Inspiring leaders toimprove children’s lives                                                              Schools and acad...
ContentsExecutive summary                                   3Introduction                                        4Family v...
Executive summaryIn an era of diminishing centralisation,accelerating the rate and depth of schoolimprovement and reducing...
IntroductionOn 18 October 1976, the UK’s Labour prime            Should we persist with these same strategies forminister ...
A similar approach is being adopted elsewhere by           4. How might the system move from wherethe College to increase ...
Family virtuesThe idea of schools working collaboratively has a       find it easier to meet the needs oflong history, but...
aid the integration of children’s                          by which the process of mutual improvement             services...
Hamel and Prahalad also observe that:                  Clusters are a critical structural building block of a             ...
The local solutions approachThe College realised that the impending crisis in      government in generating system change ...
Co-construction in family clustersFamilies of schools working on local solutions,whether it is middle leadership or succes...
Expanding system leadershipIn education, the term system leader, originally          All the projects linked to clusters e...
Beyond the self-managing schoolFor the last quarter of the 20th century, a majortask for school leaders in England was the...
Conditions of a sustainable self-improvingsystemTo create a sustainable SISS, three key questions      above, few schools ...
Not all clusters would be newly formed: many           with the College to broker clusters, then supportalready exist unde...
The skills of leading a successful school and the                      In the business world some partnerships workskills ...
the recognition that working with another              spending time understanding the culture        school is a reciproc...
This is not a simple, straightforward business 4,                   Without nodal system leaders at its heart, there is   ...
Is this task easier than it was with the first system             demonstrating leading-edge pedagogyleaders? Richard Elmo...
What inevitably began as separate strands of theCollege’s work at different levels, often indifferent locations, and led b...
Towards a mature self-improving systemOriginally the College’s task was to ensure the      Over a five-year period, the Co...
The emergent range of patterns is           support, clusters either failed toconsiderable and at first sight looks       ...
At the leading edge of suchdevelopments are clusters reaching ahigh level of maturity: they are reapingin full the benefit...
ConclusionIt has long been known that the most powerfulinfluences on teachers are other teachers, butpolicies have rarely ...
AcknowledgementsI am most grateful to Toby Greany, Maggie Farrarand Katy Emmerson, and many of their NationalCollege colle...
ReferencesBrown, J S & Duguid, P, 1991, Organizational learning      Huberman, M, 1995, Networks that alter teaching:and c...
The National College is the first           ©2010 National College for Leadership of Schoolsprofessional body uniquely ded...
Upcoming SlideShare
Loading in …5

Creating a-self-improving-school-system


Published on

Published in: Education, Business
  • Be the first to comment

No Downloads
Total views
On SlideShare
From Embeds
Number of Embeds
Embeds 0
No embeds

No notes for slide

Creating a-self-improving-school-system

  1. 1. Inspiring leaders toimprove children’s lives Schools and academiesCreating a self-improving school systemDavid H Hargreaves, July 2010 Resource “The crisis of the world is, above all, an institutional crisis demanding institutional innovation” (Peter Drucker) “The future is already here: it is just not distributed very well” (William Gibson)
  2. 2. ContentsExecutive summary 3Introduction 4Family virtues 6The local solutions approach 9Co-construction in family clusters 10Expanding system leadership 11Beyond the self-managing school 12Conditions of a sustainable self-improving system 13Towards a mature self-improving system 20Conclusion 23Acknowledgements 24References 25
  3. 3. Executive summaryIn an era of diminishing centralisation,accelerating the rate and depth of schoolimprovement and reducing the number ofunderperforming schools requires a new vision.Since the birth of school improvement in the1980s, the quality of school leadership hasincreased sharply and most schools have gainedexperience of working in partnerships andnetworks of many kinds. Increaseddecentralisation offers an opportunity for theschool system to build on these and become self-improving.There are four building blocks of a self-improvingsystem: clusters of schools (the structure); thelocal solutions approach and co-construction (thetwo cultural elements); and system leaders (thekey people). These are already partially in placebut need to be strengthened so that schoolscollaborate in more effective forms of professionaldevelopment and school improvement.This thinkpiece explores the conditions necessaryto achieve a sustainable, self-improving schoolsystem, with a particular focus on thedevelopment of school clusters and the associatedprovision of cluster leaders. A sketch of how sucha system might evolve over the next five years isoffered. 3
  4. 4. IntroductionOn 18 October 1976, the UK’s Labour prime Should we persist with these same strategies forminister James Callaghan gave a speech in Ruskin school improvement or is it time for a new vision?College, Oxford that started what became known Two important changes have occurred thatas the Great Education Debate. At this time it was suggest the need for a new direction. First, thevery unusual for a prime minister to discuss calibre of school leadership has improved, ineducation policy in public. In his near-apologetic many places to a dramatic degree, reflecting theapproach to the subject, Callaghan argued that National College’s central task of ensuring theeducation was now too important to be left provision of leaders with relevant capabilities.almost entirely to the teaching profession and Schools are more accustomed to managing theirthat many voices, including that of a prime own financial affairs and many have developedminister, needed to be heard on the purposes of sophisticated continuing professional developmentschooling and educational standards. More was (CPD) for their staff. Second, virtually everybeing demanded from schools, and core issues, school has experience of partnership with othersuch as the desirability of a national curriculum schools, and the education service is now moreand a stronger inspectorate, should be addressed. networked. School leaders are more aware of schools as a system, and the coalitionUp to this point, England had a highly government’s plans are evidently intended todecentralised education system. The Ruskin change the shape of this system. A new balancespeech marked the beginning of a new phase, is being struck between centralisation andwhich eventually led, under the Conservatives, to decentralisation, with a clear reduction inthe 1988 Education Reform Act that introduced a centralised action, at both national and localnational curriculum and a new assessment levels, and a matching increase in the powers andsystem. This was an unprecedented degree of responsibilities of schools.centralisation but it was matched by a degree ofdecentralisation that delegated new financial In this thinkpiece, I argue that increasedpowers to schools, and to their headteachers, the decentralisation provides an opportunity for a newspirit of which was neatly captured in Caldwell vision of school improvement that capitalises onand Spinks’s The self-managing school, also the gains made in school leadership and inpublished in 1988. partnerships between schools. It would usher in a new era in which the school system becomes theSignificantly, this second half of the 1980s gave major agent of its own improvement and does sobirth to the school improvement movement, at a rate and to a depth that has hitherto been nowhich was driven both centrally by a more hands- more than an aspiration. It is essential that such aon education department and some local change would enhance parental confidence in theeducation authorities as well as by more quality of schools and the effectiveness ofenterprising headteachers. Successive teachers, on both of which better educationalgovernments, both Conservative and Labour, outcomes depend. This short thinkpiece suggestshave for over 20 years pursued this combination – what could be done to realise such a vision. It isuneasy to some – of centralisation in some not a detailed policy prescription, but a sketch ofrespects and decentralisation in others. The the main lines of action that would need to beconstant challenge has been to minimise taken.variation, not just within and between schools butalso between local authorities, which has led School improvement depends on improvedcentral government to take ever greater powers leadership, but the necessary scale, speed andof intervention, backed by national field forces sustainability of leadership development cannotand strategies. School improvement has thus be achieved by centralised action alone. In thecome to be defined in terms of the processes of College’s innovative local solutions approach tointervention in schools that are deemed, by the shortage of headteachers, successionwhatever measure, to be underperforming. Much planning takes place across networks of schoolshas been achieved, yet it has to be conceded that (in the local authority or the diocese) in ways thatnot all schools have improved substantially or are responsive to local circumstances.even sufficiently over this last quarter century. 4
