Alma Harris is Professor of School Leadership and Director of the Institute of Education, University of Warwick, Coventry, CV4 7AL, UK; e-mail: alma.harris@warwick.ac.uk. Her research work encompasses organizational change, school improvement, and leadership. She is currently engaged in research exploring distributed leadership in practice in schools in England.

 JCS invites comments on this paper for publication on the journal's web site. Address comments to Ian Westbury. All such comments on this paper, and on other papers in the journal, can be accessed at the JCS website.

© 2005 Taylor & Francis Ltd. ISSN 0022-0272. Copies may be made under the normal terms of copyright law.  

The distributed perspective offers a new and important theoretical lens through which leadership practice in schools can be reconfigured and re-conceptualized. But contemporary discourse about distributed leadership theory remains a way of analysing rather than describing leadership practice. While the research suggests that distributed forms of leadership can assist capacity-building within schools, further work is needed to investigate the nature of the relationship between distributed leadership and improved school/student outcomes.

 

Despite the current fervour and enthusiasm for leadership and leadership development in education, there is still much that is not yet understood about the complex relationship between school leadership and school improvement. It is evident that a relationship exists, but its exact nature is still unclear. If school improvement is taken to be the 'mobilization of knowledge, skill, incentives, resources and capacities within schools and school systems to increase student learning' (Elmore 2000: 13), then the research literature consistently points towards leadership as a key factor in this process. But the empirical base connecting leadership to school and student outcomes is less clear about how leadership results in school and student improvement. In their recent analysis of the literature, Bush and Glover (2003: 12) identify eight models of leadership that provide a starting point for a normative assessment of school leadership, but they also point out the limited empirical support for these constructs and their relationship with school or student outcomes. Most recently in their systematic review of the leadership literature, Bell et al. (2003) highlight the limited number of studies that have investigated the relationship between leadership practice and student-level outcomes. In summary, much of the school leadership literature has tended to concern itself with the traits and characteristics of principals instead of probing the nature of the relationship between leadership and organizational change and development. It has assumed correlation or causation without the adequate empirical basis on which to rest such claims.

It is also questionable whether 'leadership' is the correct label or descriptor for the type of activity or influence that is considered to drive organizational change. Lakomski's (2004) recent work challenges the premise that leadership is a natural entity or essence within an organization, proposing instead that leadership is a distraction from exploring the real workings of organizational practice. She calls into question whether our 'taken for granted understanding of leadership ... squares with how leaders and organizations really work given what we know about human cognition and information processing' (Lakomski 2004: 139). Her work has consistently questioned how 'leadership' makes a difference when it would appear that much leadership research has produced inconclusive and sometimes conflicting results. In their review of the literature, Hallinger and Heck (1996) reinforce the limitations of the current research base by identifying certain 'blank spots' (i.e. shortcomings in the research) and 'blind spots' (i.e. areas that have been overlooked because of theoretical and epistemological biases) within the leadership field. An important blank spot centres on exactly what form or forms of leadership practice contributes to sustained school improvement.

Looking at the school improvement field it would appear that the most recent studies point towards the importance of capacity-building as a means of generating and sustaining school improvement (e.g. Fullan 2001, Harris and Lambert 2003, Hopkins and Jackson 2003). At the core of the capacity-building model, it has been argued, is 'distributed leadership along with social cohesion and trust' (Hopkins and Jackson 2003: 95). Leadership, from this perspective, resides in the human potential available to be released within an organization; it is what Gronn (2000) terms 'an emergent property of a group or network of individuals in which group members pool their expertise'. Implicit within the distributed leadership model are the leadership practices of teachers, either as informal leaders or in a formal leadership role as a head of department, subject co-ordinator, or teacher mentor (Muijs and Harris 2003). As Leithwood and Reil (2003: 3) note 'research suggests that teacher leaders can help other teachers to embrace goals, to understand the changes that are needed to strengthen teaching and learning and to work towards improvement'. The implication from the current school improvement literature is that distributed leadership can assist schools in building the internal capacity for development.

