Alternative Approaches for Studying Shared and Distributed Leadership

Fitzsimons, D., K. T. James, et D. Denyer (2011). “Alternative approaches for studying shared and distributed leadership.” International Journal of Management Reviews 13: 313-328.

Further, differing expectations and specialized expertise of knowledge workers in the post-industrial era may also contribute to the emergence of shared leadership (Carson, Tesluk et Marrone, 2007; Cox, Pearce et Perry, 2003; Fletcher et Käufer, 2003; Pearce et Manz, 2005; Seers, Keller et Wilkerson, 2003).

This increased ambiguity and complexity, it is argued, arises in the face of a number of adaptive challenges such as increased domestic deregulation, global economic integration, multiple and competing stake- holder environments, new technology and increased rates of change (Avolio, Jung et Sivasubramaniam, 1996; Houghton, Neck et Manz, 2003; Seers, Keller et Wilkerson, 2003).

In the team-based literature, the conversation was about how to develop from self-led teams (no leader) to collective leadership in the team (everyone a leader). In the education sector, the challenge was to move leadership away from the top of the organization to create leadership practices throughout the organization. These two separate conversations in separate literatures did not cross over until Gronn’s (2002) paper, by which time different evolutions of the concepts of shared and distributed leadership with different ontological assumptions had occurred.

Other early contributions such as Hodgson et al.’s (1965) exploration of ‘the executive role constellation’ of three leaders of a hospital, co-leadership (Heenan et Bennis, 1999), strategic leadership (Hambrick et Mason, 1984) and collective collaborative leadership (Dennis, Langley et Cazale, 1996) all propose a model of leadership in which leadership can come from group/organizational members other than the designated leader.

The historical roots of shared and distributed leadership

concept of ‘empowerment’ (Blau et Alba, 1982; Conger et Kanungo, 1988). However, since empowerment is based on the demarcation of leaders and followers, it can only be considered a prerequisite for the emergence of shared leadership (Pearce et Conger, 2003).

Leader Member Exchange Theory and Substitutes for Leadership (Kerr et Jermier, 1978).

Manz and Sims (1980) to propose a model of ‘self-management’ as a substitute for leadership.

Follett’s Law of the Situation (1924) states that leadership could stem from the individual with the most relevant skills in a particular situation.

In contrast, Benne and Sheats (1948) suggest that leader- ship is to do not with an individual, but with functions, and that several individuals could take up differentiated roles in relation to these functions.

Shared leadership is studied as an emergent phenomenon within what we term the team- based shared-leadership literature in this paper. With some exceptions and occasionally as a grammatical device to avoid repetition, this literature uses the term ‘shared leadership’

Similarly Gibb (Gibb, 1954, 1969) argues that leadership is best thought of as existing on a continuum from focused or individual leadership to a distributed pattern.

Stogdill (1950) suggests that leadership is based on role differentiation related to influencing the goal setting and goal achievement behaviours of others. a second strand of research activity developed primarily in the education literature concerned with distributed leadership; in this literature, the term ‘distributed leadership’ is clearly distinguished from shared leadership (Spillane, 2006: 3).

An oft-cited empirical study by Bowers and Seashore (1966) indicates that leadership can come from peers and that this could have a positive impact on outcomes.

The development of shared leadership

Research in this domain is concerned with measuring the influence of the ‘group-as-a-whole’ by taking questionnaires which refer to individual leadership styles such as ‘transformational’, ‘transactional’, ‘directive’, ‘aversive’ and ‘empowering’ (Pearce et Sims, 2002) and applying them at the group level.

In a landmark volume of Advances in Interdisciplinary Studies of Work Teams, authors expressed concern that leadership in self-managed teams was insufficiently explored (Avolio, Jung et Sivasubramaniam, 1996).

Beyerlein et al. (1996: xviii), note that ‘it was but a short time ago that many practitioners believed teams and leadership were mutually exclusive’ and that, through a process of trial and error, ‘we learned we could not ignore the need for leadership’.

shared leadership can, in part, be traced to a transition in the self-leadership, super-leadership constructs (Manz et Sims, 1989, 1991) from individuals self-leading, to the conceptualization of self-leadership at the team level (Neck, Stewart et Manz, 1996).

However, the role of formally appointed leader- ship has been hotly debated within the field (Pearce, Conger et Locke, 2008).

The domain of shared leadership is relatively small, cohesive and dominated by a small number of overlapping writing partnerships between Manz and Sims (Manz et Sims, 1987, 1989, 1991, 1993, 2001) (1987, 1989, 1991, 1993, 2001),Pearce and Sims (Pearce et Sims, 2000, 2002), Pearce and Manz (2005) and Pearce et al. (Pearce et Manz, 2009; Pearce, Manz et Sims, 2008)

Our research revealed only seven seminal empirical studies within MOS (Avolio, Jung et Sivasubramaniam, 1996; Carson, Tesluk et Marrone, 2007; Ensley, Hmieleski et Pearce, 2006; Mehra, Smith et Dixon, 2006; Pearce et Sims, 2002; Sivasubramaniam et al., 2002) Pearce et al. 2003;

In the conceptual model centred on the related constructs of super-leadership and self-leadership (Houghton, Neck et Manz, 2003; Manz et Sims, 1989, 1991, 2001) the role of the formal leader is central to the emergence of shared leadership.

