Table of Contents
Title Page
Copyright Page
Table of Figures
The Authors
Starting Points
How This Book Unfolds
A Note About Research Contexts
Fundamental Causes of School Failure
A Perfect Storm with Imperfect Solutions
Stage 1: Stopping the Decline and Creating Conditions for Early Improvement
Stage 2: Ensuring Survival and Realizing Early Performance Improvements
Stage 3: Achieving Satisfactory Performance and Aspiring to Much More
Stage 1: Stopping the Decline and Creating Conditions for Early Improvement
Stage 2: Ensuring Survival and Realizing Early Performance Improvements
Stage 3: Achieving Satisfactory Performance and Aspiring to Much More
Leadership at Rowlatts
Practices for Creating a Shared Sense of Direction
By the Numbers
Practices for Building Capacity
By the Numbers
Practices for Redesigning Schools
By the Numbers
Practices for Instructional Leadership
By the Numbers
Brookside Elementary School
Central Secondary School
Summary: Across the Cases
Stepping Back
A General Orientation to Sustaining School Turnaround
A Theory O Perspective on Achieving Satisfactory Performance and Aspiring to ...
Achieving Satisfactory Performance and Aspiring to Much More: A Case Study
Eight Lessons About Achieving Satisfactory Performance and Aspiring to Much More
Rational School Conditions
Emotional School Conditions
Organizational Conditions
Family and Community Conditions
Alignment of Conditions
Appendix: Methods Used for the Studies

Table of Figures
Figure 8.1 Leadership Effects on Student Achievement
Figure 10.1 Four Sets of School Conditions to Improve in Order to Influence Student Learning
Figure A.1 School Graphs


The Authors
Kenneth Leithwood is professor of educational leadership and policy at the Ontario Institute for Studies in Education at the University of Toronto. His most recent books include Distributed Leadership According to the Evidence and Leading with Teachers’ Emotions in Mind. Leithwood is the recipient of the University of Toronto’s Impact on Public Policy award and a fellow of the Royal Society of Canada. His current research and writing is about the influence of leadership on student learning.
Alma Harris is the director of the London Centre for Leadership in Learning at the Institute of Education, London. Her research work focuses on organizational change and development. She is internationally known for her work on school improvement, particularly on improving schools in challenging circumstances. Harris first studied as a secondary school teacher and taught in a number of challenging schools in South Wales. Her most recent work has focused on leadership and organizational change.
Tiiu Strauss is a project director working with Kenneth Leithwood in the Department of Theory and Policy Studies at the Ontario Institute for Studies in Education at the University of Toronto. She has published in the areas of leader problem solving, distributed leadership, and leadership in turnaround schools.

The aim of this book is to build on evidence currently available about how to quickly and significantly improve the performance of exceptionally underperforming schools and sustain those gains. Reflecting on the evidence about successful turnaround processes in organizational sectors other than schools, we assumed that the influence of leaders is a crucial feature of this process. This seems a safe assumption. For example, one of the eight lessons concluding Murphy and Meyer’s (2008) remarkably thorough synthesis of evidence about successful turnaround processes is that “successful turnaround schools almost always have good, if not exceptional, principals. As a common strand across successful school turnarounds, leadership is crucial. The principal typically sets the turnaround agenda, while leading teachers, involving the community, and building general capacity” (p. 321).
Starting from this general assumption, our purpose was not to make the case that leadership matters—or even estimate how much it matters. Rather, our focus in this book is on the nature of that leadership. What practices or behaviors do successful turnaround leaders exercise? So the book speaks directly to school and district leaders who are facing the task of turning around seriously underperforming schools. It aims to offer practical advice to those leaders—advice firmly rooted in good evidence about what works.
A summary of our own research providing this evidence is provided in the appendix to the book. It is sufficient to explain here that several multiyear studies, both qualitative and quantitative in design, undertaken in the Canadian province of Ontario and in several different sites in England provide the core of this evidence. In addition, we have wrapped around that information from our own evidence much of the relevant research on organizational turnarounds, school turnarounds in particular, as others have reported in the wider literature. This wider literature reflects research carried out especially in the United States; virtually all of the lessons about turning around schools that we discovered in our Canadian and U.K. research are, in our estimation, entirely applicable to schools in the United States.
