Improving leadership in higher education institutions
- 1. Improving leadership in Higher Education institutions: a distributed perspective Jitse D. J. van Ameijde Æ Patrick C. Nelson Æ Jon Billsberry Æ Nathalie van Meurs Published online: 19 April 2009 Ó Springer Science+Business Media B.V. 2009 Abstract This paper reports on a qualitative study exploring how distributed patterns of leadership manifest themselves in project teams within a Higher Education institution. The emphasis is on both the ‘what’ and the ‘how’ of distributed leadership, thus providing an account of the nature of distributed leadership in higher education and the factors which were found to enhance and inhibit its occurrence and effectiveness. The findings are presented in a model of distributed leadership which seeks to provide an integrative account and a framework for further study. The conclusions focus on both the theoretical implications for the study of distributed leadership and the practical implications for HE institutions wishing to promote effective leadership. Keywords Leadership Á Distributed Á Project teams Á Higher education Á Qualitative Introduction Higher Education institutions are facing continuous pressures for change. Over the past few decades traditional principles of academic leadership and collegial forms of J. D. J. van Ameijde (&) Á P. C. Nelson Human Resources Division, The Open University, Walton Hall, Milton Keynes, Buckinghamshire MK7 6AA, UK e-mail: [email protected] P. C. Nelson e-mail: [email protected] J. Billsberry Faculty for Business, Environment and Society, Coventry University, Priory Street, Coventry, Warwickshire CV1 5FB, UK e-mail: [email protected] N. van Meurs Middlesex University Business School, Hendon, London NW4 4BT, UK e-mail: [email protected] 123 High Educ (2009) 58:763–779 DOI 10.1007/s10734-009-9224-y
- 2. governance have been rapidly replaced by management principles adopted from the private sector. Evidence of this move towards what has been termed ‘New Public Management’ (Chandler et al. 2002) or ‘New Managerialism’ (Deem 1998) has been observed in countries all over the world. Increasingly, government policies require universities to account for the expenditure of public funds and provide evidence of ‘value for money’. In the UK context, such government pressures have become evident through mechanisms such as the Research Assessment Exercise (RAE) and Teaching Quality Assessment (TQA). Together with increased competition between universities for students and funding support, as well as expectations from the private sector around the provision of appro- priately trained graduates, these pressures have led to what has been experienced as a commoditization of knowledge work. The effect of these changes has even been observed within higher education discourse in which the academic language of deans, students, and courses has become increasingly displaced by language of line-managers, customers, and products (Gumport 2000; Parker and Jary 1995). As a result, Higher Education institutions are no longer the protected entities whose legitimacy is taken for granted, but instead are expected to face the complexity of bal- ancing the need to operate according to market pressures, teach an increased number of students despite diminishing financial means while struggling to maintain traditional academic and educational principles of quality. The mechanisms which many Higher Education institutions have implemented to deal with these pressures seem to have created an interesting paradox. The introduction of procedures around performance measurement, quality control, and audit aimed at improving the effectiveness and accountability of HE institutions have at the same time contributed to the creation of additional bureaucratic layers of control which have often been experienced as inhibiting organisational effectiveness and responsiveness. Also, several commentators have pointed out the negative effects of these measures through the pressures they have created on academic as well as non-academic staff, and the resulting tensions between management and staff in Higher Education institutions (i.e., Chandler et al. 2002; Parker and Jary 1995; Huisman and Currie 2004). However, it has been acknowledged that the transformation of higher education is irreversible, and that some of the principles which this transformation seeks to bring about serve a legitimate purpose, despite valid concerns regarding the way these changes have often been implemented (Parker and Jary 1995). Additionally, quite often the ‘solutions’ derived from the private sector which Higher Education institutions have tried to implement were based on principles and approaches derived from out-dated ideas and management fads which have already been abandoned in the sector from which they originated (Birnbaum 2000a, b). These issues are exacerbated by the fact that Higher Education institutions have unique cultural characteristics which make the translation from management principles derived from the private-sector to a higher education context problematic. Several authors have pointed out the cultural uniqueness of higher education establishments (e.g., Birnbaum 1988; Bergquist 1992) and the need to align the leadership of change to university culture (Kezar and Eckel 2002; Lueddeke 1999; Middlehurst 1997). Clashes between the principles of management and a traditional culture of collegial leadership have been widely reported on (Chandler et al. 2002), highlighting the need to adapt leadership and management principles to a higher education context. We believe that principles of distributed leadership provide a means for overcoming some of the problems. On a conceptual level, the notion of distributed leadership seems well aligned with notions of collegiality and professional autonomy which have 764 High Educ (2009) 58:763–779 123
- 3. traditionally been characteristic of higher education leadership, while also recognising the wider institutional needs for effectively managing the changes that turbulent environments impose on Higher Education institutions. Before we outline the aims of our study of distributed leadership in a higher education establishment within the UK, we will first provide an overview of the literature on distributed leadership. Distributed leadership Within the scientific leadership field, a growing body of research is dedicated to studying leadership as a shared or distributed phenomenon. In itself, the emergence of a new perspective on leadership is nothing new within the leadership field, which has traditionally been characterized by the co-existence of a rich variety of divergent theories. However, despite this divergence in theoretical development, one common assumption seems to have underpinned most traditional leadership theories: the view that leadership is a vertical process which separates leaders from followers as mutually exclusive categories (Pearce and Conger 2003, p. 1; Rost 1993, p. 70; Yukl 2002, p. 2). Distributed leadership chal- lenges this view of leadership as a solely vertical process in which an individual leader is seen as the main source of influence which shapes the emergence of collective action, and instead focuses on the mechanisms through which diverse individuals contribute to the process of leadership in shaping collective action. As such, it provides an important complementary understanding of the subtleties of leadership in real organisational settings. Several factors could be seen as contributing