All for one and one for all? Leadership approaches in complementary schools

Abstract

This article explores the nature of leadership in Brazilian complementary schools in the UK. Such schools are typically parent-driven, voluntary and financially vulnerable. Using data from a questionnaire survey (n=14; more than three-quarters of Brazilian complementary schools) and three in-depth case studies, leadership is examined in relation to five established approaches: directive, instructional, transformational, distributed and collaborative. The study found that the size of the school and the personality of the leader appeared to influence the type of leadership adopted. In terms of effectiveness, a combination of instructional leadership with an approach that motivated staff and volunteers (whether directive, collaborative or transformational, depending on the school’s circumstances) seemed most appropriate to the context of complementary schools. The research illustrates the complexity of school leadership and the overlap between different models. Leadership flexibility was important in responding to the needs of staff, students and parents. The findings are transferable to mainstream schools with contexts similar to those of complementary schools, particularly small primary schools and free schools.

Keywords

Complementary schools leadership styles and models distributed leadership transformational leadership instructional leadership directive leadership collaborative leadership

Introduction

At a time when alternative forms of schooling have been proliferating (Barth, 2013), researchers have been exploring the different ways in which leadership operates in mainstream schools, for example, collaborative leadership (Morrison and Arthur, 2013), democratic leadership (Hope, 2012), system leadership (Higham et al., 2009) or network leadership (Hadfield and Chapman, 2009). This article focuses on the leadership of complementary schools, which number up to 5000 in the UK according to a recent estimate (Evans and Gillan-Thomas, 2015). Many mainstream schools accommodate complementary schools in their classrooms outside normal school hours (Maylor et al., 2010). These schools (also named community and supplementary schools) are voluntary organisations run by migrant groups to support cultural, religious or linguistic/cultural education (Wei, 2006). The Brazilian schools, which are the focus of this research, represent the third type of complementary school, focusing on the linguistic and cultural heritage of migrants from Brazil.

The number of Brazilian complementary schools in the UK grew from one school in 1997 to 18 in 2016 (Souza, 2016), reflecting the increased migration from Brazil since the 1980s. The UK was recently the European country with the highest number of Brazilian migrants (MRE, 2016), with the largest concentration in London (Evans et al., 2015).

The nature of complementary schools presents particular challenges for leaders. These schools are often founded by parents as a means of influencing their children’s education (Francis et al., 2009). Most are staffed by volunteers, usually parents, who may not have appropriate qualifications and who are likely to leave when their children stop attending (Nwulu, 2015). Often staff have other work in addition to teaching at the complementary school (Maylor et al., 2010). Complementary school funding generally comes from pupil fees, and their finances are often precarious (Minty et al., 2008), which impacts on the recruitment, development and retention of staff.

Research into the leadership and management of complementary schools has been limited. Thorpe (2011) investigated Chinese complementary schools in south-east England and found that most of the Chinese complementary school leaders were parents who became involved with the school and did not necessarily have a background in education or educational leadership. The parent leaders saw their roles as dealing with the day-to-day running of the school and liaising with other organisations. Thorpe’s (2011) findings are similar to Souza and Gomes’ (2017) research in two Brazilian complementary schools – one in London and another in Barcelona – where the school leaders were also parents. In the two case study schools, the ability of leaders to motivate teachers was crucial, particularly in relation to curriculum development (Souza and Gomes, 2017). The schools’ stakeholders felt that institutionalising changes was important in order to ensure continuity when key staff moved on. Recent research has identified sustainability as important for complementary school leadership: there is a grave danger that the school may not survive if the head teacher leaves (Thorpe et al., 2018).

Although leadership issues may not be the same in different types of complementary schools, by focusing on Brazilian schools, this article aims to identify aspects of leadership which are likely to be similar to other complementary schools concerned with linguistic and cultural maintenance in the UK and elsewhere (since complementary schools are international phenomena and exist in many countries). After defining ‘school leadership’ the literature review analyses five leadership approaches which seem particularly relevant to Brazilian complementary schools: directive, instructional, transformational, distributed and collaborative. Directive and collaborative models link to the hierarchical and collectivist aspects of Brazilian culture; instructional leadership relates to the schools’ core purpose of learning and teaching; transformational leadership is relevant to the need to inspire and motivate ‘volunteer’ staff and students at complementary schools; distributed leadership is included because of its current importance in school leadership literature. These leadership types are subsequently explored in relation to the data.

Definition of leadership

Educational leadership has been described as ‘the capacity to inspire collective action towards shared goals within an organisation’ (Burbules, 2015: 35). This definition outlines the need for leaders to be able to motivate staff (inspiration) and emphasises the collaborative nature of leadership (‘collective action’ and ‘shared goals’). In this article school leadership will be defined as the ability to inspire staff to achieve shared learning goals in a school.

Approaches to school leadership

Numerous different approaches to educational leadership exist (Morrison, 2009) and a key question for researchers is whether particular types of leadership improve pupil outcomes (e.g. Robinson et al., 2008; Hallinger and Heck, 2010; Shatzer et al., 2014). There are difficulties, however, in isolating a particular leadership approach, such as transformational leadership, as a causal factor for pupil progress. In addition, it is questionable whether the ‘idealised forms’ of leadership types exist in reality (Morrison and Arthur, 2013). School leaders may not exhibit the ‘pure’ version of leadership as described in the literature. Indeed, ‘contingent’ leaders employ a range of different styles that vary depending on the situation and the school context (Bush and Glover, 2014).

Schools differ in relation to their history, community, demographics, location (urban, suburban or rural), structures, staff competencies, beliefs, practices and values (Hallinger, 2018; Noman et al., 2018). It would be surprising then, if a single, universal leadership model were appropriate for every school. There is also evidence that a leader’s style may change – as schools improve, for example, leaders move from a directive approach to become more collaborative (Day, 2009). After reviewing eight different leadership types, Bush and Glover (2014: 565) conclude that ‘most successful leaders are likely to embody most or all of these approaches in their work’.

Leadership styles are also influenced by national cultural contexts (Akkary, 2014; Dimmock and Walker, 2005; Hallinger, 2018), which is of particular relevance to complementary schools, most of which have a different culture to the host nation. While culture is difficult to define (despite the prosaic ‘how we do things around here’), it is generally considered to be collective, emotional, dynamic, based on values and beliefs, and often hidden or assumed, expressed in shared norms, symbols and rituals (Bush, 1998; Kezar and Eckel, 2002, Owens, 2003; Trice and Beyer, 1993).