Although we believe that adopting a wider perspective on turnaround issues can only be beneficial, it is quite uncommon. For example, the most recent and otherwise quite impressive U.S. texts on turnaround schools and their leadership (Duke, 2010; Murphy & Meyers, 2008) reflect little or no evidence collected from outside the borders of the United States.
Sometimes our own evidence is sufficient to justify the claims we make about how leaders turn schools around. But sometimes it serves primarily to illustrate and support evidence others have generated that is relevant to claims we consider important. We draw on several lines of well-developed research that are germane to the turnaround problem as well, for example, research about low-performing schools typically serving highly diverse and economically disadvantaged students (Corallo & Mcdonald, 2002; Harris, James, Harris, & Gunraj, 2006; Muijs, Harris, Chapman, Stoll, & Russ, 2004) and both school improvement and organizational change processes (Fullan, 2006).

Starting Points

We begin with eight basic understandings, assumptions, or starting points for our subsequent account of how to lead the successful turnaround of underperforming schools.
1. Turning “failing” schools around is a prominent focus of contemporary educational policy.
This book, and the research on which it is based, was prompted by what has become a ubiquitous concern across many countries: turning around “poorly performing” schools, a central component in most educational reform agendas. It is, according to one U.S. account (Calkins, Guenther, Belfore, & Lash, 2007), “the emerging response to an entirely new dynamic in public education: the threat of closure for underperformance” (p. 36). This threat has been prompted by educational policy changes over the past fifteen years, which themselves were generated by a substantial increase in concern about the performance of public schools (Anyon, 2005). Indeed, improving the performance of schools, most often those serving students in challenging circumstances, has become the main focus of school reform efforts in most developed countries. These challenging circumstances include high levels of poverty and disadvantage among the student population, which can be exacerbated by issues associated with racial, cultural, ethnic, and religious diversity among students served by the school. But they can also be compounded by a high degree of homogeneity within the student population, where the social mix becomes a barrier to raising improvement and achievement (Thrupp, 2001).
The concept of turnaround has significant roots in the corporate world and is associated with a high degree of intolerance for prolonged failure or downturn, along with an overwhelming bias for action and better results. This bias for action has been accompanied in school contexts by a significant narrowing and simplification of the criteria officially used for judging school performance—typically student scores on tests of math and language skills and formal examination results (Rogers & Ricker, 2006). One of the most important consequences of these changes has been the creation of an underclass of schools labeled and categorized as “in need of assistance,” “low performing,” “underperforming,” “in challenging circumstances,” “failing,” or “in special measures” (Mintrop, 2004).
Labels such as these quite intentionally serve as descriptors of student performance, usually measured by test scores and examination performance, but sometimes by rates of attendance, dropout, and exclusion (Holdzkom, 2001). These labels are also a call for action—a call to be “turned around” in a relatively short amount of time (Duke et al., 2005). England, many U.S. states, and the Canadian province of Ontario are examples of educational jurisdictions in which policymakers have recently made significant commitments to this end. Schools in England, for example, are placed in “special measures” or “serious weaknesses” categories when they are judged to have poor levels of relative performance against local or national standards and where there is not much chance of improving without external support and intervention. Under the No Child Left Behind Act in the United States, schools that fail to meet their annual targets face a series of progressively more serious sanctions. After five consecutive years of unsatisfactory progress, for example, these sanctions may include conversion to a charter school, replacement of staff considered responsible for the failure, hiring an external contractor to run the school, or some other equally significant alternative (Rhim et al., 2007).
2. Turning schools around is different from “simply” improving them.
Few would claim that improving schools is an easy business, confronting, as it does, a host of personal and organizational variables that often seem to spin out of control with the least provocation. “Herding cats” pretty much captures the feeling many principals experience in their efforts to make changes. But school turnaround, as most conceive it, is quite different from school improvement in both subtle and not-so-subtle ways. Whereas school improvement is typically viewed as a gradual and continuous process in which almost all schools now are expected to engage, school turnaround “focuses on the most consistently underperforming schools and involves dramatic, transformative change—change driven by the prospect of being closed if it fails” (Calkins et al., 2007, p. 17). Many U.S. schools are of this type. For example, in his recent analysis, Duke (2010) cites evidence indicating that of the 12 percent of U.S. schools identified for improvement in 2005-2006, about a quarter of them “had a history of failing to meet state standards for four to six years” (p. 41).
In contrast to the prevailing understandings of “improvement,” turnaround processes are typically restricted to a specific subset of schools. They usually garner significant additional resources, have short time lines within which to demonstrate success in reaching unusually precise and public targets, and are accompanied by sanctions for failure almost entirely missing from the environments in which other schools find themselves.