to the shift from studying leadership as a vertical influence process to studying leadership as a distributed phenomenon. First of all, the growing interest in complexity sciences has increasingly highlighted the shortcomings of traditional reductionist science in explaining many real-world phenomena. The power of paradigm in guiding the structure in which phenomena are studied and the types of questions that science can legitimately seek to answer could not be underestimated (Kuhn 1996). In the scientific study of leadership, this reductionist paradigm seems to have granted ontological status to the distinction between leaders and followers, and sought to explain leadership phenomena through simple cause and effect relationships between these two exclusive categories, with only limited emphasis being placed on the complexities in which context shapes the leadership process. Such observations have fuelled a recent shift away from studying social phenomena from a reductionist paradigm which traditionally sought to break down social phenomena into discrete components and study them in isolation, leading to atomistic explanations of leadership which have been accused of failing to explain the true complexities and subtleties leadership processes, in which influence processes are often shared between different individuals (Gronn 2002; Spillane et al. 2001, 2004). Secondly, the increased reliance of organizations on team based work has led to a growing dissatisfaction with traditional leader-centric approaches in explaining the com- plexities of leadership within such team-based settings. Many organizations increasingly depend on cross-functional, self managing work-teams to deal with the growing com- plexities of work and to sustain their competitiveness (Cummings and Worley 2004, p. 341). There is a growing utilization of project teams in managing project work involving matters ranging from new product development to implementing organizational change (Thamhain 2004), and knowledge work is becoming increasingly team-based and requires the coordination and integration of the expertise of diverse professionals from dispersed fields (Pearce 2004). However, even though the literature on team working has provided a wealth of knowledge regarding the factors that shape team success, it has still put High Educ (2009) 58:763–779 765 123
- 4. considerable emphasis on the role of formal team leaders in shaping team effectiveness (Day et al. 2006), despite the fact that the role of individual leaders in shaping collective outcomes is often greatly overestimated (Meindl et al. 1985). In line with these theoretical and practical concerns regarding traditional perspectives of understanding leadership, Gronn’s conclusion that distributed leadership is an idea whose time has come (Gronn 2002) seems to be well grounded. The concept of distributed leadership Given the fact that distributed leadership as an area of scientific inquiry is still in its infancy, a common understanding of distributed leadership has yet to be conceived of (Bennett et al. 2003; Day et al. 2004). Some authors use the term shared leadership (e.g., Pearce 2004; Pearce and Sims 2002) while others employ the term distributed leadership (e.g., Gronn 2002). At this point, there seems to be no clear conceptual differences between these approaches, and different authors use them interchangeably (e.g., Day et al. 2004). Besides the different terms employed, different authors diverge in their conceptual- izations of distributed leadership on various grounds, including the scope of the network of participating agents in the leadership process. Some authors have focused on a single team or group of people as their unit of analysis (e.g., Brown and Hosking 1986; Locke 2003; Pearce 2004), while others have taken a more open-systems approach, taking the whole organization and even constituencies beyond the organizational boundaries as their unit of analysis (e.g., Spillane et al. 2001, 2004). However, despite these differences most authors agree upon two principles as underpinning the concept of distributed leadership: 1. Leadership is a shared influence process to which several individuals contribute 2. Leadership arises from the interactions of diverse individuals which together form a group or network in which essential expertise is a dispersed quality Empirical support for distributed leadership A growing number of studies have provided empirical support for the concept of distributed leadership. A longitudinal study of inter-organisational rule-making groups conducted by Feyerherm (1994) provided evidence regarding the shared nature of leadership behaviours in teams which were facing a highly complex and interdependent task, requiring the integration of differing viewpoints and interests. In such highly interdependent situations, in which no single individual is capable of possessing all the relevant expertise for reaching a common goal, distributed leadership seems particularly useful (Pearce 2004). An equally complex and interdependent challenge was faced by top management teams in the start-up of new ventures, where Ensley et al. (2006) found distributed leadership variables to predict firm success beyond the predictive power of vertical leadership vari- ables. Within change management teams, Pearce and Sims (2002) found distributed leadership patterns lead to higher performance as compared to traditional ‘leader-domi- nated’ leadership patterns. However, subsequent research has indicated that distributed leadership can take dif- ferent forms, and only if different individuals within a group recognize each other as leaders (distributed-coordinated leadership) does distributed leadership lead to increased performance. If the different individuals within the group do not recognize each other as leaders (distributed-fragmented leadership) there seems to be no enhanced performance over traditional vertical leadership (Mehra et al. 2006). 766 High Educ (2009) 58:763–779 123
- 5. A question that rises is whether distributed leadership makes the notion of traditional vertical leadership redundant, a possibility that Gibb (1954, in Day et al. 2004) already identified decennia before distributed leadership became a field of scientific investigation. However, at this point distributed leadership is seen as an important form of leadership which can act as a complement rather than a substitute to traditional vertical leadership (Pearce 2004). Pearce states that although dominant vertical leadership styles can suppress team member participation in decision making, vertical leadership still plays an important role in team design and boundary management, two factors considered important for the ongoing success of distributed leadership. Also, critical events that disrupt team func- tioning can call for vertical leader interventions (Morgeson and DeRue 2006). Although it has been recognized that vertical leadership plays an important role to the ongoing success of distributed leadership (Pearce 2004), relatively little attention has been paid to other factors which constitute the successful practice of distributed leadership within organizational settings. Also, much empirical work on distributed leadership so far has adopted taken for granted traditional notions of vertical leadership and taken these as the basis for understanding leadership from a distributed perspective. Examples of such studies looked at traditional notions of leadership behaviours and studied to what extent such behaviours were shared amongst different individuals (i.e., Ensley et al. 2006; Fe- yerherm 1994). Although these studies have indicated that such behaviours are indeed often distributed, they might have overlooked the possibility that distributed leadership might be understood as something qualitatively different from a simple aggregation of vertical leadership behaviours. It could be argued that these studies only looked at what Gronn (2002) labelled ‘additive action’, ignoring the additional dynamic arising from the interactions of the various individuals, or what Gronn (ibid) labelled ‘concretive action.’ As one of the defining principles of distributed leadership is that it arises from the interactions of diverse individuals in a setting where expertise is a dispersed quality, our study aims to seek an understanding of the mechanisms through which dispersed expertise is integrated in the process of leadership. We argue that to gain deeper understanding of the concept and practice of distributed leadership, it is important to study the concept in settings of which the inherent complexity, interdependence, and need for integration of dispersed expertise require the input of several constituencies in the process of leadership. Project work is in this sense ideal for the study of distributed leadership. Project teams are often cross-functional, requiring individuals from a range of backgrounds and sources of expertise to integrate their collective inputs in order to reach a common goal (Pinto et al. 1993). Therefore, project teams are characterised by a high level of interdependency between members, placing high demands on the process of leadership to coordinate and integrate individual inputs (Thamhain 2004). Additionally, the complex nature of projects makes requires extended information processing, complex decision-making, and a high level of creativity of the team, conditions ideally suited for distributed leadership (Pearce 2004). The main aim of the present study is to gain an understanding of how patterns of distributed leadership manifest themselves in project teams within a Higher Education institution, and to understand whether successful project work is characterised by stronger levels of leadership distribution. Higher Education institutions have traditionally been characterised by collegiate cultures, which makes it likely that collegiate work patterns are still prevalent in the way project work is conducted. As Higher Education institutions increasingly adopt project work in order to cope with the increasing demands the higher education context poses, such an understanding is essential for yielding lessons regarding the ways Higher Education institutions can effectively operate and develop their workforce. High Educ (2009) 58:763–779 767 123
- 6. Method In scholarly literature there is a growing recognition of the need to study distributed leadership in the social context of organizations (Day et al. 2006). Therefore, a qualitative research approach was adopted in order to generate new hypotheses (Conger 1998) and to add to the understanding of distributed leadership in a realistic organizational setting. This research approach provides rich data and captures the specific context. Thus, it enables also to discover the ‘how’ of distributed leadership, as it helps to understand its underlying structures (Pettigrew 1990). The flexible character of qualitative research makes it addi- tionally capable for an exploratory investigation of new concepts (Bryman 2004). The main aim of the research at this stage was an explorative investigation of factors that influence distributed leadership both positively and negatively within an organizational context. The intended outcomes are aimed at gaining a deeper understanding of distributed leadership within an organizational context and extending the literature on this concept. Five different project teams were analysed; five of which were reported to be successful in achieving their purpose and five of which were not. This approach allowed us to compare and contrast these projects in terms of distributed leadership. A combination of causal mapping and semi-structured interviews were used to collect data on the ten dif- ferent projects. Research context We conducted this study at a large UK based university, which has not been excluded from the pressures to increase efficiency and the resulting frictions between the new managerial practices and the traditional way of running the university. The initial structure of the university relied on pluralistic decision-making, which was mainly represented by com- mittees whose members were appointed on their (academic) expertise. The growth of the university led to a significant increase in the number of committees, resulting in an impediment of decision-making. This impediment of decision-making reached a point at which it diminished, in the view of many staff members, the ability of the university to react on and adapt to environmental changes. With the decline in student numbers at the start of the new millennium, also due to increased national and international competition, a lot of activities and projects were launched in several directions to regain lost students and to secure the university’s income. The situation at present is that the university has developed different strategic priorities and is searching for a new mission that is able to capture all the priorities into an integrating whole. Furthermore, staff members perceive an ongoing struggle between the traditionally pluralistic culture, academic freedom and the increase of managerial top down practices. The present conditions of an increasingly complex environment and of a surge of projects within the university offered an ideal context to further explore the promoting and inhibiting factors of distributed leadership; a concept which the university itself wants to promote within its boundaries. Sampling Participants for the study were identified by purposive sampling and by referral sampling methods. Five individuals from a range of different positions were selected and approached for participation in the research project. The university’s staff consists of roughly 80% administrative employees and of about 20% academics. Middle and senior management is composed of about 60% male and 40% female. Therefore, five individuals occupying 768 High Educ (2009) 58:763–779 123
