Developing educational leaders: A partnership between two universities to bring about system-wide change

This study investigated a system-wide change strategy in a South African school district, which sought to build the leadership capacity of principals and district officials to improve instruction. The three-year venture was called the Leadership for Learning Programme (LLP). A distinctive feature of the LLP was that it was based on a partnership between two universities, a local one with understanding of the local context of schools, and an international institution, which brought international expertise, experience and repute/branding. Both universities had a shared vision to contribute to the ailing South African school landscape by using leadership development to leverage change. The LLP was implemented in a single school district, where the overall learner performance was unsatisfactory. A qualitative approach was used to research this change intervention. One of the main findings was that collaboration between principals collectively and district officials, as well as among principals, was lacking. It is recommended that collaborative structures such as professional learning communities, networks and teams are established to reduce isolation and fragmented work practices in the school district. This may speed up system-wide change towards improved learner performance.


Introduction
The crisis in the South African school system is well-documented (Bloch, 2009;Fleisch, 2008;Letseka, Bantwini & King-McKenzie, 2012).The crisis is also confirmed by the results of both national and international evaluations.In the Annual National Assessments (ANA) in 2011, Grade Three and Grade Six learners were unable to attain an average of more than 35% in either Literacy or Mathematics (Department of Basic Education, Republic of South Africa, 2012).South Africa was ranked the lowest amongst 50 countries in the Trends in International Mathematics and Science Study (TIMSS) undertaken in 2003 (Letseka et al., 2012).In the Progress in International Reading Study (PIRLS) testing in 2006, South African learners performed the lowest amongst students from 40 countries in reading ability (Kennedy, 2006).The recent Global Information Technology Report (Bilbao-Osorio, Dutta & Lanvin, 2014) ranked the quality of the South African educational system at 146th out of 148 countries.It is disconcerting that post-apartheid, historically disadvantaged, learners have not shown a significant improvement in academic performance (Van der Berg & Louw, 2008).These negative reports pertaining to the current South African school system do not bode well for the country"s developing economy and its future society.
In view of this predicament, initiatives to enhance South African schools are essential.An area that has come under the global spotlight is leadership development, which emanates from research that supports a positive link between high-quality leadership and successful schools (Louis, Leithwood, Wahlstrom & Anderson, 2010).The efficacy of leadership development programmes has resulted in leadership development interventions worldwide, so much so that governments are now keen to invest in leadership development for school leaders (Bush, 2009).A positive link between management ability and economic productivity (Mabey & Finch-Lees, 2008) reinforces the idea of investing in management and leadership development.
Against the backdrop of the collapsing South African school system, two universities joined forces to initiate change.The collaboration culminated in a leadership development programme aimed to develop instructional leadership.The novelty of the programme was that it was based on a system-wide approach that targeted leadership development at the district level of the school system.A system-wide approach requires the "collective capacity" of all those within the system "to communicate and connect, to drive change forward and to align effort", rather than individualised approaches, which focus on improving the "individual capacity of single schools" (Harris, 2010:197-198).Consistent with a system-wide approach, both district officials and principals from one school district were participants in the programme, named the Leadership for Learning Programme (LLP).
System-wide change has gained prominence in the past decade, as the need for large-scale change in school systems has emerged.Countries such as Canada, Finland, Hong-Kong, Singapore and England (Fullan, 2009a) have been forerunners in initiating system-wide change approaches at district, provincial (state) or national levels.Theorists such as Fullan (2009b) and Hopkins (2011), have maintained that the focus of change initiatives in schooling should be on all schools.The model of change which has focused on individual schools as the main unit of change has not achieved large-scale success in school systems (Harris, 2010).Furthermore, this model kept the pace of change slow (Harris & Chrispeels, 2006).Bearing in mind the current need for a swift transformation of the South African school system towards large-scale school success, we draw attention to a system-wide approach as a potential change strategy.Previous empirical evidence of systemwide efforts to enhance educational leadership at national, provincial or district level in South Africa is limited.Three system-wide change initiatives identified in the literature were undertaken in South Africa, but did not result in any conclusive findings.Two of these were the Systemic Enhancement for Education Development (SEED) in the Western Cape, and the Quality Learning Project (QLP) in De Aar (Fleisch, 2006).A more recent endeavour was the Gauteng Primary Language and Mathematics Strategy (GPLMS), which attempted to close the gap between performing and underperforming schools in Gauteng Province (Fleisch & Schöer, 2014).However, since the different versions of the ANA were not comparable over the years, the GPLMS research remained inconclusive (Fleisch & Schöer, 2014).This research on the LLP contributes to the global discourse on leadership and management development, since emerging economies are urgently striving to address their shortcomings in leadership and management in a bid to "catch up" with their economically developed counterparts (Mabey & Finch-Lees, 2008:6).
The aim of this study was to determine what may be learned about system-wide change from an exploration of the LLP.The specific objectives were to:


investigate the perceptions and experiences of the various actors involved in the programme;  ascertain the benefits and challenges experienced during the implementation of the programme; and  determine the potential of the programme to initiate systemic change in the school district.

System-Wide Change
Fullan (2009a) identifies three phases in the educational change journey, from school-based approaches, to system-wide approaches.The first phase, pre-1997, was marked by increasing pressure for educational change resulting in greater innovation at individual schools.In the second phase, 1997-2002, larger-scale change initiatives were implemented across schools, such as the National Literacy and Numeracy Strategy (NLNS) in England.During the third phase, 2003-2009, the emphasis was on how to achieve system-wide change at the district, provincial and/or national levels.System-wide change has escalated since 2009.A current initiative is the Tri-Level Reform Movement in Wales, which strives for systemic reform of the entire education system (Harris, 2010).
The gradual shift towards a systems view in education has meanwhile spanned five decades (Banathy, 1992).During this time, problem situations became increasingly complex, embedded in interconnected systems that operate in dynamically changing environments (Banathy, 1992).Systems theory (Banathy, 1991) gives primacy to the interconnectedness and interdependence of the elements in a system, as well as the evolutionary nature of a system (Banathy & Jenlink, 2004).Because of the system-wide approach of the LLP, we use systems theory to frame this investigation.The system of interest in this study is the school district.In order for system-wide change, also referred to as systemic change, to occur, other fundamental changes must be made simultaneously at multiple schools throughout the school district, thereby enabling a paradigm shift (Duffy & Reigeluth, 2008).In the case of the LLP, a leveraged emergent approach to systemic change was used (Reigeluth, 2006).Leadership development is considered a part of the system that exerts sufficient leverage to prevent the changed parts of the system from reverting to their previous state (Reigeluth, 2006).
The definition of leadership development used in this study is "the expansion of a collective"s capacity to produce direction, alignment, and commitment" (McCauley, Van Velsor & Ruderman, 2010:20-21).The term "collective" used here refers to any group of people who share work, such as partnerships, work groups and communities (McCauley et al., 2010).Such a general definition is apt for this study, where the LLP participants worked as a collective with the intention of expanding leadership capacity in the school district.Systems theory supports Fullan"s (2009b) claim that effective leaders working in an individualistic manner are unlikely to succeed at system-wide organisational change.In South Africa, leadership development has centred on the training of individuals, rather than collective capacity building, which encourages learning through interaction.For instance, the interactive sessions envisaged in The Advanced Certificate in Education: School Leadership (ACE) for principals, were largely unsuccessful (Bush, Kiggundu & Moorosi, 2011).Even internationally, leadership development programmes that targeted the level of the system have been found to be inadequate (Fullan, 2009b).A systemwide leadership development programme underwent research in the state of Georgia, in the United States of America (USA).This research found that in the long term, those schools whose leaders had participated in the programme achieved higher learner test scores than schools whose leaders had not been a part of the programme (Page, 2010).An investigation by Louis et al. (2010) involving 43 school districts, found that districts could have a significant impact on schools and learners by developing school leaders" collective sense of efficacy about their jobs (Fullan, 2009b).Such studies have highlighted the importance of school districts and principals collectively, as agents of systemwide change.Rorrer, Skrla and Scheurich (2008) have contended that the school district is an important agent for system change.The district office of the school district is a local government structure that holds authority over the multiple schools in its jurisdiction.As such, it is in a favourable position to initiate and sustain large-scale change (Roberts, 2001).Furthermore, the district office has an awareness of the local context in which schools operate (Taylor, Muller & Vinjevold, 2003).However, school change efforts worldwide have neglected the district office as an agent of change, placing individual schools, and even teachers, at the centre of change efforts (Chinsamy, 2002;Chrispeels, Burke, Johnson & Daly, 2008).The perspective that the district office is a catalyst for system-wide change in districts is supported empirically (Kaufman, Grimm & Miller, 2012;Zavadsky, 2012).However, more research into district change efforts are required in order to provide empirical guidance that district leaders might utilise to help them bring about system-wide change in their districts (Rorrer et al., 2008).
Principals, as vital agents of system change, will play a broader role than that currently held.This entails a new mindset, where individual school leaders stop thinking about "my school" or "my district" and start thinking about "our schools or our districts" (Fullan, 2007:63).The system leadership role is essentially one of cross-collaboration.It involves principals increasingly working "with other schools and other school leaders, collaborating and developing relationships of interdependence and trust" (Pont, Nusche & Moorman, 2008:6).Munby (2008) asserts that system collaboration arising from system leadership is a prospect for the improvement of learner performance in underperforming schools.From the perspective of systems theory, developing these leadership links between schools is crucial to advancing systemic change.Many of the complex problems faced by schools are systemic, and as such, warrant systemic solutions.System-wide change is unlikely to transpire if the collective capacity of principals in the system remains unharnessed (Harris, 2010).Fullan (2009b) contends that when a critical mass (90%) of school principals is engaged in development, such as instructional leadership, within and across schools, there is likely to be a positive effect on the system.However, a system-wide change endeavour is a mammoth undertaking, which may go beyond the capacity of the district office and school principals, requiring the support of external agencies.
Empirical evidence suggests school districts are likely to benefit from one or more external partners, whether local or international (Kronley & Handley, 2003).Houle (2006) found that the need for partnerships with underperforming schools is so critical, that it resulted in the formalisation of partnerships, even after his study was completed.Furthermore, external partners have been a catalyst for pushing school districts towards change (Barnett, Hall, Berg & Camarena, 2010).Such partners can also develop professional capacity in schools and district offices (Kronley & Handley, 2003).Programme funding is another benefit of external partnerships (Johnston & Armisted, 2007).Collaboration between school districts and universities is a common approach used to develop district-wide leadership capacity (Korach, 2011).The two universities that partnered to provide the LLP, provided these benefits to the participating school district.