3. Turning schools around is a “wicked problem.”
Initially coined by Churchman (1967), the term wicked problem is a label for problems that are especially difficult to solve and to resolve. A not-so-distant relative to the unstructured problems of interest to those who study the nature of expertise (Fredericksen, 1984), these are problems that defy routine solutions, mutate over time, and reemerge after we think we have put them to rest. In short, this means that turning around individual organizations is not an easy business. An illustration might help here.
One of the primary sources of evidence for this book is a multiyear, multiphased study of turnaround school leadership carried out in the Canadian province of Ontario (Leithwood & Strauss, 2008). Ontario was a congenial environment in which to be a public school teacher or administrator at the time the study was carried out—for those who were not wedded to the status quo. The Liberal government of the day had made public education one of its main priorities on its election about three years earlier and had demonstrated remarkable follow-through in the subsequent years. New money had been added to the system following a decade of diminishing resources and punishing criticism under a previous conservative government. This made a big difference to the job school leaders faced. Explained one secondary principal in our Ontario study:
I don’t think there is anything specific I did to raise morale in the school. The tone in education truly had an effect on this staff. We had just come through a very difficult time in education with major cutbacks and staffing cutbacks, and when I got here, things were just suddenly better. We were just richer, not in terms of money necessarily, but richer in resources, and staffing and so on. I think the general tone was on the up. So it was easy for me to tap into that and to get the staff on board.
Support for educators was apparent, and teachers were considered the solution to improving the system even more in turnaround settings than in simply “improvement” settings. This was a distinct departure from the view of the previous government, which had considered teachers the problem even though it had to depend on teachers for any changes it wished to make.
More specifically, a sophisticated balance of pressure and support had been introduced by the government to help underperforming schools gain some traction on their school improvement challenges. Pressure came in the form of quite specific provincial achievement targets widely shared by district staffs and widely communicated to parents and the general public. These were targets framed by provincial achievement tests, with the results available to anyone wishing to see them. Support took the form of additional funding for which poorly performing schools were eligible. These schools were also given access to technical and strategic assistance in the form of turnaround teams. Such teams (which we refer to frequently in this book) consisted of seconded (temporarily reassigned) teachers, principals, and central office staff, most with impressive records of achievement in their own organizations who agreed to work with several underperforming schools over the course of one or several years.1
Under these circumstances, it would not be unreasonable to assume impressive progress not only on the part of designated turnaround schools but of schools in the province more generally. Certainly the secondary school principal’s comments about improved teacher morale quoted above seem to support this assumption. As well, average student achievement scores on provincial tests did rise during this period, seeming to confirm success on a large scale.
But aggregate increases in achievement, while perfectly suitable for provincial accountability purposes, may easily mask important unresolved challenges in large numbers of individual schools, lulling us into thinking that the prevailing strategies are good enough. So we decided to test this reasonable assumption further. Restricting our attention to elementary schools, we dug out the provincial achievement scores of students in the 3,817 schools for which such data were available for the previous three years. These were aggregated reading and math scores at grades 3 and 6.2
First, we examined changes across all schools (minus the seventy-three that were part of a government-sponsored turnaround initiative, about which we have more to say later) over three years. We then examined changes in such scores separately for the seventy-three schools that had opted to be part of the province’s turnaround project and for which data were available. We were curious to know the patterns of change in achievement in both reading and math over the three years for all schools, as well as how schools in the turnaround project fared in comparison. Table I.1 summarizes these results.
As the left column of Table I.1 indicates, a school could experience an increase (I), a decrease (D), or no change (stable or S) in its achievement scores each year.3 To make the twenty-seven possible permutations more meaningful, we created six categories of achievement trajectories, described in the left column, and calculated the proportion of schools in the province on each of the six trajectories. The first number in each cell refers to all schools minus the turnaround project schools (% Reg), and the number in parentheses refers to the schools that were part of the turnaround project (% TA).
Table I.1 Three-Year Elementary School Achievement Trajectories in Regular and Turnaround Schools (2002-2005)
For the province’s elementary schools as a whole (minus turnaround project schools), Table I.1 indicates that
• The proportion of schools within each trajectory pattern was roughly the same whether the comparison was between grades 3 and 6 or between reading and math (school change seems to be a “whole cloth”).