The literature is less clear however on the exact form that distributed leadership takes. Bennett et al. (2003: 4) note there are almost no empirical studies of distributed leadership in action; hence contemporary studies of distributed leadership are not readily available and 'operational images' of distributed leadership remain rare (Hopkins and Jackson 2003). Conversely, descriptions or accounts of distributed leadership practice can be found in the broader research literatures concerning school improvement and teacher leadership. Both these literatures reinforce the centrality of distributed leadership in schools that improve, but the studies in both fields have not exclusively focused upon distributed leadership practice. This paper aims to explore, in a tentative sense, the relationship between distributed leadership and school improvement by, first, examining the concept of distributed leadership and, second, considering the empirical evidence that indicates a relationship between distributed leadership and school improvement. I highlight some of the practical implications of distributed leadership in action and conclude by reiterating the need for further research that investigates the relationship between distributed forms of leadership upon school and student outcomes.

 Distributed leadership: the theoretical case

Despite widespread contemporary interest in distributed forms of leadership, the concept itself is far from new. As far back as the late 1950s, in the first edition of the Handbook of Social Psychology, Gibb (1954) posed the possibility of leadership displaying a distributed pattern or configuration. In the 1970s, this idea was revisited by organizational theorists and researchers who paid increased attention to models of situated cognition and the inherent patterns of distribution this theory implied. Their work highlighted the difficulty of separating action from the context of action, suggesting instead that the situation is critical in constituting leadership practice. This view of 'distributed leadership' assigns a central role to the relationship between agency and structure, where structure is the medium of human interaction. It implies that to understand the human situation necessitates exploring how structure and agency interact together to construct practice, including leadership practice.

In theoretical terms, Gronn (2000) and Spillane et al. (2004) are leading the current debate on distributed leadership. Gronn (2003) clearly sees activity theory as the centrepiece of his analysis. In activity theory, the notion of activity bridges the gap between agency and structure. In Giddens' (1984) sociological theory of action, social or organizational structures can be modified by the agency of individuals by using whatever power resources are to hand. In activity theory, leadership is more of a collective phenomenon. As Gronn (2000: 331) puts it, 'the potential for leadership is present in the flow of activities in which a set of organization members find themselves enmeshed'.

Explanations based on activity theory are particularly applicable to professional contexts such as schools, because most conceptions of professionalism include the idea of autonomous judgement. In his work Gronn (2000) suggests that distributed leadership is an emergent property of a group or a network of interacting individuals. Here leadership is a form of concerted action which locates the additional dynamic that occurs when people work together, or that is the product of conjoint agency.

Spillane et al. (2003, 2004) similarly imply a way of understanding leadership that focuses upon interaction and the exploration of complex social processes. In this sense, leadership is best understood as 'practice distributed over leaders, followers and their situation and incorporates the activities of multiple groups of individuals' (Spillane 2004). It implies a social distribution of leadership wherein the leadership function is stretched over the work of a number of individuals and the task is accomplished through the interaction of multiple leaders. It also implies inter-dependency rather than dependency, embracing how leaders of various kinds and in various roles share responsibility.

Empirically, the work of Spillane et al. (2004) rests upon a qualitative study of 13 elementary schools in Chicago. The analysis is based upon vignettes drawn from their research that offer support for their argument that leadership practice has to be analysed in relation to the task and what they call 'the artifacts that represent in reified form the problem-solving initiatives of previous human action'. In contrast, Gronn's (2000) analysis of distributed is a purely theoretical exploration that draws upon a re-analysis of a range of other studies. He develops distributed leadership as an analytical tool for understanding leadership in practice as a form of concertive action rather than as a way of describing distributed leadership in practice.

