Jump to ContentJump to Main Navigation
The Politics and Governance of Basic EducationA Tale of Two South African Provinces$

Brian Levy, Robert Cameron, Ursula Hoadley, and Vinothan Naidoo

Print publication date: 2018

Print ISBN-13: 9780198824053

Published to Oxford Scholarship Online: November 2018

DOI: 10.1093/oso/9780198824053.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2019. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in OSO for personal use. date: 15 October 2019

Case Studies of School-Level Governance Dynamics in the Western Cape

Case Studies of School-Level Governance Dynamics in the Western Cape

(p.201) 8 Case Studies of School-Level Governance Dynamics in the Western Cape
The Politics and Governance of Basic Education

Ursula Hoadley

Brian Levy

Lawule Shumane

Shelly Wilburn

Oxford University Press

Abstract and Keywords

This chapter details four case studies of school-level management practices in the appointment process of school principals. The school-level processes are used as a lens through which to refract local governance dynamics, and thereby gain insight into the broader multi-stakeholder contexts within which the principal is embedded. The contrasts between the Western Cape and Eastern Cape provinces provide an ideal opportunity to explore a central theme of this book, namely how ‘good fit’ works—how preferred approaches to policy design and implementation might vary according to the contexts in which they are being undertaken. The chapter suggests that rather than viewing the interaction between hierarchical and horizontal governance as zero-sum, the task for practitioners is to find ways to make more effective the ‘both/and’ balance, with an emphasis on impersonal forms of decision-making.

Keywords:   principal selection, school-level governance, school management, resilient schools

8.1 Introduction

South Africa’s public schools fall under the hierarchical control of the country’s nine provincial departments of education, which vary widely from one another in their effectiveness. Chapters 4–7 analysed the causes and effects of this variation, focusing especially on the Western and Eastern Cape. As they show, variations in bureaucratic quality indeed affect educational outcomes. However within provinces, and controlling for the socio-economic circumstances of learners, there also are quite substantial variations in performance across schools. Why?

An extensive literature, summarized in Chapter 1 of this book, suggests that for services such as basic education where provision is dispersed, giving ‘horizontal’ actors at the service provision front line some autonomy vis-à-vis hierarchical control potentially can improve performance. South Africa’s 1996 Schools Act delegated quite substantial formal authority to the school level, and assigned a major role to school governing bodies (SGBs) to exercise that authority. The South African education system thus offers an ideal opportunity to explore how the delegation of authority to the service provision front line affects performance. This chapter explores the question through a series of in-depth case studies of school-level governance in the Western Cape province. Chapter 9 focuses on schools in the Eastern Cape.

There are at least four distinct channels through which delegation of authority to the service provision front line might improve performance. One of these involves the customization of service provision; for education, this raises questions of approaches to pedagogy, which fall outside the scope (p.202) of the present study. The other three work more directly through governance. They comprise:

  • A managerial channel—improving motivation, with a ‘zone of autonomy’ at the service provision front line hypothesized to provide the opportunity for internal leaders to motivate their teams effectively, including by fostering an environment of continuing learning on the part of staff as well as students (Wilson, 1989; Lipsky, 2010).

  • An informational channel—creating scope for the utilization of local-level information of a kind to which higher-level hierarchical authorities lack access—and thereby enhancing processes for the selection of good quality staff and leaders, and the efficacy of efforts to hold staff and leaders accountable for their performance (Sah and Stiglitz, 1986; North, 1990; and Aghion and Tirole, 1997).

  • An accountability channel—enabling developmentally oriented local stakeholders (including professionally committed teaching staff) to hold school staff accountable for making their best effort. A necessary condition for ‘empowerment’ to be effective as a means of strengthening accountability is that developmentally oriented stakeholders indeed have sufficient influence to be able to ‘trump’ predatory actors seeking to capture school-level resources (teaching and administrative positions, contracts, other discretionary resources) for private purposes (Levy, 2014).

Figure 8.1, reproduced from Chapter 1 (with one adjustment, on which more below) illustrates schematically the multiple causal mechanisms through which stakeholders potentially might influence school-level governance, and thereby affect educational outcomes. The heavy bolded arrow from the school principal (p.203) to the teaching staff reflects a central finding of empirical research on the determinants of school performance—namely that the quality of school leadership is an important proximate explanatory variable (Hallinger and Heck, 1996; Leithwood, Patten, and Jantzi, 2010). The lighter arrow in the opposite direction points to the possibility of a two-way relationship, with the teachers’ organizational culture affecting the approach to management of the school principal. The arrows linking the SGB, parents and community to the principal and teachers illustrate the potential role of local-level actors from outside the school. Although not included explicitly in Figure 8.1, the school-level research also probed the role and influence on school-level governance of the teachers’ unions (SADTU and NAPTOSA) and political parties.

Case Studies of School-Level Governance Dynamics in the Western Cape

Figure 8.1. School-level governance interactions

What differentiates Figure 8.1 from the equivalent figure in Chapter 1 is the heavily bolded lines linking the education bureaucracy to the principal and teachers. These bolded lines signify that, consistent with the in-depth analysis in Chapter 4, the Western Cape’s bureaucratic hierarchy is relatively strong. (The lines linking the education bureaucracy to schools are drawn very differently in the parallel figure in Chapter 9’s analysis of school-level governance in the Eastern Cape, where bureaucratic capabilities are much weaker.)

The contrasts between the Western Cape and Eastern Cape provinces provide an ideal opportunity to explore a central theme of this book, namely how ‘good fit’ works—how preferred approaches to policy design and implementation might vary according to the contexts in which they are being undertaken. Viewed from this perspective, the Western Cape school-level cases can be interpreted as exploring the potential role of horizontal governance as a complement to a well-functioning bureaucracy. By contrast, the Eastern Cape cases in chapter 9 explore horizontal governance’s role as a potential substitute in a setting where hierarchy is weak.

8.2 Research Context and Methodology

This section introduces the research methodology used for the Western Cape school-level case studies. Our point of departure is the finding in the education literature that the quality of school-level leadership is an important proximate determinant of school performance. The school-level processes of principal selection are used as a lens through which to refract local governance dynamics, and thereby gain insight into the broader multi-stakeholder contexts within which the principal is embedded. This section provides some background on our sample schools; on the principal selection process in the Western Cape; and on our research methodology. Sections 8.3–8.5 delve in detail into the school-specific processes. Section 8.6 concludes.