• By far the largest proportion of schools (39.4 percent) had an achievement trajectory described as “no consistent direction.”
• More than a quarter of schools (27.3 percent), while initially unsuccessful, eventually gained some traction, perhaps through their own efforts to improve student achievement, although we cannot be certain that is the explanation (see
Linn, 2003, for other explanations).4
• Continuous improvement, the holy grail of school reformers, was rare (12.8 percent). This, you will recall, was in a remarkably promising policy environment, infused with exceptionally sophisticated knowledge about promising change strategies and able to demonstrate aggregate student achievement improvements during this period.5
• Continuous failure was the trajectory of a very small proportion (2.4 percent) of schools.
These results suggest quite strongly that continuous improvement is much more an aspirational goal than something a significant proportion of schools should realistically be expected to accomplish. These results also argue, as Linn (2003) has, for judgments about achievement trends in schools to be based on time spans much longer than is typical. Indeed “typical” often means just one year, while the evidence in Table I.1 suggests that three years is likely too short.
For the largest majority of schools in difficulty, the improvement road is filled with potholes. Their common experience is to achieve some success, only to experience a subsequent downturn. They peak and then they trough, and as we will show, much of the cause of such fluctuations arises from factors outside the school. Many underperforming schools are located in areas of high social deprivation, which creates a turbulent and often volatile context for any improvement efforts to be implemented and sustained (Harris, 2009).
Table I.1 also compares schools not considered in need of being turned around (“only” improved), with schools that were more or less actively participating in the government’s turnaround schools project. Mean score comparisons reported in Table I.1 indicate that
• Slightly fewer turnaround schools, as compared with regular schools, could be described as failing initially but eventually improving (27.3 versus 22.6 percent).
• Slightly fewer turnaround schools, as compared with regular schools, could be described as initially successful but failing in the longer term (19.4 versus 16.9 percent).
• Almost the same proportion of turnaround as regular schools changed in no consistent direction (37.8 versus 39.4 percent).
• Continuous improvement was achieved by more turnaround than regular schools (19.5 versus 12.8 percent).
But this mean score comparison does not do justice to the subject-specific scores it represents. Differences between regular and turnaround schools in grade 3 reading and math and grade 6 reading all favor the turnaround schools by a large enough margin to qualify as practically significant. Participation in the government’s turnaround project, these comparisons suggest, was associated with a greater likelihood of success in improving student performance. But given the lower (much lower in many cases) average scores of students in turnaround schools at the outset, a more dramatic difference would have been a reasonable expectation. This reinforces the evidence showing that turnaround schools may still be underperforming in terms of local and national standards. Nevertheless, any success can be significant, if relative, achievement.
4. Multiple causes, slippery high ground, and issues of scale account for the “wicked” nature of the school turnaround problem.
The best available evidence about private sector turnarounds suggests about a 70 percent failure rate (Kotter, 1995), and the small amount of available evidence does not justify a more optimistic prediction for public sector turnarounds, as the Ontario illustration suggests. Underperforming schools present especially thorny turnaround challenges (Kowal & Hassel, 2005) for at least three reasons that have been identified in U.S., as well as Canadian and U.K., contexts:
Multiple and external causes of underperformance. Multiple factors outside the actual organization often influence its ability to turn around. This is especially the case for school turnarounds. In the case of schools, these external factors are often a product of students’ socioeconomic backgrounds. Such factors also sometimes include dysfunctional district policies and regulations, inadequate funding, and disincentives to the recruitment of high-quality teachers, for example (Chapter One examines these factors in more detail). Schools facing a large number of these causes of poor performance typically require very different forms of improvement strategies than schools in more favorable environments.
The challenge of sustaining improved performance. While turning around individual schools is not for the faint of heart, it is possible, at least for a short period of time. Evidence demonstrates that school turnaround even in the most difficult circumstances is possible and that schools can succeed against the odds (Harris et al., 2006). However, this evidence also warns us that turnaround can be fragile and improvement short-lived. The high ground of improved performance is a slippery slope. For every success story, many more schools return to their old ways once special resources and additional support have been removed (Duke, 2010; Gray et al., 1999).
The added complexity of attempting turnarounds on a large scale. Although turning around an individual school is no easy task, the current aspiration in many jurisdictions is to solve this wicked problem on a very large scale. For example, this is the goal of the No Child Left Behind legislation in the United States. It is also the premise of the National Challenge in England, a policy aimed at enforcing improvement in over four hundred schools deemed to be persistent underperformers (Harris, 2009). The best evidence available about what this aspiration entails suggests that neither top-down nor bottom-up approaches by themselves work, and that many millions of dollars so far thrown at the problem have largely failed (Fullan & Sharratt, 2007).
