Teacher teams and distributed leadership: A study of group discourse and collaboration

Citation Scribner, J. P., Sawyer, R. K., Watson, S. T., & Myers, V. L. (2007). Teacher teams and distributed leadership: A study of group discourse and collaboration. Educational Administration Quarterly, 43(1), 67–100. Sage Publications. Sidewiki
BibDesk PDF




author = {Scribner, Jay Paredes and Sawyer, R Keith and Watson, Sheldon T and Myers, Vicki L},
date-added = {2016-06-18 23:13:35 -0400},
date-modified = {2016-06-19 00:43:49 -0400},
journal = {Educational Administration Quarterly},
keywords = {distributed leadership},
number = {1},
pages = {67–100},
publisher = {Sage Publications},
title = {Teacher teams and distributed leadership: A study of group discourse and collaboration},
volume = {43},
year = {2007},


Links here


Purpose: p. 67

article explores distributed leadership as it relates to two teacher teams in one public secondary school p. 67

Both situational and social aspects of distributed leadership are foci of investigation. p. 67

Methods: p. 67

qualitative study used constant comparative analysis and discourse analysis p. 67

Data from field notes and video recordings of two teacher teams during one semester p. 67

Findings: p. 67

Three constructs emerged that informed our understanding of collaborative interaction within each professional learning team: purpose, autonomy, and patterns of discourse p. 67

Purpose and autonomy, manifest as organizational conditions, largely shape patterns of discourse that characterize the interaction of the team members p. 67

We argue that the nature of purpose and autonomy within a teacher team can influence the social distribution of leadership. p. 67

Conclusions: p. 67

The nature of teams in shared governance structures—the fact that teams can organize to either find or solve problems—has important implications for the creative and leadership capacity of individual teams p. 67

Thus, structures and social dynamics of distributed leadership must be attended to and not taken for granted. p. 67

Implications include p. 67

(a) conceptualizing leadership in terms of interaction p. 67

(b) needing to help teachers become aware of conversational dynamics that lead to or subvert effective collaboration p. 67

© needing to help principals become more aware of their role in helping to establish clarity of purpose and appropriate levels of autonomy, so that teams may engage in work that leads to effective and innovative problem-finding and problem-solving activities. p. 67

The distributed leadership perspective is relatively new, and has begun to influence empirical studies of school leadership (Bennett, Wise, Woods, & Harvey, 2003; Smylie et al., 2002) p. 68

To better understand how distributed leadership works, we believe that researchers must closely attend to the dialogues that take place within the organizational structures that are the artifacts of the situational distribution of leadership, such as teacher teams p. 68

The following questions guide our work: p. 68

  • Whatfactorscontributetoorinterferewithteamdecisionmaking? p. 68
  • What discursive patterns are associated with leadership within teacher work teams? p. 68
  • What organizational conditions foster or impede leadership within teacher work teams? p. 68

By applying discourse analytic methods to the dialogues of these two teacher teams, we identified two distinct patterns of discourse that we call passive discourse and active discourse. p. 68

Through the use of constant comparative analysis, we also discovered that these two patterns were related to p. 68

the surrounding organizational conditions of the two teams p. 69

Our most important finding is that both organizational context and discourse patterns were related to the leadership effectiveness of the teams. p. 69


leadership in general (e.g., Bass, 1981) have acknowledged that leadership is not solely embedded in formal roles; it often emerges from relationships between people (e.g., Crow et al., 2002; Scott, 1992) p. 69

the mid-1990s witnessed a renewal of conceptions of networked leadership that first emerged in the 1950s and 1960s (i.e., Barnard, 1968; Thompson, as cited in Smylie et al., 2002). p. 69

Networked leadership emphasizes looking at what school personnel do, more than who is doing it, and it challenges the conventional belief that leadership is associated with particular positions. p. 69

those seeking to study leadership should look for the performance of these key activities instead of assuming that watching the principal will explain how the school is managed. p. 69

Renewed interest in networked leadership has led researchers to focus on the activities that constitute leadership and the network of organizational relationships that contributes to effective school leadership. p. 69

An alternative perspective conceptualized leadership as an organizational resource (Ogawa & Bossert, 1995; Pounder, Ogawa, & Adams, 1995; Smylie et al., 2002). p. 69

In this model, leadership is embedded in the relationships between networked roles p. 69

Such networks support a multidirectional flow of influence throughout organizations p. 69

Understanding leadership thus requires an analysis of the degree of social influence possessed by individuals, groups of individuals, or the entire organization p. 69