(p.204) 8.2.1 The Sample

Our school-specific analysis builds on a prior research programme (the SPADE initiative1), which studied a stratified sample of fourteen Western Cape schools in relation to their internal governance dynamics and instructional regimes (Hoadley and Galant, 2015). For the present study we selected four schools from the SPADE sample. The schools fall within two of the Western Cape’s eight educational districts. Though the sample is small, in-depth depiction of school-level governance within these schools may be suggestive of broader patterns that prevail in schools located within lower-income communities in the Western Cape. The present study thus provides a framework and direction for the exploration of the findings in a larger sample. The intention in the sampling is to explore positive possibilities, rather than to confirm findings of dysfunction reported widely.

The four schools that constitute our cases were initially selected for the SPADE research based on their performance on the Western Cape Education Department’s (WCED) systemic tests in the early 2000s. Two schools were selected as ‘above average’ performers within their socio-economic profile; both schools had achieved an overall mean for the period that was at least 5 per cent above the predicted value, given their profile. These two schools were matched with two schools that had achieved 5 per cent below the expected value.

Two matched pairs, each with a high- and low-performing school, were thus established, one set in a former mixed-race area (coloured, in the South African vernacular), and the other located in a black township. Within each set are two differentially performing schools situated in the same community, about 2 km apart. Both communities can be described as urban, economically depressed, and affected by a range of social problems, such as violence, substance abuse, absent or young parents, and illiteracy. Each pair is thus similar in demographic composition and general functionality, but with different levels of academic performance.

The first matched pair is located in the settlement of Brandonville,2 approximately 30 km outside the city of Cape Town. The broader community surrounding these schools originated in the late 1980s and is home to 25,364 residents, according to the 2001 national census. Of the total population, 82 per cent are Afrikaans-speaking, 94 per cent are coloured, and 44 per cent (p.205) of the working age population are unemployed (Census, 2001). In the early 1990s the community’s population began to expand, necessitating the establishment of additional primary schools in the area, including School 1 and School A. Local principals were requested to select particular teachers for transfer to these schools. School 1 was established in 1993 and currently serves 1,321 learners, drawn from its immediate community. School A opened in 1995 and provides for 1,204 learners. The majority of School A’s learners reside in the local community, and about 200 of the learners are isiXhosa-speaking. Both schools offer Afrikaans and English as mediums of instruction. The broader community continues to grow steadily, with the construction of local housing projects.

The second matched pair is located in Khayelitsha. Khayelitsha Township was established in 1983, built under the principle of racial segregation executed by the apartheid government. The government envisaged Khayelitsha (meaning ‘new home’) as a relocation point to accommodate all ‘legal’ black residents of the Cape Peninsula in one new, purpose-built and easily controlled township. The government classified people as legal if they had already lived in the area for ten years. Due to the immense influx of people, it is the second biggest black township in South Africa after Soweto in Johannesburg, with a population of between 400,000 and 450,000 people. Khayelitsha is located approximately 35 km outside the city of Cape Town. Residents are 97 per cent isiXhosa-speaking, 99 per cent are black, and about 47 per cent of the working age population are unemployed (Census, 2001). Around 60 per cent of households are classified as shacks, predominantly constructed out of corrugated iron.

The two Khayelitsha Schools are located about 2 km apart. School 2 was established in 2000. The staff was largely made up of teachers who were declared excess in other schools where student numbers had declined. In 2012, School 2 had 1,175 learners. The entire student body is isiXhosa-speaking and the school has had a good reputation in the local community. School B was started in a community centre in 1991, without the formal permission of the provincial education department. In 1993 it was formally opened by the provincial department, a principal formally appointed and teachers paid. In 2012 it had 1,124 learners, all of whom were isiXhosa-speaking, residing locally. Both schools offer isiXhosa (from Grade R to Grade 3) and English (from Grades 4 to 7) as mediums of instruction.

Table 8.1. Systemic Tests—Percentage of Cohort that Meets the Grade 3 Proficiency Standard

School 1

School A

School 2

School B




































Table 8.1 details the annual average (literacy and numeracy) scores obtained by each school for the Western Cape systemic tests between 2002 and 2013.3 The SPADE research identified School 1 and School 2 as relatively high (p.206) performers, and School A and School B as relatively low performers, relative to the median for their relevant demographic cohort. In practice, as the table suggests, the performance patterns turned out to be messier than those intended in the initial research design and school selection processes.

As Table 8.2 highlights, there turned out to be a strong correlation between turnover in leadership and school-level performance over time. (To disguise identities, but facilitate narrative flow in subsequent sections, the table includes a pseudonym for each school’s principal during each period.) All four schools in our sample experienced a turnover in leadership (i.e. the school principal) over the period studied, with noteworthy consequences:

  • In both of the hitherto better-performing schools (Schools 1 and 2) the change in principal resulted in subsequent performance declines.

  • In the relatively low-performing School A, the change in principal was associated with a worsening of subsequent performance.

  • The interaction between performance and leadership in School B is complex; we postpone discussion until section 8.5.

Consistent with the above patterns, the analysis which follows focuses centrally on the interactions between leadership, leadership change, and trends in performance in each of the schools highlighted in Table 8.2.

Table 8.2. Governance Episodes Across Four Schools, 2002–14

Case Studies of School-Level Governance Dynamics in the Western Cape

8.2.2 The Principal Selection Process—De Jure and De Facto

The South African Schools Act of 1996 and the Employment of Educators Act of 1998 specify an elaborate process for the appointment of principals. In Table 8.3, we use the public governance typology introduced in Chapter 1 to characterize the official, de jure (‘ideal’) process of principal selection. As per the discussion in Chapter 1, any specific governance arrangement is likely to be a hybrid combination of the four ideal types defined by the cells; we capture this heuristically in Table 8.3, and in subsequent tables in the chapter, by allocating 100 points across the four cells. As per the table:


  • Looking down the columns, in its official, de jure (‘ideal’) form) the entire process is impersonal—decisions are made following formal rules which apply impartially to all candidates, with no scope for more personalized deals.

  • Looking across the rows, within the broadly impersonal framework there is a mix of hierarchical and negotiated processes. Higher political and bureaucratic levels set the parameters for appointments, provide some resources, and formally are responsible for actual appointments. Much of the actual decision-making is at the lower levels, in a negotiated form involving the SGB (which formally is responsible for principal selection) and the district office of the bureaucratic hierarchy.

We characterize this ‘ideal’ form in the table as 50 per cent impersonal/hierarchical and 50 per cent impersonal/negotiated.

Table 8.3. Governance of Principal Selection—The Policy ‘Ideal’

Appointment of principal x







Source: authors

The Western Cape process for appointing a principal formally involves the following steps:

  • An interview panel is constituted, consisting of the SGB, a district official and a union representative. The district official and the union representative are intended to serve only as observers. The district official observing on the interview panel acts as advisor and representative of the WCED. The district official can call the SGB to order, but cannot make recommendations on their behalf. The SGB may co-opt additional members (p.208) onto the interview committee, should they require additional expertise. Where the SGB in general lacks capacity, it is the district officer’s role to provide support.