Most top-down strategies to date poorly reflect our basic understandings of successful change processes. The almost entirely punitive nature of No Child Left Behind is the poster child for this inadequacy. The evidence from Ontario described throughout this book, however, was collected in a very different policy context, one carefully attuned to evidence about successful change processes. So one important purpose for this book is to demonstrate the viability of large-scale school turnaround under conditions that nurture schools on a carefully balanced diet of both support and pressure, as well as autonomy and central direction. That said, the results reported in Table I.1 argue for conservative estimates about what is possible on a large scale even under exceptionally favorable circumstances.
5. More ambitious estimates about what is possible on a large scale depend on a better understanding of how to turn around schools on a small scale.
As is apparent from earlier observations, we are far from having a well-codified process—or some equally useful advice—for turning around underperforming schools. There is no magic formula or sure-fire way of improving these schools; an added danger is that any improvement will turn out to be temporary. Not enough research has been done in improving schools in serious difficulty to produce a definitive model of improvement for these schools. For example, Rhim et al.’s (2007) U.S.-centric review of such evidence is based on only twenty-six studies carried out in schools, all of which used only case study designs. Evidence from such studies, although rich in detail and possible insights, provides very little certainty about what works. Furthermore, much of the research about school improvement has focused on schools that require a lift in performance rather than radical intervention. So evidence informing the turnaround process is quite limited.
Yet worldwide, large-scale government reform initiatives aspire to increases in student performance that at least implicitly assume such knowledge exists. In England, a number of large-scale attempts have sought to improve schools that continuously underperform. Initiatives like Excellence in Cities Education Action Zones and the latest suite of challenges there (London Challenge, Black Country Challenge, and the National Challenge) have all been focused on raising student achievement in these groups of underperforming schools. All of these approaches share a common model: targeted resources, prescribed interventions, compulsory staff development, constant scrutiny, endless planning processes, and continual weighing and measuring from external agencies.
Over fifteen years, this model has been followed again and again in different guises, but the net result has been the same: some schools change while the new resources remain, but no large-scale change in overall performance occurs. This is an example of the wrong model being used over and over in the vain hope that it will work next time. It is also indicative of the point that relevant evidence is not sufficiently robust to persuade policymakers to undertake a different approach. As the data in Table I.1 also imply, turning around significantly underperforming schools is likely to require something different from or additional to the improvement strategies that work in adequately performing schools.
Especially as the focus of these reform efforts shifts from raising the bar (increasing mean levels of achievement across the entire system) to closing the gap (reducing the discrepancies in achievement between students who typically succeed and those who are typically at risk), this lack of procedural know-how is an albatross around reformers’ necks, placing a low ceiling on what it is possible to accomplish and, at the same time, encouraging the replication of failed improvement approaches in different guises. Change, as it has often been observed, is a problem of the smallest unit. If we do not have a good understanding and a robust model of how to turn an individual school around and ensure that improvement stays around, the chances of turning around many schools at the same time seems remote.
6. Poorly performing schools stand virtually no chance of turning around without good leadership.
A considerable body of case-based research addresses the process of turning around failing organizations (Mellahi & Wilkinson, 2004; Murphy, 2008), although relatively little of this research has been undertaken in schools (Paton & Mordaunt, 2004). Much of the evidence can be found in the business literature where turnaround is a business in itself and the idea of turnaround specialists is commonplace. Within this literature, leadership is widely considered to be vital to private sector turnarounds. We are unaware of any documented example of a successful turnaround in business without some change in leadership, and the same seems likely to be the case in schools as well (Harris & Chapman, 2002a). In his study of schools on probation in two U.S. states, for example, Mintrop (2004) concluded, “What schools did under probation largely depended on how principals reacted to the low-performance status” (p. 13). In his truly remarkable book, Lessons Learned: How Good Policies Produce Better Schools, Whelan (2009) also concludes from several successful cases that “schools which have run into severe difficulty can be improved, often mainly through the efforts of existing staff and even when the school faced strong external pressures and challenges” (p. 96). And leadership, Whelan claims, is a central explanation for both successful school turnaround and the subsequent sustaining of improved school performance. This is the most obvious lesson from schools that have been successfully turned around.
