Deputy heads—leadership and power in change?

ABSTRACT

Much current research emphasizes the importance of leadership as an interactive activity in leading school change and improvement. In the Norwegian context, there is growing interest in redesigning the historical hierarchical leadership structure, with a single school leader at the top to a more distributed leadership model that includes several middle leader levels, including deputy heads. While research has been conducted on middle leaders and distributed leadership as an interactive activity, few empirical studies to date have investigated how changing leadership structures affect the relation between leadership and power and how this situation can influence deputy heads’ room for maneuvering. In this article we use data from a Norwegian case to investigate deputy heads’ increased responsibility toward leading school change and improvement and how these changes influence questions of power. The analysis indicates that different forms of power influence deputy heads’ room for maneuvering.

Introduction

In this article, we explore the position of deputy heads, how they create their own room for maneuvering as leaders from the middle. The middle leader position has been referred to by many concepts, Paranosic and Riveros (2017), in international literature often conceptualized as teachers complementary to, but not the same as senior school management (Grootenboer et al., 2019), where deputies are considered part of the senior leadership group (Gurr et al., 2013). Others have included deputy heads as middle leaders, something which shows that it is a fluid concept and bound by context and perspective (De Nobile, 2018). The term Assistant principal is often used in U.S.A., department head in Canada, middle leader/middle manager in Australia, Singapore, New Zealand and deputy headteacher in UK (Lillejord & Børte, 2018). In the Norwegian context deputy heads function both as department heads, part of the senior leadership group, and they often teach students. They are often described as middle leaders, situated below the principal in the hierarchical structure. We view school leadership as a distributed interactive activity with various impacts on school change and improvement (Spillane, 2006).

Influenced by international trends reflecting a growing acknowledgment of the importance of middle leadership, the Norwegian context has witnessed interest in redesigning the historical hierarchical leadership structure with a single principal on the top, to a more distributed leadership structure. This structure includes a middle management level of deputy heads which is redesigned from a traditional administrative function toward increased responsibility for leading development and change and for parts of staff. The redesign from a traditional administrative support function for the principal toward a leader role responsible for school development and change and for leading teachers’ professional learning would involve a role with trust (Paulsen, 2021) and power (Sørhaug, 1996).

Over the past 60 years, a large body of research has concluded that school leadership has the potential to positively affect change and implement improvement at schools (Day et al., 2009; Robinson et al., 2009). From a historical perspective, the educational administration (EA) knowledge base has grown dramatically since 1960, with an increasing geographic diversity occurring over the past two decades. In a bibliometric review of research on EA, Hallinger and Kovačević (2019) identified a paradigm shift from ‘school administration’ to ‘school leadership’ during the previous six decades. The researchers also identified ‘leading school change and improvement’ (LSCI) as a key theme within leadership for student learning, and they identified development as the ‘cognitive anchor’ of the intellectual structure. Another significant finding of their study was that geographic diversity in research had led to increased interest in how context influences leadership.

In an overview of research on what characterizes leadership in the development of schools, Aas et al. (2021) searched a selection of journals for peer-reviewed articles published between 2010 and 2020, which yielded 2,628 hits for school leadership development. In addition, a hand search yielded 30 hits in three Nordic journals that had their own special issues on school leadership. After a thorough quality and relevance assessment, 45 articles were read in full. Based on the articles included in their study, the consensus seems to be that the hallmark of leading school development is that leadership can be characterized as an interactive activity. Aas et al. (2021) categorized the articles into five leadership categories representing various characteristics of leadership, new leadership tasks, and various leadership expectations. A consistent finding showed a clear shift in the direction of schools being developed through collective learning and development processes, and that leading the collective processes appeared to be the most prominent leadership challenge.

So far, much research has been conducted on distributed leadership, which is based on leadership as an interactive activity, but few empirical studies have investigated how changing leadership structures affect the relation between leadership and power for deputy heads. In this article we use Norwegian data to investigate how deputy heads experience these changes. Together with the principal and assistant principal deputy heads form the senior leadership team (Brandmo et al., 2021; Hallinger & Heck, 2010; Hjertø & Paulsen, 2017). A distributed perspective on leadership, which is often connected to middle leadership, has gained favor (Liljenberg, 2016). By understanding leadership as an interactive activity (Aas et al., 2021) that contributes to changes in leadership practice, we build on a distributed perspective on leadership, where the term ‘leader plus’ involves structures, functions, and design, while the term ‘leader practice’ describes the interactions that play out within a given social context (Spillane, 2006). The ‘leader plus’ aspect directs our attention toward the formal role, such as the redesigned deputy head role from mainly administration tasks toward an emergent educational leader role (Muijs & Harris, 2003) having responsibility for staff and leading change and improvements in school, and which will influence how leadership practice unfolds. Redesigned roles implies that something will change and develop, and in this picture questions of power is of interest to investigate. The ‘leader practice’ aspect directs our attention toward the leadership practice that unfolds between deputy heads and the principal and teachers, and for investigating deputy heads’ room for maneuvering. We understand power as capacities in person and institutions which make people do things they otherwise perhaps not would have done in social situations (things, ideas, language, actions, structures and processes (Sørhaug, 1996) and use Lumby’s (2013) framework of power to explore how these changes influence questions of power.