In theoretical terms distributed leadership therefore means multiple sources of guidance and direction, following the contours of expertise in an organization, made coherent through a common culture. 'It is the "glue" of a common task or goal-improvement of instruction -- and a common frame of values for how to approach that task' (Elmore 2000: 15). The distributed perspective focuses on how leadership practice is distributed among formal and informal leaders (Harris 2002). As Bennett et al. (2003: 3) note 'distributed leadership is not something "done" by an individual "to others" ... rather it is emergent property of a group or network of individuals in which group members pool their expertise'. Distributed leadership incorporates the activities of many individuals in a school who work at mobilizing and guiding other teachers in the process of instructional change (Spillane et al. 2004). It extends the boundaries of leadership significantly as it is premised upon high levels of 'teacher involvement' and encompasses a wide variety of 'expertise, skill and input' (Harris and Lambert 2003: 16). Engaging many people in leadership activity is at the core of distributed leadership in action. Hopkins and Jackson (2003: 99) suggest it is where 'leadership and organizational growth collide and by definition, it is dispersed or distributed'.

Elmore (2000:14) suggests that the basic idea of distributed leadership is not very complicated. As he notes,

in any organized system, competency varies considerably among people in similar roles; ... organizing these diverse competencies into a coherent whole requires understanding how individuals vary ... and how the competence of some can be shared with others. (p. 15)

Elmore argues that in a knowledge-intensive enterprise, complex tasks like teaching and learning cannot be achieved without distributing the responsibility for leadership, i.e. guidance and direction. In this sense distributed leadership is essentially concerned with harnessing and enhancing the skills and knowledge of all those within an organization to create a common culture that functions positively and effectively. There is a wide range of empirical evidence that reinforces the importance of distributed leadership on subsequent school and teacher performance. Much of this work is located in the school improvement and teacher leadership fields and will be explored in the next section.

 Distributed leadership: the empirical case

The empirical evidence for distributed leadership can be found primarily in the school improvement and teacher leadership research fields. In the school-improvement literature, distributed leadership draws upon several strands. First, it draws upon the strand that makes a clear association between school culture and improvement. The school-improvement research has consistently underlined the contribution of strong collegial relationships to school improvement and change. Little (1990) suggests that collegial interaction at least lays the groundwork for developing shared ideas and for generating forms of leadership. Rosenholtz (1989) argues even more forcibly for teacher collegiality and collaboration as means of generating positive change in schools. Effective schools, she argues, have tighter congruence between values, norms, and behaviours of principals and teachers, and that this is more likely to result in positive school performance. Consequently, if we accept that a distributed perspective is one which focuses upon co-practice of routines there are some immediate links that can be made to the evidence-base pertaining to school culture and improved performance.

The second strand can be found in the composite lists of the characteristics of the 'improving school'. In their review of successful school improvement efforts, Glickman et al. (2001: 49) construct a composite list of the characteristics of what they term the 'improving school', a 'school that continues to improve student learning outcomes for all students over time'. At the top of this list appears 'varied sources of leadership, including distributed leadership'. The most recent literature on change and school improvement also suggests that the form of leadership most often identified with improved learning outcomes is one that is distributed or shared (Fullan 2001, Hopkins 2001).

The third strand of evidence can be found in research that has focused upon the process of organizational development. The work by Silins and Mulford (2002) has shown that student outcomes are more likely to improve where leadership sources are distributed throughout the school community and where teachers are empowered in areas of importance to them. In an earlier study Louis (1996) found that in schools where the teachers work was organized in ways that promoted sharing of leadership roles, there was a positive relationship with the academic performance of students. This presents a view of the school as a learning community chiefly concerned with maximizing the achievement capacities of all those within the organization.

Within the teacher leadership literature, there are a variety of studies that show clear evidence of the positive effect of distributed leadership on teachers' self-efficacy and levels of morale (MacBeath 1998, Crowther 2000). The evidence suggests that where teachers share good practice and learn together, the possibility of securing better quality teaching is increased (Little 1990, Lieberman et al. 2000). The literature also provides positive evidence about the impact of teacher leadership on student performance. Leithwood and Jantzi's (2000) study concluded that teacher leadership far outweighed principal leadership effects after taking into account the moderating effects of family educational culture. The evidence from this study suggests that principal leadership does not stand out as a critical part of the change process, but that teacher leadership does have a significant effect on student engagement. The study concluded that distributing a larger proportion of current leadership activity to teachers would have a positive influence on teacher effectiveness and student engagement (Leithwood and Jantzi 2000: 61). Other studies also report positive effects of teacher participation in decision-making, finding that teacher involvement in decision making leads to decreases in teacher absenteeism (Sickler 1988, Rosenholz 1989).