Super-leadership is often contrasted with three other leadership types – the Strongman, the Transactor, and the Visionary Hero (Manz et Sims, 1991, 2001). It is suggested that these three types of leadership can be shared among team members leaving the fourth type – the Empowering leader – to be taken up by the vertical super-leader.

All these studies attempt to measure the relative significance of vertical versus shared leadership for team outcomes using survey methods to measure leadership behaviours at the level of the team.

Avolio et al. (1996) found shared leadership in undergraduate students doing community volunteer work to be positively correlated with self-reported ratings of effectiveness.

Cox et al. (2003) studied the development and maintenance of shared leader- ship in New Product Development teams focusing on vertical leader behaviours, including boundary management and empowerment.

Seers et al. (2003) developed a model which does not include formal leaders. The model high- lights the importance of factors such as social status differentials and group member ratings of each other’s ability to contribute to task achievement, as influencing positively or negatively the likelihood of shared influence emerging (Seers, Keller et Wilkerson, 2003)

Team adaptability in the face of complex tasks and the transference of the leadership function are explored in a conceptual framework focused on shared cognition as a factor enabling shared leader- ship and team adaptability (Burke, Fiore et Salas, 2003)

Shared meta-cognition is seen as an important factor determining the development of both shared team mental models and shared situation mental models which, when combined with attitudinal factors such as open climate, and collective efficacy, influence when and to whom leadership should move (Burke, Fiore et Salas, 2003)

Distributed leadership

Prior to 2002, we found little or no cross-citation between the ‘team-based’ and education literatures until Gronn (2002) integrated these two strands of work.

Distributed leadership, for Spillane (2006) has a pragmatic focus on leadership as a practice and focuses on the school as the unit of analysis, rather than the team.

Spillane (2006) found that ‘followers’ – those not designated as holders of formal leadership roles such as teachers and other school support staff – also contribute to leadership. However, distributed leadership goes beyond acknowledging that multiple individuals are involved in leadership practice (termed ‘leader plus’ or shared leadership), by also exploring the interactions between individuals and investigating the situation in which leadership is enacted (distributed leadership).

Spillane and colleagues identify three forms of distributed leadership – collaborated, collective and co-ordinated distribution (Spillane, 2006).

Table 1. The key characteristics of shared and distributed leadership

Shared leadership Distributed leadership
Leadership often emanates from the designated leader plus other group members who share leadership roles (e.g. Strongman, Transactor, Visionary hero and Super-leader). Leadership is not only held by those with designated, formal leadership roles but is enacted by multiple individuals in the organization.
Leadership involves several individuals leading themselves and allowing others to lead them through a reciprocal influence process. Leadership practice is constituted and shaped by the interactions between leaders and followers and the organizational context.
Cognition is shared by members of the group. Cognition is ‘stretched over’ both human actors and aspects of the context they are in.
Advantage is offered through the aggregate of attributed influence in a group (collective influence). Advantage is offered by developing a capacity to act by means of ‘concertive action’, ‘co-performance’ or ‘conjoint agency’.

Alternative ontological views and leadership epistemologies

Table 2. Alternative approaches for studying shared and distributed leadership

  Relational-entity Relational-structural Relational-processual Relational-systemic
The nature of leadership Leadership is shared/distributed between discrete minds/entities (leaders) who perceive, evaluate and make decisions including how and when to act in pursuit of goals. Leadership is shared/distributed in a system or pattern of relations. These networks are cognitive structures in the minds of nodes/egos (leaders) and opportunity structures that facilitate and constrain action. Leadership is a practice constantly being constituted within the flow of a set of social processes that take place in particular cultural, historical and political contexts. Leadership is a function of a collective and involves conscious and unconscious psycho-social processes that are systemic in nature and particular to a specific context. Thus leadership is always shared or distributed.
The nature of relationship Relationships are inter-personal and formed and acted upon by knowing subjects for instrumental purposes. Relationships are real and can be measured Relationships are formed by concertive units (pairs, teams, cross-departmental groups) within ongoing social processes. Patterns of relating often reflect systemic and unconscious strategies for managing the collective anxieties associated with adaptive learning
The role of context Little emphasis since shared/distributed leadership is dependent on the skills, attributes and behaviours of individual leaders aggregated to the group level that can be applied in multiple settings. Context is important in that effective shared or distributed leadership is dependent on accurately perceiving and leveraging the structure of social ties in the organization. Strong emphasis on context since, in contrast to contingency theories, elements of the situation are considered constitutive of shared/distributed leadership practice. Strong emphasis, since no statement about leadership from this perspective is possible without reference to how group level phenomena link individuals and groups to their contexts.
How leadership is studied Variance methods are often employed to seek explanations of leadership, with independent variables acting upon and causing changes in dependent variables. Social network methods are often employed to study the individual’s position in the larger networks within which the individual is located. Eclectic designs are used to identify or reconstruct the process through which leadership emerges and changes over time. Action research and ethnographic designs are employed to explore systemic emotional dynamics and unconscious group processes and how these unfold within a particular task context.

While defining shared leadership as a ‘simultaneous, ongoing, mutual influence process within a team’ (Ensley et al. 2006, p. 218), that it represents a ‘condition of mutual influence embedded in the interactions among team members’ (Carson, Tesluk et Marrone, 2007: 1218) or that it emerges through an ‘unfolding series of fluid, situationally appropriate exchanges of lateral influence’ (Cox, Pearce et Perry, 2003), these scholars usually theorize the individual as a discrete isolate.