  • The School Management Team4 (SMT) defines criteria according to which the SGB assesses applicants.

  • An advertisement is posted by the Western Cape Education Department (WCED). The advertisement contains information relating to: i) key results areas and duties; ii) job description; and iii) job competencies and qualifications. Schools are able to add items to the job description of the principal (according to the school’s needs); however, the addition cannot be inconsistent with higher-level selection criteria negotiated and agreed upon at the national level.

  • The WCED accepts applications, and screens the applications for educators who have misconduct charges against them, those who have been fired, and those who have retired due to poor health. It also indicates which applicants have the relevant qualifications. The applications are then put into a sealed envelope and sent to the school. This is done on the basis of a collective agreement with the unions to ensure no names are added or taken out the envelope.

  • The SGB sets a date when the envelope will be opened. Unions are invited to attend. An initial screening takes place where the SGB shortlists five to six candidates. This list is sent to the WCED. Only established school principals and deputy principals are eligible to apply for principal posts. Currently equity and representation criteria are taken into account, but only at the early stages of the appointment process. At the shortlist stage, expertise, qualification and experience are the primary criteria for selection.

  • The interview committee then conducts interviews. They may give assignments to candidates to complete, and they also make use of competency tests (paid for by the WCED if conducted by an external agency).

  • Once the interviews are completed, the SGB provides a list of three candidates (in order of preference) to the provincial head of education. The provincial head of education makes their final selection from this list, although they are not compelled to select the SGB’s most preferred candidate.

In practice, formal processes may or may not play out in the ways intended by those who write the formal rules. A variety of de facto alternatives are possible, including:


  • A high quality de facto process that follows the de jure rules, with robust developmentally oriented decision-making on the part of the SGB, aligned with the WCED, and resulting in the selection of a well-qualified and committed principal.

  • A process that follows the de jure rules, but that de facto is captured by influential, non-developmental factions—resulting in the selection of a principal who lacks the commitment and/or skill to prioritize good educational outcomes.

  • A contested process, in which conflict among stakeholders entrusted with decision-making responsibilities results in a failure to agree on a candidate.

  • A process where decision-making is inconsistent with the formal rules laid out above—perhaps because school-level stakeholders act outside the formal structures (this could be for developmental or predatory reasons), or perhaps because WCED intervention supersedes the formal rules. (In these instances, a variety of alternative possible outcomes are possible, each paralleling those listed above.)

Understanding which of the above processes of principal selection played out in each of our schools—and why—is a central goal of the present paper. Since these processes do not play out in a vacuum, we also examine the processes of decision-making that prevailed within each of the school in the period preceding the selection of a new principal—as shaped by the organizational culture established by the ‘period 1’ principal.

8.2.3 Research Methodology

Our research method is what George and Bennett (2005) refer to as ‘process tracing’. Process tracing focuses on a very specific set of decisions. It ‘attempts to trace the links between possible causes and observed outcomes’ (p. 4). Often used to test the hypotheses of a theory of causation, process tracing considers the sequence and values of intervening variables in a case ‘to see whether the causal process a theory hypothesizes or implies in a case is in fact evident in the sequence and values of the intervening variables in that case’ (p. 6) The focus in gathering data, then, is on sequential processes within a particular historical case, not on correlations of data across cases. The aim is to achieve ‘high internal validity and good historical explanations of particular cases versus making generalizations that apply to broad populations’ (p. 22).

To learn about these decision processes, semi-structured, in-depth interviews were conducted with a number of key informants in each of the schools. A total of eleven in-depth interviews were conducted across the four schools, (p.210) lasting between two and four-and-a-half hours each. These interviews were supplementary to the in-depth knowledge the researchers already had of each of the schools as a result of prior rounds of interviews and engagement in the context of the earlier SPADE research. Detailed field notes were kept by at least two interviewees on each occasion. These notes were integrated and a comprehensive record of each interview constituted the data for analysis. Responses between different informants were triangulated, and contradictions in accounts were identified, and examined further in subsequent interviews.

The interviews aimed at identifying dominant and influential stakeholders in the school, and mapping stakeholders in relation to the achievement of developmental goals (in the case of the school, improved student learning). In considering ‘multi-stakeholder governance’, we considered those setting the goals of school management and overseeing performance and the recruitment and management of staff in the school (school management team (SMT); school governing board (SGB); district administration). To understand the workings of the accountability channel, we were also interested in the existence of predatory and trumping coalitions and how these played out in the history of the schools. Section 8.3 focuses on these leadership dynamics in Schools 1 and 2, where performance initially was relatively strong. Section 8.4 seeks to account for the consistently weak performance in School A, notwithstanding a shift in leadership. Section 8.5 explores some of the more paradoxical leadership dynamics that underpinned School B’s performance patterns over time.

8.3 Schools 1 and 2—Brittle Strengths

While Schools 1 and 2 had been included in the initial SPADE sample on the basis of their exemplary performance, as Table 8.3 signals both schools have seen their systemic test results decline radically subsequent to 2011. In both schools, a likely contributor to this decline was the replacement of an effective principal with a weak successor. The failure to appoint strong successors did not occur in a vacuum. Some of the reasons may be found in the ways in which, in the earlier period, the successful principals went about the tasks of school governance. So it is there that we begin.

8.3.1 Two High-Performing Principals in Action

Both Schools 1 and 2’s episodes of relatively strong performance were characterized by a disproportionate emphasis on hierarchical modes of governance. Further, as Table 8.4 signals, using the heuristic device introduced earlier, in (p.211) both schools the hierarchical pattern took a very specific form, in which the personalized and impersonal dimensions of hierarchy were wholly intertwined. Each case was characterized by a principal who personally was strongly committed to achieving strong performance in their school—and leveraged the impersonal-hierarchical framework of rules provided by the WCED as a way of safeguarding the educational mission of the school from efforts at capture.

Table 8.4. ‘Intertwined’, Predominantly Hierarchical Governance








Source: authors

For School 1, interviewees attributed its steady and relatively good performance prior to 2013 to the principal at the time, Mr Smit, and his ‘systems’. Smit was well-liked, respected, and extremely vigilant with respect to attendance and latecoming. He monitored teachers and ‘drove performance’. Interviewees reported that:

Mr Smit had a vision for the school…He knew what was going on in the classes and knew the curriculum. Teachers also did not have to fill in forms all the time, compared to now.