Under the right conditions, the positive benefits of distributed leadership are clearly identified within the teacher-leadership literature. Griffin (1995) found that distributed leadership resulted in positive effects on pedagogy, on school culture, and on educational quality. In her study of a schools where distributed leadership was being implemented, Ovando (1994) found that time to meet was a central component of success and in schools that were improving teachers were given dedicated time to collaborate with one another. The research evidence would also suggest that the success or otherwise of distributed leadership within a school can be influenced by a number of factors, such as relationships with other teachers and school management. These practical factors are considered in the next section.

 Distributed leadership: the practical case

While the findings from the teacher leadership and school improvement fields highlight the advantages of distributed forms of leadership, there are inevitable and inherent difficulties associated with its widespread adoption and adaptation within schools. It would be naïve to ignore the major structural, cultural, and micro-political barriers operating in schools, at least in the UK, that make distributed forms of leadership difficult to implement. Clearly schools as traditional hierarchies with the demarcations of position and pay-scale are not going to be instantly responsive to a more fluid and distributed approach to leadership. Furthermore, there are inherent threats to status and the status quo in all that distributed leadership implies. First, distributed leadership requires those in formal leadership positions to relinquish power to others. Apart from the challenge to authority and ego, this potentially places the head or principal in a vulnerable position because of the lack of direct control over certain activities. In addition, there are financial barriers as formal leadership positions in schools carry additional increments. Consequently, to secure informal leadership in schools will require heads to use other incentives and to seek alternative ways of remunerating staff who take on leadership responsibilities.

Second, the top-down approaches to leadership and internal school structures offer significant impediments to the development of distributed leadership. The current hierarchy of leadership within both elementary and secondary schools means that power resides with the leadership team, i.e. at the top of the school. In addition, the separate pastoral and academic structures in schools, the subject or department divisions plus the strong year-groupings present significant barriers to teachers working together. These structures can actively prevent teachers attaining autonomy and taking on leadership roles within the school as they demarcate role and responsibility.

Finally and most importantly, distributed leadership poses the major challenge of how to distribute development responsibility and authority and, more importantly, who distributes responsibility and authority. It is clear that a top-down approach to distributed leadership is possible, and that giving improvement or development responsibilities to teachers offers a means of empowering others to lead. But it will be important to ensure that distributed leadership is not simply misguided delegation. Instead, it implies a social distribution of leadership where the leadership function is stretched over the work of a number of individuals and where the leadership task is accomplished through the interaction of multiple leaders (Spillane 2002: 20). It implies inter-dependency rather than dependency, embracing the ways in which leaders of various kinds and in various roles share responsibility. As Bennett et al. (2003: 10) highlight, there may be both 'institutional and spontaneous' forms of distributed leadership. There may be a long-term institutional form of distributed leadership through team structures or working groups and there may be ad hoc groups offering a more fluid and immediate response to the change and development needs of the school. It is clear that certain tasks and functions would be have to be retained by those in formal leadership positions but that the key to successful distributed leadership resides in the involvement of teachers in collectively guiding and shaping instructional and institutional development.

The research evidence would also suggest that the success or otherwise of distributed leadership within a school can be influenced by a number of interpersonal factors, such as relationships with other teachers and school management. The importance of these is evident, both with respect to teachers' ability to influence colleagues and with respect to developing productive relations with school management, who may in some cases feel threatened by teachers taking on leadership roles. There may also be conflicts between groups of teachers, such as those that do and do not take on leadership roles, which can lead to estrangement among teachers. Recent research has shown that colleagues can at times be hostile to distributed leadership because of factors such as inertia, over-cautiousness, and insecurity (Harris and Muijs 2004). Overcoming these difficulties will require a combination of strong interpersonal skills on the part of the 'teacher-leader' and a school culture that encourages change and leadership from teachers.