Our study’s research question is as follows:

How do Norwegian deputy heads experience leading and room for maneuvering within changing structures towards leading school change and development?

In the following, we present a review of school leadership for change and deputy role before outlining the study’s theory and methods. We then present, analyze, and discuss the findings. Finally, we draw conclusions and point to several implications.

Literature review

Leadership from a distributed and democratic perspective

Leadership as an interactive activity reflects a distributed perspective on school leadership and a research interest in studying middle leaders at different levels. The concept of distributed leadership is sometimes described as a characteristic of democratic leadership. Harris (2013) argues that in Scandinavian countries, distributed leadership is typically associated with more democratic school forms. Woods (2004), however, emphasizes that democratic leadership is only part of distributed leadership practices. Equating these concepts obscures the relationship between a democratic commitment and, in contrast, the apparent lack of trust from a governance perspective, as evidenced by the increase in control of schools (Hatcher, 2005). As the ten articles in Aas et al. (2021) review in the category ‘democratic leadership’ show, when the expectations of stronger accountability meet the democratic traditions of the Nordic countries, several leadership dilemmas arise. With reference to the various leadership dilemmas that Aas et al. (2021) highlighted, one commonly repeated feature appears to be that the democratic leadership tradition is based on the idea of a short distance between leader and teachers as well as communication characterized by respect, trust, and recognition. Democratic leadership is often linked to the concept of the Nordic leadership profile, a concept that tries to capture the special characteristics of school leadership in the Nordic countries. Based on these articles, a Nordic leadership profile seems to combine responsibility and demands at the system level with a democratic leadership based on equality and cooperation between school leaders and teachers (Aas & Törnsén, 2016).

Middle leaders – servants for the principals?

Leading schools is a complex subject, and considerable research has been conducted on middle leaders at different levels, spanning from the early 2000s up to the present. Middle leaders at different levels in schools are increasingly acknowledged as important figures for leading instructional work and for student learning, and they have experienced the dual pressures of accountability and responsibility from the levels below and above them (Bennett et al., 2007). Bearing an increasing responsibility for LSCI has resulted in the redesign of middle leadership roles in many contexts, such as within areas including strategic, supervisory, staff development, and administrative roles and toward a student-focused role (De Nobile, 2018). Researchers have concluded that middle leaders at different levels matter: they are important connectors to structural ideas about leading from the middle, they build strong professional learning, and they have a positive influence on teachers’ classroom practice (Bennett et al., 2007; Harris & Jones, 2017). The middle leader position of department heads can contribute to build relationships with teachers and other departments, and to have influence in collaboration with other school leaders and the principal (Leithwood, 2016). Deputies play an essential role in a school’s improvement process (Tahir et al., 2019), and department heads in well-functioning departments contributions toward improvements in teaching and student learning are likely greater than the influence of the principal and schools as a whole (Leithwood, 2016). This increased responsibility has led to a growing number of teacher middle leaders internationally (Gurr et al., 2013).

At the same time that middle leaders are increasingly considered to be important, research also shows that they are an underutilized as leaders (Harvey, 1994; Leithwood, 2016; Muijs & Harris, 2003). They lack clear job descriptions, and their work often consists of delegated tasks dependent on the principal’s needs (Lillejord & Børte, 2018). In addition, they are expected to help teachers in facilitating practices (Abrahamsen & Aas, 2019). Leithwood (2016) argues that one way to increase the potential for middle leaders’ leadership capabilities is for principals to provide structures for sharing decision-making and for providing considerable responsibility to department heads, in other words to provide room for influence and power for deputy heads in the senior leadership team.

The middle leader role and practice develop differently in different cultures and traditions, and we often overlook the importance of context in understanding variations in responsibility and difficulty in defining middle leaders’ roles, tasks, and responsibilities (Busher et al., 2007; Lipscombe et al., 2021). No organization can be understood properly without its wider social and cultural context being included (Scott, 1995). Nordic school leadership has long been situated within a context focused on equity and democratic participation (Imsen et al., 2017; Oftedal Telhaug et al., 2006). In recent decades, such leadership has also been influenced by neoliberal reforms in terms of decentralization, local autonomy, accountability, effectiveness, and centralized standards for assessment (Aas & Törnsén, 2016; Krejsler & Moos, 2021) as well as an emphasis on ‘what works’ in schools (Gunnulfsen & Møller, 2021).

Redesigning leadership – a question of power?