  5. 5. A similar approach is being adopted elsewhere by 4. How might the system move from wherethe College to increase the provision of middle it is now to becoming a self-improvingleaders through local clusters of schools as well as system? Do the College’s currentin City Challenge. In this sense, the College is achievements (including those notedacknowledging changes in the system and then above) contribute to such a system? Whatdeveloping them further in the interests of better additional action might be needed?leadership provision. 5. What would make a fully-fledged self-Scaling up such local solutions necessarily entails improving system robust and self-new ways of deploying the headteachers of sustaining?successful schools, who accept responsibilitiesbeyond the boundaries of their own schools and The language around the concept of a self-are prepared to help other schools. The College’s improving system of schools (henceforward aaction with such headteachers – in the form of SISS) is confusing. Associated terms, such as anational leaders of education (NLEs) and local self-managing system or self-developing system,leaders of education (LLEs) – runs parallel with are used interchangeably despite variablethe emergence of larger groups of schools in connotations of the terms. At its core, the notionforms such as federations and chains (Hill, 2010), of a SISS assumes that much (not all) of thein addition to clusters of schools serving a wide responsibility for school improvement is movedvariety of functions, all of which is altering the from both central and local government and theirshape of the school system. agencies to the schools. An obvious forerunner in England is local management of schools (LMS),The College’s work on the provision of school the delegation of financial responsibilities toleaders has thus evolved from centralised schools in the 1980s, which is generally regardedprovision to the point where the goal is making as a world-leading success story. However, a SISSleadership development a largely self-generating is not merely the sum total of self-improvingenterprise, grounded in networks of schools. So schools. The system element in a SISS consists ofcan the changed strategy of leadership clusters of schools accepting responsibility fordevelopment become the basis for a largely self- self-improvement for the cluster as a whole. Aimproving system? Is it possible to move from a SISS embodies a collective responsibility in a waycentralised model of driving every individual that neither school improvement nor LMS hasschool to improve itself to a process of systemic ever done. In effect this involves the creation of aself-improvement that matches the new model of new intermediary body between the individualleadership development? Indeed, do changes in school and the local authorities, which are usuallyleadership development and school improvement seen as the middle tier between centralnecessarily have to be aligned? government and the individual school.In addressing these issues, this thinkpiece poses The architecture of a SISS rests on four mainfive linked questions to frame the argument: building blocks: 1. What would a self-improving school capitalising on the benefits of clusters of system look like and what would be its schools defining features? adopting a local solutions approach 2. In what ways would a self-improving system be an advance on our current stimulating co-construction between system? schools 3. What would be the system’s building expanding the concept of system blocks and to what extent is that leadership architecture already in place? 5
  6. 6. Family virtuesThe idea of schools working collaboratively has a find it easier to meet the needs oflong history, but recently this has become more every student since the range ofcommonplace as a result of government initiatives provision, including curricular and 14-19(eg, leadership incentive grants), the needs of provision, is much greater than that of astudents (eg, post-16 provision, small A-level single school, and students can easily beoptions), the attractions of formal association (eg, moved within the familyfederations, trusts), the outcome of critical Ofstedreports (eg, NLEs), as well as projects aimed deal more effectively with specialdirectly at fostering inter-school collaboration (eg, education needs, especially when athe College’s networked learning communities special school is a family member and(National College, 2006a), some of which professional expertise in particularcontinue to this day). So few schools lack aspects of such needs is shared betweenexperience of partnership, though the character schoolsand quality vary considerably, from a relativelyshallow, short-term relationship affecting limited find it easier to meet the needs offunctions and few people (a loose partnership) to every staff member since staff cana deep, enduring relationship that affects most job-rotate or be offered freshfunctions and most people in the schools (a tight opportunities between schools withoutpartnership). Very few groups of schools are at changing jobs, and school-basedthe tight extreme, with common governance and professional development, enriched bya collective strategy. the resources of several schools, replaces out-of-school coursesVarious names are used for these partnerships: support new leaders since the existingthe most common are cluster, network, chain and headteachers and leaders in the familyfamily. Agreement on what might be a generic cluster are at hand to support theterm is lacking, so for the purposes of this newcomerthinkpiece I shall use the term family cluster,because of its organic associations and build leadership capacity and boostimplications. The name has been used within City succession planning since staff areChallenge to identify schools with statistically interchangeable within the family ofsimilar intakes in terms of various contextual schoolsvariables, including prior attainment. Each schoolcan then examine how it compares with others in protect their members, for while eventhe family – to a maximum family size of 22 the most successful schools are, likeschools – in relation to student attainment and businesses (Collins, 2009) vulnerable torate of progress. A member of staff from each crisis and failure, if this happens to aschool in the family joins a meeting once or twice school in a strong or tight family cluster,a term with others to share ideas and materials as other members get an early warning –well as encourage mutual visiting. The aim is to earlier than Ofsted – and intervene withshare good practice and in particular help low- immediate support without provokingachieving schools to improve their performance. defensive resistanceIn terms of the continuum mentioned above,many of these partnerships are loose, though distribute innovation by sharing thesome are developing into tighter ones. costs, in time and resources, of new developments, and by working with otherI use the term family cluster in a stronger sense partners, such as business and furtherto indicate an organic and sustainable relationship educationof a relatively small number of schools, between 3and 12 per cluster. Considerable benefits transfer professional knowledgepotentially accrue to family clusters, which: more readily through joint professional development and the ease of mentoring and coaching 6