- 7. middle and senior management positions were selected, four of which had administrative positions and one which had an academic position Two of the participants were female and three male. All of the five individuals had an organizational tenure of at least 2 years and had been involved in a range of projects. To gather participants for the surrounding semi- structured interviews to every case, each individual was asked to name five other indi- viduals who could provide additional insights in the identified cases. As it proved to be difficult for most of the initial individuals to name five additional individuals, a total of 20 (3–4–4–4–5) additional participants were interviewed. Data collection With each of the initial five individuals, a separate session was held in which they were asked to describe two projects they had been involved in; one project they regarded as successful, and one project they regarded as unsuccessful. During these sessions, we constructed causal maps together with the individual for both the successful and the unsuccessful projects, focusing on the factors which caused the projects to be successful or unsuccessful. At the end of these sessions, the individuals ware asked to name up to five additional people who were involved in the projects they had described. These people were subsequently interviewed using a semi-structured interview technique to provide more data on the leadership of these projects. A total of 20 additional participants were interviewed, making 25 in total. Data analysis All the causal mapping sessions were audio recorded and verbatim transcribed. The causal maps themselves were imported into Decision Explorer, a software program to analyze causal maps. The interviews were audio recorded and then verbatim transcribed. The transcripts of both the causal mapping sessions and the interviews were then imported into Nvivo7, a computer based program for qualitative data analysis. The transcripts were used as the main source of data analysis. The causal maps served as complementation and verification for the transcripts of the mapping sessions. The transcripts were coded on the basis of factors perceived as influencing distributed leadership. During this coding, the focus was laid on the team being the unit of analysis. Therefore, no difference was made between the different members of a project team, regardless of their formal status within the team. Content analysis was used to identify recurrent themes, by then classifying these recurrent themes into categories which were perceived to be factors influencing distributed leadership. The factors were then discussed to reach consensual inter-subjectivity. Consensual inter-subjectivity is the consensus that is reached between researchers on the interpretation of data, and is regarded as a main tool in qualitative research to measure reliability (Smaling 1992). On the basis of a randomly chosen transcript an interrater reliability of 87% was found, using the percentage agreement method. Results Through the iterative process of analyzing and coding the transcripts, several factors were identified which were thought to relate to distributed leadership in project teams. We divided the emerging factors according to the level at which they operate. At the organizational level, several factors around external activities and processes were identified. These factors High Educ (2009) 58:763–779 769 123
- 8. involved the way in which the team interacted with and related to important constituencies outside of the team. These factors were found to influence important outcomes, which in turn served as inputs for continuing distributed leadership at the team level. At the team level, several factors were found which relate to the way the team itself operates. These factors were divided into critical internal conditions and critical internal processes. By comparing successful and unsuccessful projects, it became apparent that the factors we identified at the organizational and team levels were important in shaping the success or lack thereof of projects. We have attempted to encapsulate these factors in a model of distributed leadership as presented below. However, it is important to note here that although most of the factors seemed to play a role in all of the projects, the relative importance and the specific configuration in which the factors interacted varied between the projects. One aspect that seemed to particularly influence the configuration of the way the different factors influenced project outcomes was the nature of the project as emergent or top-down initiated. Top-down versus emergent projects An important distinction that emerged in the analysis of the ten different projects evolves around the nature of the projects as either top-down driven or emergent. Four of the analysed projects were distinctively emergent in the sense that they were initiated by a group of two or more individuals without any formal leadership status and without formal appointment in setting up the concerned projects. Five of the analysed projects were distinctively top-down in nature, and were formally initiated by individuals or groups holding senior management positions. The final project seemed to have emergent as well as top-down properties, as the idea for this project emerged from an individual initiative, yet the subsequent project set-up was initiated through formal management structures. The distinction between emergent and top-down projects was seen as important as the nature of leadership within these projects as well as the challenges the project leadership had to overcome seemed to show important differences. Some of these differences will be highlighted as we describe the different factors below. However, first we will introduce our model of distributed leadership, as depicted in Fig. 1. Distributed leadership factors We will now provide a more in extensive description of the factors depicted in Fig. 1. As mentioned before and presented in the model, the factors we identified as relating to distributed leadership in project teams were divided into factors relating to external activities (organizational level) and factors relating to internal team activities (team level). The factors at the team level could subsequently be divided into critical internal conditions and critical internal processes. We will first outline the factors relating to the inputs to and outcomes of external activities (organizational level). Then we will turn to the factors relating to team functioning (team level), Organisational level factors From our data, it became clear that none of the projects could be analysed in isolation from the wider institutional context in which they were embedded. In each of the projects, leadership was not solely confined to facilitating the internal processes of the project 770 High Educ (2009) 58:763–779 123
- 9. teams, but had to deal with challenges which reached beyond the direct boundaries of the projects. Involvement of and negotiation with external stakeholders, gathering and dis- seminating important information regarding the projects, and facilitating an alignment between the projects and the wider organisation context were regarded as vitally important in determining the success of the various projects. Critical external processes Tailoring message to the receiver was reported to be an important aspect contributing to project outcomes. Each of the studied project teams had to engage in communication exchanges with stakeholders external to the team itself. In such cases it often proved necessary to present information in such a way as to meet the requirement of the specific stakeholder group being addressed. Sometimes such tailoring was political in nature, presenting certain aspects that were thought necessary to persuade external decision makers or organisational members to buy into an initiative or to provide needed resources. However, it also often constituted the presentation of information in a format under- standable by the recipient by omitting or translating technical jargon or unnecessary details. Another external activity project teams often had to engage in constituted the feedback of progress. Such feedback served to inform external stakeholders of any successes or obstacles faced by the project team and was often necessary to ensure continuing support and involvement of external stakeholders. Also, such feedback was often found important Fig. 1 Model of distributed leadership in project teams High Educ (2009) 58:763–779 771 123