Redesigning the middle leader role in the Norwegian context is part of this dual discourse of school leadership, with increasing responsibility and accountability and a belief in school leaders’ importance for leading change and improvement. Research on the redesigned middle-leader-level deputy heads in the Norwegian context has revealed that room for maneuvering can expand within tensions between influence, support, and control and can connect middle leadership with an instructional leadership role, distributed leadership, and questions of power (Abrahamsen, 2018; Abrahamsen & Aas, 2019). Norwegian school leaders are generally expected to be explicit leaders while at the same time distributing leadership within a context with considerable autonomy and trust in an environment of workplace democracy with a ‘flat’ hierarchy (Abrahamsen & Aas, 2019). By increasing the deputy head’s responsibility for improvement work and for staff, the assumption is that such change will influence the processes in the leadership team, and that the previous ‘solo principal on top’ scenario is changing toward a ‘leadership team on top’ scenario, although the principal still remains the overall strategic and responsible leader.

Whereas research on middle leaders has been connected to a distributed perspective on leadership (De Nobile, 2018; Grootenboer, 2018; Lárusdóttir & O’Connor, 2017), researchers have also criticized the distributed perspective on leadership for having less concern with questions of power (Hall, 2012; Lumby, 2013). Diamond and Spillane (2016), with reference to Weber, argue that leadership practice is often rooted in influence relations involving belief in the legitimacy of the leader and voluntary compliance. Others argue that the concept of leadership itself has a power function, and the acceptance of leaders as a concept provides leaders with enhanced agency (Courtney et al., 2021) and more power or authority to have their will enacted in the organizational hierarchy (Heffernan, 2018). Lumby (2013) claims that questions of power must be addressed, because leadership will inevitably include a focus on power, since organizations are fields of power, and power relations are always at work in decision-making and information-generating.

In Fivelsdal and Østerberg (2000) Weber distinguishes between power tied to the personal characteristics of individuals or groups, and authority linked to social positions. The individual school leader with gifts and traits who has a formal leadership role can exercise legitimate power and authority. But power is also insolubly attached to the concept of trust, and organizing is based on power and trust, both of which assume and threaten each other (Sørhaug, 1996). The question of trust thus is important when investigating relations of power in the redesigned middle leader role within a frame of distributed perspectives.

This review shows that considerable research has been conducted on middle leaders’ roles, functions, vertical and horizontal structuring, tasks, and responsibilities, however less research has focused on deputy heads’ room for maneuvering within a distributed leadership perspective and on questions of power – something this article aims to respond to. Inspired by a three-dimensional power analysis (Lukes, 2021; Lumby, 2013) and by Scott’s (1995) three pillars of institutions in the present article we aim to investigate redesigned deputy heads’ room for maneuvering in the Norwegian context, from mainly having administrative roles and functions toward being middle leaders with responsibility for leading teachers’ professional learning, improvement work, and change within a distributed perspective on leadership. Our purpose is to contribute with an analysis of middle leaders’ room for maneuvering beyond the Norwegian context and toward broader international research on different middle leaders’ levels.

Theoretical framework

An institutional view of school leadership

The concept of the institution carries many approaches and different foci, and the literature on institutions varies in relation to different researchers’ elements of interest. Building on Scott’s work (1995), institutions are carriers of cultures, structures, and routines that provide stability, meaning, and legitimacy to social relations and situations: a condition that reflects congruence with rules and laws, normative support, or cultural alignment. Scott (1995) describes this scenario through the regulative, normative, and cognitive pillars. The regulative pillar, which emphasizes conformity to legal systems as a basis of legitimacy, is generally understood to be the school’s organizational infrastructure, such as when a middle leader role is redesigned from administrative toward improvement work. The normative pillar is typically understood to be legitimacy through morals or norms, such as how leadership should be performed when redesigned toward change and improvement work. The cognitive pillar focuses on how meanings and ideas are constructed in meeting redesigned middle leader roles (Scott, 1995). In this article, we assume that the structural change of the deputy head role will lead to a cognitive and normative change as the role develops through middle leaders’ purposive actions (Lawrence & Suddaby, 2006).

Three dimensions of power

In investigating deputy heads’ room for maneuvering connected to a redesign of their role, we are inspired by Lumby’s (2012) three-dimensional power analysis connected to a distributed perspective on leadership. Leaning on Lukes’s work (2021), one-dimensional power, or episodic agency, describes observable and specific episodes where power is exercised by an individual who can make another person or group of people do something or to prevent them from action (Lukes, 2021). Principals who make solo decisions in the leadership team and expect the deputy heads to fulfill their decisions in the departments and toward teachers are performing one-dimensional power. Two-dimensional power, which is not as visible, is exercised through social processes that control agendas and establish structures of information flow in the organization (Lukes, 2021). Lumby (2013) argues that two-dimensional power is about avoiding conflicts because fear of transgressing acceptable boundaries or wishing to be rewarded silences people and leads to their acceptance of power. For example, the principal can decide to provide information to one deputy head that the others do not receive, decide which cases should be discussed, and decide what is excluded from the agenda. Three-dimensional power is more complex because it describes people who are socialized into accepting dominant interests as their own without being conscious about it: the preferences of the dominant group appear so normal that they are not experienced as being contestable at all (Lukes, 2021). Lumby (2012) argues that one example of three-dimensional power connected to a distributed leadership perspective is ‘the belief that everybody can lead’.