  7. 7. aid the integration of children’s by which the process of mutual improvement services because external agencies find occurs. Family members both challenge one it more efficient to work with a family another and support one another, and then cluster than with separate schools celebrate their individual and collective achievements. become more efficient in the use of resources because schools share both There is a powerful next step: competition material resources, (eg expensive between family clusters. This has yet to develop technology or sports facilities) and in our education system, though the phenomenon human resources (eg, business and is well-established in the business world, where financial services), especially in primary such clusters would be called strategic alliances or schools coalitions. Hamel and Prahalad (1994) highlight one problem in the business world: Many of the College’s NLEs and LLEs have discovered these benefits, sometimes as an “Almost every large company has a unexpected effect of emergency action, where a spaghetti bowl of alliances, but there is family relationship originates in a crisis and an seldom an overall logic to the set of NLE assumes a role of responsibility for a school partnerships in that there is no distinctive, in difficulties. However, these are potential underlying point of view about industry benefits. To my knowledge, no family cluster, future and no conscious attempt to even a tight one such as a federation, has yet assemble the companies that have reaped all these benefits in full. The best clusters complementary skills to turn that have partially secured some of them, but full conception of the future into reality. Thus, benefits await cluster maturity. although many companies have a wide variety of partnerships, the individual Several schemes 1 have demonstrated that pairing partnerships are often disconnected, each a high-performing school with a weaker one acts serving an independent and unrelated as a positive force for improvement. One purpose. By way of contrast, what we have unanticipated consequence is that the high- in mind are multilateral partnerships that performing school actively gains from the pairing. possess a clear ‘cumulative logic’.” There is, of course, a cost involved, but this is offset by the boost to morale and the professional Hamel & Prahalad, 1994 skills of the lead school’s staff that arise from the help they offer to schools in difficulties. In the This is precisely the problem in many school event, both schools improve. System-motivated partnerships too. Many loose clusters are simply altruism pays rich dividends. too superficial to yield much in the way of family benefits. Tighter clusters in a SISS ensure that The more family-like the cluster arrangement, I the different strands of partnership explicitly suggest, the greater the chance that more of the share a ‘cumulative logic’, the core purpose of benefits will be realised and the more likely it is which is the joint improvement of teaching and that all member schools will improve. Cluster learning. arrangements do not preclude competition between members, but combine it with co- operation. This is often the case with business firms: ‘Co-operation is ceasing to be the opposite of competition and is becoming, instead, one of its preferred instruments’ (Deering & Murphy, 2003). The consequential benefits are the means1 Examples are City Challenge, Leading Edge, and the raising achievement transforming learning (RATL) programme of the SpecialistSchools and Academies Trust (SSAT). 7
  8. 8. Hamel and Prahalad also observe that: Clusters are a critical structural building block of a SISS, but three others are also essential, the first “Competition for the future often takes of which is breaking free from a dependency place between coalitions as well as culture in which the solutions to school problems between individual firms… Managing are thought to lie somewhere beyond the schools coalitions thus often entails a careful themselves. balancing of competitive and cooperative agendas over time. Coalition members must be careful to keep their competitive instincts in check or run the risk of undermining the partnership prematurely.” Hamel & Prahalad, 1994Competition between school clusters similarlydrives the mutual improvement within andbetween clusters to the next level, but it takesskilful leadership to know when to build oncollaboration by the introduction of the friendlycompetition that drives up standards in theinterests of collective achievement.Schools do, of course, form clusters on avoluntary or self-selected basis, without anexplicit aim of school improvement. The College’scluster-based middle leadership developmentprogramme (MLDP) is a projected alternative tothe centralised provision of training for middleleaders that simply cannot cope with the numbersneeded annually. In the new model, clusters ofschools work together, with trained facilitators, toprovide on-the-job professional development,supported by College-provided materials. Inparallel, the Training and Development Agency forSchools (TDA) has developed continuingprofessional development (CPD) clusters, showinghow school-focused CPD can be locally providedin families. All such schemes have a beneficialimpact beyond their stated goal: in particular,they help to foster and embed a culture ofprofessional learning within and between schools,an advance that is critical in moving from a self-improving school to a self-improving system.The challenge is whether clusters whose originslie in issues other than improvement can makethe transition to inter-school support where themain rationale of partnership becomes thecumulative logic of joint improvement. 8
  9. 9. The local solutions approachThe College realised that the impending crisis in government in generating system change bythe supply of headteachers could not be averted supporting local solutions in place of top-downby means of a conventional, centralised model of prescription. As Bunt and Harris (2010) put it:succession planning; and a solution washampered by the perception that headteacher “Government has traditionally found itsupply was the College’s responsibility and each difficult to support genuine localschool had to fend for itself in a competitive solutions while achieving nationalmarket. In reality, the detailed nature of the impact and scale… Centrally drivensuccession problem, the kinds of organisations initiatives have struggled to make annecessarily involved, and the particular kinds of impact on many of the complex issuesaction demanded, all varied from place to place. confronting us today…. [This] requiresSo the College mapped the national landscape for not only action from government, butsuccession planning, provided relevant data and engagement and local knowledge fromevidence, and set in place the overall strategy and citizens. But despite support fromsupport. The solution, however, was determined across the political spectrum, genuineand driven locally, tailored to local circumstances localism is something governments findand resources. This local solutions approach difficult to achieve. What makes ‘localinvolves local self-evaluation, local objectives and solutions’ effective is their locallocal action plans. It means that, with College specificity, and the ability of groups tohelp, problems have to be diagnosed and owned tailor solutions to local contexts. Locallocally, and the commitment and creativity for groups are also best placed tosolutions also generated locally. encourage community engagement on a social issue, through access to localThe local solutions approach builds the culture of networks and existing relationships.a SISS, because it necessitates the acceptance by There is therefore an inherent tensionschools of three related ways of thinking about between the factors for successfultheir condition and what to do about it. localism and the impulse to achieve impact nationally… Policymakers need Schools take ownership of problems and an alternative that combines local reject the notion that the school itself can action and national scale – an effective do little or nothing because it is approach to ‘mass localism’. somebody else’s responsibility to provide a solution. Mass localism depends on a different kind of support from government and a Solutions are seen to be available from different approach to scale. Instead of within the school system, provided assuming that the best solutions need schools work together to diagnose the to be determined, prescribed, driven or problems and devise solutions in their ‘authorised’ from the centre, mutual interests. policymakers should create more opportunities for communities to The school system is not simply an develop and deliver their own solutions amalgam of isolated schools but a and to learn from each other. It is not collection of groups of schools that enough to assume that scaling back sometimes need to collaborate in order to government bureaucracy and control get better. will allow local innovation to flourish.”The local solutions approach also involves a Bunt & Harris, 2010recognition by central government that thecentralised and clumsy one-size-fits-all approach The work of the College has demonstrated thethat ignores local contexts is becoming less and power of such ‘mass localism’ in education andless appropriate as the local solutions approach is how it is an essential ingredient of a SISS.embedded, and indeed impedes that process.What the College has done is very much in linewith new approaches adopted by the new 9