- 10. in ensuring alignment between project teams and wider organisational needs. Additionally, such feedback was often seen as building on important relationships with external con- stituencies needed for the continuing success of distributed leadership. Involving key people was found to be a third important external activity. This involvement was found to be especially essential in cases where certain critical expertise was missing within the team. In these cases, such expertise had to be ‘borrowed’ from individuals external to the team boundaries in order to integrate their contribution in the process of leadership. Furthermore, the involvement of people external to the team often helped building valuable informal relationships around the organization, enhancing the project’s future success. At other times, such involvement was found to be essential to ensure the support of important external decision makers who had a stake in the project outcomes. In a number of projects failure to involve such key external stakeholders was reported to severely diminish the success of the respective projects. In most projects, external activities were not just conducted by one specific individual. Although most top-down initiated projects did have one or a number of designated team members formally responsible for conducting some of the external activities such as reporting to higher management, in most projects external activities were conducted by a number of different project team members, depending on the specific situation or skills required. Especially in emergent projects, conducting external activities seemed to be quite organic in nature, often involving informal networks of various team members. In such cases, the distributed nature of the leadership process seemed particularly emphasized. Critical external conditions Successfully conducted external processes were reported to shape several valuable con- ditions which enhanced the continuing success of distributed leadership of project teams. When successful, such processes led to the support of external constituencies consisting of members of the wider organisational community and important decision makers, which in turn increased the likeliness of the project team to acquire valuable information, resources, and expertise needed for the project, as well as maintaining alignment with the wider organisational needs and building trust. Most projects that reported problems in conducting such external activities faced problems which in some cases even resulted in project failure. The outcomes of external activities often served as important inputs for continuing successful distributed leadership within the teams, such as trust from external decision makers leading to an increased level of autonomy for the team in conducting their activities. Support from members of the wider organisational community was important in most projects and essential especially in projects which were aimed at managing organisational change. In these cases, continuing involvement of and communication with the wider community was seen as key in determining project outcomes. Support from decision makers outside of the team was reported as important for ensuring a flow of information and resources needed by the team, as well as gaining their trust and willingness to delegate autonomy, making it possible to make important decisions at the team level. Information gained through conducting external activities was often reported to serve an essential role in making important decisions within the team and aligning team activities to the wider organisational operations. A number of teams which reported problems in gaining external information were found to lack the ability to make important decisions or identify issues and obstacles, in some cases resulting in interventions from external 772 High Educ (2009) 58:763–779 123
- 11. decision makers or cancellation of the project. Effectively conducting external activities provided teams with timely and accurate information enhancing the effectiveness of dis- tributed leadership within the team. Another important outcome of external activities consisted of resources. Resources included aspects like funding, time, facilities and equipment necessary for effectively conducting project activities. Although most resources were gained through key external decision makers, often important resources were gained through informal networks, especially in cases where time was too critical to acquire resources through formal organisational channels. The final important outcome of external activities consisted of externally acquired expertise. As distributed leadership is underpinned by the notion of expertise as a dispersed quality which needs to be integrated in the process of leadership, this outcome of external activities is considered especially important. Although most project teams possessed most of the required expertise internally, there were often moments where certain essential expertise could not be found within the team boundaries. In these cases, it proved essential to integrate externally acquired expertise in the process of leadership. In some projects, such external expertise was acquired through external consultants, but more often it was acquired through informal organisational networks of team members. In several projects, a failure to gain access to expertise external to the team led to stifled team progress. Team level factors The second aspect important for distributed leadership is the internal functioning of the team. Several factors influencing distributed leadership at the team level were identified. They were subdivided into critical internal conditions and critical internal processes. Below, we will first describe the critical internal conditions, followed by the critical internal processes. Critical internal conditions Critical internal conditions constitute the enabling factors that need to be in place in order for successful distributed leadership to occur. Most of these conditions can be influenced through team design. Together, these conditions allow for the emergence of the critical internal and external processes necessary for effective distribution of leadership. Six critical conditions were derived from our case study data. Autonomy was found to be an important factor in enabling teams to engage in dis- tributed leadership. Teams which had a high degree of autonomy were more flexible in making decisions and coordinating their activities, as they did not rely on the approval of external decision makers. Autonomy also seemed to grant a degree of ownership in team activities and outcomes. It was found however that autonomy was not a given for most project teams, and often a team needed to build an extent of trust with external decision makers in order to be allowed the autonomy needed to influence the leadership within the team. This seemed particularly important in emergent projects. However, too much autonomy was also found to lead to isolation, highlighting the need to balance the amount of autonomy with a level of alignment with the wider organisational context. A clearly defined goal was another condition which was found important in order to enable effective distributed leadership to emerge. Teams which had a clear goal were found to be more successful in coordinating their activities and focus their energies in a common direction. A clearly defined goal was also found to enhance a team’s capacity to High Educ (2009) 58:763–779 773 123