Material and methods

Research design

This qualitative study draws upon a wider study of school leadership as educational discourse in an age of increased accountability, which investigates changing policy expectations on school leaders in policy documents and at local schools in the Norwegian context in the years after the introduction of a governance reform starting in 2006. This reform (known as LK06) focused on deregulation, efficiency competition, learning outcomes, and accountability and called for strong leadership for improvement. The reform also resulted in the redesign of the role of deputy head from having an overall administrative function toward an instructional leader function responsible for improvement work in school and with more responsibility toward staff (Abrahamsen & Aas, 2016). This Norwegian context has enabled researchers to examine school leadership in moments of transition.

The study spanned over four years with empirical data from five lower secondary schools. All the informants from the strategic sampling worked under the same reform policies and discourses, which means that the Local Education Authorities (LEA) in their municipalities had made changes in the deputy head role and were concerned with the importance of middle leaders for quality improvement. In these municipalities LEA described themselves as development-oriented and had put extensive efforts into improving the quality of their schools, including the development of school leadership for the schools in their community. All the five schools were considered as good schools in their local communities. We contacted the municipalities and asked them to choose which schools to visit.

A strategic sample of five principals and 15 redesigned deputy heads were interviewed during nine visits to the five lower secondary schools (students in the 8th to 10th grades, from 13 years to 16 years), with between 450 and 600 students at each school. In this paper we focus on the interviews and observations with the deputy heads. Eight contextual observations (Hultman, 2008) from the senior leadership team meetings with the principal and the deputy heads before we conducted the focus group interviews allowed for bringing events from the observations into the interviews that followed. This setup enabled a dialogue between observed situations and retrospective interpretations (Hultman, 2008). In the focus group interviews with the deputy heads, the researchers endeavored to bring forward different meanings on the topics under discussion, as well as events from the previous observations (Hultman, 2008), by facilitating a non-intrusive interview situation in which discussions among participants were allowed to unfold (Krueger & Casey, 2009). Topics for discussion were connected to people’s experiences and reflections as redesigned deputy heads in the leadership team and from leading teachers. We asked questions about how they experienced being responsible and how they were expected to lead improvement work in the department, and how this responsibility affected their leadership relations toward the principal and their functions in the leadership team within various decision-making processes. Table 1 gives an overview of schools and informants.

Table 1. Municipalities, schools and informants.

Schools	School A (560 students)	School B (587 students)	School C (480 students)	School D (532 students)	School E (524 students)
Geographical location	Western part	Western part	Eastern part	Eastern part	North Western part
About Schools	Increased responsibility of deputy heads	Increased responsibility of deputy heads	Increased responsibility of deputy heads	Increased responsibility of deputy heads	Increased responsibility of deputy heads
Deputy Heads and years of experience	Sarah 4 years Frank 9 years Rolf 3 years	Jim 7 years Linda 4 years Mark 3 years Diana 3 years	Carl 4 years Sam 2 years Ingrid 5 years	Rolf 9 years Susan 1 year Roy 2 years	Anne 6 years Robert 10 years Frida 7 years

Drawing upon the data, this paper provides insights into how deputy heads choose to interact with the principal and the teachers at their schools and how their room for maneuver changes (or does not) according to how they choose to influence, support, and control the principal and the teachers.

Analysis

The analysis process was divided into three levels: descriptive, thematic, and analytical (Richards, 2009). We focused on describing the findings in the material, identifying themes, and conducting further analytical work where interpretation was the core focus. This analytical process involved working iteratively, shifting back and forth between theoretically driven and data-driven coding. In the early descriptive and thematic processes of the analysis, the purpose was to identify and describe common focuses across the individual interviews and across the focus group interviews. In the next analytical stage, our aim was to interpret meanings about leading from the middle and, through this interpretation, develop the analysis from description to interpretation and power analysis; we also sought to present perspectives on the phenomenon under investigation (Kvale & Brinkmann, 2009). The case study approach enables in-depth exploration and is appropriate for research that seeks to take context into account (Yin, 2009).

Ethical considerations

Ethical questions were considered at all stages of the research process (Kvale & Brinkmann, 2009). We have followed ethical guidelines according to the National Committees for Research Ethics in Norway (NESH). The project was approved by the Norwegian Data Protection Office for Research (NSD). The participants were informed about the study, that they could withdraw at any time and that we at all time would secure confidentiality. All names, locations and municipalities have been excluded and replaced by fictitious names or by numbers.