  10. 10. Co-construction in family clustersFamilies of schools working on local solutions,whether it is middle leadership or successionplanning, share a common feature: their capacityto stimulate co-construction among theparticipants. The term co-construction hasrecently come into widespread use to refer to theway the partners agree on the nature of the task,set priorities, co-design action plans, and thentreat their implementation as a co-production. Insome schools, co-construction is also well-developed between students and teachers in theco-design of aspects of learning and is associatedwith the growth of mentoring and coachingamong students. Co-construction is the actiontaken to ensure ‘what works’ in specific contextswith particular people; it is about adapting andadjusting the practices of teaching and learning tosecure the promised outcomes.Co-construction does more than get results.Through its processes, social capital (trust andreciprocity) within and between schools is built upand then fostered by the extent and depth ofmentoring and coaching that is easier to achievewithin a family of schools. The enriched socialcapital generated by these organic relationshipsenables the member schools’ intellectual capital(knowledge and skill, core competences) to beexploited more fully. Schools that offer deepsupport to other schools, such as staff in nationalsupport schools working with their NLEheadteacher, repeatedly insist that they too havegained from the partnership. The activities of co-construction lead to the co-evolution of theschools as effective organisations.Family clusters provide the basic units of a SISS;the local solutions approach combined with co-construction provides its collaborative culture. Thecomplexities of school systems mean that many ofthe family benefits arising from schemes of schoolimprovement and professional development arebeing secured as a by-product of action with amore limited aim. It is opportune to consolidatewhat began as separate developments in a waythat reaps the benefits of clusters. But for this toamount to a SISS, its fourth building block iscritical. 10
  11. 11. Expanding system leadershipIn education, the term system leader, originally All the projects linked to clusters entail forms ofintroduced by Michael Fullan (2005) has now distributed leadership. Because professional work inattracted various definitions. They have in common clusters necessitates a system view and the threethree core features, all of which reflect a deep moral core features of system leadership noted above, itpurpose: should be recognised as system leadership now being distributed to all levels. Teachers are, from a value: a conviction that leaders should early in their professional development, being strive for the success of all schools and their progressively inducted into the knowledge and skills students, not just their own that will be required of system leaders at the higher levels. Individual professional development and a disposition to action: a commitment to organisational development are becoming work with other schools to help them to inextricably interwoven. Teaching and leading go become successful hand in hand and acting on this helps to build leadership capacity within and between schools in a frame of reference: understanding one’s the family. Unless the ideas and implications for role (as a person or institution) as a servant action of system leadership are widely diffused, the leader for the greater benefit of the teaching profession and its leaders will not take education service as a whole collective responsibility both for the success of all schools in the system and for ensuring theThe term is already expanding, despite being so new development of system leaders.and relatively little known or understood. Originallythe term was most often applied to headteachers In short, the College’s work on succession planningready to work with other schools in difficulties – thus and middle leaders has, along with parallelNLEs and LLEs. It is now applied more generally to developments elsewhere in the education service,heads working to support schools other than their created new structures and cultures that are leadingown and to school improvement partners (SIPs). many teachers, and especially senior school leaders,The College’s role in the development and provision to adopt a systemic perspective on their work and aof system leaders in England has been substantial commitment to system improvement. In the best(National College, 2006b; Carter & Sharpe, 2006; current practice, students too are adopting a systemColeman, 2008). view by offering support to students in schools other than their own. The leadership building blocks for aRecent and rapid changes in leadership SISS are already being put in place.development, including the College’s projects onsuccession planning and middle leadership, indicatethat the numbers of system leaders at headteacherlevel need to be increased and an understanding ofsystem leadership needs to be extended to staff atevery level. We need also to go beyond the need forsome very good schools to intervene in failingschools to a position where good schools can learnwith and from one another so that they becomegreat schools. Happily, the evidence is that manyheadteachers are interested in some kind of systemleader role; most teachers on leadership coursesnow want to take a big-picture or systemic view ofschooling, not merely a narrow preparation for aparticular role; and most do, or want to, spend timein schools other than their own. 11
  12. 12. Beyond the self-managing schoolFor the last quarter of the 20th century, a majortask for school leaders in England was thedevelopment of the self-managing school, and inthis England has led the way internationally. Asschools became more self-managing over sometwo decades, they were enabled to become moreself-improving – when they were well led. Today’ssystem leaders are a direct product of successfulleadership of self-managing schools. A major taskfor school leaders in the first quarter of the 21stcentury may be the development of the self-managing school system. Achieving this status islikely to be a precondition of becoming a self-improving system.Central to the success of such a mission would bean increased capacity of schools to improvethemselves. For many years, the process ofschool improvement was led, even determined, bycentral and local government intervention,because most schools had not reached the levelof self-management to be able to move to self-improvement. Today’s outstanding school leaders,who masterminded the powerful co-evolution ofself-management and self-improvement, haveoften become NLEs or LLEs helping other schoolsat the same time as becoming the entrepreneurialleaders of established, longer-term family clustersof schools, which is a new organisational form. A SISS depends on the creation of familyclusters, but ones of the right kind will notemerge unless they are led, initially at least, bythe headteachers of highly successful schoolswilling to be system leaders. What more, then,needs to be done, by the College and by otheragencies, to pave the way for a self-managingand self-improving school system and to effectthe transition from where we are now to a robustand sustainable SISS? 12
  13. 13. Conditions of a sustainable self-improvingsystemTo create a sustainable SISS, three key questions above, few schools will opt for either extremeneed to be answered. position – the near-permanent tie of federations or trusts at the tight end and the shallow What sorts of family clusters are commitments at the loose end – preferring the needed in a SISS and what action is flexibility and moderate constraints of more needed to create them to scale? central positions.Many of the clusters in National College schemes Most clusters will be geographically local, sinceare not newly formed, but based on existing two key features of close collaboration are ease ofclusters, such as SCITT and EBITT (school- face-to-face contact and mobility of staff andcentred and employment-based initial teacher students. Some existing clusters have memberstraining respectively) clusters, as well as some distance apart, even in different localfederations and trusts. Collaborative clusters have authorities, which have boundaries that are oftenover the years taken many forms: some senior arbitrary. Distant family clusters might laterstaff retain fond memories of TVEI collaboration dissolve, with each member starting a new localin the 1980s. cluster. Whilst local clusters will probably become the dominant type, some of which will span localWhat types of family cluster are likely to populate authority boundaries, other types, for instancea SISS? The most common, I suspect, will be a not-so-local clusters of faith schools, may thrive.homogeneous family cluster, either from the samephase (eg, a cluster of primary schools) or same Some headteachers, and even more governingfaith (eg, a group of Catholic schools). Most bodies, are wary or even sceptical about familiescurrent ones are of this type. of schools. Indeed, some governors find it difficult to think beyond the individual school that theyThere are also heterogeneous family clusters, for may have loyally served over many years, and soexample comprising one secondary school with its are more resistant to new partnerships