- 12. monitor and feed back on progress, as well as enabling them to engage in adaptive behaviours in case of changes in the team environment. In a number of projects, the lack of clarity regarding the team goal resulted in divergence of activities and a lack of integration, which stifled team progress and frustrated team members. Especially in emergent projects, where team goals were not given by external decision makers, a clearly defined goal played an important role in managing external relationships and securing external support. A shared internal support for the team goal was found to be another factor influencing the emergence of successful distributed leadership. Team members who believed in the team goal were more likely to actively contribute to team leadership and engage in adaptive behaviours if a present course of action proved to be ineffective. In some projects, the lack of internal support for the team goal was found to result in team members ‘just doing their job,’ even though it was clear that the present course of action was clearly ineffective. Most emergent projects were characterized by a high support for the team goal, as team members took part in such project on their own initiative. However, some top- down initiated projects lacked the support of team members, resulting in stifled progress especially if barriers were encountered. Clearly defined responsibilities were found to be another factor important for successful distributed leadership. If responsibilities were clearly defined, teams were found to be more effective in coordinating their collective inputs. Also, responsibilities needed to be linked with individual expertise in order for individuals to successfully contribute to the leader- ship process. Especially in some top-down projects, certain individuals were given responsibilities which lied outside of their area of expertise or motivation, resulting in poor team performance. A lack of clearly defined responsibilities for individual members often resulted in conflicts between team members who felt others encroached on their perceived responsibilities. Also, if clearly defined responsibilities were lacking, this often led to duplication of activities, or essential activities not being performed. However, a certain balance between clear individual responsibilities was needed to be offset by a sense of shared responsibility for team outcomes. Some of the projects lacked the capacity to adapt to changing circumstances due to the fact that dealing with changing circumstances was not perceived by any individual member to be ‘their responsibility.’ In such cases, a shared support for the goal was found to be important in motivating team members to redefine responsibilities in a flexible way. Another interesting observation was that responsibilities often seemed more ‘organic’ in emergent projects, where the boundaries between responsibilities were often less well defined than in top-down initiated projects. Key internal expertise was another essential factor for the emergence of distributed leadership. Successful distributed leadership depends on the integration of diverse sources of expertise in the process of leadership. Although some expertise can be ‘borrowed’ through external activities, it was found to be important that most of the essential expertise needed for the team task needed to be available within the team. Such expertise did not only constitute knowledge and skills associated with a certain professional field, but also encompassed cultural, political, and procedural knowledge of the organisation. Team size was found to be another important aspect influencing the emergence of successful distributed leadership. When teams were too small, they lacked the necessary internal expertise to successfully deal with the complexity of the team task. However, teams that were too large were often found to face difficulties in integrating and coordi- nating the activities of the individual members, including all members in the process of leadership, or to reach joint decisions. Therefore it seemed important to seek a balance between a team large enough so that all necessary expertise needed for managing the task 774 High Educ (2009) 58:763–779 123
- 13. complexity was available and a team small enough to make coordination and integration of team effort possible. Critical internal processes Critical internal processes constitute the team processes which were found to underpin the emergence of distributed leadership. Their emergence seemed to rely on the critical internal conditions described above. However, in contrast to the critical internal conditions, critical internal processes could not directly be influenced by team design, but rather needed to be fostered and developed within the team. Five critical internal processes were identified. Information sharing within the team was found to be essential for team members to develop a common understanding of the team task and the issues faced by the team, which in turn enabled members to participate in shared decision making and leadership. Openly sharing information within the team seemed to foster a transparency in decision making and the development of trust between team members, fostering an atmosphere in which all team members could contribute to the leadership process. Mutual performance monitoring involves team members taking account of each other’s activities and offering feedback, help, or suggestions when needed. Through mutual per- formance monitoring, teams seemed better able to diagnose their situation and surface emerging issues before they became problematic. Mutual performance monitoring also seemed to help teams develop an awareness of individual strengths and weaknesses, which allowed them to build on each others’ strengths and complement each others’ weaknesses. Coordinating activities constitutes the process through which team members organise and plan their individual activities in order to gear them to a collective team goal. This process was often found to be based on negotiation and relied heavily on the exchange of information between team members. Project teams which did not specifically address the coordination of their activities were often found to work ineffectively as certain work was duplicated or not addressed at all. Adaptive behaviours involve team members changing their approach in case of altered requirements due to changes in the team’s environment or when established approaches turn out to be ineffective in achieving team goals. Some teams were found to continue on an established course which no longer aligned with changing organisational needs, resulting in top-down interventions to halt the project or force a change of course. Other teams were actively engaged in assuring that changing circumstances were incorporated in team conduct, thereby establishing continuous realignment in dynamically changing circumstances. Inclusiveness constitutes team members actively seeking to involve one another in the process of sense-making and decision-making. Inclusiveness did not only facilitate the inclusion of all available expertise in the process of leadership, thereby making sure team activities were monitored from a variety of perspectives and interests, it also created an atmosphere in which team members experienced a sense of ownership and being valued. Such ownership ensured that team members remained actively involved and supportive of team goals. Discussion Leadership remains an elusive and enigmatic concept, and studying leadership from a distributed perspective does not mitigate the complexities of the phenomenon. However, High Educ (2009) 58:763–779 775 123