Smit concentrated on two issues: the appointment of strong staff, particularly in management positions, and the establishment of strong bureaucratic systems in the school. He sought to promote expertise in those managing the school, and was described as unafraid of challenges to his own authority. He developed and relied on clear systems and principles, which had bite at the level of staff appointment and management, particularly in relation to teacher performance. He set up school-level processes to ensure rigorous appointment processes, co-opting the circuit (WCED) in this regard to support decisions taken in the school. He also established a strong management committee (SMT) within the school, allowing for the establishment of rule-setting. He also addressed staff underperformance. Teachers were dealt with individually, and where problems arose, individual strategies were developed to deal with these. Smit developed strong administrative procedures for all activities in the school, and a filing system that kept a careful record of policies, decisions and processes. He also had a close relationship with the circuit office and with a professional network of teachers in the Brandonville area. According to interviewees, under Smit the SGB appeared to be entirely compliant with respect to the principal’s directives.

In School 2, paralleling Mr Smit, Mrs Komape also laid down explicit rules and procedures for resolving disputes and making decisions. She used these to deal with a number of inherited disputes and contestations around teacher (p.212) contract posts. She also inherited an SGB heavily involved in local politics and with strong influence over the former principal. Komape disciplined the SGB, thwarting a number of attempts by the SGB to capture school funds. In her words: ‘If you create a space for your SGB to mess with you, you will lose control as principal’. Slowly the SGB was brought in line, and co-operated with the rules laid down by Komape.

Mrs Komape actively pursued attempts to make processes transparent. She spent a great deal of energy educating other school actors in legitimate processes and rules. She put in place strict observation of school hours—for both students and teachers; in her words, instilling ‘a culture of diligence’ in the school. She took the same transparent, bureaucratically driven approach to teacher underperformance. ‘Progressive discipline’ as recommended by the Department was followed. The tabling of a systematic record of the teachers’ conduct, as well as regular meetings with the teachers (‘the teachers were welcome to bring their union representatives to the meeting so that they didn’t feel they were being victimized’) provided a systematic basis on which to address underperformance and come to a mutual agreement on an improvement plan for the situation.

Mrs Komape acknowledged that there were risks associated with taking a strong developmental path: ‘You do not know what will happen tomorrow.’ She said it was always a possibility that unions or staff members could use parents to initiate an investigation of ‘mismanagement of school funds’ against a principal. It is for this reason that Mrs Komape used established bureaucratic processes to perform her role. As she put it, she did her tasks ‘according to the requirements of government circulars’. According to her, this limited the points where fault could be found. She considers herself lucky to have never experienced such intimidation.

Again paralleling Mr Smit, Mrs Komape drove a deliberate, merit-based staff appointment process in the school. She invested substantial time and effort in educating the SGB in interview processes, including assisting in preparing questions and suitable answers ahead of time, candidate scoring procedures and minute keeping. Her own records of appointment processes were impeccable, anticipating the possibility of contestation of an appointment. In these ways, she attempted to safeguard the process from capture. While unions were a real threat to rule-bound school-level governance, Mrs Komape argued that they could only have a negative effect if the space was created for them to capture the decision-making processes.

The recruitment of an HOD position offers a striking example of how Mrs Komape leveraged formal rules—including the backing of the WCED hierarchy—to prevail in the face of an attempt at the capture. The position drew the interest of a branch chairperson of the South African Democratic Teachers Union (SADTU). The successful candidate would be required to teach (p.213) music at the school. The branch chairperson was escorted to his interview by his vice chairperson. The union observer selected by the candidate to be part of the interview panel was the SADTU branch treasurer. In anticipation, the principal established one of the interview questions as requiring that the candidate play the melodica and ‘teach’ the panel a musical piece. It became evident that the branch chairperson had never been involved in music. He could neither play the instrument nor read the music piece. He ended up singing the piece incorrectly. Since the criteria and the questions were carefully established, the non-appointment of the SADTU candidate could not be contested.

8.3.2 Principal Succession—Things Fall Apart

In their efforts to achieve results, both Mr Smit and Mrs Komape relied on a combination of charismatic leadership and formal rules. But in neither case was the strengthened governance sustained once they exited. In both cases, things rapidly fell apart.

In 2009, Mr Smit retired from School 1 and Mr Jooste, who had been one of the deputy principals at School 1, was appointed acting principal, and subsequently principal. There were strong indications across the interviews of discontent about Jooste’s appointment. At the time of the interviews, it was clear that the school had become split between those who supported and those who opposed Mr Jooste’s principalship. Several argued that he had not been the best candidate in the application pool for the principal position. Rather, as Table 8.5 illustrates, the outcome appears to have been the result of personalized deal-making involving Mr Jooste, the SGB and circuit-level staff within the WCED.

Table 8.5. Predominantly, Personalized Deal-Driven Governance









Source: authors

According to interviewees, three SGB members were actively courted by Mr Jooste while he was acting principal. He granted rights to one member to sell food on the school property, and supported another’s career progression in the school. In the interviews, it was claimed that Jooste intimidated some members of the SGB and co-opted others, such that they ‘would never go against the principal’.

In addition, Jooste could rely on a historically established professional network in the Brandonville area. Jakobs, the circuit manager, had been a principal at one of the primary schools in the Brandonville area prior to taking (p.214) up the job as circuit manager, and he had been friends with Mr Jooste for years. One interviewee claimed that Jakobs persuaded the SGB to appoint Jooste, another claimed that Jakobs influenced the interview process by ‘assisting’ the interview committee to craft questions that would favour Jooste.

An HOD, as union observer, wrote a report on the appointment process, arguing that two of the external candidates were better qualified for the post. Knowledge of Mr Jooste within the tight teacher professional network in the Brandonville area also prompted the report: ‘We came together from [one of the more established primary schools in the Brandonville area]. I know his record. I know what he is like.’ The report that was submitted by the HOD was never consulted, as this happened only in the case of a dispute, and none was formally declared in this appointment.

Interviews consistently described Mr Jooste’s leadership at School 1 once he became principal as ‘hands off’, taking no action in relation to increasing underperformance and absenteeism of teachers. He was reported to comply strictly with bureaucratic procedure, but without consultation and negotiation with other staff. Relying on the systems and good reputation of the school established by Smit, he was not perceived as contributing to developing the school. Rather, he undermined it by eliminating strong teachers who challenged anti-developmental practices within the school. He was reported to have co-opted both the circuit and the SGB in supporting his decisions in the school. The negative consequences for the educational mission of the school were reflected in the declining test scores shown in Table 8.3.