than theirfeeder primaries, and perhaps a special school. own headteacher. Much the same may be said ofThese are particularly suitable for rural areas, parents, only a small minority of whom havewhere the single secondary school’s intake comes experience of a family cluster. But clusters cannotmainly from local primary schools. be imposed on unwilling schools: that would undermine a SISS. It would be essential toMixed family clusters are a third type, for instance harness the support of headteachers, governorswhen a cluster of maintained sector schools and parents by making them more aware of theincludes just one faith school or an independent many benefits of family clusters. Some start-upschool, or when schools of different faiths, such additional funding might be a necessary incentiveas Christian, Jewish and Muslim schools, form a until the benefits, including cost-saving ones, aremixed-faith family. recognised. Schools in mature family clusters happily pay into the cluster as a recognisedSome existing federations, of either schools or investment (Hill, 2010).academies, were formed with a business, charityor academic sponsor. In the case of what arepopularly called hard federations, where thegoverning bodies of more than one schoolamalgamate, the ties could be difficult to dissolve.Some family clusters created by NLEs started, asit were, as an equivalent to a merger oracquisition in the business world. I suspect thatmost family clusters in a SISS will be brokeredand essentially voluntary relationships, with amore flexible, less permanent tie than that of thehard federation or trust. In terms of the loose–tight continuum of partnership I proposed 13
  14. 14. Not all clusters would be newly formed: many with the College to broker clusters, then supportalready exist under a variety of names reflecting and monitor their self-improvement, in place ofdifferent purposes and origins, including some direct provision for school improvement.recent ones, such as national challenge trustschools and federations, and accredited school Will there be enough system leadersgroups or providers. New clusters may well arise to take to scale the number of clustersfrom policies adopted by the new government. to make effective family clustersSome outstanding primary schools will be sustainable?reluctant to lose the support their local authorityoffers by becoming a lone academy, but might NLEs, concludes a recent review (Hill & Matthews,choose to become one within a self-managing 2010), are in the vanguard of transformingfamily cluster with shared administrative support. England’s education system, and:This would reduce back-office costs and minimisethe burden on individual headteachers. As they “the successful recruitment,work with underachieving schools, a family of deployment and expansion of a cadreacademies would become self-improving. of schools capable of sharing their excellence with other schools and,Some rationalisation of clusters could forestall where necessary, taking over andunnecessary overlap and undesirable rescuing failing institutions, introducesbureaucracy. It is possible to start with small a powerful lever for change into theclusters of three or four schools that could, with school system. By showing that theyexperience, expand into larger families. can bring about change in the most intractably underperforming orFor the system to become self-improving, it is not challenging schools, NLEs havenecessary for every school to join a cluster. demonstrated their capacity as agentsFreestanding schools can, as now, be self- of change. They and their schools relishevaluating and self-improving units. Indeed, this such work; their governors areis how many schools have achieved outstanding persuaded of the mutual benefits; andstatus. There may be good reasons why a school tens of thousands of children andshould not join a cluster and could continue as at young people are getting a better dealpresent either within a local authority or as an as a result.”academy, free school or trust. A balance wouldneed to be struck between offering incentives to Hill & Matthews, 2010:116schools to join clusters and acknowledging thatthis would not always be the right way for some Capturing the knowledge and skills of theseschools. exceptional pioneers of system leadership and transferring it to leaders who follow in the wake ofWere many, even most, schools to join family the trailblazers is now the task of the College.clusters, this would herald changes for localauthorities and their relationships with schools. As The leaders of outstanding schools fall into twoschools became self-managing, they became less categories: those who want to be or have becomereliant on the local authority: the transition was system leaders working with other schools, andnot always easy. As schools become self- those with little interest in system leadership.improving, the transition will again be one that Among the latter are those who have simply notlocal authorities must decide to support or resist. been given or have not availed themselves ofHitherto, the local authority has been the middle opportunities for system leadership and thosetier between central government and the who may fear that a close partnership with otherindividual school, but clusters are now an schools will jeopardise their achievements andemergent kind of middle tier. Some local reputation. The task is to persuade many of thisauthorities have been active in the promotion of group to join the former group, as could happenclusters: they are well-placed to phase out their in the new government’s policy that outstandingown school improvement arm and transfer self- schools may become academies on the conditionimprovement responsibility and activity to family that they work with at least one other school.clusters. In a SISS, the local authority would work 14
  15. 15. The skills of leading a successful school and the In the business world some partnerships workskills of helping another school to become equally and others fail, and it will be much the same withsuccessful are not, however, coterminous. This schools. Recent research suggests three corewas the mistake made with Beacon schools, features of inter-firm partnership competence:introduced in 1998. Some schools knew how tomake effective partnerships with other schools, co-ordination: building consensus onand improved their skills in so doing. But others partnership goals, ways of working, rolesdid not. It takes talent to be a successful head, of and responsibilitiescourse, but that talent is not enough for thehighest forms of system leadership. (As some communication: being open andheadteachers in their second headship know to honest, sharing information fully and withtheir cost, having run one school successfully accuracy and in a timely waydoes not in itself guarantee one can replicate thisin a different school.) For a school to achieve an bonding: creating trust and ensuringOfsted grade of ‘good’ or ‘outstanding’, the that people get pleasure from workingheadteacher must be expert at what in the togetherbusiness world would be called the corecompetences 2 that underpin success. In The distinctive constituents of partnershipschooling, the most critical core competences are: competence in education are becoming clear. Among them I would include: the relentless focus on learning and teaching, and the conviction that the best the conception of the school as a learning teaching and learning yield high community, with the expansion of school- examination and test results and rounded based CPD to embrace mentoring and persons with the right qualities for a coaching, teachers’ observations of one successful life in the 21st century another at work and the co-construction of better professional practice ensuring order, attendance and good behaviour as a precondition of the investment in innovation in teaching improvement in learning and teaching and learning (‘doing things differently in order to do them better’), arising fromKnowing how to lead a high-performing school is CPD that is school-based and classroom-a necessary but not always a sufficient condition focused, with ‘learning as the bridgeof knowing how to help another school to between working and innovating’ (Brownsucceed. NLEs and LLEs and similar cluster & Duguid, 1991) so that improvementleaders are successful in what they do because becomes an inherent part of teacherthey possess some additional competences. In the professionalismbusiness world, the similar notion of alliancecapabilities and how to develop them has been distributed leadership, with an emphasiswell-researched 3, but its equivalent in the on preparing leaders at every level,education field is at a much lower level. The including pupils, by identifying talent andability to forge partnerships with other schools is empowering the taking of responsibilitynot yet one of the core competences of highly and initiativesuccessful schools. Creating a SISS entailsensuring that what one might call partnershipcompetence becomes a core competence of allheadteachers. 2 Hamel & Prahalad (1994) define a core competence as ‘a bundle of skills and technologies that enables a company to provide a particular benefit to customers.’ 3 Robert Spekman & Lynn Isabella, 2000, Alliance Competence, John Wiley; James Austin, 2000, The Collaboration Challenge, Jossey- Bass; Deering & Murphy, op.cit; Koen Heimericks, 2008, Developing Alliance Capabilities, Palgrave Macmillan, where alliance capability is defined as an organisation’s ‘ability to capture, share, disseminate, internalise and apply alliance management know-how and know-why.’ 15