Spillane, Halverson, et al.’s (2001, 2004) model of distributed leadership brings together ideas from these perspectives, as well as from other sources. p. 69

functionalist approaches p. 69

Spillaneet al.associateleadershipwithactivityandarguethattounderstand school leadership, we should focus on activities and tasks rather than on the p. 69

behavior of individuals formally identified as the leaders p. 70

The focus is on leadership practice, which “is distributed over leaders, followers, and the school’s situation or context” (Spillane et al., 2004, p. 11). p. 70

Borrowing from distributed cognition and activity theory, this model locates leadership practice within a networked web of individuals, artifacts, and situations p. 70

this approach allows researchers to distinguish between a school’s officially stated theories of practice and what really happens in practice. p. 70

Once the actual practices are identified, an important challenge is to connect them to the specific tasks facing school leadership (Spillane et al., 2004). p. 70

From a distributed leadership framework, interaction between individuals plays a central role in accomplishing effective leadership (Gronn, 2000). p. 70

To identify the distributive dimensions of leadership, researchers must pay close attention to the interdependencies between activities—whether those interdependencies are pooled, sequential, or reciprocal (Spillane et al., 2004; Thompson, as cited in Smylie et al., 2002). p. 70

Decisions are not made by a single individual; rather, decisions emerge from collaborative dialogues between many individuals, engaged in mutually dependent activities. p. 70

These collaborative dialogues are a key component of what Spillane et al. have defined as the social distribution of leadership p. 70

To understand what is truly distributed about socially distributed leadership thus requires an empirical focus on interactioncollaboration, dialogue, and communication p. 70

no study has yet closely examined the moment-tomoment interactions between individuals that actually constitute socially distributed leadership p. 70

our goal is to extend the power of the approach by integrating discourse analysis with ethnographic and observational methods and by placing our primary focus on interaction (Sawyer, Scribner, Watson, p. 70

& Myers, 2005). p. 71

The case study focuses on the interactions of two professional learning teams (PLTs) that are part of the school’s broader improvement efforts. p. 71



Without the interaction between individuals on the team, however, leadership could not be effectively distributed. p. 73

Because teacher teams meet face to face, the primary medium of interaction for this social distribution of leadership is conversation. p. 73

Leadership can thus be viewed as an emergent activity (Gronn, 2003), partially constituted via social interactions evidenced by the artifact of conversation. p. 73

Interactions within teams are evidence of the constitution of distributed leadership at yet another conceptual level, which Spillane et al. (2004) refer to as the social distribution of leadership. p. 73

This intragroup level of distribution p. 73

has not yet been researched in terms of focusing on the specific artifact of talk p. 74

Other artifacts of these interactions, such as documents and policies, have been studied (e.g., Goldstein, 2003; Spillane, Diamond, et al., 2001). p. 74

We identify these as designed artifacts, artifacts that are more structural in nature and that exist at some distance from acts of individual agency. p. 74

Designed artifacts are the products of socially distributed leadership manifested in particular situations; yet once created, these artifacts become structuring forces (Watson, 2005). p. 74

Interactions are the bridge between the collective agency of the collaborative group and the new structural forms they produce. p. 74

This research has shown that the most effective teams manage themselves through an improvisational and emergent process (Crossan & Sorrenti, 1997; Moorman & Miner, 1998; Weick, 2001). p. 74

Method p. 75

One of the unique features of this study is that we use the methodology of discourse analysis (Sawyer, 2006) p. 76

We use the term discourse analysis broadly to refer to a wide range of approaches that sometimes go by the names conversation analysis or interaction analysis. p. 76

Findings p. 78

Three constructs emerged that informed our understanding of collaborative interaction within each professional learning team: purpose, autonomy, and patterns of discourse. p. 78

Patterns of Discourse p. 87

We found that conversation in the two groups was different in content and form. Comparison of the two groups revealed that the building PLT engaged in patterns of passive discourse, whereas the instructional PLT engaged in active discourse. p. 87

We hypothesize that the purpose and autonomy of a team partly determined which pattern of discourse was manifest in collaborative interaction p. 87

The influence of passive discourse on group communication. The meetings of the building PLT were predominantly characterized by the speech acts that Searle (1976) would label as representatives and expressives. These utterances are used to convey factual information and feelings p. 87

The influence of active discourse on group communication p. 89


First, teachers and administrators working in team-based governance structures should consider how the scope and nature of a team’s challenge and charge can influence team functioning within the group and in relation to the greater organization p. 95

Second, the nature of teams in shared governance structures— the fact that teams can organize to find problems or solve problems—has important implications for the creative and leadership capacity of individual teams. p. 95

problem-finding teams cannot be treated the same as problem-solving teams. p. 95

the performance of a problem-finding team might be enhanced if granted more autonomy or, at a minimum, provided clear parameters of what autonomy the team does have p. 95

Finally, the structures and social dynamics of distributed leadership must be attended to continuously and not taken for granted. p. 95