School 2’s process of principal selection was even more fragmented than that of School 1, along lines suggested by Table 8.6. In 2007, following seven years of strong management, Mrs Komape was seconded to WCED district-level administration to provide governance support across a number of schools. As an interim measure, School 2’s SGB and principal made a decision for the school’s two deputy principals to alternate in performing principal responsibilities on a quarterly basis. The deputies were aligned with the two phases in the school—a female deputy in the foundation phase (Grades R to 3) and a male deputy in the intermediate phase one (Grades 4 to 7). This was a temporary solution. In 2009, Komape was appointed formally at the district and vacated her position at the school.

Table 8.6. Fragmented Governance









Source: authors

Following Mrs Komape’s formal appointment, the SGB held an internal application process for appointing an ‘acting’ principal, and the female (p.215) deputy principal was appointed for the rest of 2009. The appointment process resulted in divisions in the staff, based on vested interests of individuals in line for promotion in the different phases. If the deputy principal from the intermediate phase was appointed as principal, this would open up promotion posts (potentially a deputy and HOD position) in that phase. The same would apply if the foundation phase deputy were to be appointed. A new SGB was appointed in 2010 that supposedly favoured a male candidate.

The conflict between the acting principal and the other deputy worsened, and came to a head when the acting principal (the female foundation phase deputy) reported the situation to the district office. There, she discovered that the male deputy had been compiling a case against her. The case was around alleged mismanagement of funds. The story given by the then acting principal (female deputy) was as follows:

The school’s choir had the opportunity to travel to Joburg for a competition. The SMT went to the school’s book supplier and got 10 per cent of their payment back for books. This was done ‘behind my back’. The SMT then used this money to buy air tickets for their travel.

When I found out, I did not report this, even though I now see I should have. I was afraid that reporting them would have implications for their jobs (‘their bread’). I also felt that I would start a battle that I had no chance of winning.

The district intervened in 2010, in order to address the conflict at the school and drive the appointment of a permanent principal. The acting female head applied for the post and the circuit manager, together with the SGB, managed the appointment process. After the interview process, the SGB declared a dispute, saying they had not been adequately trained to appoint the appropriate candidate. Some interviewees claimed that the SGB had been progressively captured by the male deputy. There were also claims that SADTU had become involved, and that the male deputy, one of the HODs, and the SADTU branch chairperson who had been an observer in this round, were influencing the SGB. The female deputy also stated that there were rumours at the time that the SMT had purchased a cell phone and groceries for the SGB chairperson, in order to receive information on the interviews.

During this process, conflict between the male deputy and the female head worsened, until finally they agreed to the appointment of a new acting principal, who stepped in for two years from 2010 to the end of 2011. The female deputy again applied for the job when it was advertised in 2011, but lost out to an external candidate, Mrs Madolo. In 2012, the new principal took over. From 2009 until 2012, then, a failed principal selection process resulted in School 2 experiencing troubled and disruptive governance and, as per Table 8.3, a collapse in its hitherto exemplary scores on systemic tests. As of (p.216) the time of our interviews, there was little evidence that the new principal had been able to reverse the decline.

8.4 School A—Persistent Low-Level Equilibrium

In School A, as the systemic test scores in Table 8.3 show, things went from bad to worse, with a leadership transition exacerbating rather than reversing an earlier period of relatively weak performance.

The roots of School A’s weaknesses can be traced to the way in which it was started. Interviewees gave two reasons given for teachers coming to the school at its start-up. One was that there were promises of opportunities for promotion. The other was that principals in neighbouring schools used the opportunity to rid themselves of teachers regarded as ‘lazy’ or as ‘troublemakers’. As one interviewee put it, ‘The problem cases landed at [School A]’.

Under the first principal, Arendse, there was a series of contestations around promotion posts at the school. Leaks from selection processes, suspicions of undue influence of the SGB, and relations of patronage were reported across the interviews. For example, in 1996 the appointment of an HOD was contested. The appointment process was carried out a second time and a different person was then appointed. Around 2005, a friend of Arendse was appointed into an HOD position. The process of shortlisting and interviews was undertaken without informing a potential candidate on the staff who had indicated interest in the position.

More generally, interviewees claimed that the best person for the job was not always appointed. Arendse had strong personal connections to his management team. Appointments were made according to family and friendship networks. One member, who had a significant drinking problem, remained in his post despite this problem, as he was a rugby referee who supplied tickets to major games to Arendse. Attempted capture of the SGB by potential appointees to posts occurred regularly. Interviewees pointed out parents’ vulnerability towards influence, given their low literacy levels and poverty. Suggestions that bribes could be paid (though no direct evidence or cases were reported) were made. In sum, Arendse’s approach to leadership was disproportionately personalized, anchored in horizontal deal-making along the lines illustrated by Table 8.5. In Mr Arendse’s case, these deals had little developmental purpose—but rather (paralleling the selection of Jooste as School 1 principal) were predominantly centred around individual objectives.

In 2006, after twelve years of tenure, Arendse was removed following the bringing of criminal charges against him. Following Arendse’s removal, the WCED played a central role in the principal appointment process which followed. It appointed a circuit manager, Mr Damonse, to serve as acting (p.217) principal. Further, because of the school’s history of contested appointments and a dysfunctional SGB, the WCED intervened to oversee the appointment of new deputies and a permanent principal. The WCED organized and chaired the interview process, including some parents in the process.

But the WCED’s direct involvement did little to transform the prevailing culture in the school. School A, like School 1, is in an area characterized by longstanding personalized ties between WCED circuit staff and school staff. Given these ties, the likelihood was high from the first that the new principal would be hired from within School A’s existing staff. At the time of Arendse’s removal, there were two deputies—Poole and Arendse’s nephew. The latter died of a heart attack shortly after the aforementioned criminal investigations of his uncle. A strong relationship developed between Damonse and Poole, which had historical roots, but which strengthened during Damonse’s acting headship, with Damonse mentoring Poole to take over. In sum, School A’s process of principal selection was overwhelmingly personalized—along the lines of Table 8.5, though more hierarchical.

Poole took over the school in 2008. He soon faced problems with the SGB and community over alleged misuse of school funds. An audit was held, and the main ‘troublemakers’ who had instigated the inquiry in the school left. The teachers involved in the incident were eventually charged with inciting and were fined. Two ‘camps’ have endured in the school, those for and those against the principal. In 2010 there was conflict between Poole and the community over the appointment of so-called ‘mommy teachers’—parents who were brought in to supervise classes, given the high teacher absentee rate. Parents blockaded the entrance to the school and demanded the removal of the principal. They demanded to know who approved the employment of temporary educators whom they believed to be unqualified. A new SGB was appointed in 2012, with careful oversight from the principal. From interviews, it appears that the SGB currently functions to rubber stamp the principal’s decisions.