Evidence from several other recent lines of school turnaround research also indicates a central role for leadership. For example, research in England by Chapman and Harris (2004), and Clarke (2004) highlights the centrality of a change in leadership as the main lever for turnaround and improvement in schools that underperform. This finding is reflected as well in Rhim et al.’s (2007) review of primarily U.S. evidence. Leadership, in sum, is a largely uncontested linchpin in both accounting for an organization’s failure and returning the organization to a stable state (Kanter, 2003; Kowal & Hassel, 2005).
7. We know what almost all successful leaders do.
Two particular claims about leadership are quite central to the general claims we make in this book. The first of these is that there is a common core of leadership practices used by successful leaders in almost all contexts. Chapters Five through Eight describe these practices in some detail and summarize evidence about their effects. As you will see, these practices aim to accomplish four purposes critical to leadership success:
• Create a widely agreed-on sense of direction for the organization (Chapter Five).
• Help develop the capacities of organizational members to move the organization in that direction (Chapter Six).
• Redesign or restructure the organization to support people’s work (Chapter Seven).
• Manage the “technical core” of the organization. In schools this will be teaching and learning processes or, to use Raudenbush’s (2009) term, the “shared instructional regime” (Chapter Eight).
We spend considerable effort justifying our claim about these categories of leadership practices in Chapters Five, Six, Seven, and Eight. For now we simply note that they are almost equally important, for example: in both large and small schools; in schools serving either diverse or homogeneous student populations; schools in Canada, the United States, England, and most other developed countries; and not only in schools but also in non-school organizations. The four categories include fifteen to nineteen more specific leadership practices depending on which of our studies is referred to (Leithwood & Jantzi, 2005; Leithwood, Day, Sammons, Harris, & Hopkins, 2006).
We use the term practice as a synonym for the term function. Neither the four categories nor the specific practices within them actually describe behavior or activity. Rather, the practices signify the outcome of such behavior. These are proximal or short-term outcomes in a longer-term effort toward organizational improvement. So these practices do not actually capture how leaders do their work, but they do capture what leaders intend to accomplish through that work.
8. We have almost no knowledge about how successful leaders do their work in turnaround contexts.
I would have to say that to be active in the learning process would probably be one of the best things I’ve done. Particularly it helped because I was a new principal. It was only my second year when we started the turnaround. I told the teachers, “Honestly, you know I’m new at this too. I don’t know where we’re going with this exactly, but I do know that I will be a better principal by the end of this, and if I’m a better principal, that should mean that you feel more support, you feel you’re better teachers, the kids will have a better quality of education, and you’ll have a stronger skill set when we’re done. And if that means that I have to be uncomfortable, then I’m fine with that.” Because it was not always comfortable. There was criticism. There were questions. It was humbling. There were lots of questions that for a principal would take you outside your comfort zone and called you to task, to step up your game and be more aware of things. I think I’ve gotten better at most of those things—not all of them. If there’s things I’m not good at, I think I know what they are now. I think it’s better to at least know where your gaps are than to kind of stumble around. —Elementary principal
We began the discussion of starting point 7 (above) by noting that we make two claims about successful leadership which are quite foundational for this book. Starting point 7 was the first claim (successful leaders in almost all contexts engage in a common core of practices). The second claim is that, such general application notwithstanding, successful leaders are also exquisitely sensitive to the contexts in which they find themselves. Large schools present much different leadership challenges than do small schools. The challenges facing leaders of inner city schools are typically quite different from those facing leaders of schools in the leafy suburbs. And so on.
Now, having a core set of common practices on which to draw (we described them as intended proximal or short-term outcomes) is extraordinarily helpful for leaders, otherwise they are without intention. Furthermore, having such proximal intentions in mind to guide what leaders do still leaves considerable flexibility for leaders to decide how to accomplish or enact those intentions in their own specific contexts. But this raises a “level of specification” question for leadership research—or in the case of this book, an account of what works based on leadership research. What level of detail should descriptions of successful leadership practices provide in order to be most helpful to school leaders and their collaborators?
Our answer to this question is to give descriptions of how leaders successfully work in contexts that are shared by many. Examples of such widely shared but unique contexts are large or small schools, elementary or secondary schools, schools serving culturally diverse or culturally homogeneous student populations and—of course—turnaround versus “simply” improving schools. Accounts of how leaders work in such contexts does not acknowledge the even more unique features faced by each leader. But research cannot be (nor should it be) in the business of trying to describe successful leadership practices as enacted in response to such unique contextual differences as this school versus that school, or differences in leaders’ relationships between this teacher versus that teacher.