 Commentary

While the evidence about the potential of distributed leadership to contribute to school improvement is compelling, it is clear that more research is needed that focuses directly on this relationship. Thus Harris and Muijs (2004) concluded that (1) distributed forms of leadership are evolving in schools but that it is far from the dominant institutional structure; (2) that distributed leadership means sharing the most important tasks, which means those associated with instructional improvement; and (3) that distributed leadership can co-exist with, and indeed depends upon, more formal leadership structures. This study found that distributed leadership is unlikely to flourish unless those in formal leadership positions positively promote and support it.

This study also highlighted a number of benefits and challenges associated with distributed leadership. In terms of benefits, it showed that distributed knowledge and expertise is less vulnerable to organizational shifts and changes. In other words collective expertise is less likely to be destabilized by external or internal changes as the knowledge is broadly based. The study also highlighted how distributed leadership ensures that professional development is naturally built into the system: the overlapping areas of expertise and collaborative ways of working create a powerful learning environment. Finally, it showed that distributed leadership is more likely to result in long-term system stability and continuity of school performance: there is an inherent mechanism of adjustment and adaptation built in through members sharing expertise. In terms of the challenges, as noted earlier, there are obvious threats to formal leadership role and responsibilities within the system. Plus there are significant challenges to individualized instructional practice, to dependency, and what Judith Warren Little (1990) calls the 'privacy of practice'. Teachers will need to share, reveal, and expose their practices if expertise is to be distributed within a school. Finally, policy directives aimed at radical systemic change will be less powerful if organizations and systems embrace distributed leadership. The reason for this resides in the fact that complex, distributed systems are also powerful cognitive and computational systems able to resist and safeguard against unnecessary interference (Hutchins 1995)

More research is needed that focuses on distributed leadership in schools: to elucidate different models, approaches, and forms of distributed leadership in practice; to investigate the relationship between distributed leadership and organizational improvement; to provide case-study exemplars of distributed leadership in action; and to identify the conditions in which distributed forms of leadership can flourish and grow. As Lashway (2003) notes,

[the] research base for distributed leadership is still embryonic. While there is considerable theory, we have relatively little empirical knowledge about how, or to what extent, principals actually use distributed leadership. And evidence that firmly links distributed leadership to student achievement is still far in the future.

Woods (2004: 453) argues that future research needs to 'encompass the interplay of the structural and agentive dimensions of distributed leadership' and that a sound research-base is needed to assess the effectiveness of distributed leadership strategies in enhancing positive educational outcomes.

The ascendancy of distributed leadership as a powerful concept and a theory represents a significant shift in thinking about leaders, leadership, and leadership development. It not only challenges the mythology of individualistic leadership but also reclaims leadership for teachers and others working in schools. Undoubtedly, more research is needed to give this new leadership perspective greater legitimacy, but as the association between distributed leadership and improved student and school performance becomes clearer (Leithwood and Jantzi 2000, Silins and Mulford 2000, Fullan 2001) so the opportunities for more innovative, flexible, and responsive forms of leadership increase.

The empirical evidence presented in this paper suggest that successful leaders are those who distribute leadership, understand relationships, and recognize the importance of 'reciprocal learning processes that lead to shared purposes' (Harris and Lambert 2003: 7). Essentially, these leaders are more connected to people than those holding the 'traditional' forms of leadership, i.e. the lone chief atop a pyramidal structure -- they distribute leadership in order to generate organizational development and change (Murphy et al. 1993, Greenleaf 1996). Yet, as Hopkins and Jackson (2003: 17) note,

despite more than two decades of writing about organizational development we are still in a position of needing to develop understandings about what leadership really involves when it is distributed, how schools might function and act differently, and what operational images of distributed leadership in action might look like.