The identification of cultural ‘dimensions’, initially by Hofstede (1980), has assisted understanding of national cultural differences, although there have been criticisms that Hofstede’s (1980) original research was based on an unrepresentative group (IBM employees), used an inappropriate method (questionnaires) and that globalisation has led to more complex, emerging cultures (Litz and Scott, 2017). Nevertheless, two of Hofstede’s dimensions, later developed by Dimmock and Walker (2005) in relation to educational leadership, are relevant to Brazilian complementary schools. The first is ‘power-distance’, or ‘power-distributed/power-concentrated’ (Dimmock and Walker, 2005), which indicates the extent to which people accept hierarchical structures and inequitable distributions of power. The second is collectivism/individualism, or ‘group-oriented/self-oriented’ (Dimmock and Walker, 2005), which specifies whether it is acceptable for people to prioritise their own needs over those of the group to which they belong. Brazil was identified as having high power-distance, meaning that schools are likely to have strongly hierarchical structures, and relatively low individualism, so that commitment to the group is likely to outweigh individual needs (Hofstede et al., 2010).

National cultural differences have implications for school leadership approaches. For example, distributed, authentic and transformational leadership may be inappropriate in countries with highly centralised, hierarchical education systems (Hallinger, 2018). Litz and Scott’s (2017) research in the United Arab Emirates found that the theory of transformational leadership needed to be adapted for Islamic cultures, while Akkary (2014) noted that managerial, rather than instructional leadership was in evidence in Lebanese schools. As will be seen later, cultural dimensions influenced the choice of leadership approaches to analyse in our research into Brazilian complementary schools.

These complexities lead to a ‘so what?’ question about leadership styles and models. Is there any point in distinguishing between different types of leadership if head teachers use a variety of approaches or change their approach, depending on the school’s improvement trajectory? Even if there is evidence that certain types may improve pupil outcomes – for example, distributed leadership (Hallinger and Heck, 2010), instructional leadership (Shatzer et al., 2014), a combination of transformational and instructional leadership (Day and Sammons, 2013) – there is no guarantee that head teachers can adopt what are considered to be effective styles if that approach does not suit their personality. Different national contexts may also constrain leadership approaches: education policy-makers are urged to resist ‘borrowed’ leadership models (Noman et al., 2018). Moreover, even within the same country, a leadership approach that is effective in one school context may not work so well in another, so head teachers should tailor their strategies to the needs of their school, irrespective of their preferred model of leadership (Day and Sammons, 2013).

Nevertheless, school leadership models ‘help to reflect, and to inform, changes in school leadership practice’ (Bush and Glover, 2014: 553). Knowledge of different types of leadership may assist in identifying ways of leading ethically across a range of different contexts, including complementary schools. Since complementary schools operate across national cultural boundaries – supporting learning within one culture while operating in another – research into their leadership approaches could help to inform conceptions of cross-cultural leadership. In applying theories about leadership models and styles to complementary schools, this article focuses on the five approaches most relevant to Brazilian complementary schools – directive, instructional, transformational, distributed and collaborative – in order to develop a better understanding of school leadership in such schools.

Directive leadership

Directive, top-down or hierarchical leadership is primarily located in the person at the top of the school. This type of leadership is common in South America (Newland, 1995), which might explain why, although Brazil’s educational policies encourage democratic management of schools, involving the school community, there is evidence that participation is limited (Mariano et al., 2016). Directive leadership might be deemed to be culturally appropriate within the complementary schools of the Brazilian community because of its high-power distance (Hofstede et al., 2010; Walker, 2003).

Directive leadership appears to have lost popularity in state schools in England, where the ‘heroic’ leader of bygone days no longer seems appropriate to current educational settings, which have a focus on democratic communities (Gumus et al., 2018). Yet there remains an assumption that the head teacher is the person in control (Bush, 2011) and head teachers are able to exercise considerable power, even if they choose to eschew top-down power over others in favour of power with others that is shared and cooperative (Woods, 2016).

Instructional leadership

Complementary schools exist to support learning and teaching relating to language, religion and/or culture (Wei, 2006). As a result, it might be anticipated that instructional leadership would be the preferred approach.

While all the leadership models in education impact on learning and teaching to some degree (Morrison, 2009), instructional leadership focuses primarily on teaching and learning, with perhaps a greater emphasis on teaching (Bush, 2013). Instructional leadership has been directly linked to school improvement (Bush, 2008; Mooney and Mausbach, 2008; Robinson t al., 2008).

Murphy et al.’s (2016) review of over 1000 articles on school leadership noted that instructional leaders maintain their identities as teachers and are deeply committed to learning and teaching, manifested, for example, in their direct involvement in the classroom, in evaluating teaching, in the professional development of teachers and in extending their own knowledge of learning. Instructional leaders influence teachers’ practices by modelling teaching and learning, providing feedback on lessons and creating an appropriate environment (Heck and Hallinger, 2014; Murphy et al., 2016).

In instructional leadership, the link between leadership, teaching and learning is a linear one: from head teachers to middle leaders to teachers and finally to learners (Morrison, 2009). This top-down approach has led to the criticism that instructional leadership focuses on the principal at the expense of other leaders in a school (Hallinger, 2003).

Transformational leadership

The concept of transformational leadership originated outside schools: Burns (1978) suggested that transformational business leaders motivate their staff to embrace the organisational vision to the extent that it supersedes their personal goals. Leithwood (1994) identified the benefits of transformational leadership in the context of school restructuring and it has since become a popular leadership theory in education (Bush, 2014). Recent research indicates, however, that instructional leadership appears to be more effective than transformational leadership in improving student achievement (Robinson et al., 2009; Shatzer et al., 2014).

Transformational leadership combines vision and direction-setting, charisma and creativity, inspirational motivation, and the ability to respond to individual needs, particularly developing people and re-designing the organisation (Leithwood and Jantzi, 2006; Shatzer et al., 2014). Burns (1978) suggested that transformational leaders need to be able to ‘sell’ their vision to subordinates, but in education there has been more emphasis on creating a shared vision and empowering others (Cummings, 2008). This may be due to the potentially thin line between a charismatic, transformational leader and a manipulative despot.

Complementary schools share a common, transformational aim: to engage students in developing a language, culture or religion that is not mainstream in state schools. Moreover, the schools are often reliant on volunteer staff so the transformational leader’s ability to motivate and inspire is important. There is a danger, however, that the departure of a charismatic complementary school leader leaves a void that is difficult to fill (Thorpe et al., 2018).

Distributed leadership

Distributed leadership is both a highly influential (Harris, 2010) and the most studied (Gumus et al., 2018) leadership model. Leadership is considered to be ‘distributed’ when it is dispersed through a school instead of residing with the head teacher (Gronn, 2010a; Harris, 2010).