Poole’s management style is described at times as divisive, at other times as autocratic, but never as focused on issues of instruction. Significant problems of teacher absenteeism and large classes remain unaddressed in the school. There is distrust between management and teachers. Of the teachers, the principal says: ‘They mainly come to earn a salary. This is their main driver.’

8.5 Unexpected Resilience—The Case of School B

Compared with the other Western Cape case study schools, School B is an outlier. Its patterns of governance have been participatory and personalized—along the lines illustrated in Table 8.5. But unlike the other instances of (p.218) Table 8.5-style governance, noted earlier, in the case of School B, participatory governance turns out to have been a source of resilience.

The school began as a community centre with seven ‘volunteer’ teachers. It consisted of ten rooms, no blackboards, and each teacher had a class of 160 learners. Mrs Somana began her tenure as the first principal of the school at this time. In 1993, the school was opened formally by the WCED, and they began providing teaching posts and funding.

Somana served as principal of School B for nearly twenty years. Interviewees repeatedly referenced her kind-hearted character towards students and their parents, and her positive impact on the community. Her management style was informal, and oriented towards a culture of ‘looking after one’s own’. Interviewees asserted that during her tenure as principal, the filling of promotion or senior posts (e.g. HOD, deputy principal) did not often follow bureaucratic procedure. In general, external candidates were not appointed. One interviewee put it thus in relation to an advertised HOD post:

Some of the external applicants didn’t attend their interview…they assumed an internal candidate would receive it. Appointments are up to the SGB and external candidates don’t usually receive posts.

Viewed through a lens of rule-boundedness as a desirable pattern of governance, School B’s relatively low scores in the initial systemic tests are unsurprising. Indeed, using the lens of the typology framework, School B’s personalized and participatory governance patterns superficially are similar to the School A pattern. In School A, as we have seen, this pattern was associated with a persistent low-level equilibrium of mediocre performance. But what happened in School B once the initial systemic test results were released was very different, and within two years the school showed significant improvement in its test scores.

These test score gains were attributed in the interviews to the instituting of a number of developmentally oriented strategies: an afterschool programme, NGO involvement, home visits by Somana when learners had been absent, and support structures for orphans and vulnerable children (initiated by a parent). Although tentative, the data suggest that Somana’s personalized leadership, embedded positively in the community, may have provided a ‘floor’ of sorts, constructively responsive at key moments.

Informality also had another consequence. In the later stages of Mrs Somana’s tenure (2006–09) issues of financial mismanagement were brought to light. Teachers began to notice the poor condition of the school (e.g. no toilet paper, leaking taps, etc.); some did not receive salaries; and the prospective Grade R facility was at a standstill. Eventually a service provider and a number of teachers reported non-payment of funds to the WCED. The capturing of school funds threatened the school’s developmental stability. In early 2010 (p.219) the WCED launched a formal investigation. Department officials conducted an audit as well as a formal Whole School Evaluation. A few months later, Mrs Somana submitted a letter to the SGB and WCED for ‘early retirement’. From the interviews, it did not emerge clearly who had been implicated in the financial mismanagement.

In selecting a successor to Mrs Somana, School B’s legacy of strong community involvement and a developmentally oriented SGB turned out to be a source of resilience. Following Somana’s departure, the SGB requested the WCED’s assistance in selecting an acting principal so as not to negatively impact on the school’s performance. Shortly thereafter, the department appointed a ‘caretaker’ principal—a coloured man who was at the time awaiting the outcome of his application for a permanent post elsewhere. An SGB member described the situation thus:

He was a good guy. Things improved, but the teachers had a negative perception of him. They thought he wouldn’t understand the challenges of the school. [The SGB] feared that his life was in danger. You see, he was good, but it was a cultural issue.

These serious threats in the broader school community led to the caretaker principal’s departure; he took a permanent post at another school. The SGB decided to appoint one of the deputies as acting principal. After deliberation amongst parents and the SGB, the most senior or long-standing deputy, Mr Mayila, was appointed in January 2011.

Over the next few months, the permanent principal post was advertised publicly with clear criteria determined by the SGB, which focused on the development of the school. Because Mr Mayila, the acting principal at the time, was a candidate for the permanent position, a local high school principal oversaw the appointment process. Another internal staff member (an HOD) was considered for the post, as well as two external candidates, one male and one female. The primary stakeholders throughout the process were SGB members, the local high school principal, a union representative, and the circuit manager (the latter two as observers only). The process strictly followed the WCED’s established policies. The appointment process was described as ‘harmonious’ and ‘professional’, with ‘no discrepancies’. ‘By the book’ transparent processes (which closely approximated the principal selection ‘ideal’ of Table 8.3) allowed for the two most suitable (high-scoring) candidates to be shortlisted. The WCED made the final decision and offered Mr Rala, an external candidate, the principalship.

8.6 Patterns and Implications

Relative to other provinces in South Africa, public schools in the Western Cape are well-governed, and generally show better results. As Chapter 4 (p.220) detailed, by and large the WCED hierarchy delivers effectively on the things hierarchies are expected to deliver. However, there are continuing challenges to improvement, including the hugely difficult socio-economic setting faced by many children in the Western Cape, a delayed effective curriculum regime, and continuing weaknesses in teachers’ instructional capacity.

Might there also be some ‘micro-governance’ reasons? In this final section, we draw on our school-level case studies to reflect more broadly on the ways in which de facto hierarchical and horizontal governance arrangements might help explain why the effort to improve outcomes continues to be enormously challenging.

8.6.1 Hierarchical Governance

The WCED’s well-functioning hierarchy is an important asset. Getting textbooks delivered; ensuring that teaching posts are filled with teachers who meet a minimum set of criteria; ensuring an optimal balance between personnel and non-personnel expenditure; tracking how schools use resources (including trends in performance); getting funding to the right places at the right times; pro-actively trying to fill leadership positions with the right people for the job—in contrast to many other departments of education in South Africa and elsewhere, the WCED does all of these things well. These are important strengths.

The focus of our research, though, has been on narrower micro-governance concerns. Our interest has been to understand at school-level (along lines suggested by Figure 8.1) both the potential benefits and the limits of a relatively well-performing hierarchy. We have focused especially on the position of principal—both how principals choose to exercise their authority, and the processes of principal selection.

Our case studies identified three distinct ways through which principals exercise their authority—each with different implications as to the influence of hierarchy on school performance:

i. Developmentally oriented governance through top-down leadership, underpinned by rules—illustrated by the leadership styles of Smit in School 1 and Komape in School 2.