So our two claims about leadership are: most successful leaders use a common core of practices, but they enact those practices in ways that are sensitive to the contexts in which they find themselves. A significant part of the success of most leaders depends on their creating a sense of shared purpose in their organizations, for example. This is a core leadership practice. But successful leaders of schools in danger of going out of business in six months if dramatic improvements do not take place often “sell” their own views of where their schools should be headed to their staffs and communities. In contrast, successful leaders of schools in no such danger are much more likely to build a sense of shared purpose through highly collaborative and inclusive processes. A major purpose of this book is to uncover and describe how successful school leaders enact the common core of successful practices in turnaround school contexts.
Kowal and Hassel (2005) have claimed that while the “evidence is strong that a school’s leader makes a big difference in student learning in all school settings . . . understanding of the characteristics that distinguish high-performing school leaders from the rest is very limited. In addition, there is limited research that describes how the characteristics of high-performing leaders differ in emerging school contexts such as start-up and turnaround schools” (p. 17). Comparable evidence indicates furthermore, that successful turnaround leadership practices vary in response to the unique challenges encountered at each stage in the turnaround process (Slatter, Lovett & Barlow, 2006). Turnaround schools, it would seem, are unique contexts for leadership and require unique enactments of successful leadership practices.

How This Book Unfolds

The ten chapters of the book are organized into three thematic parts. The chapters in Part One address the causes of school failure and declining school performance (Chapter One); a staged conception of school turnaround processes, which serves as an important framework for our subsequent examination of turnaround leadership practices (Chapter Two); and a case study illustration and elaboration of some key ideas associated with those stages from one secondary school in England (Chapter Three).
Part Two consists of five chapters. Four of these chapters provide an extensive description of how each of four categories of leadership practices is enacted by successful turnaround leaders: direction setting is examined in Chapter Four, developing people in Chapter Five, redesigning the organization in Chapter Six, and managing the instructional program in Chapter Seven. Chapter Eight illustrates these core leadership practices as they were enacted by successful leaders in one elementary and one secondary turnaround school in our Ontario study.
The two chapters in Part Three speak to two outstanding challenges. How to sustain a failing school’s performance once it has been significantly improved is the first of these challenges (Chapter Nine). The limited evidence available about this challenge indicates that improved performance is usually not sustained, thus calling into question the wisdom of expending the often enormous efforts required for the initial turnaround. This challenge, touched on in earlier chapters, deserves the much further unpacking provided in Chapter Nine. Finally, Chapter Ten describes just how those leadership practices outlined in earlier parts of the book eventually influence student learning. This chapter offers advice to turnaround leaders, based on recent syntheses of research, about how to select those conditions for improvement in their schools most deserving of their time and attention—a major challenge for school leaders confronted with a potential blizzard of things to do but time to do only a few things well.

A Note About Research Contexts

The original evidence for this book was provided by research carried out in England and Canada. While we believe our results are of direct use to leaders in the United States and most other developed countries, a brief description of some of the key similarities and differences among the English, Canadian, and American school systems is provided here for those wishing to draw their own conclusions.
Major responsibility for K-12 education rests at the subnational levels in both Canada and the United States. Canada’s constitution cedes responsibility to provinces, and states hold this responsibility in the United States. Unlike Canada, however, the U.S. federal government wields considerable influence on schools through major policy initiatives that schools must implement if they are to have access to the funding these policies provide. As one example, the U.S. No Child Left Behind (NCLB) legislation has been a major influence on how schools and districts think about curriculum, testing, and accountability. The achievement targets set for schools to remain eligible for federal funding have strongly influenced which schools are considered to be in need of turnaround, as well as some of the most common strategies for accomplishing school turnaround. The Canadian federal government, in contrast, has no such influence, and provinces guard their jurisdiction over primary and elementary schools closely.
Districts or boards of education are a key feature of both the Canadian and American school systems, although the exercise of local authority is a stronger feature of the American system. In general, provinces exercise more authority over Canadian districts than states exercise over American districts. In the case of the research carried out for this book, the strong authority of the provincial government was an important part of the story about turnaround schools in Ontario. The authority of states is likely to have less influence in U.S. districts. Nonetheless, the nature of the provincial authority in Ontario was largely based on expertise, not position. So expert-based authority on the part of states in the United States might yield similar results under favorable circumstances.