Distributed leadership has been associated with the positive attributes of empowering staff, giving them at least a feeling of ownership and developing organisational learning (Hartley, 2010). The involvement in leadership enhances teachers’ self-efficacy and motivation (Day et al., 2009). Distributed leadership may also increase students’ learning (Hallinger and Heck, 2010), but the research evidence has not yet established causal links (Hartley, 2010).

There have been criticisms that distributed leadership focuses on ‘the instrumental tasks and targets set by officialdom’ (Hartley, 2010: 281), without staff playing a significant role in decision-making. School leaders are held accountable for school outcomes, so distributed leadership may make them vulnerable if progress is adversely affected (Harris, 2004). Lumby (2013) argues that distributed leadership may simply be another means of increasing teachers’ workloads without additional pay.

Despite its popularity, it is not clear how far distributed leadership has travelled beyond the US and European contexts. Yet this leadership style has the potential to assist the leaders of complementary schools by sharing leadership tasks and developing skills that could ensure the sustainability of the school if the head teacher leaves.

Collaborative leadership

Alongside a preference for more directive leadership approaches, Brazilian culture appears to value collectivism more than individualism (Hofstede, 2010); in other words, being part of a cohesive group is more important than individual achievement. This might suggest that collaborative leadership styles would be attractive to Brazilian complementary school leaders. Certainly, in Souza and Gomes’ (2017) research study, collaboration was seen as essential to ensure the implementation of innovations in complementary schools.

There are similarities between distributed and collaborative leadership: both could be located on a spectrum of ‘shared’ leadership approaches (Crawford, 2012). In collaborative leadership, teachers are involved in decision-making about curriculum development, and parents and governors participate closely in the development and evaluation of the school’s vision and purpose (Hallinger and Heck, 2010), whereas studies of distributed leadership focus on teachers and rarely include parents or pupils (Woods et al., 2004). Collaborative leadership, then, appears to involve greater empowerment of all stakeholders in the school. Hallinger and Heck (2010) conclude that collaborative leadership makes a positive contribution to school improvement, measured through progress in reading. Their research was limited, however, by the lack of a measure of classroom teaching and the difficulty of proving causal connections in the complex and ever-changing context of schools.

Potential problems with collaborative leadership are that it can be cumbersome and time-consuming, while consensus among staff may not be possible or even desirable (Morrison and Arthur, 2013). Like collegiality, collaboration can be ‘contrived’ (Hargreaves, 1994) by leaders who wish to disguise more hierarchical approaches.

Collaborative leadership may nevertheless offer an attractive model for loose-knit, community-focused, parent-organised complementary schools. Such schools may be strengthened by the shared purpose, participation and empowerment offered by collaborative leadership.

In addition to the commonalities between collaborative and distributed leadership, there are similarities between several other leadership approaches. Transformational leadership, for example, can be distributed through a school, creating a link between distributed and transformational leadership (Leithwood and Jantzi, 2006), and there is substantial overlap between transformational and instructional leadership (Hallinger, 2003). These commonalities seem inevitable, given the complexity of school leadership and the likelihood that leaders display more than one model (Bush and Glover, 2014). Our analysis of the data reflects these complexities, while focusing on the ‘purer’ versions of the five leadership approaches as a means of explaining the findings. The data analysis follows the methodology section.

Methodology

A mixed methods approach was adopted for this research study. This enables triangulation of the data (Onwuegbuzie and Teddlie, 2003) that contributes to reliability and validity. Data was gathered between September 2016 and July 2017, after Oxford Brookes University’s Research Ethics Committee had given ethical approval. Participants were provided with details about the study, such as the purpose, benefits, funding and findings dissemination as well as participants’ confidentiality and withdrawal rights. All participants have been anonymised in this article; pseudonyms have been used for the schools and school leaders.

The study had two phases. In the first phase, an online questionnaire was emailed to the head teachers of all the Brazilian complementary schools in England (17 at the time of the research study). The schools (and their email addresses) were identified from a list available on a public website held by ABRIR, the Brazilian Association for Educational Projects in the UK. The questionnaire included questions on leaders’ activities, challenges, skills and backgrounds. In the second phase, three schools were selected for more in-depth, case study research. The case studies comprised one-to-one, semi-structured interviews with the school leader, focused on their views of their role, activities and challenges, as well as focus group interviews with (separately) teachers and teaching assistants, covering their role in the school, motivation, decision-making and their perceptions of the leader. Triangulation took place through comparing the results of the questionnaire survey with the more detailed picture presented from the individual and face-to-face interviews, as well as the literature on school leadership approaches.

Fourteen questionnaires were completed, representing more than three-quarters of Brazilian complementary schools. Most of the schools were relatively small as indicated in Table 1.

Table 1.

Schools and the number of students.

Band	Number of students	Number of schools
1	0–20	6
2	21–40	5
3	41–60	0
4	61–80	1
5	81–100	0
6	Over 100	2

For the second phase, three of these schools were selected, according to size and location. Thus the case study schools included a large school (from Band 4), a medium-sized school (from Band 2) and a small school (from Band 1). In terms of geographical location, since 53% (i.e. 9 out of 17) of Brazilian complementary schools in the UK are located in London, two of the three case study schools were based in London. Table 2 provides a profile of these schools.

Table 2.

Profile of the case study schools.

School and head teacher pseudonyms	Students	Salary	Staff	School hours	Location
Araponga: Sofia	61–80	Yes	6 teachers 7 TAs	2 hours on Saturday	London
Arara: Isabela	21–40	Yes	1 teacher 2 TAs	2 hours on Mon, Tues, Thurs, Fri	London
Siriema: Eliane	0–20	No	2 teachers 3 TAs (1 per session)	2 hours on Saturday	Out of London

This article focuses mainly on the three case studies. Mindful of the possibility that head teachers’ assertions of their leadership model may be interpreted differently by their staff (Burns, 1954), this research study included separate focus group interviews with teachers and teaching assistants at two of the schools, as well as individual interviews with school leaders. In this way a more holistic understanding of the head teachers’ leadership styles was gained. (At Siriema School, the three teaching assistants contribute on different Saturdays of the month so it was not possible to hold a focus group of teaching assistants at this school.) Thus the trustworthiness of our findings was supported through triangulation of the data from questionnaires and interviews and through comparison of accounts from school leaders and their staff.

The questionnaires were originally designed in English, translated into Portuguese and the results were then translated back into English. Similarly, the interviews were originally designed in English, carried out in Portuguese, transcribed and then translated into English.