Results

In this part of the article, we present excerpts from the deputy heads’ statements as examples of different dimensions of power, which will be discussed in the next part of the article. Because the power dimensions are analytical dimensions, some overlap exists between the dimensions, especially between one- and two-dimensional power exertion as practices, and information is closely connected. Despite the danger of overlap, the analytical processes can help to reveal how different practices can be understood on their own while also being connected to each other. The focus of the analysis is what the deputy heads said about being middle leaders in the leadership team, with the principal as the overall strategic and responsible leader, and what they said about being leaders for improvement work and teachers’ professional learning. The quotes in this paper have been lightly edited for clarity in English.

One-dimensional power

All the principals in the research process welcomed the regulative change and expressed that they wanted the redesigned deputy heads to be responsible leaders, which means that the deputy heads were intentionally given influence and power toward the teachers, and also toward the principal. But this situation was not necessarily what the deputy heads experienced. They spoke of principals who made decisions on their own, which may be characterized as one-dimensional power. On questions of how the deputy heads responded appeared as two different sets of practice responses toward the principal: one as acceptance of the one-dimensional power exertion from the principal, and one as non-acceptance where the deputy heads themselves exerted one-dimensional power toward the principal.

Acceptance of principals’ one-dimensional power exertion

Deputy heads told us of situations where ‘their’ teachers had skipped addressing them and instead had gone directly to the principal with problems or wishes in cases where the deputy heads were the ones who should have been contacted. When the principal let that happen and made decisions directly with the teachers and only informed the deputy heads afterward, the deputy heads experienced that the principal had reduced their autonomy and room for action: something that was necessary to be acknowledged as a real leader with decisional power. As deputy head Carl from school C expressed:

One of the greatest challenges for me as a deputy head is that the principal has talked with one of ‘my’ teachers, who suggests something concerning his or her teaching or other things, and the principal decides on something without asking me as a deputy head.

Deputy head Sam (school C) agreed and said, ‘Yes … the principal is very good at making decisions very rapidly and without discussing [things] with us’.

Deputy head Anne from school E said, ‘There are things that the teachers address directly to the principal, and they skip us’ [the deputy heads].

At these schools, the deputy heads experienced the principal as a leader who left no doubt about who was the leader in the school and could make decisions without discussing and collaborating with the deputy heads.

Deputy heads exerting one-dimensional power toward the principal

The other type of response was illuminated through how the deputy heads from school A exerted one-dimensional power toward the principal. They decided that they would confront their principal together by making explicit that they did not accept the principal making decisions without consulting them. As deputy head Sarah from school A said: ‘No, we told the principal that he had to talk with us first in cases concerning the teachers, something which he accepted, and he changed his practice’.

This excerpt reveals a group of deputy heads who argued against the principal making observable decisions alone; by doing so, they themselves exerted one-dimensional power toward the principal, with the purpose of changing the principal’s one-dimensional power practice. In this school, the principal chose to listen and acknowledge the deputy heads as leaders who were responsible for ‘their teachers’ in ‘their departments’. He changed his leadership practice and made structures for sharing decisions in the leadership team together with the deputy heads.

As deputy head Frank (school A) noted, ‘I’ve experienced that there’s now much more room [in the leadership team] for making decisions together’.

Whereas the deputies in school A collectively exerted one-dimensional power by being explicit toward the principal about their wish for shared decision-making processes, the deputy heads in school C and E accepted the one-dimensional-power leadership practice from the principal without taking up the matter. In these schools, the change of practice did not happen within the data-collection period.

Two-dimensional power

Principal two-dimensional power

When constructions of school leadership change from the individual principal with observable power toward ideas of distributed leadership practice in a culture where leadership is shared among several leaders the principal can still control the processes of sharing information and retain control of the agenda. In one of the schools, the deputy heads told us how they experienced a situation where they were not informed of a decision concerning the whole school the principal had made:

Deputy head Jim said, ‘It was the principal who made the decision; it was presented [to the teachers] without us [the deputy heads being] informed about it’.

Deputy head Linda responded, ‘Yes, and the principal told us what we were expected to do, and when the teachers asked where this decision had come from, I didn’t know what to answer’.

Deputy head Mark said, ‘It’s very important that being given more responsibility as deputy head leaders doesn’t become a new role by name only’.

This example shows what Lumby (2012) describes as two-dimensional power, through choosing what information is shared, with whom, when, and how. It also shows the close connection with one-dimensional power, since making observable one-dimensional decisions in this case was closely connected to making decisions about when and how information was shared with the deputy heads.

