthday: “Don’t spend it all in one place!”

In the course of cutting and pasting a standard list of arguments against standardized testing, Richard Rothstein seems to have lost track of the topic at hand. If it is, in fact, true that test scores are "increasingly inflated," then it seems fair to assume that schools reporting rock-bottom test scores despite that inflation are likely to be low-performing. There are public schools, open today, that have been identified as low-performing by three or four successive accountability regimes dating back to the mid-1990s, schools where the majority of children fail to graduate or advance, schools with depleted enrollment, crumbling facilities, and demoralized faculty. Rothstein's notion that we don't actually know which schools need to be turned around flies in the face of all evidence, test-based and otherwise.

Trading good seats for bad seats is the most effective strategy--closing bad schools and opening good school in roughly the same proportion. As Hoxby has pointed out, this is working well in Harlem. However, as pointed out yesterday, RttT and School Improvement Grants require school specific interventions. Green Dot’s takeover at Locke High School in LA is a promising example of closing and reopening with much the same group of students. Close and restart has been happening successfully in New York since Julia Richmond High School was replaced by four new schools in 1993. In this decade, Evander Childs, South Bronx, and Morris High Schools were closed and replaced with new district schools. Graduation rates have improved significantly and academic achievement is improving. New York:s close/replace strategy has been aided by a full system of choice and public transportation.

Improving existing high schools may be less disruptive for families but it is much more difficult and less successful in cases of chronic failure. A couple providers have tried (e.g., IRRE, Talent Development, America’s Choice) with modest success. MLA Partner Schools took over Manuel Arts and is building a Promise Neighborhood in west LA—it will get better but it will take time. Unless there is the opportunity to change fundamental conditions (structure, staffing, schedule, and leadership) there’s little hope of more than incremental improvement. The Department attempts to require changing conditions for 'turnaround' schools. State leadership will be key to making new conditions possible and making turnarounds successful.

District 'transformation' efforts (the light touch model sanctioned by the Department) are likely to be most successful in RttT tops that have adopted high standards and are improving human capital and data systems. But even with an improving reform context, every state will consider strategies for building restart, turnaround, and transformation capacity.

One answer to the capacity question, especially in rural settings, is blended restart charter management organizations that combine the best of online and onsite learning. Most of these blended schools will use an online curriculum as core and will supplement with projects, community connections, tutoring, guidance and extra-curricular activities.. Combine proven instructional delivery with high quality school operators and you get a scalable restart capacity.

Despite a strong federal school accountability framework, most states have allowed chronic failure to fester for the last 8 years. RttT and SIG provide a strong framework and big budget to support aggressive action to improve at least the worst 5% of schools in every state. States with strong education leadership will take full advantage of this opportunity and will produce dramatic improvement. The other half (that don't get RttT funds) will show incremental progress with their SIG grants. The net result will be thousands of new and improved schools and a lot more knowledge about and capacity for turnarounds.

Race to the Top is placing its bets (and more than $4 billion) on a risky gamble: that schools will get better if they are "turned around" or closed. Unfortunately, there is no sure strategy for turning around a low-performing school other than tossing out all the kids and replacing them with higher-performing ones. Earlier this year, the U.S. Department of Education released a booklet offering advice on turning around low-performing schools and admitted that its recommendations had "low" evidence.

There is also no guarantee that a new school opened in place of a "failing" school will be any better. Chicago has not had any success with that strategy. Charter schools may or may not be better. The latest NAEP scores for mathematics once again show that charter schools do no better than regular public schools.

In a recent Brown Center Letter on Education, "Don't Forget Curriculum," "Russ" Whitehurst at Brookings pointed out that changes in curriculum and selecting effective instructional programs were far more effective than charter schools, merit pay, or any of the other strategies embedded in the Race to the Top.

Race to the Top is not based on evidence or solid research, but on the hunches of the people in charge. It appears that Race to the Top is promoting expensive bandaids that will do little to improve American education.

Diane Ravitch

The turnaround component of Race to the Top is deeply problematic. Rather than focusing Race to the Top on stripping away barriers that impede school improvement and creative problem-solving, the turnaround strand is one that encourages grandiose and ill-conceived efforts. With intense focus, political willpower, and sufficient resources, it is probably possible for some states to effectively turn around a handful of schools. But, in throwing a big slug of federal dollars and moral support behind self-promoters promising to turn around lots of schools, and the public officials who sign on for the ride, the administration is setting itself up to undermine a reasonable idea, ensure that the “turnaround” strategy comes to be viewed as a failure, waste hundreds of millions of dollars, and sidetrack more promising efforts.

That said, given the good intentions of those championing turnarounds, it is hard not to root for them. But, while the phrase "turnaround" may be relatively new to those in education, the practice has been around for decades in other sectors. Its track record suggests a need to avoid grand promises and for steely-eyed realism. Even in the business world, where management enjoys many more degrees of freedom, turnarounds are an iffy proposition. Peter Senge, director of the Center for Organizational Learning at the MIT Sloan School of Management, has observed, "Failure to sustain significant change recurs again and again despite substantial resources committed to the change effort (many are bankrolled by top management), talented and committed people 'driving the change,' and high stakes. . . . There is little to suggest that schools, healthcare institutions, governmental, and nonprofit institutions fare any better.”

Today, much of what experts know about turnarounds comes from experience in the private sector, where two dominant approaches to organizational reform have prevailed for several decades--Total Quality Management (TQM) and Business Process Reengineering (BPR). TQM was first introduced by Japanese firms in the 1950s. It seeks to remove waste at every stage of the production process. Dishearteningly, research suggests that TQM has been largely ineffective at spurring successful corporate transformation. Two high-profile studies on the effectiveness of TQM, one conducted by Arthur D. Little and the other by McKinsey & Company, concluded that out of the hundreds of TQM programs studied, about two-thirds "grind to a halt because of their failure to produce hoped-for results."