Context of complementary schools: Questionnaire findings

As indicated in the literature review, complementary schools are vulnerable, both to the risk of their leader’s departure and the low enrolment of children. Financial resources are limited and many complementary schools rely on volunteers, who may not be easy to manage. In some cases the teachers lack educational qualifications, so pedagogical leadership is vested heavily in the head teacher. Parents play a key role in complementary schools – they are often the founders of the schools and are closely involved in its activities. Children attend voluntarily and so need to be enthused by the learning experience.

Bearing in mind these characteristics, it is perhaps not surprising that the questionnaires indicated the leaders’ main activities were ‘liaising with parents’ (14/14), ‘leading by example’ (13/14), ‘planning regular activities’ (12/14), ‘organising’ (12/14) and ‘safeguarding students’ (12/14). Most of the respondents ticked many of the available options, perhaps because as leaders they are responsible for almost all the activities in the school. When they were asked to rank the activities, however, the highest ranked were ‘leading by example’, ‘leading learning and teaching’ and ‘developing a vision for the school’, which might indicate that instructional or transformational models would be preferred. Respondents were asked about their main challenges and the ones that were ranked highest were ‘leading learning and teaching’ and ‘fundraising’. These activities are not dissimilar from the key dimensions of successful school leadership identified by Day and Sammons (2013), which include vision-building, improving teaching and learning and highlighting common values.

Case study findings

The three case study schools have been named for the purposes of this article Araponga, Arara and Siriema. The largest of these schools is Araponga, which falls in the 61–80 bracket in terms of numbers of pupils. Its head teacher is, for the purposes of this article, called Sofia. (Names of the schools and head teachers have been changed to preserve anonymity.) Arara is a medium-sized school (21–40 pupils), whose head teacher is Isabela, and Siriema is a small school (20 or under pupils), whose head teacher is Eliane. As can be seen in Table 2, Siriema is run on a purely voluntary basis while the other two schools charge fees and pay salaries to the staff. Araponga and Siriema are both Saturday schools, running for two hours every Saturday whereas Arara runs in the afternoon on Mondays, Tuesdays, Thursdays and Fridays. Two of the schools – Araponga and Arara – are based in London; Siriema is outside London.

In relation to fulfilling the definition of school leadership, all three head teachers inspired their staff to achieve shared learning goals, whether these related to the Portuguese language or Brazilian culture. In addition, each head played a key role in their school and without them it seemed unlikely that the school would continue to function; for example, ‘Today I cannot even be off sick. I cannot be absent because it is only me’ (Sofia, Araponga School).

The head teachers of the case study schools demonstrated a range of the models identified in the literature review, but the degree to which they exhibited each model varied, depending on the needs of the school and the background and personality of the head teacher. One style tended to dominate, however, with Sofia having a predominantly directive approach (with some elements of distribution), Isabela transformational and Eliane collaborative. These are described in more detail below, while the extent to which there are common features in complementary school leadership models is considered later in the discussion section.

Sofia: Directive and distributed leadership

Sofia was the head teacher whose personality seemed most suited to top-down leadership. She described her leadership style in directive terms: ‘I love the management side of it.…I love to boss around. I love to call the shots…I command and the others follow my commands.’ The directive style is not necessarily unpopular with Sofia’s staff – as one of the teaching assistants put it, ‘I am very good at being second. I am that kind of person that is behind someone else’ (Araponga Teaching Assistant (TA) 3). As a directive leader, Sofia protected her staff; for example, ‘She fights when there’s a need for fighting – mainly when she needs to defend us’ (Teacher (T) 1). One teacher’s description of Sofia’s personality as ‘determined’ and ‘problem-solving’ (T4) attests to this leadership style.

The hierarchical approach in Sofia’s school was evident in the teaching assistants’ roles. The teaching assistants are not involved in decision-making about the lessons: ‘we have to organise the toys, help the teacher…The only spontaneous thing I do is ask the children if they want to go to the loo’ (Araponga TA5). This was confirmed by Araponga TA2: ‘[The teacher] prepares everything. She decides what she will give during the lesson. She comes and tells me.’

It is clear from the combined views of Sofia, her teachers and the teaching assistants that Sofia is a strongly directive leader, but she was also concerned about both the staff and the students: ‘She looks after us a lot and she looks after the students’ (T1). This reflects a high ‘power-distance’ culture in which hierarchical authority is linked with a paternalistic/maternalistic care for staff (Hofstede et al., 2010). Although Sofia may appear to have ‘power over’ rather than ‘power with’ her staff (Woods, 2016), the warmth of the relationships between them indicates a more complex set of hierarchical relations. Sofia describes, for example, how one of the staff felt able to criticise her: ‘“You fail in this and that. You’re not very well organised. You need to improve.” That’s great! And I say I want to hear much more advice.’ Some of her ideas were not accepted by the staff and had to be abandoned, such as the end-of-term school assemblies, to which the teachers objected: ‘because they had to spend the whole term rehearsing the children and it would affect the lessons. And I had to put my hands up and say: “You’re right”.’

Despite Sofia’s preferred style being directive, she was obliged to distribute responsibility for pedagogical leadership to a pedagogical coordinator, Nilce, because she does not have a background or qualification in teaching. Sofia felt that the pedagogical coordinator might make a better leader than her: ‘she would know what this school needs much more than I do’. She also felt unable to guide teachers on how to improve: ‘I don’t know what makes a good lesson…I have my own personal view but I never know if it is the right one.’

As a result of her lack of teaching skills, Sofia relies on her pedagogical coordinator to lead teaching and learning at the school: ‘And it is Nilce who trains them. Nilce helps them. Nilce offers her own house and they do everything together.’ Nilce described what amounted to a distributed leadership system at the school in relation to the curriculum: ‘the school runs around an annual theme…. So the class project has to be linked to that theme, but each class is free to develop the project they want, with their own objectives, to adapt everything to that class’. T4 agreed: ‘it is very open for the teacher to decide what to work on’. The teachers felt that it was not possible to have a more centrally organised system of teaching because each class had its own identity. One teacher described the benefits of being able to decide what and how to teach: ‘this is also a factor that motivates us because when a person trusts you, you do your job better’ (T3). Nevertheless, Sofia’s directive style impacted on the more distributed approach taken by the pedagogical coordinator. Nilce expressed concern that the other teachers were forced by Sofia to adopt her approach to teaching – she felt that this put her in an awkward position with the other teachers.