In addition, despite a wealth of school-improvement literature advocating more collaborative, democratic, and distributed forms of leadership, clear links with improved student outcomes have yet to be established.

Most recently Bell et al. (2003: 4) suggest in their systematic review that 'distributed forms of leadership among the wider school staff is likely to have a more significant impact on the positive achievement of student/pupil outcomes than that which is largely or exclusively top-down'. For this reason we undoubtedly need to understand much more about effective distributed leadership practice, and we need to how it can be nurtured, supported, and developed. We need more empirical studies of distributed leadership that interrogate the relationship between distributed leadership and school improvement. But, most importantly, we need to know whether distributed forms of leadership contribute to improved student outcomes and, if so, in what form? While distributed leadership theory offers a dynamic and alternative way of understanding leadership practice, without an associated empirical base it faces the danger of becoming yet another leadership theory that proves to be misleading.

Bell, L., Bolan, R. and Cubillo, L. (2003) A systematic review of the impact of school headteachers and principals on student outcomes. In Research Evidence in Education Library (London: EPPI-Centre, Social Science Research Unit, Institute of Education). Available at: http://eppi.ioe.ac.uk/EPPIWeb/home.aspx?page=/reel/review_groups/leadership/review_one.htm (accessed December 8, 2004).

Bennett, N., Wise, C., Woods, P. and Harvey, J. A. (2003) Distributed Leadership (Nottingham, UK: National College for School Leadership). Available at : http://www.ncsl.org.uk/index.cfm?pageID=literaturereviews (accessed December 14, 2004).

Bush, T. and Glover, D. (2003) School Leadership: Concepts and Evidence (Nottingham, UK: National College for School Leadership). Available at : http://www.ncsl.org.uk/index.cfm?pageID=literaturereviews (accessed December 8, 2004).

Crowther, F., Hann, L., McMaster, J. and Fergurson, M. (2000) Leadership for successful school revitalization: lessons from recent Australian research. Paper presented at the annual meeting of the American Educational Research Association.

Elmore, R. F. (2002) Bridging a New Structure for School Leadership (Washington, DC: Albert Shanker Institute). Available at: http://www.shankerinstitute.org/education.html (accessed November 11, 2004).

Fullan, M. (2001) Leading in a Culture of Change (San Francisco, CA: Jossey-Bass).

Gibb, C. A. (1954) Leadership. In G. Lindzey (ed.), Handbook of Social Psychology, Vol 2 (Reading MA: Addison-Wesley), 877&endash;917.

Giddens, A. (1984) The Constitution of Society: Outline of the Theory of Structuration (Cambridge: Polity Press).

Glickman, C. D., Gordon, S. P. and Ross-Gordon, J. M. (2001) Supervision and Instructional Leadership: A Developmental Approach (Boston, MA: Allyn and Bacon).

Griffin, G. A. (1995) Influences of shared decision making on school and classroom activity: conversations with five teachers. Elementary School Journal, 96 (1), 29&endash;45.

Greenleaf, R. K. (1996) On Becoming a Servant-Leader, ed. D. M. Frick and L. C. Spears (San Francisco, CA: Jossey-Bass).

Gronn, P. (2000) Distributed properties: a new architecture for leadership. Educational Management and Administration, 28 (3), 371&endash;338.

Gronn, P. (2003) The New Work of Educational Leaders: Changing Leadership Practice in an Era of School Reform (London: Paul Chapman).

Hallinger, P. and Heck, R. H. (1996) Reassessing the principal's role in school effectiveness. Educational Administration Quarterly, 32 (1), 5-44.

Harris, A. (2002) Distributed leadership in schools: leading or misleading? Keynote Address to the Annual Conference of the British Educational Leadership, Management & Administration Society (BELMAS). Available at: www.shu.ac.uk/bemas/harris2002.html (accessed December 8, 2004).

Harris, A. and Lambert, L. (2003) Building Leadership Capacity for School Improvement (Maidenhead, UK: Open University Press).