  16. 16. the recognition that working with another spending time understanding the culture school is a reciprocal process, because and working methods of partners and there is always something to be learned using differences as a spur to learning both from and with others rather than conflict local knowledge and the ability to adapt having open communication between whatever arises, from necessity or partners, covering performance data and, preference, to the immediate context, as they arise, differences and changing with its distinctive history and culture circumstancesThe first three are prerequisites for the whole developing strong links betweenstaff of the lead school, not just the headteacher, organisations at all levels so thatto engage professionally with the whole staff of a partnership is supported by a dense webpartner school; the fourth removes the suspicion of interpersonal connectionsby staff in the partner school that they are beingtreated as inferiors without any worthwhile spending as much time on building upqualities and something is being done to them, commitment to collaborative activitynot with them; and the last is the sensitivity to within an organisation as on buildingcontext, including the personalities and cultures in relationships with partnerspartner schools, to support making the rightdecisions in the right way at the right time. using interim or input measures to assess a partnership’s early progress before theSystem leaders build such competences into their full value of a partnership comes throughown school, so they are well-placed to transferthem to a less successful one. It is never just a agreeing a clear status and remit andmatter of what the headteacher does. It is decision processes for the collaborationbecause the rest of the staff, some of whom mayhave more experience and insight than their It is system leaders with partnership competenceheadteacher into the art of working laterally who should take the lead in family clusters.rather than vertically, are also able to transfer Without them, teachers may busily share goodtheir values and ways of working – their shared practice, that is, talk about what they do, butpartnership competence – to another school in without any significant change in their practice. Astheir cluster. A precondition of being able to Michael Huberman (1995) puts it, ‘There is atransfer professional knowledge and skills into “discussion culture” among teachers…another school is the honed experience of so interspersed with timid attempts at the level ofdoing in one’s own school. This is precisely what actual implementation… To get from a peerdistributed system leadership means in practice. discussion to its enactment in one’s classroom is aThe exemplary school has to replicate its culture, phenomenal leap.’ What we are after isnot just some of its practices, in the less knowledge transfer, by which I mean that asuccessful one if the relationship is to be teacher successfully puts into practice somethingtransformative on a sustainable basis. new that has been learned from another teacher.Drawing on the business literature as well asdirectly from school partnerships, Robert Hill(2008; 2004) has provided a high-quality guidefor school leaders. Key lessons include: ensuring that collaboration fits with the objectives of all the partners so that everyone involved wants to make the partnership work 16
  17. 17. This is not a simple, straightforward business 4, Without nodal system leaders at its heart, there is but conditions that facilitate it obtain in schools a serious risk that a cluster-based system would that distribute system leadership and work in a merely recycle mediocrity and affirm cluster. Huberman’s model for effective complacency, or would simply collapse because knowledge transfer, which he calls the ‘open the headteachers lack partnership competence collective cycle’, is based on two premises: the and/or there is too little pay-off from the group of teachers shares the same subject or partnership. Embedding the ambition, drive and discipline but its members come from different know-how for deep school improvement in schools; the cycle of professional development is clusters is vital for a robust and sustainable SISS. managed by this group, not a consultant. Family Many clusters contain good schools that have the clusters are the ideal location for Huberman’s potential to be outstanding. The Good to Great approach. (G2G) programme in London (Matthews & McLaughlin, 2010) has demonstrated that with It will take some time to generate a the right kind of inter-school support, based on comprehensive, evidence-based understanding of mentoring and coaching, good schools can indeed partnership competence within clusters of schools. be raised to the level of outstanding, and that In the interim, school leaders can profit from must be one of the purposes of family clusters. experience in the business sector 5, where there is extensive advice on the selection of partners and Ideally for a SISS, then, there should be enough making the partnership effective. One notion nodal system leaders to ensure one per cluster. At education could usefully borrow from business this stage, of course, it is impossible to estimate comes from Hamel and Prahalad (1994). In their how many clusters of what size might arise over study of business coalitions, they talk about a what timescale 6. The College would need to nodal company at the centre of the network or determine the knowledge, skills and experience coalition with a large share of influence within it: appropriate for a nodal system leader and then recruit and prepare them to the desired scale. “Nodal firms must accept that all coalition This is an intellectual and practical challenge for partners may not have the same level of the College, but one of its most urgent tasks. The commitment to the concept. Partners College should use its current experience of exhibit a wide variety of interests and brokering and supporting clusters, including its varying levels of commitment… Nodal firms new approaches to succession planning and need to have this perspective and middle leadership development as well as the understanding in order to manage each experience gained from City Challenge, to provide partner appropriately… Influence within the written guidance and training on best practice for coalition comes from an ability to recognize self-improving clusters. and then exploit, or redirect and frustrate, the differing agendas and concepts of self- interest possessed by the various partners… Any one company’s capacity to motivate, direct and manage the coalition derives not from legal control and unilateral dependency, but from political skills, possession of critical competences, a clearly articulated and inspiring point of view about the future, and a track record of honouring commitments to partners.” Hamel & Prahalad, 19944 Mark Ackerman, Volmar Pipek & Volker Wulf, 2003, Sharing Expertise, MIT Press; Gabriel Szulanski, 2003, Sticky Knowledge, Sage; PaulHildreth & Chris Kimble, 2004, Knowledge Networks, Idea Group Publishing.5 See footnote 3 above.6 By May 2010 the College had recruited, designated and deployed 431 NLEs and 825 LLEs. At the same time there are 5,000 accreditedSIPs and 600 professional partners (experienced headteachers who provide structured mentoring and support to newly appointedheads). 17