Sofia mentioned several times the problem of staff turnover, which impacts on the extent to which she is able to distribute leadership: ‘we are managing to have a certain continuity. But then one might leave, and another, and another. And this kind of knowledge you do not hand on overnight.’ Nilce also mentioned teacher turnover: ‘People come and go and sometimes they don’t know what the identity of the [school] is, and this is important to Sofia, not to lose the identity.’

Thus Araponga School demonstrates an interesting mixture of directive and distributed leadership. The head teacher has considerable power in the school and her preferred style appears to be directive, but in relation to learning and teaching, it is clear that a distributed approach dominates, thanks to the pedagogical coordinator. This type of distributed leadership could not be described as ‘holistic’ (Gronn, 2010b) or ‘emergent’ (Bolden et al., 2009), however, in which leadership develops bottom-up from groups of staff. Rather, pedagogical leadership has been delegated to a single member of staff and she has adopted a distributed style. This does not extend to the teaching assistants: they are clearly at the bottom of the hierarchy and have no influence over what is taught. Yet there are still elements of Woods’ (2016) ‘power with’. The teaching assistants felt that they were treated courteously – ‘everybody treats everyone with respect. This is what makes it great’ (Araponga TA3) – and their voices were heard: ‘Sofia always asks us for our opinion during our meetings’ (Araponga TA5).

Isabela: Transformational leadership

Transformational leadership combines establishing a clear vision, improving the programme of instruction, inspirational motivation, responding to individual needs and empowerment of others (Burns, 1978; Cummings, 2008; Leithwood and Sun, 2012; Shatzer et al., 2014). Isabela’s leadership style could be seen as transformational in her approach to creating a shared vision at the school. Her goal was to move beyond literacy: ‘My job is not to teach the children to speak Portuguese. My job is to go much further.’ One of the two teaching assistants at the school described how she shared the school’s vision: ‘as I joined Arara, I could see Isabela’s project, I was captivated’ (Arara TA2). Isabela hoped to be able to publish her teaching resources, and this was an objective that was shared by her teaching assistants, as Arara TA1 commented: ‘her dream became our dream too’.

Isabela demonstrated the ability to empower others. Arara TA2 felt that she was treated as an equal: ‘She doesn’t treat me as an intern. She treats me as a teacher, a partner.’ She felt that they stood ‘shoulder-to-shoulder’: ‘Shoulder-to-shoulder means working together. She doesn’t want…to work separately. She wants to be together’ (Arara TA2). TA1 said that although the partnership is still at an early stage, Isabela is ‘more of a friend’ who is ‘very open to our ideas’. Arara TA2 was grateful for the opportunity to develop her skills: ‘They should get paid for allowing me to be there, you know? To learn from them.’

Transformational leaders are able to motivate their staff partly through their charismatic personalities (Shatzer et al., 2014). One of Isabela’s teaching assistants described her as follows: ‘Isabela’s name should be “motivation”…she has a good vibe, she is always so positive, she is always winning children’s hearts because she is very charismatic indeed. She is captivating’ (Arara TA2).

Isabela demonstrated her concern for individuals by acknowledging the constraints facing the teaching assistants, who both have other jobs and only limited availability: ‘I’m the one who prepares the activities, supplies the material, gives the content to them and explains how they will work, also, because they don’t have the time.’

This also affects the extent to which Isabela is able to distribute leadership: ‘I’d even like that they’d prepare some content so that they can gain experience, but they work too much, so they have only two hours.’ Arara TA1 described the pressures on her time, as a single parent with a full-time job and acknowledged that she did not have time for lesson planning. TA2 would have welcomed more distributed leadership and to be given more responsibility: ‘I want to be an intern that could be, for example, if Isabela was off, that I could be responsible for the class.’ She recognised that she did not have sufficient skills to do that at present, however.

This indicates two further problems with distributed leadership in complementary schools, in addition to the issue of staff turnover mentioned by Sofia: it may not be possible for the school leader to distribute tasks to staff who cannot commit the time to complete the activity; and the staff may not have the right skills for more responsible roles. Often the leaders of complementary schools are dependent on parent volunteers, who may not have appropriate teaching expertise.

Eliane: Collaborative/collegial leadership

Eliane’s leadership was collaborative. She described her school as ‘communitarian’ and said that she was ‘trying to give responsibility to the others and make everyone see it is a community project’.

One of the reasons why collaborative leadership is important in Eliane’s school is that she works with volunteers. This presents leadership challenges. Eliane has to be ‘very tactful. It is voluntary work and how do you manage a voluntary worker? It is already difficult to manage a paid worker.’ Eliane is not able to recruit the best teaching assistants because it depends on which mothers volunteer. She talked about how to motivate volunteers: ‘It is love for the project, you see? That’s the thing. It’s about writing to people, speaking to people, explaining what it is. And then the person starts helping too.’ She recognised the importance of empathy: ‘I think that my skill is to understand, to put myself into someone else’s shoes.’

Eliane meets with the teaching assistants regularly to plan and monitor the teaching: ‘We do everything together.’ She listens to their ideas: ‘it’s great to meet with them because they bring excellent things…they bring many ideas’. Some do not contribute so much to the meetings, however: ‘I’m not sure if it is because certain people are more creative than others…or if they still don’t feel at ease to speak up.’ The parents also contribute ideas. Eliane described them as, ‘working like we were kind of the same family’.

Eliane sees herself as both teacher and coordinator: ‘I am teaching every week. I am there every week. And the coordination. I am the one who does the biggest part of the coordination.’ Her collaborative approach is partly intended to address the potential problem of succession planning: ‘I said, “Girls, we have to find other teachers somewhere. A real teacher, because I’ll leave one day”.’

Eliane demonstrates all of Hallinger and Heck’s (2010) aspects of collaborative leadership: collaborative decision-making, empowerment of staff, the inclusion of parents, participation in reviewing the school’s aim and purpose. She does not appear to have ‘contrived’ (Hargreaves, 1994) the collegiality at her school. There seems, however, to be a power imbalance. Eliane is the driving force behind the school and the person with teaching skills and qualifications. As a result, while she wishes to take a collaborative approach, she recognises the need to find a ‘real teacher’ to replace her in due course. This raises a potential difficulty with a leader taking a ‘power with’ (Woods, 2016) approach: shared co-operation may be less meaningful if there is an unequal level of qualifications and experience.