Harris, A. and Muijs, D. (2004) Teacher Leadership: Improvement Through Empowerment: A Review of the Literature (Nottingham, UK: National College for School Leadership). Available at: http://www.gtce.org.uk//pdfs/research/Teacher_Leadership_litreview.pdf (accessed December 13, 2004).

Hopkins, D. (2001) School Improvement for Real (London: RoutledgeFalmer).

Hopkins, D. and Jackson, D. (2003) Building the capacity for leading and learning. In A. Harris, C. Day, M. Hadfield, D. Hopkins, A. Hargreaves, and C. Chapman (eds), Effective Leadership for School Improvement (London, RoutledgeFalmer), 84&endash;105.

Hutchings, E. (1995) Cognition in the Wild (Cambridge, MA: MIT Press).

Lakomski, G. (2004) Managing Without Leadership: Towards A Theory of Organizational Functioning (Oxford: Pergamon/Elsevier).

Lashway, L. (2003) Distributed leadership. Research Roundup, 19 (4) (Eugene, OR: University of Oregon, Clearinghouse on Education Policy and Management). Available at: http://eric.uoregon.edu/publications/roundup/Summer_2003.html (accessed December 13, 2004).

Leithwood, K. and Jantzi, D. (2000) The effects of transformational leadership on organizational conditions and student engagement. Journal of Educational Administration, 38 (2), 112&endash;129.

Leithwood, K. A. and Riehl, C (2003) What We Know about Successful School Leadership (Philadelphia, PA: Temple University, Laboratory for Student Success). Available at: www.ncsl.org.uk/mediastore/image2/randd-leithwood-successful-leadership.pdf (accessed 9 December, 2004).

Lieberman, A., Saxl, E. R. and Miles, M. B. (2000) Teacher leadership: ideology and practice. In The Jossey-Bass Reader on Educational Leadership (San Francisco, CA: Jossey-Bass), 339&endash;345.

Little, J. W. (1990) The persistence of privacy: autonomy and initiative in teachers' professional relations. Teachers College Record, 91 (2), 509&endash;536.

Louis, K. S. (1996) Teachers' professional community in restructuring schools. American Educational Research Journal, 33 (4), 757&endash;798.

MacBeath, J. (1998) Effective School Leadership: Responding to Change (London: Paul Chapman).

Murphy, J. and Beck L. G. (1993) Understanding the Principalship: Metaphorical Themes, 1920s&endash;1990s (New York: Teachers College Press).

Muijs, D. and Harris, A. (2003) Teacher Leadership: A Review of Research (Nottingham, UK: National College for School Leadership). Available at: http://www.ncsl.org.uk/mediastore/image2/randd-engaged-harris.pdf (accessed December 13, 2004).

Ovando, M. N. (1994) Effects of teacher leadership on their teaching practices Paper presented at the annual meeting of the University Council of Educational Administration, Philadelphia, PA. ERIC ED 380 446.

Rosenholz, S. J. (1989) Teachers' Workplace: The Social Organization of Schools (White Plains, NY: Longman).

Sickler, J. L. (1988) Teachers in charge: empowering the professionals. Phi Delta Kappan, 69 (5), 354&endash;356, 375&endash;376.

Silins H. C. and Mulford, W. (2002) Leadership and school results. In K. Leithwood and P. Hallinger (eds), Second International Handbook of Educational Leadership and Administration, Part 2 (Dordrecht, The Netherlands: Kluwer), 561&endash;613.

Spillane, J. P., Diamond, J. B. and Jita, L. (2003) Leading instruction: the distribution of leadership for instruction. Journal of Curriculum Studies, 35 (5), 533&endash;543.

Spillane, J. P., Halverson, R. and Diamond, J. B. (2004) Towards a theory of leadership practice: a distributed perspective. Journal of Curriculum Studies, 36 (1), 3&endash;34.

Woods, P. A. (2004) Democratic leadership: drawing distinctions with distributed leadership. International Journal of Leadership in Education, 7 (1), 3&endash;26.

   

---

Publisher's information |  | Other Op-eds and Commentary