Background to the LLP
A South African doctoral student in the School of Education at the international university that became involved in the LLP, elicited the School"s interest in contributing to the improvement of schooling in South Africa.Hearing about the possibility of a project with the international university, a team from the local university visited the international university to explore the possibility of a partnership.The international university reciprocated by visiting the local university.In subsequent discussions, the academics from both universities agreed upon a system-wide venture, which would focus on leadership development in a school district.Thereafter, a memorandum of understanding between the two universities was drawn up, and funding was sought for the intervention.Stakeholder buy-in from the teacher unions and the provincial MEC (Member of the Executive Council for Education-parliamentary stature), was sought.Buy-in from the unions was not achieved.However, the MEC agreed to the implementation of the project, and selected a school district in Gauteng Province where learner performance in the Senior Certificate Examination (Grade 12) had been inadequate for several years.Two significant factors contributed to the possibility of co-ordinating a venture of this magnitude throughout the school district.The first was the establishment of a leadership institute by the local university to steer the project.The second was extensive funding secured from external sources.These allowed the selected school district to participate in the programme, and sustained the procurement of all the required human and material resources to offer the LLP.After extensive discussions between the two universities, a programme was designed.In view of the particular context of the school district, fraught with its many challenges, it was decided that the LLP would follow an organic evolution for its three-year duration (2010)(2011)(2012).This approach provided flexibility in addressing the emergent needs of the participants.At this stage, it was decided to investigate the initial needs of the LLP participants (principals and district officials).Focus group interviews with the participants generated themes providing topics for the LLP start-up.An LLP framework (Figure 1) was then developed to conceptualise the intervention.
At the core of the framework (Figure 1) was improved academic outcomes in the school district.This was dependent upon the quality of teaching, the successful learning attained by learners, and school and classroom factors promoting improved academic outcomes.Four areas, namely leadership and management, data analysis and utilisation, systems thinking and systemic support, were identified as being significant contributors to effective teaching and learning.Systems-thinking is understood as a conceptual framework for seeing cohesive wholes, rather than merely parts of a system in isolation; interrelationships between rather than the autonomy of individual elements; and for seeing patterns of change instead of static snapshots (Senge, 2006).