Combined school leadership models

Although each head teacher had a dominant leadership style, all of them also demonstrated other models of leadership, particularly instructional leadership. Sofia, for example, claimed not to be involved in pedagogy, but she wanted her school to ‘make learning fun’ and for teachers to adopt an appropriate teaching style for Brazilian children brought up in the UK. This apparently simple aim illustrates a key issue for complementary schools: the children who attend may be used to more active engagement when learning in local schools, and this needs to be taken into account in the different cultural context of the complementary school. Isabela, too, demonstrated aspects of instructional leadership, through modelling teaching and learning to her staff, as indicated by Arara’s TA1, who described how, through Isabela’s demonstration of how to teach, she had absorbed what needed to be done: ‘she gave it to us at the beginning, right? And afterwards, of course, with the daily routine of going there we end up by understanding it.’ Eliane said that the aim for her school is: ‘the children learn how to read and write and learn the Brazilian culture’. Literacy is ‘the central piece…the piece that is the foundation of it all’. She saw her main responsibility as the pedagogical coordination of the school. Thus managing the instructional programme and promoting a positive learning climate – aspects of instructional leadership (Gumus et al., 2018) – were evident in all the three schools.

While the most directive leader was undoubtedly Sofia, the other two head teachers also showed some aspects of the directive approach. For example, Isabela was described as the ‘big boss’ of her school (Arara TA2), which was seen as positive: ‘as a leader she is very peaceful to me…She organises everything’ (Arara TA2). Eliane, too, played a key role in her school and found the ‘huge responsibility’ tiring. While committed to collaboration she also recognised that it was her role to reprimand staff when necessary. This is clearly a more directive, than collaborative, aspect of her leadership. This supports the argument that managerial, directive approaches are deeply embedded within the role of school leader (Cuban, 1988).

Discussion

The previous descriptions indicate a different emphasis from each of the head teachers: directive and distributed; transformational; collaborative. These approaches seemed to be partly dependent on the school context, with a more directive style in the largest school and more collaborative in the smallest. School size was identified as a contextual influence in a survey of principals’ leadership behaviour in Germany: the size of the school had an effect on the value profiles of the principals (Warwas, 2015). Although Warwas’ (2015) research was relatively small scale (involving 56 German schools), our research supports her findings, since in these complementary schools, size seemed to influence the leadership approach adopted.

The overall aim of the school also seemed to influence the leadership model; for example, collaboration was more appropriate in a school intended to be a community project. In addition, the head teacher’s personality appeared to impact on their leadership: Isabela had the charismatic and inspirational personality of a transformational leader, whereas Sofia’s declared bossiness suited a more directive approach. The range of factors that might influence school leadership models identified earlier – from the school’s history to its demographics to the location and national culture (Day, 2009; Dimmock and Walker, 2005; Hallinger, 2018; Noman et al., 2018) – do not include the school leader’s personality, although the charismatic element of transformational leadership is linked to personal characteristics, as are the honesty and integrity of moral and ethical leaders (Gumus et al., 2018). This study indicates that a leader’s personality may play an important role in determining his or her leadership approach.

In addition to the influence of the school’s aims and context – as well as the leader’s personality – on the leadership approach chosen, there was a clear overlap between different leadership models, with instructional and directive leadership evident in all the schools, together with distributed, transformational and collaborative approaches. This supports the contentions that contextual factors play an important part in school leadership approaches (Hallinger, 2018; Noman et al., 2018), that school leaders draw on a range of different models (Bush and Glover, 2014), that leadership practices are culturally contingent (Hallinger, 2018) and that traditional models may need to be adapted to meet the needs of different cultures (Litz and Scott, 2017).

The experience of complementary school leaders raises questions about distributed leadership, which are relevant to mainstream schools. It is clearly difficult for head teachers to distribute leadership in schools where teacher turnover is high and teacher skills are relatively low; for example, in state schools in challenging areas, where a higher proportion of staff are newly qualified (Allen et al., 2012). This was the case for the Arara and Siriema schools where the head teachers were considerably better qualified than the teaching assistants working with them.

The distribution of leadership in the case study complementary schools was undertaken in order to promote the future sustainability of the school – as a means of succession planning and development – or where the head teacher needed to transfer responsibility for activities for which she felt under-qualified. This seems different from the way in which distributed leadership is used in mainstream schools as a means of relieving head teachers from some of their burdens of responsibility while helping to develop teachers’ leadership skills as their careers progress (Day et al., 2009; Hartley, 2010).

Sofia found the greatest challenge of her role was her lack of skills and qualifications in teaching. This prevented her from being able to judge teaching and learning effectively. Some years ago, a report by PriceWaterhouseCoopers (2007) recommended enabling suitably qualified professionals from outside the school sector to take on leadership roles in schools. Sofia’s experience may help to explain why this has not happened. Such leaders are unable to judge teaching and learning from an expert perspective. Yet Sofia clearly had relevant approaches, both to behaviour management and to teaching and learning. Children attend complementary schools voluntarily, so the strong emphasis on making learning fun at Sofia’s school was entirely appropriate to this context.

Conclusion

From the findings presented here, it is difficult to identify how far different leadership approaches are culturally and historically situated, and how far they are due to the nature and context of individual schools. For example, where staff are volunteers or difficult to recruit, key skills are likely to be negotiation, cajoling and motivating others, so styles such as transformational and collaborative may be preferred. The size of a school may make a difference, too, since it is potentially more difficult to achieve collaboration with a larger group of staff; hence a directive style may be more appropriate in a larger school. It is clear that personality also plays a role, with some leaders feeling more comfortable with a collaborative approach while others enjoy the sense of being in command as a directive leader.

Recent research on leadership models has focused on the relationship between different models of leadership and pupil outcomes. This research study was too small to be able to determine whether a particular leadership style enabled better student progress in complementary schools. It does, however, point to the complexity of school leadership, the overlap between different models and the importance of leadership flexibility in responding to the needs of staff, students and parents. The leader’s personality and the school’s context, especially its size, were two factors that seemed to influence preferred leadership models. While it is not possible to conclude that a particular type of leadership is more suited to complementary schools, it is clear that a combination of instructional leadership with a style that motivates staff and volunteers (whether directive, collaborative or transformational, depending on the context) is appropriate to the nature of complementary schools with their fragile staffing, financial vulnerability and dependence on parents.

In light of the overlap between different leadership models, Leithwood and Sun (2012) suggested that there should be more focus on effective leadership practices rather than models. In order to build on our findings, we would recommend further research examining leadership practices and models in relation to a range of contextual factors, including national cultures, school size and the personality of the school leader (although we acknowledge the difficulties of assessing the latter in a research study). It is important to recognise the complexity of national cultures, indicated in this study by two very different models of leadership (directive and collaborative) being accepted within complementary schools from the same national culture. We would also recommend further research into leadership in complementary schools, comparing a broader range of complementary schools representing different nationalities in the UK and also British and other complementary schools abroad. Souza and Gomes (2017) found similar issues faced the leaders of Brazilian complementary schools in London and Spain, but this international dimension could be explored further. Interestingly, one of the case study schools in our study had previously drawn pupils from the wider Portuguese-speaking community, but then divided on national lines, with pupils and teachers from Portugal separating from the Brazilians. With a greater range of schools it would be possible to research leadership approaches in different types of complementary schools and compare these with approaches in both the ‘home’ country and the ‘host’ country, in order to identify possible cultural differences in leadership.