BPR, developed in 1990 by Michael Hammer and James Champy, is more aggressive. It differs from TQM in that it concentrates on tearing down and rebuilding the business process as a whole, rather than tweaking the functional tasks that comprise it. Evidence suggests, on the whole, that BPR has fared about as poorly as TQM in spurring organizational improvement. Most articles, including some by BPR's founders, estimate a success rate of approximately 30 percent. John Kotter of Harvard Business School, in a study of one hundred BPR efforts, concluded that less than half survived their initial phases.

If the administration were saying, “Let’s provide some support to those states that have a clear, coherent, and politically feasible plan to attempt some targeted turnarounds,” that would seem a reasonable investment. But the wholesale, punch list-style mandate that states pursue turnarounds is likely to prove wasteful and self-defeating.

Andy Smarick is partly correct. While he does not cite the work of Heinrich Mintrop, both reach the same conclusion: the practice of sanctioning schools has provided little concrete benefit. The lesson that Smarick draws is that we should simply shut those schools down and replace them with new schools. The problem is that there is no research that such an approach will work any more frequently than reorganizations and turnaround initiatives. There is little evidence that Chicago's close-and-reopen strategy is a game-changer, and the evidence on charter schools is mixed over the same period of time that Smarick identifies as the era of the school turnaround.

The central problem is that most efforts to change dysfunctional schools are driven by either ideology or seat-of-the-pants initiatives, and that there is little research on what is necessary either to turn around a school or to open enough successful new schools to justify a close-and-reopen strategy. Thus far, the research has focused on whether probation, turnover, and turnaround strategies work, not what might make such efforts more successful.

Fortunately, the Race to the Top initiative is more flexible than the question implies. Yes, states could apply for funds to implement turnaround strategies and only turnarounds. But we will have to see what states propose before any of us can conclude that the program will lead to states' proposing largely failed ideas.

It is an admirable goal to "turn around" low-performing schools. But before attempting this, we need to ensure that we have accurately identified which schools are low-performing. It would be tragic if we aggressively intervened in (or even closed) schools that were, in fact, better performers, while ignoring schools that were worse.

This is the fundamental flaw in Arne Duncan's proposal. We don't, in fact, have any good ways to identify low-performing schools, so any turnaround efforts are likely to include considerable misdirection.

Indeed, as I have written in a recent Policy Brief for the Economic Policy Institute, the assumption that we presently know how to identify low-performing schools is inconsistent with principles of educational accountability that Arne Duncan has himself articulated.

At present, the only tool being used to identify low performing schools is the percentage of students who pass (i.e., are deemed "proficient" on) state standardized tests of math and reading, required by the No Child Left Behind law.

Because these tests are so flawed, they cannot accurately identify high- or low-performing schools.

Here is what we know about these tests:

a) they narrow and distort the curriculum by giving schools with low test scores incentives to abandon other aspects of a well-rounded curriculum – the arts, science, history, social studies, health and physical education, character development. Survey data confirms that such narrowing is in fact taking place.

b) they create incentives to reduce instruction to the most basic skills, because states can reduce the cost of testing by eliminating open-ended questions that more easily assess critical thinking.

c) they spur teachers and schools to focus intensive instruction on students whose past performance indicates they are almost ready to pass the test, while paying less attention to students whose past performance indicates they will easily pass. A school following this strategy (and again, there is considerable evidence, both qualitative and statistical, that many are doing so) can have rising percentages of students proficient, while its average scores are stagnating or even declining.

d) their scores are reported by subgroups that are not comparable. For example, "low-income" students are defined as those who receive free or reduced-price lunches, but this category includes students whose families are very poor, and those whose families have incomes up to 185% of the poverty line.

e) their scores are themselves increasingly inflated, as successive test years reproduce similar questions (again, to avoid the expense of designing entirely new tests from year to year) and teachers learn to focus instruction on those basic skills likely to appear on the test, while giving less emphasis to those unlikely to appear. State test scores are almost universally climbing, while scores on the National Assessment of Educational Progress are relatively flat. This is a good indication of such inflation.

Not every school responds to the perverse incentives of No Child Left Behind in the same way. A school that sacrifices some gains on test scores of math and reading in order to give children access to a broader curriculum may falsely appear to be lower-performing than schools that do not make this choice. A school that sacrifices some gains on test scores of math and reading in order to balance instruction in basic math and reading skills with opportunities for critical quantitative and literary thinking may falsely appear to be lower-performing than schools that do not make this choice. A school that sacrifices some test gains on "percent proficient" in order to ensure instruction for children at all points in the achievement distribution may falsely appear to be lower-performing than schools that do not make this choice. Schools whose students come from extremely distressed families and communities may have lower test scores, but falsely appear to be lower-performing than schools that contribute less "value-added" but whose students come from stable low-income families. And a school that sacrifices some test score gains because its teachers devote more time to instruction, and less to test preparation and drill, may falsely appear to be lower-performing than schools that do not make this choice.

If we want to turn around low-performing schools, the first task should be to ensure we are identifying these schools accurately. Such identification requires much more than test scores. It requires expert human judgment, with qualified experts visiting schools to interpret test scores and evaluate the overall quality of instruction. I have described the need for such an accountability system in Grading Education and the Broader, Bolder Approach to Education campaign has proposed such an evaluation system to replace the flawed identifications produced by the testing mandates of No Child Left Behind. Without such an evaluation system, the Race to the Top ambition will, in retrospect, turn out to have been yet another blunder in our test-obsessed school accountability policy.