Figure 1 The Leadership for Learning framework
The LLP comprised four, week-long (28 hours per week) contact sessions, held on the university campus during the school holidays.The first session focused on instructional leadership; the second on effective communication, leadership values and collaboration; the third on leadership tools and strategic planning; and the fourth on the topics data-wise, charting the course and instructional rounds.Reflection and review by the academics took place after each session, in order to decide on the themes for the following session.On average, 85 participants attended each session, or part thereof.Some were absent due to prior commitments to the Gauteng Education Department.Between the contact sessions, a team of contracted facilitators provided on-site support at both the district office and the participating schools.In addition, regional cluster groups were established, for on-going collaboration between sessions.Academic staff were responsible for coordinating the sessions, and were assisted by an administrator.Presenters with the required expertise, including the academics involved, facilitated the sessions.Funding enabled 54 participants to attend a six-day leadership development programme at the international university.
Researchers from both universities were assigned to investigate the LLP.The lead researcher of this paper, who was responsible for the data collection and analysis, was not actively involved in decision-making regarding the design and implementation of the LLP.The data was analysed on completion of the LLP and did not have an action research orientation.

Method
The study used a generic qualitative approach, as we intended to uncover and understand the perspectives of the participants involved in the LLP (Merriam, 1998).The sample consisted of 11 academic staff members, one administrator, seven facilitators, 101 principals, and 44 district officials.Simple random sampling was used to select district officials, academic staff of the universities, and the programme facilitators for the interviews.Stratified random sampling enabled one principal from each of the five regional clusters of the school district to be represented in the sample.
Data was collected by means of participant observation and semi-structured, individual interviews.The participants were observed during the LLP contact sessions.Principals were further observed in three cluster group meetings.Field notes were used to keep track of observations, encounters and decisions made during the study.On conclusion of the programme, individual interviews were conducted with one female and four male principals; four female district officials; one female academic, three male academics and two female facilitators.The interview schedule was piloted among each of these groups to strengthen validity.
Tesch"s method (1990) was employed to analyse the data.In the coding process, the raw data were read, and units of meaning were identified and labelled.Codes were grouped into categories and synthesised into sub-themes and themes.Triangulation was applied across the four data sources, namely principals, district officials, facilitators and academics, to search for convergent evidence.Furthermore, triangulation was used to compare the participant observations with the interview findings.Triangulation can be useful in facilitating cross-data validity checks (Patton, 2002).Ethical research procedures included receiving approval from the relevant authorities, acquiring written informed consent from the participants, anonymity of places and people, and confidentiality of sources.

Findings
The emergent themes and sub-themes are presented in Table 1.A discussion of the themes and sub-themes ensues using the following codes: principal (P), district official (DO), facilitator (F) and academic (A).Numbers are used to indicate the different interviewees as follows: P1 refers to the first principal who was interviewed, P2 refers to the second principal interviewed, and so on.The same system of numbering applies to the other groups interviewed.

Interrelationships
Both the participant observation and the interview data revealed the relationships between principals and district officials to be strained.Tensions mounted during the second contact session of the LLP, as principals and district officials blamed each other for malfunctions in the school district.The interview findings confirm the strained relationships.An academic recounted: There was this [sic] clash between district office and principals.It seemed to me that there was a 'blame game ' [sic] going on both sides.For example, if a school would say, 'We didn't get the books that we needed on time.'And the central office people would defensively say, 'You all didn't send in the numbers when we asked' (A2).
A facilitator confirmed the poor interrelationships between principals and district officials, stating: "so it was a lot of conflict there … there was mudslinging, but from both sides" [sic] (F1).Ineffective communication also played a role in the strained relationships, as is evident in the next remark:

You're [district officials] saying, 'I sent out a letter telling you what to do. Why didn't you do it?' And principals say that's not a way to communicate. I think a lot focused on: 'how do we communicate more effectively so that we can be more effective leaders?' [sic] (A2).
The hierarchical structure of the school district, which reinforces the positional authority of district officials over principals, further contributes to the poor interrelationships.A facilitator remarked: It was an attitude adjustment that was most needed -that we are equals.We [district officials] are equal to the teachers and the principal.We are not their seniors.Because, for as long as you believe that you are superior to another person, you speak down to that person … They [principals] wanted the district to speak to them on level terms (F1).
Academics also alluded to the management style of the district officials as being authoritarian.An academic stated: There should be more thinking that we are a team, we are one system … perhaps the district officials feel that they lay down the law and you dance to it [sic] (A3).
Poor interrelationships also emerged between the district officials themselves at the second contact session, where, during the breakaway sessions, when district officials worked as a group, conflict arose.There was discontent over the allocation of new cars to district officials and the lack of consultation by senior management in decisionmaking.An academic explained:

We did the difficult conversation and I think that was the most difficult session we had, because a lot of issues were placed on the table … The district office had major issues to address and … a lot of participants walked out there feeling that, 'I have vented my frustration and someone listened to me' [sic] (A1).
Findings from the observation data reveal that the conflict was well managed.The presenters used protocols, for example, taking the participants down "The ladder of inference" in order to challenge assumptions that were placed at the top of the ladder.
Data triangulation from all four sources, namely principals, district officials, facilitators and academics, confirmed that the LLP enabled the building of more collegial relationships between principals and district officials.A district official elaborated: What was very interesting for me is the fact that we could have sessions with principals, have a sense of what principals are feeling, and their frustrations.Principals have a sense of what our frustrations were as a district office … Also to get us to a point that we understand each other, there is not 'us' and 'them' [sic].But we actually need to work together for it to work (DO4).

Contextual Challenges
A principal (P1) was frustrated that district officials merely enforce policies of the department without addressing the implementation difficulties that they face in schools.Other principals concurred:

Lots of us [principals]
were dissatisfied with the way the higher educational authorities operated.There were challenges on the ground that we felt they didn't know much about, and there wasn't much consultation with schools at that level.Funding was a problem that [sic] all schools experienced difficulty, especially when it came in to school fees (P2).
Another contextual challenge was that racial integration in the school district appeared to be inadequate.It was observed that the participants seated themselves according to their race groups.However, the racial divide extended beyond the seating.An academic elaborated:

It was as if somebody from one group, when they got up to ask a question, a clarifying question, the people in their own group acknowledged the same question. And you might not have seen the same acknowledgements in the other groups. It made very clear to me what I had been told about the history of separations, you know, planned separations and apartheid … at this point there was a lack of trust across the divides, I felt (A2).
By interacting with one-another in the LLP, racial barriers were broken down.A principal stated: In the past we didn't have much communication between Afrikaner people and Indian people and Coloured people and Black people.And then once we had settled we found ourselves forming friendships.We became so comfortable with each other (P2).
A contextual challenge that most principals were vocal about was the lack of parental support to learners.A principal stated that many parents are "illiterate" (P2).Another principal reported that 80% of his learners were orphans, had single parents or came from child-headed homes (P4).

Experiences encountered during the Implementation of the LLP
One of the strengths of the LLP was the organic nature of the course it assumed.By adapting the LLP to the needs of the participants, the LLP adopted a context-specific focus.An academic elaborated:

I see this whole project as one in which we constantly had to scaffold the learning. You could and every time we went too far we had to recalibrate, because if people weren't ready to receive it, then you were wasting your breath (A2).
A further strength was the partnership with the international university.The expertise of the presenters had a profound effect on the participants.An academic noted: They were all speaking from their own experience, their own research, so they weren't speaking from book knowledge; so in that sense, it had emotive value, rather than purely cognitive value (A4).

A district official concurred as follows:
Our district is classified as an underperforming district.So, I was interested in listening to Lesley [pseudonym] when she was presenting on being [the leader] of an underperforming district as well … I was saying, 'we are not alone'.If we stay focused, if we have our vision, we understand our vision and we run with our mission on daily basis, we can turn around the district as well (DO1).
A participant commented on the interactive pedagogy used by the presenters: Its the way they give you a thing to study, and then they have this interactive growth, where you talk and where they facilitate.And when I studied, how lecturers in South Africa do it, they talk and you listen.Whereas, these people gave you things to work on, and then they facilitate [sic] (DO2).
The observation findings indicate that, in order to promote greater interaction among the participants, the venue for the contact sessions was reviewed.The long, step-down lecture theatres used initially in the programme were replaced in favour of a hall, with the participants working in roundtable.
Challenges encountered during the course of the LLP included long gaps between contact sessions, and the need for mentors.Holding the programme during a school holiday also meant that some of the principals did not attend the LLP.There were contextual problems at various schools that affected the implementation of the programme at the classroom level.One principal explained: "we were in the process of merging … I couldn't implement some of the things" (P5).A facilitator expressed concern that some principals were not sharing new learning from the programme with their staff (F2).Another difficulty experienced was that the district office had not committed to the programme prior to its commencement.One academic stated: "we need to ensure that we have buyin from the beginning.We have to let them [district] know they are equal partners, because I got the feeling [sic] initially that to them it seemed as if we are being imposed on them" (A1).