Footnotes

Acknowledgements

The authors are indebted to Professor Graham Butt (Oxford Brookes University), Dr Anthony Thorpe (Roehampton University) and Ms Pascale Vassie (National Resource Centre for Supplementary Education) for their support and very useful advice as members of the Advisory Group for the study. The authors also deeply thank the leaders, teachers and TAs of the Brazilian complementary schools who kindly contributed to this study and to the discussions in this article.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The authors are very grateful to BELMAS (the British Educational Leadership Management and Administration Society) for their funding of the Leadership development and challenges in complementary schools study (RDG/16/16), on which this article draws.

ORCID iD

Linet Arthur

References

Akkary

(2014) The role and role context of the Lebanese school principal: toward a culturally grounded understanding of the principalship. Educational Management Administration and Leadership 42(5): 718–742.

Allen

Burgess

Mayo

(2012) The Teacher Labour Market, Teacher Turnover and Disadvantaged Schools: New Evidence for England. Bristol: The Centre for Market and Public Organisation.

Barth

(2013) The time is ripe (Again). Educational Leadership 71(2): 10–16.

Bolden

Petrov

Gosling

(2009) Distributed leadership in higher education. Rhetoric and reality. Educational Management Administration and Leadership 37(2): 257–277.

Burbules

(2015) What is the relationship between authority and educational leadership? In: Griffiths

Portelli

(eds) Key Questions for Educational Leaders. Ontario, Canada: Word and Deed Publishing, pp. 35–38.

Burns

(1978) Leadership. New York: Harper and Row.

Burns

(1954) The directions of activity and communications in a departmental executive group. Human Relations 7(1): 73–97.

Bush

(1998) Organisational culture and strategic management. In: Middlewood

Lumby

(eds) Strategic Management in Schools and Colleges. London: Paul Chapman Publishing, pp. 32–46.

Bush

(2008) Leadership and Management Development in Education. London: Sage.

10.

Bush

(2011) Theories of Educational Leadership and Management. London: Sage.

11.

Bush

(2013) Instructional leadership and leadership for learning: Global and South African perspectives. Education as Change 17(1): 5–20.

12.

Bush

(2014) Instructional and transformational leadership: Alternative and complementary models? Educational Management Administration and Leadership 42(4): 443–444.

13.

Bush

Glover

(2014) School leadership models: What do we know? School Leadership and Management 34(5): 553–571.

14.

Crawford

(2012) Solo and distributed leadership: Definitions and dilemmas. Educational Management Administration and Leadership 40(5): 610–620.

15.

Cuban

(1988) The Managerial Imperative and the Practice of Leadership in Schools. Albany, NY: SUNY Press.

16.

Cummings

(2008) Only Connect: Using a Critical Incident Tool to Develop Multi-Agency Collaboration in Two Children’s Centres. Nottingham: NCSL.

17.

Day

(2009) Capacity building through layered leadership: Sustaining the turnaround. In: Harris

(ed) Distributed Leadership. Dordrecht: Springer, pp. 121–137.

18.

Day

Sammons

(2013) Successful Leadership: a review of the international literature. Reading: CfBT Education Trust.

19.

Day

Hopkins

Harris

Ahtaridou

(2009) The impact of school leadership on pupil outcomes. Final report: Department for Children Schools and Families. Available at: https://dera.ioe.ac.uk/11329/1/DCSF-RR108.pdf (accessed 16 May 2020).

20.

Dimmock

Walker

(2005) Educational Leadership: Culture and Diversity. London: Sage.

21.

Evans

Gillan-Thomas

(2015) Complementary Schools - Descriptive analysis of Complementary School Pupils’ Characteristics and Attainment in Seven Local Authorities in England, 2007/08—2011/12. London: Paul Hamlyn Foundation.

22.

Evans

Dias

Martins

Souza

Tonhati

(2015) Diversidade de Oportunidades: Brasileir@s no Reino Unido [Diversity of Opportunities: Brazilians in the UK]. London: GEB/Goldsmiths/Queen Mary/Oxford Brookes.

23.

Francis

Archer

Mau

(2009) Language as capital, or language as identity? Chinese complementary school pupils’ perspectives on the purposes and benefits of complementary schools. British Educational Research Journal 35(4): 519–538.

24.

Gronn

(2010a) Where to next for educational leadership? In: Bush

Bell

Middlewood

(eds) The Principles of Educational Leadership and Management. London: Sage, pp. 70–85.

25.

Gronn

(2010b) Leadership: Its genealogy, configuration, and trajectory. Journal of Educational Administration and History 42(4): 405–435.

26.

Gumus

Bellibas

Esen

Gumus

(2018) A systematic review of studies on leadership models in educational research from 1980 to 2014. Educational Management Administration and Leadership 46(1): 25–48.

27.

Hadfield

Chapman

(2009) Leading School-based Networks. London: Routledge.

28.

Hallinger

(2003) Leading educational change: reflections on the practice of instructional and transformational leadership. Cambridge Journal of Education 33 (3): 329–352.

29.

Hallinger

(2018) Bringing context out of the shadows of leadership. Educational Management Administration and Leadership 46(1): 5–24.

30.

Hallinger

Heck

(2010) Leadership for learning: Does collaborative leadership make a difference in student learning? Educational Management, Administration and Leadership 38(6): 654–678.

31.

Hargreaves

(1994) Changing Teachers. Changing Times: Teachers’ Work and Culture in a Postmodern Age. London: Cassell.

32.

Harris

(2004) Distributed leadership and school improvement: Leading or misleading? Educational Management and Administration 32(1): 11–24.

33.

Harris

(2010) Distributed leadership: Current evidence and future directions. In: Bush

Bell

Middlewood

(eds) The Principles of Educational Leadership and Management. London: Sage, pp. 55–69.

34.

Hartley

(2010) Paradigms: How far does research in distributed leadership ‘stretch’? Educational Management, Administration and Leadership 38(3): 271–285.

35.

Heck

Hallinger

(2014) Modeling the effects of school leadership on teaching and learning over time. Journal of Educational Administration 52(5): 653–681.

36.