  18. 18. Is this task easier than it was with the first system demonstrating leading-edge pedagogyleaders? Richard Elmore (2008) argues that onceteachers adopt a system perspective, they treat exemplifying high-quality assessment fortheir knowledge and skills not as professional learningprivate property, but as a collective good thatshould be shared with colleagues. He also providing teams of experts (eg, advancedquestions the common view of leaders as people skills teachers) and enquiry and researchwith highly unusual personal attributes who are, teamsalmost by definition, in a permanent minority. Inplace of this ‘essentialist’ definition of leadership, The extensive professional development of middlehe argues that what outstanding leaders have is leaders and aspirant headteachers in recentknowledge and skill of particular kinds. When years, as well as the College’s work on successionsuch leaders first emerge, they are indeed planning, is producing a cohort of betterunusual, as was the case with NLEs. The prepared, system-orientated leaders at just belowimperative task is to separate what they do headteacher level, some of whom will be ready(leadership) from who they are (leaders) so that for more active system leadership in the neartheir practices can be captured, codified and future. As elements of system leadership becometaught to others, often through mentoring and an inherent part of the training of leaders at levelscoaching. The task, in other words, is to ensure other than headteacher – middle leaders whothe wider distribution of partnerships competence, have day-to-day responsibility for much inter-of which we now have a better grasp, through an school mentoring and coaching, and the facilitatorexpanded cohort of system leaders. The College roles in the middle leadership developmentand its partners have to prepare system leaders programme – headteacher leaders of clustersslightly ahead of the pace at which family clusters have a higher level of human capital on which toof schools are formed so that every family can rely. Indeed, in best current practice leadershipinitially be led by a prepared system leader. At the talent is being spotted and developed in initialsame time, many clusters do not contain teacher training and early years within themembers who are in, or close to, special profession. It is now possible for the mostmeasures, and so the high-level skills of NLEs and talented to become an assistant headteacherLLEs would not always be so critical to self- within three years.improving cluster success. Hitherto much leadership development has beenThe pool from which system leaders can be to increase the organisational capacity ofrecruited and trained is itself being enlarged. New autonomous schools. All the above demonstratesforms of school-based leadership development are how the College’s most recent focus on systema rich breeding ground for future system leaders, leadership is geared to enhance the systemsuch as the facilitators in the College’s middle capacity on which a self-improving systemleadership development clusters and those in key critically depends. The way forward for theroles in the national teaching schools, which have College has two obvious elements:devised ways of: scaling up the recruitment and training of engaging in effective school-to-school system leaders support at middle leadership level that focuses on building sustainable bringing together into a more coherent leadership capacity whole its wealth of experience of relevant work at various levels of leadership other developing expert practitioners able to than the headteacher, for here lies the deliver high-quality coaching and essential complementary support that teaching programmes would make self-improving clusters effective providing professional development in leading teaching and learning 18
  19. 19. What inevitably began as separate strands of theCollege’s work at different levels, often indifferent locations, and led by different people,should now be aggregated into a compellingvision of a SISS, with a consolidation of recentprojects and a specification of the further workneeded to realise the vision. How can the process of self- improvement be assured?Although both schools and the inspection systemhave in recent times increased the importance ofschool self-evaluation, guidance has mainly beendirected to the individual school rather than afamily of schools, with the exceptions of aNational College publication (NCSL, 2006c) andthe TDA’s development of a benchmarking tool foreffective practice in CPD clusters. At present,many headteachers complain in vain that Ofstedignores cluster membership, even when itevidently contributes to better teaching andlearning. Ofsted should formally assess the qualitynot only of the individual school, but also of thecluster of which it is a member, including theextent to which the cluster realises the familybenefits noted above. Ideally, Ofsted would alsoreport on student performance and progress atthe cluster level as well as at individual schoollevel, which would show how family clusters raisethe achievement of all students. Clusters need tobe accountable for what they do and for theiradded value.The most detailed knowledge about what makesan effective family cluster, including how well itmanages continuous self-improvement, is likely tocome from those who take the lead in brokeringand supporting such partnerships, namely theCollege and the local authority. All three parties ofCollege, local authorities and Ofsted play a role inidentifying dysfunctional clusters (for some wouldundoubtedly come to light) and in shapingremedial action. As knowledge of what makeseffective nodal school leaders and self-improvingclusters sharpens, and it becomes easier toidentify them, there is considerable scope forexemplary clusters to work laterally to supportnewly-formed clusters and those in difficulties. 19
  20. 20. Towards a mature self-improving systemOriginally the College’s task was to ensure the Over a five-year period, the Collegecontinued supply of school leaders. As the recruited, prepared and designatedconcept of school leaders has broadened and the sufficient system leaders for nodalscale of provision has grown, the College’s focus schools and, with the local authorities,has turned to problem prevention, creating sensitively brokered clustersystem-based means by which schools and local arrangements. Many schools nowauthorities take ownership of leadership belong to two clusters:development and devise sustainable ways ofidentifying, preparing and supporting leaders at a homogeneous family of schools ofmany different levels. Leadership development is the same phase/type (primary,now conceptualised as a progressive trajectory secondary, special, faith etc) tosupported throughout a teaching career. As in ensure improvement of phase-turn the teaching profession progressively specific mattersdevelops a more sophisticated conception ofleadership and its role in school improvement, the a heterogeneous family of mixedbuilding blocks of a self-improving system are put phases/types, the most common ofin place, ready for consolidation, expansion and which would be a geographically localfurther development. mix of primary, secondary and special schools, at the heart of whichAn explicit intention to move to a SISS over the is a secondary school with its feedernext five years would be an ambitious but primary schoolsattainable goal, fully in line with Secretary ofState Michael Gove’s declaration at the College’s In both of each school’s clusters, thenational conference in June 2010 that ‘At the content and timing of professionalheart of this government’s vision for education is development are aligned acrossa determination to give school leaders more member schools so that closepower and control; not just to drive improvement collaboration is common. Staff andin their own schools, but to drive improvement students move between schools inacross our whole education system.’ accordance with needs and opportunities. A few schools are in veryKey policy decisions to help progress towards a tight clusters in the form of federationsSISS would include: or chains, with an executive headteacher. At the other extreme, confirmation of the College’s direction of some schools have chosen not to travel on leadership development, local belong to a cluster. Whilst they mainly solutions and school-to-school support stand alone, they network with clusters as the need or inclination arises. Most a requirement that the College family clusters fall between these strengthen the building blocks of a SISS, extremes. As headteachers change, so especially the provision of system leaders do some cluster arrangements. It is the and leaders of family clusters to greater voluntary membership and flexible ties scale, as well as written guidance on best that make the family cluster so practice for self-improving clusters attractive and effective. the support for new roles and The NLEs, whose specialist skills of responsibilities for the key agencies (the working with failing schools are not College, TDA, local authorities and needed by most nodal school leaders, Ofsted) continue their work. Where a failing school has not been a member of aWhat, then, would the landscape of a maturing cluster, joining one and developingself-improving school system look like? A short partnership skills is part of thespeculation is in order: what could be achieved by remediation.2015? 20