Capacity Building
Learning and growth occurred during the LLP.A principal commented: "the instructional trips that I take into the classes is something that I pick up in the programme … It's something that I have been doing ever since" (P3).Another principal elaborated:

Before[hand], when I used to go to a teacher's class, I use to look at all the negative things … you have to go and look at what is wrong in the class to go and fix it … but now I have changed in terms of … going in and looking at what teachers need and what they [are] doing, and what they [are] doing right (P1). i
There was evidence of a changed attitude, of being more consultative, and of improved communication from a district official: I think I've learned to become a little bit less defensive … I've learned to get more involved with my people … And a very important thing is not to be up there and talk down [sic] (DO2).
The inadequate induction of principals into the principalship emerged in the interview data.A newly appointed principal commented: the induction was more focused on a normal functioning system.It was 'one size fits all' … it did not address what to do when you encounter challenges.The programme [LLP] focused on specific challenges (P4).

Working in Isolation
Interviews revealed that principals had not worked collaboratively, but in isolation.Two principals used metaphors to describe their isolated work practices.The first stated: "you sit like an island when you [are] a principal, and you don't know what's going on in other schools" (P1).ii The second remarked: "it was a course that really would help us to develop and it would take us out of our cocoon [sic]" (P4, L27).There exists a lack of alignment between the different units in the district office, where an official remarked: "our main problem is everybody is doing 'his own thing ' [sic].We are actually not there for each other.I've got no idea what [the] curriculum is doing.[The] curriculum has got no idea what I'm doing" (DO4).iii The inadequate co-ordination among units results in confusion at schools: the problems that would come up is aligning of diaries … there were instances like three different officials, district officials, who'd rock up at the same school to see the same principal for three different things (F1).
The interactive nature of the LLP promoted collaboration, such as networking, professional learning communities (PLCs), and system leadership.A principal expressed: "It [LLP] also gives us the platform to liaise and network with colleagues from various social cultures whereby there was that divide" (P4).A facilitator reported the emergence of a PLC, where a system leadership mindset is evident: "the group grew to the extent that they were working as a team, even supporting one another, even addressing their issues and trying to assist where they could" (F2).