- 14. by embracing the complexities of leadership as a distributed phenomenon and studying it in real organizational settings, we have tried to understand some of the factors that influence the occurrence and success of distributed leadership. Due to the inherent com- plexity and interdependence of project work, studying the factors that influence the success or failure of projects and analyzing these factors through a distributed leadership lens allowed us to gain an understanding of the processes through which leadership is dis- tributed over a range of individuals. By taking a qualitative approach aimed at identifying the factors that influenced the success or failure of projects and looking at the processes which gave rise to collective activity within these projects, we have avoided adopting a specific predefined conception of distributed leadership. Despite the exploratory nature of our study, our data support some of the previous findings from the fields of distributed, shared, and team leadership. Pearce (2004), for instance, highlighted the importance of expertise, the allocation of responsibilities, optimal team size, and a clearly defined goal or vision as factors essential to the development and successful continuity of shared leadership. Day et al. (2004) found that adaptability and mutual performance monitoring were necessary for effective team leadership, a finding supported by our study. Additionally, their concept of team orientation shows strong resemblance with what we have termed inclusiveness, another aspect found to be important for the development of distributed patterns of leadership. Burke et al. (2006) described empowerment as an enhancing factor for successful teamwork, in line with our finding that a level of autonomy is essential for a team to engage in effective distributed leadership. Additionally, our notion of inclusiveness as important to distributed leadership seems to be similar to Burke et al.’s concept of consideration. Besides supporting these previous findings and grounding them in qualitative empirical data, we hope that our study has contributed to an integration of these previous findings in a field which is still characterized by theoretical and conceptual pluralism. The model of distributed leadership as we propose it is by no means meant to be a final framework, but rather a device to structure our understanding of and inquiry into the concept of distributed leadership. Implications for the field One thing that our study highlights is the importance of engaging in external activities as a necessary condition for the continuous success of distributed leadership. Other researchers have already identified such external activities—often termed boundary management—as important aspects of effective team work (e.g., Burke et al. 2006; Elkins and Keller 2003; Pearce 2004). However, most of these authors have placed considerable emphasis on the designated team leader as playing a vital role in such external activities, something which was not reflected in our cases. From our data, it became obvious that different team members were responsible for conducting activities related to the wider organizational context, either because of their specific expertise or because of their particular informal networks. Also, previous research has focused on boundary management as a means for securing resources and information from the wider organizational context (e.g., Brown and Ei- senhardt 1995; Pearce 2004). What our study adds to this is an understanding of boundary management as a means to integrate certain vital expertise not available within the team, as well as its role as a mechanism for ensuring continuous alignment between a team and the wider organizational context. This finding also carries implications for conceptual considerations around the notion distributed leadership. As noted in the introduction, authors in the field of distributed 776 High Educ (2009) 58:763–779 123
- 15. leadership differ in terms of how they define the scope of participating agents in the process of distributed leadership, where on one end the focus lies solely on the team level, and on the other end the organization as a whole as well as agents outside of the organ- isational boundaries are seen as participants in the process of leadership. Based on our findings, we suggest that the network of participation in leadership is essentially a boundary judgment which needs to be made based on the context of the situation. Also, we believe that such boundaries are essentially fluid, and can be extended or limited depending on the present needs of the team or organization. Implications for practice and leadership development in higher education As noted in the introductions, Higher Education institutions seem to struggle in dealing with the tensions between traditional collegial notions of leadership and the introduction of management principles derived from the private sector. We believe that the principles of distributed leadership show promise in overcoming some of these tensions and help Higher Education institutions deal more effectively with the pressures of adapting to ever increasing rates of environmental change. In practice, we have seen this in a number of our cases where academic and adminis- trative staff have transcended the cultural frictions and successfully engaged in initiatives where each complemented rather than opposed the other, even leading to successful attempts to start linking in-house academic research to leadership and management practice within the organisation. However, some of our cases also highlighted the fragility of effective distributed leadership. Examples of this include cases where team members reported ceasing to actively contribute to the process of leadership due to other—more senior—team members exerting authoritarian influence over decisions that concerned the team as a whole. In other cases, repeated obstructions to team progress resulting from bureaucratic organisational procedures or powerful groups within the organization caused teams to give up their efforts to actively contribute to project outcomes. In order for distributed leadership to work effectively, it seems that organizations need to approach leadership development in a different way than traditional conceptions of leadership would direct. Instead of focusing on the development of the leadership capa- bilities of an organization’s designated leaders, focus would shift to investing not only in developing leadership skills of the workforce as a whole, but also to facilitating the conditions conductive for the emergence of successful distributed leadership and the for- mation of informal networks of expertise. Developing the leadership skills of the work- force without facilitating the conditions for distributed leadership to thrive would quite likely lead to frustrations and inhibited effectiveness and engagement, whereas the facil- itation of the necessary conditions without development of the required skills would likely lead to confusion and misalignment of teams with the wider organisational context. Conclusion Distributed leadership could play a major role in the future of our knowledge-based society as it combines the strengths of various individuals and balances their weaknesses. It might well be that in a world where work is increasingly team-based and no single person can hold all the relevant knowledge to make the right decisions, there is a growing belief that the competitive advantage of organizations will increasingly depend on their ability to High Educ (2009) 58:763–779 777 123