Deputy heads exerting two-dimensional power

The interviews revealed that the deputy heads often constructed a common understanding of themselves as a group who shared experiences and reflected on everyday challenges in the departments and by following up on the teachers’ instructions. Together they viewed themselves as being able to influence and support the principal through the process of sharing information and knowledge about the departments and about the school: something the principal would need to lead the school. But they chose what to bring into the leadership team. As such, the middle leader level could, to a certain extent, develop both autonomy and power through controlling information sharing. As deputy head Rolf said, ‘Together we deputy heads are strong. The principal is dependent on the information we as a leadership group give him in leadership meetings’

Deputy heads leading teachers

How did deputy heads exert one- and two-dimensional power toward the teachers? When asked how they were leading improvement work with the teachers, the deputy heads expressed that their leadership practice was characterized by a questioning approach and by being close to the teachers’ practices in the classrooms. Most of the deputy heads told that they ‘controlled’ the teachers by knowing what was going on in the classrooms, through the questioning approach and by observations in the classrooms. Frida from school E said: ‘We know our teachers and their instructional practices, and we ask questions to follow up’. So, being explicit and making observable individual decisions toward the teachers in the departments was not a preferred leadership strategy. But the respondents stressed the importance of being part of the senior leadership team, and that decisions involving leading teachers should be anchored in the leadership team before being brought to the teachers. They experienced that it was easier to inform the teachers about decisions that had been made in the leadership team than for the teachers to consider them individually responsible for making decisions. Another reason for this scenario was connected to developing a school with shared visions. As deputy head Susan noted:

It’s very important that we agree in the leadership team before we meet with our teachers at the departments, also to prevent one department from developing its own culture and other departments developing another culture.

Deputy head Roy followed up:

Yes, if there’s something concerning the teachers, we can back each other up, and therefore it’s important that we have the leadership team … it’s easier to say that the leadership team’s decided this than to say that I’ve decided this.

These expressions also provide an example of how tightly one- and two-dimensional forms of power are connected. By choosing to inform others about decisions made in the leadership team, outside the teachers’ arena for discussions, deputy heads exerted a form of collective and indirect one-dimensional power through two-dimensional information-sharing.

Three-dimensional power

Leadership changes within the leadership team?

Regulative changes (Scott, 1995) that include increased responsibility for deputy heads with expectations about leading teachers and improvement work can implicitly lead to the belief that the senior leadership team is an arena for distributed activities where everybody is a leader on equal terms and where people collaborate, trust each other, and share decisions. Regulative change toward more formal responsibility can lead to the belief that distributed leadership means sharing decision making. Deputy head Robert expressed his reflections on the redesign and what ideas on leadership the redesign had resulted in:

Earlier I didn’t have any defined leader responsibility—you weren’t responsible. Back then, loyalty conflicts between the principal and us [previous administrative role] could arise. I think we’re much more defined as a leadership team now—something that was also the idea behind the redesign into the deputy head leader role. I think it’s easier now.

Although Robert experienced a positive change, other deputy heads’ excerpts which we showed in the one-dimensional and two-dimensional parts of this article, also show that regulative changes does not always lead to cognitive and normative changes, toward the belief on distributed leadership. This argues for investigating beliefs.

Leading teachers through a questioning and distributed approach

Deputy heads who are leading situated between the principal and teachers can contribute to building the belief in leadership as a distributed practice. Constructing a responsible leader function in between the principal and the teachers may provide the belief that this is bringing the leadership team closer to the teachers. All the deputy heads in this project expressed that they wanted to be leaders and lead the teachers as a result of been given responsibility and autonomy from the principal. When asked how they wanted to lead the teachers, they expressed that they wanted to lead through collaborating with the teachers in teams and through shared decision-making. One way of exerting control of the teachers, they noted, was to know what was going on in their teachers’ classrooms and to follow up their instructional practices by asking questions. They wanted to build trust and collaborate with the teachers. Deputy head Sarah represents what her fellow deputy heads said about being leaders of teachers: ‘If you as a deputy head don’t manage to play on the same team with the teachers, it’ll be more difficult for the teachers’ [something the deputy heads did not want]. Deputy head Jim said:

We gave the responsibility to a group of teachers, and things have functioned much better. One of my teachers and I agree that she takes the responsibility for a certain case, and then she leads, and my function is to be a participant.

The belief in distributed leadership and that ‘everybody can lead’ as a preferred leadership practice is difficult to overlook in this case and illustrates an acceptance of the idea of sharing leadership through distributed practice.

Discussion

Different types of practices seem to be represented in the findings from the previous section, and which reveals the complexity of understanding leadership in context (Aas & Törnsén, 2016; Blossing et al., 2014; Busher et al., 2007; Lipscombe et al., 2021; Scott, 1995). The findings show that leadership practice can be characterized as a mixture of individual ‘principal on top’ leadership and practices involving many participants in distributed activities. These different types of actions seem to be intertwined in a variety of ways. One such way and point for discussion can be described as the interdependency between individual leadership and a distributed perspective on leadership. This notion is about exerting observable power, with the aim of developing a distributed leadership practice characterized by mutual dependency and collaborative decision-making. This can be described as the dual function of explicitness. A second point of discussion directs our attention to the intertwining between visible middle leadership and the leadership team as a configuration of a school’s leadership enactment.