  21. 21. The emergent range of patterns is support, clusters either failed toconsiderable and at first sight looks collaborate in sufficient depth or simplychaotic when compared with the collapsed. In a few cases, the Collegeisolated schools of the old local had to de-designate a nodal school andeducation authorities with their strict its leader.boundaries. This is a natural andinevitable consequence of local One example of the impact of the newsolutions: different kinds of cluster are arrangements on teaching and learningappropriate to different areas, urban is the provision for middle leaders inversus rural, and to local contexts and primary and secondary schools. Eachcultures. Moreover, some local cluster runs a middle leadership courseauthorities responded positively to their on a regular basis: every middle leaderbrokering role, but others did not, has, as an entitlement, access to suchespecially where clusters straddled local a course, which builds on work underauthority boundaries, necessitating way in each cluster’s routineintervention from the Department for professional development. As part ofEducation. As in chaos and complexity the course, middle leaders engage intheory, however, below the surface is a development tasks that contribute tonew kind of order in which schools the improvement plans of their ownworking together in networks have school as well to those of the otheraligned their continuing professional schools in the cluster. The impact ofdevelopment and their leadership the course on teaching and learning isdevelopment, and woven these into a key criterion of course effectiveness.their school development and The strongest impact has been onimprovement plans, both for each secondary school subject departmentsschool and for other schools in the of just one or two teachers, where thecluster. ability to work with their equivalents in other schools has come as a boon.It was quickly realised that the majority Middle leaders now grow fast in theirof school leaders lacked in-depth understanding of, and contribution to,experience of working with other system leadership.schools, and their partnershipcompetence was over-estimated: A second example is the revisedenthusiasm outpaced skill. Progress in implementation of assessment forcluster formation and development was learning (AfL), an evidence-basedaccelerated in two ways. First, as the approach to enhancing teacher actionCollege prepared more leaders of nodal in classrooms that raises test scoresschools, experienced clusters and their and students’ meta-cognitive skills. Theleaders at various levels worked conventional way to induct teacherslaterally with new clusters. Some into AfL was by teacher attendance at aschools in well-established clusters left day’s course. Although teachers weretheir cluster to start a new one. strongly attracted to the ideas, inSecond, local authorities provided practice they found them simply toofacilitators for cluster development or difficult to implement. Under the familyoutsourced the task to those with the cluster system, one school took severalnecessary expertise – the College’s role months to embed AfL throughbeing to quality assure and accredit sustained professional development,training programmes. Mentoring and and then used its expertise to transfercoaching has not been a normal part of the practice to other schools in theteacher training, and only with help family through mentoring andcould many teachers develop the coaching. What was once a widespreadconfidence and skills to ensure the failure has now become a successfultransfer of best practice between model of effective professionalschools. In the absence of such development. 21
  22. 22. At the leading edge of suchdevelopments are clusters reaching ahigh level of maturity: they are reapingin full the benefits of familyarrangements. Criteria for clustereffectiveness have been devised anddisseminated. More schools areattracted to join clusters as the benefitsbecome better known.The key metrics for the effectiveness ofcluster arrangements are theproportions of schools in difficulties orspecial measures and those ratedoutstanding by Ofsted. Clusters havedemonstrated their power by thepreventative action that has reducedthe number falling into difficulties and,where this has happened, by speedingrecovery. The number of schools ratedoutstanding has risen. Most importantlyof all, student performance asmeasured by examination and testresults is again rising steadily: theplateau effect of the previous era ofschool improvement has beenovercome in the new era of systemicself-improvement. 22
  23. 23. ConclusionIt has long been known that the most powerfulinfluences on teachers are other teachers, butpolicies have rarely built on the fact. The best wayof exploiting this phenomenon is through regular,face-to-face encounters among professionals thatfocus on the improvement of teaching andlearning. Under the direction of system leaders,clusters of schools are the simplest way ofmaximising inter-school professional developmentas the main driver of a SISS. Once established, aSISS potentially reduces the need for extensivebureaucratic, top-down systems of monitoring tocheck on school quality, the imposition ofimprovement strategies that are relativelyinsensitive to local context, with out-of-school in-service courses not tailored to individualprofessional needs, and external, last-ditchinterventions to remedy schools in difficulties, allof which are very costly and often only partiallysuccessful. In a self-improving school system,more control and responsibility passes to the locallevel in a spirit of mutual aid between schoolleaders and their colleagues, who are morallycommitted to imaginative and sustainable ways ofachieving more ambitious and better outcomes.England is part way there. Will it now decide totravel the rest of the journey? 23
  24. 24. AcknowledgementsI am most grateful to Toby Greany, Maggie Farrarand Katy Emmerson, and many of their NationalCollege colleagues, as well as to those whoattended invited seminars, for their constructivecomments and suggestions on earlier drafts ofthis paper.For further information about other researchavailable from the National College please 24
  25. 25. ReferencesBrown, J S & Duguid, P, 1991, Organizational learning Huberman, M, 1995, Networks that alter teaching:and communities of practice: toward a unified view of conceptualizations, exchanges and experiments,working, learning and innovation, Organizational Teachers and teaching: theory and practice, 1(2)Science, 2(1) Matthews, P & McLaughlin, C, 2010, Up for it?Bunt, L & Harris, M, 2010, Mass Localism, London, Evaluation of the London Leadership Strategy Good toNational Endowment for Science, Technology and the Great Programme, Nottingham, National College forArts (NESTA) Leadership of Schools and Children’s ServicesCaldwell, B & Spinks, J, 1988, The self-managing National College, 2006a, Networked learningschool, London, Falmer Press publications directory, Nottingham, National College for School Leadership. Available atCarter, K & Sharpe, T, 2006, School leaders leading the system leadership in perspective, Nottingham, e=learning-networks-publications-directory.pdfNational College for School Leadership. Available at [accessed 5 August 2010] [ National College, 2006b, System leadership in action,accessed 5 August 2010] Nottingham, National College for School LeadershipColeman, A, 2008, Trust in collaborative working: the National College, 2006c, Getting started with networkedimportance of trust for leaders of school-based school self-evaluation, Nottingham, National College forpartnerships, Nottingham, National College for School School LeadershipLeadership. Available [accessed 5August 2010]Collins, J, 2009, How the Mighty Fall, London, RandomHouseDeering, A & Murphy, A, 2003, The PartneringImperative, London, John WileyElmore, R, 2008, Leadership as the practice ofimprovement, in B Pont, D Nusche & D Hopkins (eds),Improving School Leadership, 2, Paris, OECDFullan, M, 2005, Leadership and Sustainability, NewYork, Corwin PressHamel, G & Prahalad, C K, 1994, Competing for theFuture, Boston, MA, Harvard Business School PressHill, R, 2010, Chain reactions: a thinkpiece on thedevelopment of chains of schools in the English schoolsystem, Nottingham, National College for Leadership ofSchools and Children’s Services. Available [accessed 5 August 2010]Hill, R, 2008, Achieving more together: adding valuethrough partnership, London, Association of School andCollege LeadersHill, R, 2004, What does a network leader do?Nottingham, National College for School LeadershipHill, R & Matthews, P, 2010, Schools leading schools II:the growing impact of National Leaders of Education,National College for Leadership of Schools andChildren’s Services. Available [accessed 5 August 2010] 25
  26. 26. The National College is the first ©2010 National College for Leadership of Schoolsprofessional body uniquely dedicated and Children’s Services – All rights reserved. No part of this document may be reproducedto developing and supporting aspiring without prior permission from the National College.and serving leaders in schools, early To reuse this material, please contact the Marketingyears settings and children’s services. and Communications Team at the National College or email College gives its members theprofessional development and recognitionthey need to build their careers andsupport those they work with. Membersare part of a community of thousands ofother leaders – exchanging ideas, sharinggood practice and working together tomake a bigger difference for childrenand young people.Triumph RoadNottingham NG8 1DHT 0845 609 0009F 0115 872 2001 PB741E