Higham

Hopkins

Matthews

(2009) System Leadership in Practice. Maidenhead: Open Univesity Press.

37.

Hofstede

(1980) Culture’s Consequences. London: Sage.

38.

Hofstede

Minkov

(2010) Cultures and Organizations: Software of the Mind. London: McGraw-Hill.

39.

Hope

(2012) Small and perfectly formed? Is democracy an alternative approach to school leadership? School Leadership and Management 32(3): 291–305.

40.

Kezar

Eckel

(2002) The effect of institutional culture on change strategies in Higher Education. The Journal of Higher Education 73(4): 435–460.

41.

Leithwood

(1994) Leadership for school restructuring. Educational Administration Quarterly 30(4): 498–518.

42.

Leithwood

Jantzi

(2006) Transformational school leadership for large-scale reform: Effects on students, teachers, and their classroom practices. School Effectiveness and School Improvement 17(2): 201–227.

43.

Leithwood

Sun

(2012) The nature and effects of transformational school leadership: A meta-analytic review of unpublished research. Educational Administration Quarterly 48(3): 387–423.

44.

Litz

Scott

(2017) Transformational leadership in the educational system of the United Arab Emirates. Educational Management Administration and Leadership 45(4): 566–587.

45.

Lumby

(2013) Distributed leadership. the uses and abuses of power. Educational Management, Administration and Leadership 41(5): 581–597.

46.

Mariano

Costa e Silva

(2016) Brazil: leadership in Brazil. In: Arlestig

Day

Johansson

(eds) A Decade of Research on School Principals: Case studies from 24 countries. London: Springer, pp. 445–470.

47.

Maylor

Glass

Issa

, et al. (2010) Impact of Supplementary Schools on Pupils’ Attainment: An Investigation into what Factors Contribute to Educational Improvements. London: Department for Children, Schools and Families.

48.

Minty

Maylor

Issa

, et al. (2008) Our Languages: Teachers in Supplementary School and their Aspirations to Teach Community Languages. London: Institute for Policy Studies in Education, London Metropolitan University.

49.

Mooney

Mausbach

(2008) Align the Design: A Blueprint for School Improvement. Alexandria: Association for Supervision and Curriculum Development.

50.

Morrison

(2009) Leadership and Learning: Matters of Social Justice. Charlotte: Information Age Publishing (International perspectives on curriculum).

51.

Morrison

Arthur

(2013) Leadership for inter-service practice: collaborative leadership lost in translation? An exploration. Educational Management Administration and Leadership 41(2): 179–198.

52.

MRE (2016) Estimativas Populacionais das Comunidades Brasileiras no Mundo [Estimate of Brazilians around the World], Brasília: Itamaraty Available at: http://www.brasileirosnomundo.itamaraty.gov.br/a-comunidade/estimativas-populacionais-das-comunidades/Estimativas%20RCN%202015%20-%20Atualizado.pdf (accessed 2 January 2020).

53.

Murphy

Neumerski

Goldring

Grissom

Porter

(2016) Bottling fog? The quest for instructional management. Cambridge Journal of Education 46(4): 455–471.

54.

Newland

(1995) Spanish American elementary education 1950-1992: bureaucracy, growth and decentralisation. International Journal of Educational Development 15(2): 103–114.

55.

Noman

Awang-Hashim

Shaik-Abdullah

(2018) Contextual leadership practices: The case of a successful school principal in Malaysia. Educational Management Administration and Leadership 46(3): 474–490.

56.

Nwulu

(2015) Beyond the School Gates: Developing the Roles and Connections of Supplementary Schools. London: Royal Society of Arts.

57.

Onwuegbuzie

Teddlie

(2003) A framework for analyzing data in mixed methods research. In: Tashakkori

Teddlie

(eds) Handbook of Mixed Methods in Social and Behavioural Research. Thousand Oaks: Sage Publications.

58.

Owens

(2003) Organizational behavior in Education: Leadership and School Reform. Boston MA: Allyn and Bacon.

59.

PriceWaterhouseCoopers (2007) Independent Study into School Leadership. London: Department for Education and Skills. Available at: https://webarchive.nationalarchives.gov.uk/20090323025702/http://www.dcsf.gov.uk/research/data/uploadfiles/RR818A.pdf (accessed 2 January 2020).

60.

Robinson

Hohepa

Lloyd

(2009) School Leadership and Student Outcomes: Identifying What Works and Why. Best Evidence Syntheses Iteration (BES). New Zealand: Ministry of Education [online] Available at: http://www.educationcounts.govt.nz/publications/series/2515/60169/60170 (accessed 2 January 2020).

61.

Robinson

VMJ

Lloyd

Rowe

(2008) The impact of leadership on student outcomes: An analysis of the differential effects of leadership types. Educational Administration Quarterly 44(5): 635–674.

62.

Shatzer

Caldarella

Hallam

Brown

(2014) Comparing the effects of instructional and transformational leadership on student achievement: Implications for practice. Educational Management, Administration and Leadership 42(4): 445–459.

63.

Souza

(2016) Is Brazilian Portuguese being taught as a community or heritage language? Language Issues 27(1): 21–28.

64.

Souza

Gomes

(2017) Innovations in the teaching of Portuguese as a heritage language: The case of Brazilian complementary schools in London and in Barcelona. In: Kagan

Carreira

Chik

(eds) A Handbook on Heritage Language Education: From Innovation to Program Building. London: Routledge.

65.

Thorpe

(2011) Leading and managing Chinese complementary schools in England: leaders’ perceptions of school leadership. Paper presented at the British Educational Research Association Annual Conference, Institute of Education, University of London, 6-8 September.

66.

Thorpe

Arthur

Souz

(2018) Leadership succession as an aspect of organisational sustainability in complementary. Schools in England Leading and Managing 24(2): 61–73.

67.

Trice

Beyer

(1993) The Cultures of Work Organizations, Englewood Cliffs: Prentice-Hall.

68.

Walker

(2003) School leadership and management in the Asia Pacific region. In Keeves

Watanabe

(eds) The Handbook of Educational research in the Asia-Pacific Region. Dordrecht: Kluwer, pp. 973–987.

69.

Warwas

(2015) Principals’ leadership behaviour: values-based, contingent or both? Journal of Educational Administration 53(3): 310–334.

70.

Wei

(2006) Complementary schools, past, present and future. Language and Education 20(1): 76–83.

71.

Woods

(2016) Authority, power and distributed leadership. Management In Education 30(4): 155–160.

72.

Woods

Bennett

Harvey

Wise

(2004) Variabilities and dualities in distributed leadership. Educational Management Administration and Leadership 32(4): 439–457.