The dual function of explicitness—observable power to develop distributed practice

Through changes in regulative patterns (Scott, 2008), redesigned deputy heads are given increased responsibility, where they are expected to be leaders of departments’ development work and of the teachers. They are supposed to lead the school together with the principal. According to the interviewees, this situation seems to either happen or not and appears to be characterized by choice: the principal’s choice and/or the deputy heads’ choice. The principal either chooses to provide room for maneuvering for the deputy heads or chooses to make decisions alone, thereby exerting one-dimensional power (Lukes, 2021; Lumby, 2013).

Those deputy heads who experienced little room for maneuvering seemed to have two choices: either to accept the situation without taking action, or to be explicit toward the principal about their wish to share decision-making and to be given considerable responsibility (Leithwood, 2016). Those deputy heads who explicitly tried to increase their room for maneuvering toward the principal exerted one-dimensional power, with the aim of developing more distributed practices (De Nobile, 2018; Grootenboer, 2018; Lárusdóttir & O’Connor, 2017; Spillane, 2006). This shows that being explicit, exerting one-dimensional power, can open the way for change in cognitive and normative structures (Scott, 1995), thus leading to increased influence with room for maneuvering among deputy heads (Lawrence & Suddaby, 2006). Choosing not to be explicit seemed to lead deputy heads to accept the principals’ decisions they made alone, exerting one-dimensional power, and no change happened. Fear of transgressing boundaries toward their own leaders may have silenced some of the deputy heads (Lumby, 2012).

Following this line of argument, regulating power between the principal and the deputy heads can help to reduce unclear middle leader role descriptions (Lillejord & Børte, 2018) and to achieve a more distributed practice and shared decision-making in the leadership team, something the idea of middle leadership has often been connected to (De Nobile, 2018; Grootenboer, 2018; Lárusdóttir & O’Connor, 2017; Spillane, 2006). Within this perspective, one-dimensional power exertion and distributed leadership practices seem to be intertwined.

Towards the teachers, however, the deputy heads, revealed a different leadership practice than in the leadership team, and characterized by an inquiry-oriented and facilitative leadership practice concerned with building trust (Paulsen, 2021). The teachers’ voices were given considerable weight. ‘Everybody’ could influence and lead in the departments during improvement work in the departments, and the deputy heads could function in a participant position, as a supporter (Abrahamsen, 2018; Abrahamsen & Aas, 2019) and follower (Spillane, 2006). In other words, a distributed and collaborative leadership enactment was present toward the teachers, and the need for being explicit did not seem to be necessary. In the wake of this scenario, distinguishing between teachers’ leading and middle leaders’ leading can become difficult, despite their accountability responsibilities. Exerting one-dimensional power by being explicit, making decisions and exerting control in the department as a deputy head seems more difficult to do. Following this line of reasoning, deputy heads may experience tension between an inquiry-based democratic leadership style in a context focused on equity and democratic participation (Imsen et al., 2017; Oftedal Telhaug et al., 2006) and leadership demands for increased accountability and control influenced by neoliberal reforms (Aas & Törnsén, 2016).

But being leaders with responsibility also includes making and following up on decisions. The department heads seemed reluctant to exert one-dimensional power toward the teachers, instead preferring to use strategies of passing on information about decisions that had been made in the leadership team. Creating more distance between decision-making processes and the individual deputy head situated close to the teachers seemed to make it easier to lead in a ‘distributed way’ and to build trust (Sørhaug, 1996).

Two-dimensional power was at play because exerting one-dimensional power does not align with the inquiry-oriented leadership strategy deputy heads typically prefer. A preference for informing teachers about collective leadership team decisions can reduce the idea of making observable decisions alone in the department and can be more aligned with equity and a democratic leadership style (Harris, 2013), and short distance between the deputy heads and the teachers (Aas & Törnsén, 2016). The exertion of one-dimensional power by individual department heads may create tension between a Nordic culture of equity and democratic participation and neoliberal reforms (Aas & Törnsén, 2016) and a focus on ‘what works’ (Gunnulfsen & Møller, 2021). Within this context, two-dimensional power seemed to function as an indirect and collective form of one-dimensional power exertion from the senior leadership team.

Visible leadership team—the configuration of the school’s leadership enactment

The empirical material used in this project showed that structural redesign can lead to cognitive and normative change (Scott, 2008) if the principal acknowledges a deputy head role with real influence in the school. Deputy heads who are explicit about how they want to fill the middle leader role using purposive actions (Lawrence & Suddaby, 2006) can become influential (Abrahamsen & Aas, 2019) and reduce the lack of clarity in their roles (Lillejord & Børte, 2018; Muijs & Harris, 2003). While this idea does not mean that a process toward shared decision-making will happen in every leadership team, it does show that such change can happen. Towards the teachers the deputy heads exerted leadership as an inquiry-based practice, and they preferred to pass on decisions that had been made in the leadership team to the teachers more than they wanted to make individual decisions in the departments. They clearly expressed a preference to exert two-dimensional power (Lumby, 2012).