Both principals gave strong emphasis to putting in place a framework of rule-boundedness within their schools. In doing so, both benefited hugely from confidence that the rules would indeed be enforced at higher levels of the WCED’s bureaucratic ladder. Both used this platform of credible rules as a key buttress against pressures to act in ways that were inconsistent with the school’s educational mission.


ii. ‘Isomorphic mimicry’—the use of leadership authority to establish a seemingly desirable form (in this case hierarchical governance), but without the substance (accountability for performance) which the form is intended to deliver.

As recent work has explored globally, this pre-occupation with form, rather than the pursuit of concrete development results, is especially prevalent where ‘entities are highly dependent on getting greater legitimacy from external constituencies in which ‘best practices’ are highly defined’ (Andrews, 2013; Andrews, Pritchett, and Woolcock, 2012). In our case studies, School 1 under Jooste, and School A under Poole provide two examples of low-level equilibria of rule-following mediocrity along these lines.

iii. Participatory leadership—in which a principal governs the school by actively fostering a sense of participation and teamwork, underpinned by shared commitment to a framework of rules which supports co-operative decision-making.

While none of the schools in our sample provided an unequivocal example of ‘good practice’ along these lines, the participatory approach through which School B was governed (though without formal rules) provided a partial illustration. (School A1 in the Eastern Cape case study in Chapter 9 provides a good example of how participatory governance can work.)

Note that both (i) and (ii) are wholly consistent with institutional arrangements where schools are embedded within strong organizational hierarchies. Only (iii) requires for its effectiveness the presence of a ‘zone of autonomy’ at school level, which principals potentially can use to motivate teachers.

Given the centrality of school-level leadership in shaping outcomes, the process of principal selection offers one seemingly straightforward way of improving quality. Here our case studies are sobering (although it is important to qualify what follows by noting both that our sample size is too small to serve as a basis for generalization and that, as detailed below in our final sub-section, a variety of recent initiatives are under way within the WCED to improve principal selection). In three of the four cases in this paper (School B was the exception), the process of principal selection turned out to be retrogressive. In both of the initially high-performing schools (Schools 1 and 2), leadership transitions resulted in a clear subsequent decline in performance. In a third (School A), a change in leadership did nothing to disrupt a low-performance equilibrium. Though the specifics of why principal selection was so difficult varied across School 1, 2 and A, the case studies suggest three underlying patterns.

First, a key driver fuelling contestation in all three cases was the presence of in-house candidates for principal (i.e. from the incumbent deputy principals). In the culture prevailing in the schools, length of service and the occupation (p.222) of a particular post are regarded as a natural conduit to promotion. Over 55 per cent of principals nationally are promoted from within schools (Wills, 2015). Further, when an internal candidate is promoted, it opens up a whole set of potential promotion posts below this position. In School 1, the presence of a well-networked internal candidate resulted in complaints (which were never formally followed through on) that better-qualified external candidates were passed over. In School 2, contestation for the top position between two competing deputies resulted in the process dragging on for almost four years.

Second, in two of the cases (both cases were in the Brandonville area, where, as noted, there were close linkages between school staff and officials in the WCED circuit office), the relevant WCED officials appear to have abetted an insider-driven and only partially competitive process. In one case, interviewees suggested that the circuit staff-person steered the SGB interview process towards a preferred, insider candidate. In the other, the circuit staff actively mentored an internal candidate, and then took direct leadership of the interview process, which resulted in the mentee’s selection.

Third, in neither School 1 nor the 2007–10 contestation in School 2 did the SGB function as an impartial judge and overseer, with the best interests of the school at heart. Instead, the SGB became a focal point for lobbying by insider candidates, with multiple allegations from interviewees of efforts by candidates and their supporters to informally influence the SGB decision processes.

8.6.2 Horizontal Governance

As the evidence on principal selection signals, in three of our four case study schools (School B again being the exception), the patterns we observed provided little evidence that horizontal governance played a positive role. On the contrary, in these three schools, SGBs (school governing bodies) more often were sites for political contestation and personalized favour than they were part of the solution (though we feel it necessary to note that, contrary to a familiar narrative, we found very little evidence that contestation and the pursuit of favour were driven by teachers unions).5 On occasion, developmentally oriented principals turned to non-governmental organizations outside their immediate communities for support, but mostly the involvement of these outside organizations was quite superficial. Indeed, in some of our cases, any positive potential of horizontal governance was confounded by the (p.223) strength of predatory influence networks. In such circumstances, neither the motivational nor the informational rationales for horizontal governance arrangements can have much, if any, positive effect.

While we recognize that our sample is small, and thus that our findings could be an artifact of sample selection, broader research (for the specific Western Cape demographic profile which is our focus) suggests that the pattern of the principal driving school performance, with relatively limited constructive input from the SGB, communities, or other non-governmental actors is a more general one (Hoadley, Christie, and Ward, 2009). In relatively affluent and stable communities with high social capital, negotiated governance could indeed be prevalent, and associated with strong performance. (Indeed, in such settings this may be the normatively preferred mode of governance.) However, where social capital is weaker and conflict over resources is acute, the absence of strong hierarchical governance could render a school especially vulnerable to predation. These patterns accord with recent work on management in resilient schools in South Africa: performance is driven from within, without reliance or support from external agents (Chikoko, Naicker, and Mthiyane, 2015).

Yet, for all of these evident weaknesses in horizontal accountability, our research cautions against focusing on hierarchical performance measures to the exclusion of the development of more sustained, horizontal relations between stakeholders at the school and community level. For reasons set out in the introduction to this chapter, the ability of any bureaucracy to exert strong control at the micro level is inevitably limited. Our results point to the real danger that surface compliance, or ‘isomorphic mimicry’, can mask underperformance, making the necessity and means for intervening in a school more opaque. And even where performance is good, insofar as it is dependent on top-down leadership from an incumbent principal, as our case studies of Schools 1 and 2 suggest, the risk of performance reversal is especially acute at moments of succession from one principal to another.

Against this backdrop, the patterns we observed in School B are striking. Though tentative, School B possibly indicates the potential of strong school-community ties to support developmentally oriented decision-making. This relationship, between the school and community, as a ‘floor’ or support for enhanced decision-making has been raised by Hoadley, Christie, and Ward (2009), though they argue that it derives from a support for, rather than direct action in, decision-making processes in the first instance.

8.6.3 Some Policy Implications

The evidence from our case studies raises a troubling dilemma. On the one hand, our results are consistent with a pattern that is evident in many parts of (p.224) the world (Pritchett, 2013)—the reality of dysfunction beneath the surface of seemingly well-organized bureaucratic processes. The difference between a high-performing bureaucracy and ‘isomorphic mimicry’ can be difficult to discern. On the other hand, our results also are consistent with broader research which suggests that, for the specific Western Cape demographic profile which is our focus, the absence of constructive input from the SGB, communities, or other non-governmental actors is the norm, rather than the exception (Lewis and Naidoo, 2004; Karlsson, 2002).