- 16. integrate the widely dispersed knowledge and skills of their entire workforce in the process of leadership. References Bennett, N., Harvey, J. A., Wise, C., & Woods, P. A. (2003). Desk study review of distributed leadership. Nottingham: National College for School Leadership. Bergquist, W. H. (1992). The Four Cultures of the Academy. San Francisco: Jossey-Bass Publishers. Birnbaum, R. (1988). How colleges work: The cybernetics of academic organization and leadership. San Francisco: Jossey-Bass. Birnbaum, R. (2000a). Management fads in higher education: Where they come from, what they do, why they fail. San Fransisco: Jossey-Bass. Birnbaum, R. (2000b). The life cycle of academic management fads. The Journal of Higher Education, 71, 1–16. doi:10.2307/2649279. Brown, S. L., & Eisenhardt, K. M. (1995). Product development: Past research, present findings, and future directions. Academy of Management Review, 20, 343–378. doi:10.2307/258850. Brown, M. H., & Hosking, D. M. (1986). Distributed leadership and skilled performance as successful organization in social movements. Human Relations, 39, 65–79. doi:10.1177/001872678603900104. Bryman, A. (2004). Qualitative research on leadership: A critical but appreciative review. The Leadership Quarterly, 15, 729–769. doi:10.1016/j.leaqua.2004.09.007. Burke, C. S., Stagl, K. C., Klein, C., Goodwin, G. F., Salas, E., & Halpin, M. (2006). What type of leadership behaviors are functional in teams? A meta-analysis. The Leadership Quarterly, 17, 288– 307. doi:10.1016/j.leaqua.2006.02.007. Chandler, J., Barry, J., & Clark, H. (2002). Stressing academe: The wear and tear of the new public management. Human Relations, 55, 1051–1069. doi:10.1177/0018726702055009019. Conger, J. A. (1998). Qualitative research as the cornerstone methodology for understanding leadership. The Leadership Quarterly, 9, 107–121. doi:10.1016/S1048-9843(98)90044-3. Cummings, T. G., & Worley, C. G. (2004). Organization development and change. OH: South-Western. Day, D. V., Gronn, P., & Salas, E. (2004). Leadership capacity in teams. The Leadership Quarterly, 15, 857– 880. doi:10.1016/j.leaqua.2004.09.001. Day, D. V., Gronn, P., & Salas, E. (2006). Leadership in team-based organizations: On the threshold of a new era. The Leadership Quarterly, 17, 211–216. doi:10.1016/j.leaqua.2006.02.001. Deem, R. (1998). ‘New Managerialism’ and higher education: The management of performances and cultures in universities in the United Kingdom. International Studies in Sociology of Education, 8, 47–70. doi: 10.1080/0962021980020014. Elkins, T., & Keller, R. T. (2003). Leadership in research & development organizations: A literature review and conceptual framework. The Leadership Quarterly, 14, 587–606. doi:10.1016/S1048-9843(03) 00053-5. Ensley, M. D., Hmieleski, K. M., & Pearce, C. L. (2006). The importance of vertical and shared leadership within new venture top management teams: Implications for the performance of startups. The Lead- ership Quarterly, 17, 217–231. doi:10.1016/j.leaqua.2006.02.002. Feyerherm, A. E. (1994). Leadership in collaboration: A longitudinal study of two interorganizational rule- making groups. The Leadership Quarterly, 5, 253–270. doi:10.1016/1048-9843(94)90016-7. Gibb, C. A. (1954). Leadership. In G. Lindzey (Ed.), Handbook of social psychology (Vol. 2, pp. 877–917). Reading, MA: Addison-Wesley. Gronn, P. (2002). Distributed leadership as a unit of analysis. The Leadership Quarterly, 13, 423–451. doi: 10.1016/S1048-9843(02)00120-0. Gumport, P. J. (2000). Academic restructuring: Organizational change and institutional imperatives. Higher Education, 39, 67–91. doi:10.1023/A:1003859026301. Huisman, J., & Currie, J. (2004). Accountability in higher education: Bridge over troubled water? Higher Education, 48, 529–551. doi:10.1023/B:HIGH.0000046725.16936.4c. Kezar, A., & Eckel, P. D. (2002). The effect of institutional culture culture on change strategies in higher education: Universal principles or culturally responsive concepts? The Journal of Higher Education, 73, 435–460. doi:10.1353/jhe.2002.0038. Kuhn, T. S. (1996). The structure of scientific revolutions. Chicago: University of Chicago Press. Locke, E. A. (2003). Leadership: Starting at the top. In C. L. Pearce & J. A. Conger (Eds.), Shared leadership: Reframing the hows and whys of leadership (pp. 271–284). Thousand Oaks, CA: Sage. 778 High Educ (2009) 58:763–779 123
- 17. Lueddeke, G. R. (1999). Toward a constructivist framework for guiding change and innovation in higher education. The Journal of Higher Education, 70, 235–260. doi:10.2307/2649196. Mehra, A., Smith, B. R., Dixon, A. L., & Robertson, B. (2006). Distributed leadership in teams: The network of leadership perceptions and team performance. The Leadership Quarterly, 17, 232–245. doi:10.1016/ j.leaqua.2006.02.003. Meindl, J. R., Ehrlich, S. B., & Dukerich, J. M. (1985). The romance of leadership. Administrative Science Quarterly, 30, 78–102. doi:10.2307/2392813. Middlehurst, R. (1997). Reinventing higher education: The leadership challenge. Quality in Higher Edu- cation, 3, 183–198. doi:10.1080/1353832970030208. Morgeson, F. P., & DeRue, D. S. (2006). Event criticality, urgency, and duration: Understanding how events disrupt teams and influence team leader intervention. The Leadership Quarterly, 17, 271–287. doi: 10.1016/j.leaqua.2006.02.006. Parker, M., & Jary, D. (1995). The McUniversity: Organization, management, and academic subjectivity. Organization, 2, 319–338. doi:10.1177/135050849522013. Pearce, C. L. (2004). The future of leadership: Combining vertical and shared leadership to transform knowledge work. Academy of Management Executive, 19, 47–57. Pearce, C. L., & Conger, J. A. (2003). Shared leadership: Reframing the hows and whys of leadership. Thousand Oaks, CA: Sage. Pearce, C. L., & Sims, H. P., Jr. (2002). Vertical versus shared leadership as predictors of the effectiveness of change management teams: An examination of aversive, directive, transactional, transformational, and empowering leader behaviors. Group Dynamics: Theory, Research, and Practice, 6, 172–197. doi: 10.1037/1089-2699.6.2.172. Pettigrew, A. M. (1990). Longitudinal field research on change: Theory and practice. Organization Science, 3, 267–292. doi:10.1287/orsc.1.3.267. Pinto, M. B., Pinto, J. K., & Prescott, J. E. (1993). Antecedents and consequences of project team cross- functional cooperation. Management Science, 39, 1281–1297. Rost, J. C. (1993). Leadership for the twenty-first century. Westport, CT: Praeger. Smaling, A. (1992). Varieties of methodological intersubjectivity: Relations with qualitative and quanti- tative research and with objectivity. Quality & Quantity, 26, 169–180. doi:10.1007/BF02273552. Spillane, J. P., Halverson, R., & Diamond, J. B. (2001). Investigating school leadership practice: A dis- tributed perspective. Educational Researcher, 30, 23–28. doi:10.3102/0013189X030003023. Spillane, J. P., Halverson, R., & Diamond, J. B. (2004). Towards a theory of leadership practice: A distributed perspective. Journal of Curriculum Studies, 36, 3–34. doi:10.1080/0022027032000106726. Thamhain, H. J. (2004). Linkages of project environment to performance: Lessons for team leadership. International Journal of Project Management, 22, 533–544. doi:10.1016/j.ijproman.2004.04.005. Yukl, G. (2002). Leadership in organizations (5th ed.). New Jersey: Prentice-Hall. High Educ (2009) 58:763–779 779 123