Discussion
Systems theory is concerned with the interrelationships among the elements in a system.The elements are nonlinear, dynamic and mutually interactive (Razik & Swanson, 2010).Therefore, "when the demands on one part of the system are linked to the demand of other parts, those parts will only perform well if they are connected together" (Bar-Yam cited in Duffy, 2010:3).In this study, it was found that the poor working relationships among people within the system, contributed to disconnections within the system.When people are unaware of systems thinking, they ignore their interconnectedness, and in so doing, miss the bigger picture (Reynolds & Holwell, 2010).This refers to the systemic outlook which places emphasis on the concept of the whole.Since they are part of one system, district leaders and principals must come to understand that there can be no external agents to blame when something goes wrong (Senge, 2006:67).Instead, the solution can be found in the relationship among the various elements within the school district (Senge, 2006: 67).Hence, attention needs to be given to the development of relationships within institutions and among institutions.When relationships improve, the culture of a school district is bound to change (Fullan, 2007).When parts of a system are connected, synergy can be fostered (Duffy & Reigeluth, 2008).However, the findings revealed that the hierarchical structure of the school district reinforces power relationships and threatens collegial relationships.If a view of "leadership as practice" is emphasised over "leadership as position" (James, Connolly, Dunning & Elliott, 2007:576) perhaps more collegial working relations between district officials and principals might ensue.
The findings point to a lack of collaborative structures for principals to engage in collective capacity building.The LLP provided a unique forum which brought principals and district officials together for collective capacity building.Harris and Jones (2010) emphasise that systemwide change is unlikely to succeed without structures for collaboration.Initiating networking and collaboration among the participants, many of whom were working in isolation in their own schools, was a benefit of the LLP.Networking and collaboration are features of PLCs, where "practice is developed and refined through the collaboration of groups of people who share a concern, a set of problems, or a passion about a topic, and who deepen their knowledge and expertise by interacting on an ongoing basis" (Wenger cited in Harris & Jones, 2010:175).Collaboration was further enhanced in the LLP by using a mode of presentation and facilitation that was highly interactive.Bush (2008) notes that in this era, a shift is occurring from content-based to process-based programmes, which espouse learner-centred learning, action learning and open learning.In this regard, the organic design of the LLP was a strength that enabled the organisers to address the needs of the participants at various stages during the course of the programme.
While the LLP promoted collective capacity building, an aspect that required greater thought is how to transfer what was learnt in the programme to the school context.Pegg (cited in Rhodes & Brundrett, 2009) points out that to know how educational leaders learn, researchers must observe the way in which leaders put into practice the theories learnt from training programmes.Fullan (2007) explains that the change process consists of three phases, namely: the initiation of change; the implementation of change; and the institutionalisation of change.The implementation phase is important, as it will influence whether the change is ultimately successful or not.It may be speculated that if there were buy-in from the senior leadership of the school district from the onset of the LLP, it would have strengthened the implementation of the LLP.
Various challenges arising from the systemic context of the school district were highlighted in the findings.Issues pertaining to policy implementation, racial integration and parental involvement are some of these.Taylor, Fleisch and Shindler (2007) identify policy implementation in South African schools as a core weakness in the school system.The lack of parental support for schools, fuelled by socio-economic challenges, is another daunting area system stakeholders grapple with.New principals have reported on the failure of induction programmes to prepare them to manage these contextual challenges.Even the most efficient school leaders will flounder against adverse contextual realities (Bush et al., 2011).Using a systems lens, the contextual challenges that emerged during the LLP remind us that the school district is a complex system, where the problems are multidimensional, characterised by uncertainty and contradictory perspectives on the problem at hand (Reynolds & Holwell, 2010).It is imperative, however, to address these systemic problems that impact on systemic performance due to the fact that "their consequences will be felt in the economy and society for years to come if they are not addressed" (Christie, Butler & Potterton, 2007:29).

Conclusion and Recommendations
Communication and collaboration in the school district can be enhanced by establishing collaborative structures, such as PLCs, networks or teams for principals.District officials can play a new role in co-ordinating collaborative teams, monitoring the work of these teams and providing support, resources and training.Education systems moving towards system-wide change should endeavour to reduce isolation in work practices.An understanding of systems thinking is recommended for all education leaders in a school district.
A recommendation is to establish a District Co-ordinating Committee comprising representatives from the district office, principals, teachers and unions.Such a committee could strive to better understand the complex challenges facing schools, and engage in problem-solving strategies.This ought not to be a controlling body but rather an advisory or collaborative body.A further recommendation is to institute an Annual Leadership Development Forum for principals and district officials premised on collective capacity building.Collective capacity is likely to generate greater emotional commitment and technical expertise towards system-wide change than that gained from developing individuals.In undertaking this venture, the school district can enlist the support of uni-versities and the expertise of other agencies.Developing the capacity of leaders and managers for the future is said to lead to financial and economic gains for governments (Mabey & Finch-Lees, 2008), which is of significance for both emerging and developed economies.
Based on the findings in this investigation, it is theorised that a programme such as the LLP does have the potential to initiate systemic change in the school district, subject to the following conditions.Firstly, the necessary structures for meaningful collaboration within the school district need to be established, developed and maintained.Secondly, systems theory posits that the participation of all the schools in the school district is essential for a paradigm shift.Thirdly, buy-in from all stakeholders is important in a system-wide change initiative.Finally, formulating a comprehensive plan, which outlines the implementation of the programme at the different levels of the district, is essential.These levels include schools, classrooms and the district office.If these conditions are met, it is likely that a programme such as the LLP could achieve system-wide change, due to some of its strong features.These include the provision of a dedicated forum for collective capacity building among district officials and principals; the flexibility to adapt to the participants" needs instead of having a one-size-fits-all programme; processbased programme delivery that facilitates interactive learning; and utilising persons with expertise to build the required capacity.

Notes i.
Verbatim quotation was edited for the publication.ii.
Verbatim quotation was edited for the publication.iii.
Verbatim quotation was edited for the publication.

Table 1
Research themes and sub-themes