Given these findings, one temptation for policymakers (at least in settings such as the Western Cape, where bureaucratic quality is relatively strong) is to try and ‘double down’—to eliminate performance shortfalls by the introduction of seemingly more and more robust tools of top-down performance management. Our cases suggest the limitations of this.

What, then, is to be done? As discussed further in Chapter 10, we propose pragmatism and incrementalism—foreswearing bold reform initiatives in favour of relatively modest tweaks capable of achieving seemingly small (but potentially far-reaching in their consequences) improvements in the functioning of both hierarchical and horizontal systems of governance.

Our case studies suggest that the developmental returns may be especially high from an intensified focus on the selection of school principals. In the episodes of principal selection examined in our case study schools, neither hierarchical action by the WCED nor participatory engagement by SGBs was able systematically to assure the recruitment and placement of good principals. A better balance between hierarchical and horizontal governance is needed—one which is better able to leverage the strengths of each, while limiting the risks of local capture or of isomorphic mimicry in the face of the inevitable limitations of higher levels of the bureaucracy in accessing local-level information.

Part of the solution may lie in the WCED’s recent intensification of efforts to influence principal selection, detailed in Chapter 4. Our case studies suggest that in settings such as the Western Cape, where a platform of capable bureaucracy is in place, pragmatic managerial interventions along these lines have the potential to yield substantial improvements in the process of principal selection.6 Along with the ongoing intensified focus on putting in place strong, developmentally oriented school leaders, renewed focus on the structure of the relationship between SGB, principal and district (circuit), and what functions they should serve would be helpful.

(p.225) Excluding SGBs entirely from the processes of principal selection may not be an ideal solution. As our case study of School B (and our school-level case studies in the Eastern Cape) suggest, some involvement of SGBs can help limit the risks of capture, while maintaining a floor of support for developmental decision-making. But all too often the current relationships do not work well; and there also is a need for systemic support to enable SGBs to better play their developmental role.

In our view, rather than viewing the interaction between hierarchical and horizontal governance as zero-sum, the task for practitioners is to find ways to make more effective the ‘both/and’ balance, with an emphasis on impersonal forms of decision-making. Our cases have shown that effective hierarchical modes have the potential to create the conditions for fostering local initiative and developmental practice by the school to augment the work of the state. There is also a strong suggestion that informational and other inputs from developmentally oriented local stakeholders have the potential to contribute to the principal selection process—as long as the door is not opened for predatory capture. Finding a better balance is a fundamental challenge for practitioners—but one which, if addressed successfully, appears from our case studies to offer real opportunity for achieving quite substantial short- to medium-term gains in educational outcomes.


Bibliography references:

Aghion, P. and Tirole, J. (1997) ‘Formal and real authority in organizations’, Journal of Political Economy 105, 1–29.

Andrews, M. (2013) The Limits of Institutional Reform in Development: Changing Rules for Realistic Institutions. New York: Cambridge University Press.

Andrews M., Pritchett, L., and Woolcock, M. (2012) ‘Escaping capability traps through problem-driven iterative adaptation’, Working Paper No. 299. Washington DC: Center for Global Development.

Chikoko, V., Naicker, I., and Mthiyane, S. (2015) ‘School leadership practices that work in areas of multiple deprivation in South Africa’, Educational Management Administration Leadership 43(3), 452–67.

George, A. L. and Bennett, A. (2005) Case Studies and Theory Development in the Social Sciences. Cambridge, MA: MIT Press.

Hallinger, P. and Heck, R. H. (1996) ‘Reassessing the principal’s role in school effectiveness: A review of empirical research, 1980–1995’, Educational Administration Quarterly 32(1), 5–44.

Hoadley, U., Christie, P., and Ward, C. (2009) ‘Managing to learn: Instructional leadership in South African secondary schools’, School Management and Leadership 29(4), 373–89.

(p.226) Hoadley, U. and Galant, J. (2015) ‘Specialization and school organization: Investigating pedagogic culture’, British Journal of Sociology of Education 37(8), 1187–210.

Karlsson, J. (2002) ‘The role of democratic governing bodies in South African schools’, Comparative Education 38(3), 327–36.

Leithwood, K., Patten, S., and Jantzi, D. (2010) ‘Testing a conception of how school leadership influences student learning’, Educational Administration Quarterly 44(4), 455–7.

Levy, B. (2014) Working with the Grain: Integrating Governance and Growth in Development Strategies. New York: Oxford University Press.

Lewis, S. G. and Naidoo, J. (2004) ‘Whose theory of participation? School governance policy and practice in South Africa’, Current Issues in Comparative Education 6(2), 100–12.

Lipsky, M. (2010) Street-level Bureaucracy: Dilemmas of the Individual in Public Service. New York: Russell Sage Foundation.

North, D.W. (1990) Institutions, Institutional Change and Economic Performance. New York: Cambridge University Press.

Pritchett, L. (2013) The Rebirth of Education: Schooling Ain’t Learning. Washington DC: Center for Global Development.

Sah, R. K. and Stiglitz, J. E. (1986) ‘The architecture of economic systems: Hierarchies and polyarchies’, American Economic Review 76(4), 716–27.

Wills, G. (2015) ‘A profile of the labour market for school principals in South Africa: Evidence to inform policy’. Working Papers 12/2015, Stellenbosch University, Department of Economics.

Wilson, J. Q. (1989) Bureaucracy: What Government Agencies Do and Why They Do It. New York: Basic Books.


(1) The SPADE (Schools Performing Above Demographic Expectations) project is interested in the factors that account for primary schools in disadvantaged communities performing above expectations. The focus of the SPADE project was on internal governance, pedagogy and home-school instructional practices and their contribution to differential performance.

(2) The name of the area is a pseudonym, given that it is a relatively small area, which may render the schools recognizable. The information, is however, provided for the actual area.

(3) The scores comprise the percentage of children who achieve a passing score, averaging across annual WCED, externally administered literacy and numeracy systemic tests.

(4) The SMT consists of the heads of department, the deputy principals and the principal (or one fulfilling this role).

(5) Although more present as a potential agent in the Khayelitsha context than in the other area studied, in none of our sample schools were unions found to be instrumental in contributing to or predating on school resources. In one case, School 2, attempted capture of the appointment process of an HOD was thwarted by commitment to official procedure.

(6) Note that in settings where bureaucracies are weak and/or captured (which, as Chapter 5 details, is the case in the Eastern Cape), initiatives to strengthen the authority of the bureaucracy in appointing school principals may simply shift the basis of contestation over capture to different terrain, with very uncertain consequences in terms of overall impact.