England has a national educational system with authority strongly centralized and a recent history of significant top-down initiatives such as the National Literacy and National Numeracy Strategies. Although local educational authorities (LEAs) were a powerful source of influence on schools, the Thatcher government and then the Blair government diminished their authority. More recently, subnational restructuring has expanded the responsibilities of what are now called local authorities to include most government services designed to assist children, including education, health, and other social services. Each school has its own elected board of governors with responsibilities parallel to those enjoyed by elected boards of trustees for districts in Canada and school boards in the United States. This more extreme form of site-based management, however, is strongly restrained by a national system of school inspections, public reports of league tables, and centralized responsibility for leadership development in the form of the National College for School Leadership. The Department for Children, Schools and Families has been responsible for the funding of school turnaround initiatives in England and has sponsored particular programs aimed at turning around failing schools.
Key Points
• Turning “failing” schools around is a prominent focus of contemporary educational policy.
• Turning schools around is different from “simply” improving them.
• Turning schools around is a “wicked problem.”
• Multiple causes, slippery high ground, and issues of scale account for the “wicked” nature of the school turnaround problem.
• More ambitious estimates about what is possible on a large scale depend on a better understanding of how to turn around schools on a small scale.
• Poorly performing schools stand virtually no chance of turning around without good leadership.
• We know what almost all successful leaders do.
• We have almost no knowledge about how successful leaders do their work in turnaround contexts.

Part One

You have to understand the context we work in. This is an area of high social deprivation where daily existence is difficult enough. Simply getting to the school gate is an achievement for many of our students. Poverty is not an excuse for underachievement, but it is a powerful influence.
Secondary school principal
The reasons for school failure are almost as complex as are the reasons we are unable to turn around underperforming schools in vast numbers. These reasons are multifaceted and interrelated, compounding and exacerbating the problem of school failure. Whether in Canada, England, the United States, or any other part of the Western world, there are common factors that make turnaround difficult and render some schools, in certain contexts, less able to raise the performance of their students. Many approaches aimed at improving underachieving schools have served to further disadvantage them, largely by failing to take adequate account of their context and by locating the blame for failure squarely within the school. But reasons for school failure are rarely one-dimensional or singular. Consequently, the polarized debate of “school’s fault” versus “society’s fault” for failure does not take us very far. Attributing blame detracts from solving the problem.
As always, the truth lies somewhere in between. In some cases, schools are at the heart of the problem. Poor teaching is condoned, weak leadership is tolerated, and the dominant view is, “What can you do with these kids?” driven by low expectations. On the other side of the coin, it is undeniable that the relationship between poverty and underachievement is powerful. The consequences of growing up poor affect millions of young people worldwide (Berliner, 2006). Poverty remains a global, social, and economic issue, and the educational reform agenda in many countries reflects a renewed interest in addressing the relationship between poverty and underachievement. The gap in achievement between children from low-income families and their more affluent peers persists, and in the majority of cases, it actually increases throughout schooling. Although social disadvantage is not an excuse for poor achievement in academic terms, it certainly is a powerful explanatory factor. It remains the case that many failing schools are located in high-poverty contexts.
A substantial corpus of international research into the relationship between poverty and education demonstrates that while the general attainment levels of poor children have improved over time, the gap between the majority of children from low-income families and their more affluent peers has widened (Knapp, 2001; Thomson & Harris, 2004). Children attending high-poverty schools are not likely to achieve as well as their peers in more favorable school contexts. The net effect of poverty on educational achievement is stark. Educational outcomes in deprived areas are worse than those in nondeprived areas, whether they are measured in terms of qualification, attendance, exclusions, or retention rates. Inner-city areas, in particular, are associated with low educational outcomes. Socioeconomic status or family background typically explains more than half the variation between schools in pupil achievement, and low family income in childhood years makes a significant difference to subsequent educational outcomes. Chudgar and Luschei (2009) provide new support for this well-known finding in their recent analysis of international achievement tests across twenty-five countries.
Part of the reason for the decline in social mobility in many countries is the strong bond between low levels of family income and subsequent educational attainment. Underachievement and levels of deprivation continue to be strongly and powerfully linked. Furthermore, poverty continues to be a chief explanatory factor for the persistent low levels of attainment for certain groups of young people.