In being explicit about how they expected their roles to be formed, middle leaders seemed to be able to choose to develop two different leadership strategies: one upward, toward the principal, and one downward, toward the teachers. They could create their own room for leadership practice toward the principal by contributing within collaborative practices and shared decision-making in the leadership team (Leithwood, 2016) by being explicit about what they wanted. They could become visible leaders in the leadership team. Toward the teachers, however, an inquiry-based leadership practice and participant role in development processes in the departments was the more typical leader strategy. By becoming more individually visible in the leadership team, a collectively visible leadership practice could be enacted toward the teachers.

Concluding remarks—leadership and power in change?

Although generally considered important (Bennett et al., 2007; Harris & Jones, 2017), middle leaders such as deputy heads have often been viewed as part of the principal’s team, and not as individual senior middle leaders situated between the principal, teacher middle leaders on different levels, and the teachers. So how do Norwegian deputy heads experience leading and room for maneuvering within changing structures toward leading school change and development?”

The empirical data of the present project shows that this position can change, depending not only on the principal providing structures for sharing decision-making and providing considerable responsibility to middle leaders (Leithwood, 2016), but also on how middle leaders themselves choose to fill their role between influence, support, and control (Abrahamsen, 2018; Abrahamsen & Aas, 2019).

This study’s power analysis helps to illuminate which kinds of practices can contribute to increase deputy heads’ room for maneuvering in leading school change and development. The findings in this research project show that deputy heads can exert visible middle leadership by being explicit in the leadership team, and by being explicit from ‘a distance’ in the departments through sharing information about the leadership team’s decisions. Not only the principal influences middle leaders’ room for maneuvering; regulative changes (Scott, 2008) also open middle leaders’ own possibilities to influence, support and control (Abrahamsen, 2018; Abrahamsen & Aas, 2019), and expand their room for maneuvering, something which can give further insights not only for the deputy heads, but for middle leadership at different levels. Knowledge about beliefs show that power comes in different forms and are connected to changes, influence and control. Following this reasoning, an interesting question to ask is whether the principal as the top leader of the school becomes more invisible as the deputy heads become more visible, and if so, how this scenario will influence the idea of leadership in schools.

Limitations and future studies

Through power analysis, this study has zoomed into how deputy heads can influence their own room for maneuvering from the middle. A limitation is that the current empirical research does not tell us anything about what the principals nor the teachers experienced when the deputy head position changed toward leading development in departments. The present study has zoomed into the deputy heads own reflections on developing room for maneuver. Further investigation into how the principals and the teachers have experienced the deputy head leadership practice and power exertion is of interest as this would contribute to more knowledge about how the different layers in school experience to interact with the deputy heads. More knowledge about how power, trust and room for maneuvering between the different layers in school is developing when leading improvement work in school is of interest to investigate.

From the current study we find two implications. First, a practical implication of the research shows that deputy heads need training in identifying and situate their room for maneuver as leaders under the principal, toward the principal as the top leader and toward the principal and for leading professional improvement work among teachers. This can contribute to address the uncertainty and under-utilized position of deputy heads (Lillejord & Børte, 2018; Muijs & Harris, 2003) and illuminate their room for maneuver. Another practical implication which the present research identifies is that principals also should focus on investigating interactions and power within the senior leadership group. A second implication which this research has shown is that there is a need for more research on the middle leader position situated directly under the principal, such as the deputy heads. There is little research on this particular group of middle leaders. The present shows that questions of power between leaders at different levels in the school should be investigated.

Another interesting approach for future research would be to investigate how department heads as senior middle leaders and teacher middle leaders at different levels are connected through having different levels of responsibility for instructional leadership and school improvement, and what implications such connections may have for the principal within the challenging work of leading collective processes (Aas et al., 2021).

Theoretical implications

In times of regulative changes different forms of distributed leadership practice evolve. Our research has investigated how deputy heads interact and develop leadership as a collective practice at the same time as functioning at different hierarchical levels. Investigating combinations of solo leadership of the principal on top with different forms of distribution of leadership practices and how analysis of power exertion contributes to reveal complexity in understanding leadership, can contribute to acquire more knowledge to the school leadership research field. Of special interest is how one-dimensional power and two-dimensional power can enhance our understanding from a distributed and democratic perspective.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Notes on contributors

Hedvig Neerland Abrahamsen

Hedvig N. Abrahamsen is associate professor at the Department of Teacher Education and School Research, University of Oslo, Norway. Formerly a school leader her research interests are in the fields of leadership and school development, professional development of school leaders, and the changing nature of leadership in times of reform.

Marit Aas

Marit Aas is professor in the Department of Primary and Secondary Education Teacher Education, OsloMet – Oslo Metropolitan University, P.O. Box 4 St. Olavs plass, NO-0130 Oslo, Norway. Her research interests are in the areas of leadership and school development, professional development of school leaders and group coaching.